yyifuuyf

Click Here To Continue

Chapter 11

Oil Storage William E. Roof. C-E Ndco *’

Types of Storage Tanks Every facility involved in the production of petroleum and related products reqmres some type of storage. Thts chapter discusses the types of storage commonly used and also provides general guidelines to aid selection of the proper type of storage for a particular application.

References to various codes. standards. and rccom- mended practices supplement the material provided in this chapter. Manufacturers also should be consulted for specific design information on a particular type of storage.

During the early day\ of oil production, the method of storing was almost exclusively white-pine wooden tanks. which were followed by cypress tanks. and then redwood tanks. However. because of the constant and steep rise in the cost of redwood lumber and the diminution of skilled erectors required, the installation of new wooden tanks is nearly nonexistent. The bolted-steel tank was developed next and virtually replaced the wooden tank.

Bolted-Steel Tanks

Bolted tanks are designed and furnished as segmental elc- mcntc assembled on location to provide complete vertical, cylindrical. abovcground. closed- and open-top steel storage tanks. Standard API bolted tanks are available in nominal capacities of 100 to 10,000 bbl, and are designed for approximately atmospheric internal pressures. Bolt- ed tanks offer the advantage of being easily transported to desired locations and erected by hand. To meet changing requirements for capacity of storage, bolted tanks can be easily dismantled and re-erected at new locations. If a tank dev)elops a hole from corrosion or becomes damaged. a single sheet or more may be replaced. A complete tank bottom may bc replaced in the field without dismantling the tank. Also. a section may be removed from the tank. a new connection installed in the sheet. and the section replaced without danger. This is not true

of any other type of steel construction. No special cyuip- ment (cranes. etc.) is required for the crcction of bolted tanks. These tanks are erect& by nonspecialized crews using hand tools and usually an impact wrench.

Bolted tanks are available with painted. galvanized. and special coatings. including factory-baked coatings. Paint- ing on both sides of the sheets during fabrication gives the inside of the tank some corrosion protection. Galvaniz- ing the sheets and all tank parts by the “hot-dip” process or applying a factory-baked coating affords high corrosion protection. The component parts of a typical bolted tank are shown in Fig. 1 I. I and partial API spccifica- tionc in Table I I I. ’

Generally. bolted tanks are fabricated from I?- or IO-gauge steel and, if not galvanized or furnished with a protective coating for corrosion protection, they do not have the expected life of the welded-steel tanks. which are usually constructed of heavier steel

Welded-Steel Tanks

Shop-fabricated welded, cylindrical-shape tanks are available in a large variety of sizes as shop-fabricated items. The API-12F specifications’ for vertical shop-welded tanks (Fig. 1 I .2 and Table I I .2) list standard sizes for nominal capacities of 90 to 500 bbl. Shop-welded tanks fabricated to API specifications provide the oil production industry wjith tanks of adcquatc safety and reasonable economy for USC in the storage of crude petroleum and nthcr liquids commonly handled and stored by the production segment of the industry. Shop-welded tanks are usually fabricated from ‘/;,-in. or heavier steel and. therefore. will permit internal pressures up to 16 oz. The heavier steel also affords a corrosion allowance. Shop fabrication permits testing in the shop for leaks and also provides immediate storage. Tanks arc merely up-ended from a truck on the location.

11-2 PETROLEUM ENGINEERING HANDBOOK

Deck

Fig. 11 .I-Typical bolted tank.

Flat-Sided Tanks (Non-API)

Although cylindrical-shape tanks may be structurally best for tank construction, rectangular tanks frequently are preferred. When space is limited, such as offshore, requirements favor flat-sided tank construction because several cells of flat-sided tanks can be fabricated easily and arranged in less space than other types of tanks. Flat-sided or rectangular tanks normally are used as atmospheric- type storage.

Field-Welded Tanks

Field-welded tanks provide large storage capacities in a single unit. API Spec. 12D lists standard sizes ranging from 500- through lO.OoO-bbl nominal capacity. Although the sizes set forth in this specification are closely paralleled by bolted tanks, these field-welded tanks are of heavier- gauge steel with a minimum thickness of ‘/ in. for the tank bottom and xh in. for the shell and deck.

Larger field-welded tanks providing storage capacities of 150,000 bbl or more have become quite prevalent for use in the storage of oil and petroleum products. Field- welded tanks, particularly those larger than IO.000 bbl. frequently are designed and erected in accordance with API Standard 6.50. This standard covers material, design, fabrication, erection, and testing requirements for welded- steel storage tanks. It also includes an alternative basis for shell design, as well as one for calculating tank-shell thickness. The API Standard 650 also may be used to govern the design and fabrication of the smaller shop- welded tanks.

Fixed Roof

Fixed roofs are permanently attached to the tank shell. Welded tanks of 500.bbl capacity and larger tnay be provided with a frangible roof (designed for safety release of the welded deck-to-shell joint in the event excess in ternal pressure occurs). In this case, the design pressure should not exceed the equivalent pressure of the dead weight of the roof including rafters, if external.

Floating Roof

Storage tanks may be furnished with floating roofs whereby the tank roof floats on the stored contents. This tank type is used primarily for storage near atmospheric pressure. Floating roofs are designed to move vertically within the tank shell to provide a constant minimum void between the surface of the stored product and the roof. Floating roofs normally are designed to provide a constant seal between the periphery of the floating roof and the tank shell. They can be fabricated in a type that is exposed to the vveather or a type that is under a fixed roof. Internal floating-roof tanks, with an external fixed roof, are used in areas of heavy snowfall since accumulations of snow or water on the floating roof affect the operating buoyancy. These can be installed in existing tanks as well as new tanks. Both floating roofs and internal floating roofs are used to reduce vapor losses and to aid in conservation programs. Fig. 11.3 is a schematic of a typical internal floating-roof tank.

Cone-Bottom Tanks

The cone bottom in either the bolted or the welded tank offers a means of draining and removing water. or water- cut oil. from only the bottom of the tank, leaving the marketable oil above. The drain line from a sump-equipped cone bottom must be equipped with a vortex breaker to drain off most of the water without coning oil into the drain. With a flat-bottom tank, some of the marketable oil must be removed if all the water is removed from the tank. Corrosion on the tank bottom is kept to a minimum by keeping all water removed. A cone bottom can be kept clean without having to open the tank if I or 2 bbl are drained off once or twice weekly and pumped back through the treating system. If this is not done and the bottom solidifies, the tank must be opened. The conc- bottom tank can be cleaned without entering. A water hose, handled just outside the cleanout opening. can be used to flush the solids to the center of the cone and drain connection.

Pipe Storage

Pipe that is used specifically for storing and handling liquid petroleum components should be designed and constructed in accordance with applicable codes. Pipe storage consists of any number of sections of line pipe laid parallel to each other and interconnected to operate as a single unit. The size and length depend on the capacity required and economics. The exterior of buried-pipe storage should be coated and wrapped for corrosion protection. It also is recommended that any coated, wrapped, and buried carbon steel pipe be protected cathodically against the possibility of eventual holidays (imperfections) in the coating. Aboveground pipe storage should be protected against the

OIL STORAGE 11-3

TABLE ll.l-PARTIAL API DIMENSIONAL SPECIFICATIONS FOR BOLTED STEEL TANKS’

Capacity

Nominal

Capacity (42-gal

barrel)

100

200

300 250

500 high

Actual Roof and

Capacity Bottom

Level Bolt

Full ID' ClEkS

(42.qal (Nominal) (dtameter) Height

bar&) (ft)

95.80 9

191.64 9

28746 9 266.28 15

53256 15

750 798.84 15

500 low 52201 21

1.000 high 1.044.02 21

1.500"

1.000 low

2.000

3.000

5.000

t 0,000

1,56603 21

993.53 29

1.98706 29

2.980 59 29

5.03745 38

10.21849 54

m ) (W 2% 9

2% 9

2% 9 4% 15

4% 15

4% 15

6% 21

6% 21

6'h 21

85/, 29

as/8 29

8% 29

75/e 38

11% 55

(in ) (ft) (in )

43/4 8 '12

4% 16 1

4% 24 1% 6% 8 '12

6% 16 1

6% 24 1%

8'12 8 ‘12

8% 16 1

8% 24 1 'h

1078 8 '/2

105/8 16 1

10% 24 1%

9% 24 1%

1 '14 24 2

Shell

Per Ring

6

6

6 10

10

uss Gauge

12

12

12

12 12

12 12

10

14

14

14

20

20

20

26

37

12 12 10

12

12 12

12 12

la

12

12

12

12 12 10

12

10 10

to 10

%6 In

Rowsof

Bolts

1

1

1

1 1

1 1

1 1 1

1

1 1

1

1

2t

2

2 2

2 2 2

2

2 2

2 2

3

Bolt Size

JQ

‘/2

‘h

‘12

‘12

‘/2

‘/2

%

% ‘12

‘/2

‘/2

‘12

'12

'12

'12 %

‘12

‘12

'12

%

‘h

‘12

‘/2

‘12

‘12

‘12

‘12

‘h

Bottom

uss Gauge

12

12

12 12

12

12

12

12

12 12

12

12

12

10

10

Cone Roof

Gauge

12

12

12 12

12

12

12

12

12 12

12

12

12

12

12

Chime Seam

(In.) (In.)

‘12 %

‘12 ‘12

‘12 ‘12

‘12 ‘12

‘/2 ‘12

‘12

'12

‘12

‘12

‘12

%

‘12

'12

‘12

‘12

'h

‘12

'12

‘12

'12

‘12

‘12

‘12

%

%

1 l-4 PETROLEUM ENGINEERING HANDBOOK

< 36” CLEANOUT

ME PLATE TANK

*/ ‘$‘, / / \ 10%” BOLT CIRCLE _ -, ,, . _. _ : r8”9 HOLE

THIEF-HATCH CUTOUT PIPE-LINE CONNECTION [c-61

PLAN

CTION

TI

’ LOUTSIDE EDGE OF TANK

DETAIL -iIEF-HATCH CUTOUT

ATT-ACH NAME PLATE TO BRACKET WITH DRIVE SCREWS OR USE NAME-PLATE HOLDER AND WELD TO BRACKET

DETAIL OF NAME-PLATE MOUNT

DETAIL WALKWAY BRACKET

LUGS

1/4” MIN

-SHELL PLATE

4

ELEVATION

Fig. 11.2-Tank dimensions. See Table 11.2.

clcmcnts with paint or other approved coating material. In some cases, pipe storage may require insulation. The individual storage pipes are manifolded together for filling and emptying at pipeline transfer rates. The pipe storage must bc protcctcd from ovcrprcssurc just like any other storage vessel.

Tank Corrosion Protection Coating Specifications3

The primary use of internal coatings is to protect the in side surface of the tank against corrosion while also protecting the stored contents from contamination.

A coating specification should be a clearly defined list of particulars or instructions. Just as a drawing must give exact dimensions. so must a coating specilication state the exact system to he used. In the preparation of such a apec- ification. consideration must be given to such factors as (I) types of coatings available. G!) types of surfaces to he coated. (3) compatibility of coatings. and (4) numhcr of coats required on the various types of surfaces for maximum protection. To secure high-quality coatings. con sideration must he given to the following factors.

Compatibility. 3 In the broadest sense, any discussion of compatibility should include a consideration of the age- old problem of heredity and environment. Environment calculates the compatibility of coatings when applied to various types of surfaces and the operating conditions to which such coatings will hc subjected. Heredity concerns itself with the birth of the coating: formulation. The ha- sic raw materials used in formulating and the art of formulation itself dcterminc whether two paints will he “capable of existing together.”

Film Thickness.” Coating film thickness is now widely recognized as one of the most important factors in obtaining desired performance from a coating system. The required thickness of a coating system will vary, dcpcnd- ing on (I) generic properties of the coating. (2) the type of substrate to which it is applied. and (3) the severity ofthe environment to which the coating is exposed. Film thickness for most protective paints and coatings is generally measured in mils; I mil is % 000 in.

OIL STORAGE 1 l-5

TABLE 11.2-PARTIAL API SPECIFICATIONS FOR SHOP-WELDED TANKS-TANK DIMENSIONS

Nominal Pressure

Capacity (o&q in.)

WY Pressure Vacuum

90 16 ‘h 100 16 ‘12 150 16 %

200 16 ‘12 210 16 ‘h 250 16 ‘h 300 16 ‘/2 400 16 % 500 8 ‘/2

Tolerance (all sizes)

Height of Overflow

Connection*

(ft) (in.)

9 6 7 6

11 6 9 6

14 6 14 6 14 6 19 6 15 6

f l/a in.

Height of Walkway

Location of Fill-Line

Lugs

(ft) (in.)

7 7 5 7 9 7 7 7

12 7 12 7 12 7 17 7 13 7

f Ysin.

Approximate Working

Capacity” WI)

72 79 129 166 200 224 266 366 479

(in.)

14 14 14 14 14 14 14 14 14

*f/8 in.

OD

(ft) (in.) - -

7 11

i 6 6 12 0 10 0 11 0 12 0 12 0 15 6

* 1/8in.

Height (ft)

10 8 12 10 15 1.5 15 20 16

k3/ain.

Size of Connections

(in.)

Cl ,2,3,7 C4.5,6 ____ ~ 3 3 3 3 3 3 3 4 3 4 4 4 4 4 4 4 4 4

‘VISCOUS 011 optlon-when so speclbed on the purchase order, tanks shall be lurnlshed for YISCOUS oil s?rwa? On such tanks, Dlmenslan C of the overflow-lme connecnons shall be 6 m less than shown in Cal 6. and DImewon E of the 11Mne connection shall be 6 m *l/g in

‘The approxmate workmg capacittes shown in Cal 3 apply to flat-bottom tanks Type A (unsklned) cone-bottom tanks have 6 m more working height than the correspondmg flat-bottom tanks The approxunate mcrease IS 4 bbl for the 7-11 11~1” ~damxer

tanks. 6 bbl for the 9.ft 6-1” -dnmeter tanks. 7 bbl for lhe IO-ft.diameter tanks. 8 bbl for the II-ft-diameter tanks. IO bbl for the 12.ft.dmmefer tanks. and 17 bbl for the 15.lt S-in -diameter tanks

Type B (sklrted) cone-bottom lanks have 8 an less working height than the correspondmg II&bottom tanks. The approximate decrease in capacity IS 6 bbl for lhe 7.ft 11.1~ diameter tanks, 8 bbl for the 9.fl 6.in -diameter tanks. 9 bbl for the IO-ft-dlameler tanks. 11 bbl for the 1 I-ft-diameter tanks, 13 bbl for Ihe 12.It-diameter tanks. and 15 bbl for the 15.fl 6.ln -diameter tanks

Surface Preparation.’ The importance of surface preparation would seem so fundamental that it would not deserve mention in specifications; however, poor surface preparation is a major contributing factor of many coating failures. Detailed instructions should be given all along the line and steps taken to see that they are carried out properly. Basically, no coating can be better than the surface over which it is applied. If that surface is dirt, grease, moisture, mill scale, rust, concrete dust, or any other for- cign or intcrfcrence material, failure can be expected. These substances, forming a film between the surface and the coating, soon break down and fall away, taking the coating with them. Such failures cannot be called coating failures. The type of surface preparation required on various surfaces is determined by (I) the nature of the surface itself. (2) the operating conditions to which such surfaces will be subjected, and (3) the type of coating to be applied to the surfaces. As a general rule. metal surfaces that are to be submerged require more thorough surface preparation than those areas that will be

nonsubmerged. The more severe the corrosive atmospheric elements will be, the more thoroughly surface preparation must be carried out. Certain coatings have a better bonding quality than others. Once recognition is given to the unequalness of bonding qualities, it is then a relatively simple matter to be certain that the correct type of surface preparation is carried out, as required, for the various coatings.

Coatings Types. Many types of internal coatings are available for numerous protection requirements. Because of the unlimited types and applications, only a few are described here.

Coal Tar. Among the oldest and most reliable coatings, coal tar has extremely low permeability. protects the sur- fact by the mechanical exclusion of moisture and air, is extremely water resistant, and resists weak mineral acids, alkalis. salts, brine solutions, and other aggressive chemicals well.

1 I-6 PETROLEUM ENGINEERING HANDBOOK

Fig. 11.3-Typical arrangement of internal floating roof

Epoxy Resin. Epoxy resin gives excellent adhesion, toughness, abrasion resistance, flexibility, high gloss and durability, and good chemical and moisture resistance. Typical applications include linings for sour-crude tanks, floating roof tanks, solvent storage tanks, drilling mud tanks. and pipelines.

Rubber Lining. Rubber lining is used as internal lining for storage tanks that are subjected to severe service, such as elevated temperatures, or for protection from extremely corrosive contents such as concentrated chlorides, and various acids, such as chromic, sulfuric, hydrochloric, and phosphoric.

Galvanized. Galvanizing (zinc coating) is highly resistant to most types of corrosion. Bolted steel tanks are ideally suited for galvanizing since all component parts are galvanized by the hot-dip process after fabrication but before erection. Galvanized bolted tanks are recommended where sulfur oil is produced and associated with hydrogen sulfide gas. Galvanizing is also very effective against corrosion in seacoast areas where atmospheric conditions present difficulties in maintaining tank life. Fig. 11.4 shows the expected service life of galvanized coatings in different environments for given thicknesses of galvanizing.’

External. The basic needs for external coatings are protection against weathering exposure and appearance. Many types of external coatings are available, ranging from basic one-coat primers to primers with one or more top coats. Environmental conditions usually dictate the extent of coating applied. Offshore and coastal installations require more extensive coatings compared with inland locations

Cathodic Protection

Cathodic protection can be applied to control corrosion that is electrochemical in nature, whereby direct current is forced to flow onto the entire surface area of the steel structure making it cathodic and thus in a noncorroding state. Self-contained sacrificial anodes are recommended for protecting the interior of tanks and vessels. An impressed-current system is recommended for pipe storage. pipelines, casing in producing wells, etc. In this

00

70

00

150 i 3 40 a E $30

20

IO

I I I I I I II II .25 so .75 1.00 1.15 1.50 ,.,5 2.00 2.25 2.50 2.75 s.00

oz. d zinc/sq. Ft. Of sulfnu

0.4 0.8 1.3 1.7 2.1 2.5 3.0 2.4 3.a 4.2 4.6 5.0 rhkkms4 ol zl"f I" YIIS

Fig. 11.4-Expected service life of galvanized coatings

system, the current is furnished by an AC power system, then rectified to DC current and fed to the structure by the use of a semipermanent anode.

Appurtenances Storage tanks can be provided with any number of appurtenances, depending on the appropriate design codes and user requirements. A tank may be fitted with mix- ers, heaters, pressure/vacuum relief devices, platforms and ladders, gauging devices, manways, and a variety of other connections. Tanks may also be equipped with sumps, inlet and outlet nozzles, temperature gauges, pressure gauges, vents, and blowdowns.

Venting Atmospheric and Low-Pressure Storage Tanks The many abnormal variables that must be considered in connection with tank venting problems make it imprac- ticable to set forth definite simple rules applicable to all locations and all conditions. Larger vents may be required on tanks in which oil is heated, on tanks that receive oil from wells or traps, and on tanks that are subjected to pipeline surges. Similarly, the use of flame arresters or other restrictions that may build up pressure under certain conditions may require the use of larger vents on tanks. The following recommendations for nonrefrigerated aboveground tanks are from API Standard 2000 and set forth determining factors relative to tank venting and pressure/vacuum release requirements. ’

Nonrefrigerated Aboveground Tanks

Determination of Venting Requirements. Conditions for which venting requirements have been set forth include (1) inbreathing resulting from maximum outflow of oil from the tank, (2) inbreathing resulting from contraction of vapors caused by maximum decrease in atmospheric temperature, (3) outbreathing resulting from maximum inflow of oil into the tank and maximum evaporation caused by such inflow, (4) outbreathing resulting from expansion and evaporation that result from maximum increase in atmospheric temperature (thermal breathing), and (5) outbreathing resulting from fire exposure.

OIL STORAGE 11-7

Requirements for Normal Venting Capacity. The normal venting capacity shall be obtained without exceeding the pressure or vacuum that may be applied intermittently to a tank without causing physical damage or permanent deformation to the tank.

The total normal venting capacity shall be at least the sum of the venting requirements for oil movement and thermal effect. *

Inbreathing (Vacuum Relief). The requirement for venting capacity for maximum oil movement out of a tank should be equivalent to 560 cu ft/hr of free air for each 100 bbl (4.200 gal)/hr of maximum emptying rate, including the gravity flow rate to other tanks, for oils of any flash point.

The requirement for venting capacity for thermal inbreathing for a given tank capacity for oils of any flash point should be at least that shown in Col. 2 of Table I I .3.

Outbreathing (Pressure Reliefi. The requirement for venting capacity for maximum oil movement into a tank and the resulting evaporation for oil with a flash point of 100°F or above should be equivalent to 600 cu ftihr of free air for each 100 bbl (4,200 gal)/hr of maximum filling rate.**

The requirement for venting capacity for maximum oil movement into a tank and the resulting evaporation for oil with a flash point below 100°F should be equivalent to 1,200 cu ftihr of free air for each 100 bbl (4,200 gal)ihr of maximum filling rate. ’

The requirement for venting capacity for thermal outbreathing, including thermal evaporation, for a given tank capacity for oil with a flash point of 100°F or above should be at least that shown in Col. 3 of Table 11.3.

The requirement for venting capacity for thermal outbreathing, including thermal evaporation, for a given tank capacity for oil with a flash point below 100°F should be at least that shown in Col. 4 of Table 11.3.

Requirements for Emergency Venting Capacity. When storage tanks are exposed to fire, the venting rate may exceed the rate resulting from a combination of normal thermal effects and oil movement. In such cases, the construction of the tank will determine whether additional venting capacity must be provided.

Tanks With Weak Roof-To-Shell Attachment. On fixed-roof tanks with a roof-to-shell attachment (maximum %,-in. single-fillet weld) as described in the “Roof De- sign” section of API Standard 650, Weld& Steel Tank.~ fix Oil Storage, the roof-to-shell connection will fail preferentially to any other joint. and the excess pressure will be relieved safely if the normal venting capacity should prove inadequate. In tanks built to these specifications, consideration need not be given to any additional requirements for emergency venting.

Tanks Without Weak Roof-To-Shell Attachment. When a tank is not provided with a weak roof-to-shell attachment as previously described, the following procedure shall govern in evaluating the required venting capacity for fire exposure.

‘However the requrec! casncW ma” be reduced for products whcse volatlktv IS such that vapor gener.&on or &d&sat& wlthm the per&stbleoperal~ng range bl vessel pressure wll provide all or part of the vent!ng requrements. In cases m which non- condenslbles are present. this should be taken mto account

“For protect!on agamst hquid OverMing, refer to Sec. 6 05 of API Standard 620, Rec. ommended Rules for Design and Consrn~cbon of Large, Welded, Low-Pressure Storage Tanks

TABLE 11.3-REQUIREMENTS FOR THERMAL VENTING CAPACITY a

1

Tank Capacity

WI) (gal) 60 2,500

100 4,200 500 21,000

1,000 42,000 2,000 84,000 3,000 126.000

4,000 188,000 5,000 210,000

10,000 420,000 15,000 630.000 20,000 840,000 25,000 1.050,000 30,000 1,260.OOO 35,000 1,470.000 40,000 1,680,OOO 45,000 1,890,OOO 50,000 2,100,000 60,000 2,520.OOO 70,000 2,940.000 80,000 3,360,OOO 90,000 3,780,OOO

100,000 4.200,OOO 120,000 5,040,000 140,000 5,880.OOO 160,000 6,720,OOO 180,000 7,560,OOO

Thermal Venting Capacity (cubic feet of free airb per hour)

Outbreathing (Pressure)

2c 3d 4e

Inbreathing Flash Point Flash Point (vacuum) 2 100°F < 100°F

60 40 80 100 60 100 500 300 500

1,000 600 1,000

2,000 1,200 2,000 3,000 1,800 3,000 4,000 2,400 4,000 5,000 3,000 5,000

10,000 6,000 io,ooo 15,000 9,000 15,000 20,000 12,000 20,000 24,000 15,000 24.000 28,000 17,000 28,000 31,000 19,000 31,000 34,000 21,000 34,000 37,000 23,000 37,000 40,000 24,000 40,000 44,000 27,000 44,000 48,000 29,000 48,000 52,000 31,000 52,000 56,000 34,000 56,000 60,000 36,000 60,000 68,000 41,000 68,000 75,000 45,000 75,000 82,000 50,000 82,000 90,000 54,000 90,000

%terpolate for mtermediate tank SIZBS. Tanks wlh a capaaty of more than 180.000 bbl reqwe ~nd1vldua.t studv.

‘At 14 7 ps,a and 60DF ‘For tanks with a capac~tv of 20,000 bbl or more. the rec~u~rements for Ihe vacuum cond,t,on

are very close to thk thewet~cally computed value of 2 cu ft of alrlhr-sq ft of total sheit and roof area. For tanks with a Capacity Of less than 20,000 bbl. lhe requirements for the vacuum condltrx have been based on 1 cu It free alrihr-bbl of tank capacity This IS substantially

,equlvalent to a mea” rate of vaporspace-temperature change of lOOoF per hour For stocks with a flash point of lOOoF or above, the outbreathing requwement has been assumed to be 60% of the mbreathmg requirement The tank roof and shell temperatures can”01 use as rapldly under any condition as they can drop, for example, during a sudden cold ml”

‘For stocks wth a flash point below ICWF. the outbreathlng requirement has been assumed to be equal to the mbreathlng requ,rement to allow for vaponzat~on at the llquld surface and for the higher specllic gravity of the tank vapors

For tanks designed for pressures of 1 psig or below, the total rate of venting shall be determined in accordance with Table 11.4. (No increase in venting is required for tanks with more than 2,800 sq ft of wetted surface area.4)

For tanks and storage vessels designed for pressures of more than 1 psig. the total rate of venting shall be determined in accordance with Table 11.4. However, when the wetted surface area is more than 2,800 sq ft, the total rate of venting shall be calculated by the equation:

q,.=l,l07Ao.**, . ..I. ..(I)

where q,, =venting requirement, cu ft of free air per hour (at 14.7 psia at 60”F), and A=wetted surface area, sq ft.*

‘This formula IS based on 0 = 21 ,OOOA” 82 as given I” API Recommended Praclw 520, Desiqn and Installation of Pressur&elievino &stems 1” Rehnerres. Part I-Dewan The totalheat absorbed, 0, IS I” Btulhr. The c&s&t 1.107 IS derived by convert~g the heat input value Of 21,000 Btulhr-sq 11 to SC1 Of free ar by “se of the late”, heat of vaporwation at 60°F and the molecular weight of hexane


TABLE 11.4-TOTAL RATE OF EMERGENCY VENTING REQUIRED FOR FIRE EXPOSURE VERSUS WETTED SURFACE AREA (NONREFRIGERATED

ABOVEGROUND TANKS)’

Venting Requlrement Venting Requirement Wetted Area* * (cu ft free Wetted Area’ l (cu ft free

wl ft) air+/hr) (sq f0 air+/hr)

20 21,000 350 288,000 30 31,600 400 312,000 40 42,100 500 354,000 50 52,700 600 392,000 60 63,200 700 428,000 70 73.700 800 462.000 80 84: 200 900 493,000 90 94,800 1000 524,000

100 1 05,000 1200 557,000 120 26,000 1400 587,000 140 1 47,000 1600 614,000 160 68,000 1800 639,000 180 190,000 2000 662,000 200 211,000 2400 704,000 250 239,000 2800 742,000 300 265,000 > 2800f

‘Inkrpolak for intermediate values. The total surface area does not Include the area of ground plates bul does include roof areas less than 30 ft above grade

“The wetfed area of ihe fank or storage vessel shall be calculated as iollows For spheres and spheroids. Ihe wetted area IS equal to 55% of the total surface area or the surface area to a height of 30 ft. whichever IS greater For horizontal tanks. the welted area IS equal lo 75% of the total surface area For verhcal tanks the wetled area IS equal to the total surface area of the shell wlthin a maximum height of 30 ft above grade

; A, 14 7 ps,a and 60°F For wetted surfaces larger than 2.800 sq ft. see sect1011 on tanks without weak roof-to-shell altachmenl

The total venting requirements, in cubic feet of free air, determined from Table 11.4 and Eq. I are based on the assumption that the stored liquid will have the characteristics of hexane, since this will provide results within an acceptable degree of accuracy for most liquids encountered. However, if a greater degree of accuracy is desired, the total requirement for emergency venting for any specific liquid may be determined by the following equation for cubic feet of free air per hour:

1.337 T y,,=v

L&i J -& ,.__..,..........,...

where V = cubic feet of free air per hour from Table

11.4 or from Eq. I.

L = latent heat of vaporization of the specific

liquid, in Btu/lbm, M = molecular weight of the specific liquid. and

T = temperature of the relief vapor, “R.

Full credit may be taken for the vent capacity provided for normal venting, since the normal thermal effect can be disregarded during a fire. It can also be assumed that there will be no oil movement into the tank.

If normal vents are inadequate. additional emergency vents shall be provided so that the total venting capacity is at least equivalent to that required by Table I I .4.

The vent size may be calculated on the basis of the pressure that the tank can withstand safely.

When additional protection is provided, the total rate of emergency venting determined at the beginning of this section may be multiplied by (1) a factor of 0.5 when

drainage away from the tank or vessel is provided, (2) a factor of 0.3 when a l-in. thickness of external insulation is provided, (3) a factor of 0. I5 when a 2-in. thickness of external insulation is provided, or (4) a factor of 0.075 when a 4-in. thickness of external insulation is provided.*

Water films covering the metal surfaces can, under ideal conditions. absorb substantially all of the incident radiation. However, the reliability of effective water application depends on many factors. Freezing weather. high winds, clogging of the system, unreliability of the water supply, and tank surface conditions are a few factors that may prevent adequate or uniform water coverage. Because of these uncertainties, the use of an environmental factor other than I .O for water spray is generally discouraged.

Means of Venting. Normal vents. Normal venting shall be accomplished by a pilot-operated relief valve. a pressure-relief valve, a pressure vacuum valve, or an open vent with or without a flame-arresting device in accordance with the following requirements.

If a pilot-operated relief valve is used, it shall be designed so that the main valve will open automatically and will protect the tank in the event of failure of the pilot valve diaphragm or another essential function device. Relief valves equipped with a weight and lever preferably should not be used.

A pressure-relief valve is applicable on tanks operating above atmospheric pressure; in cases in which a vacuum can be created within a tank, vacuum protection may be required.

Pressure vacuum valves are recommended for use on atmospheric storage tanks in which oil with a flash point

‘The values for msulatlon are based on an arbitrary thermal conductlvily af 4 Btulhrisq ft/(°F/ln of thickness) The msulat~on shall res,st dlslodgment by fwhose strums and shall be nancambustlble

OIL STORAGE 11-9

below 100°F is stored and for use on tanks containing oil that is heated above its flash point. A flame arrester is not considered necessary for use in conjunction with a pressure vacuum valve because flame speeds are less than vent velocities through pressure vacuum valves. (See API Petroleum Safety Data 2210, Flume Arresrcrs for

Trrnk Verm. )

Open vents with a flame-arresting device may be used in place of pressure vacuum valves on tanks in which oil with a flash point below 100°F is stored and on tanks containing oil that is heated above its flash point.

Open vents may be used to provide venting capacity for tanks in which oil with a flash point of 100°F or above is stored, for heated tanks in which the oil’s storage temperature is below the oil’s flash point, for tanks with a capacity of less than 59.5 bbl (2.500 gal) used for storing any product. and for tanks with a capacity of less than 3,000 bbl (126,000 gal) used for storing crude oil.

In the case of viscous oils, such as cutback and penetration-grade asphalts, where the danger of tank collapse resulting from sticking pallets or from plugging of flame arresters is greater than the possibility of flame transmission into the tank, open vents may be used as an exception to the previously outlined requirements for pressure vacuum valves or flame-arresting devices.

Emergency Vents. Emergency venting may be accomplished by use of (I) larger or additional open vents as limited by normal vent requirements, (2) larger or additional pressure vacuum valves or pressure relief valves, (3) a gauge hatch that permits the cover to lift under abnormal internal pressure, (4) a manhole cover that lifts when exposed to abnormal internal pressure. (5) a connection between the roof and the shell that is weaker than the weakest vertical joint in the shell or the shell-to-bottom connection. * and (6) other forms of construction demon- strably comparable for the purposes of pressure relief.

Vent Discharge. For tanks located inside a building, discharge from the vents shall be to the outside of the building. A weak roof-to-shell connection shall not be used as a means for emergency venting a tank inside a building.

Materials of Construction Metallic

Shop- and field-welded, and bolted storage tanks are customarily fabricated from mild-quality carbon steel. Most common for welded tanks are A-36 structural steel and A-283 Grade C structural-quality carbon steel. Sheet- gauge steel for bolted tanks is of commercial quality having a minimum tensile strength of 52,000 psi. For hydrogen sulfide crude storage, aluminum bolted tanks or aluminum decks only are often used. Various API codes (listed in General References) to which the storage tank is fabricated set forth the welding procedures. inspection procedures. and testing requirements.

Nonmetallic

Nonmetallic tanks customarily are constructed from plastic materials. These have the advantage of being noncorroding. durable, low-cost, and lightweight. Plastic materials used in the construction are polyvinyl chloride,

‘A tank wh a roof.trxhelf attachment (maximum %.-m smgle-ftllet weld) as described I” the “Roof Desngn” sectm of API Standard 650 is recognwd as hawng a weak- sea,” CO”“~C,K,” and will therefore not require emergency “e”ts

polyethylene, polypropylene, and fiberglas+reinforced polyesters (FRP’s). The FRP tanks are available in the larger sizes and are the most common.* FRP tanks are suitable for outdoor as well as indoor applications. Aboveground vertical FRP tanks can store 24.000 gal and more, depending on the shell construction.

The temperature limits of plastic tanks are approximately 40 to 150°F. Color must be added to the outer liner for protection against ultraviolet radiation. The inner liner must be selected for compatibility with the product stored. Protection from mechanical abuse such as impact loads is necessary. Good planning dictates that plastic storage should not be located next to flammable storage tanks. Special attention should be given to local codes, or- dinances, and provisions for insurance relative to storing a flammable product in a flammable container. All plastic tanks used for storage service should be equipped with pressure-relief devices if designed for relatively low- pressure storage.

Production Equipment Tank-Battery Connections

The suggested setting and connection plan for a typical tank battery is shown in Figs. 1 I .5 and I I .6. The pipeline connection in the tank should be located directly below the thief hatch and a minimum of 12 in. above the tank bottom. It should be equipped with a valve and sealing device immediately adjacent to the tank. Pipeline valves should be checked frequently for leaks.

Inlet connections preferably should be located in the deck of the tank and should have a valve located near the inlet capable of closing off against pressure.

Drain connections should be located immediately above the tank bottom in the side of the tank or in the tank bottom immediately adjacent to the side. They should be equipped with a valve and sealing device located next to the tank. Drains from all tanks in a battery should be connected together and piped well away from the tanks.

Equalizer or overflow connections should be installed below the deck in the tank shell. A valve and sealing device should be installed immediately ad,jacent to the tank if more than two tanks are in the battery and should be connected in such a manner that any two tanks can be equalized together.

Vent connections should be installed in the center of the tank deck and all tanks connected to a common line. This line should have a pressure-vacuum valve installed in the line or on the end of it. The line should be sloped to prevent accumulation of liquids in it or in the valve.

The use of gas to roll stored products is usually considered poor practice, and should be restricted to temporary or emergency use. If a roller line is used, it should enter the tank through the deck and be equipped with a valve next to the tank.

Tank-Battery Installation and Hookup

A tank battery should contain at least two tanks and usually have a capacity equal to 4 days’ production. All tanks should be level with each other and have a minimum spacing of 3 ft between tanks. Local codes or specifications may require a firewall and different spacing.

‘An application for approval of fIberglass tanks was submltted during 1984 and the flnal draft IS now pendlng approval by the API general membershlp

II-10 PETROLEUM ENGINEERING HANDBOOK

Fig. 11.5-Schematic of lease tank battery installation.

Tank Battery for Hydrogen Sulfide Crude Storage

Constant attention should be given to the hazardous condition created by iron sulfide deposits. These occur most frequently within the vapor space and particularly on the underneath exposed side of the deck. These iron sulfide deposits generate severe corrosion that can go unnoticed when deck conditions are observed from the topside only. When sour crude is stored, all openings on the tanks should be kept closed since hydrogen sulfide is poisonous. This can be accomplished by equipping the tanks with some type of ground-level gauging and thermometers located in the tank shell. Gauges and temperatures then can be read from the ground without the tank being opened. These gauging devices usually require approval by the crude purchaser. Ground-level sampling also can be accomplished by installing pipes that extend into the tank

at any desired level and to any desired distance. Valves are located at a convenient level to permit sampling on the ground without the tanks being opened. If available. a small amount of sweet gas should be fed into the top of the tank continuously to establish a “gas sweep.” This will ensure positive pressure within the tank at all times and will prohibit air from entering the tank, thereby greatly reducing corrosion. It is advisable to extend the tank

vent line well beyond the tank battery and to use a backpressure valve and flash arrester in the vent line to burn the vapors.

Maintenance and Operation of Tank Batteries*

Steel tanks should be kept clean and free from spilled oil or other material. They should be kept painted and all water or accumulated dirt should be removed from around

EOUALIZER AND PIPE-LINE-OUTLET TANK CONNECTIONS TO HAVE VALVE ADJACENT TO EACH TANK AND ACCESSIBLE FOR SEALING

Fig. 11.6-Plan view for lease tank battery installation

OIL STORAGE

the bottom edge of the tanks. Thief hatches and vent-line valves should be kept closed and inspected periodically for proper operation and gasket condition. Should any leaks occur, they may be repaired temporarily with lead sealing plugs or toggle bolts. These leaks should be repaired permanently as soon as possible.

When a closing gauge is taken, and before the tank is filled again, the pipeline valve should be sealed closed, the drain valve checked to ensure that it is closed and the seal removed, and then the seal from the equalizer-line valve removed. Before the tank is accepted by the crude purchaser, the water should be drained from the tank if necessary and the valve sealed closed. All other valves should be sealed closed except the vapor-recovery-line valve if such a system is in use. The pipeline valve is then unsealed and opened for delivery to the purchaser.

Tank Grades

Selection. Selection of the proper location on the lease for storage tanks is of prime importance. The location should provide good drainage and be on well-packed soil-not a fill-if possible. The tank foundation or grade should be slightly elevated, level, and somewhat larger in diameter than the tank itself. For steel tanks, either bolted or welded, the best grade is one made of small gravel. crushed rock, etc., held in place by steel bands. This type of grade allows no water to stand underneath the tank and provides air circulation. If the tank is to be set directly on the ground, felt tar paper should be applied to the grade first and the tank set on this. If concrete is used for the grade, it should be slightly larger in diameter than the tank and have shallow grooves on the surface to provide air circulation. Many codes. standards, and specifications regulate the location, design, and installation of storage tanks dependent on their end use. Selecting the proper specification and providing adequate fire protection for the installation may lower insurance rates over the life of the installation.

11-11

Firewalls or Dikes. Dikes are provided to contain the volume of a certain portion of the tanks enclosed depending on the tank contents. They are used to protect surrounding property from tank spills or fires. In general, the net volume of the enclosed diked area should be the volume of the largest tank enclosed (single-failure concept). The dike walls may be earth, steel, concrete, or solid masonry designed to be watertight with a full hydrostatic head behind them. Local codes and specifications may govern construction. If more than one tank is within the diked area, curbs or preferably drainage channels should be provided to subdivide the area to protect the adjacent tanks from possible spills.

Vapor Losses Vapors emitted from the vents and/or relief valves of a storage tank are generated in two ways: (I) they are forced out of the tank during filling operations and (2) they are generated by vaporization of the liquid stored in the tank. The total vapor produced in these two instances is the volume that would be available for recovery.

Filling Losses

Vapors that are forced out of the tank are generally called “filling losses.” A storage tank generally is not pumped completely dry when emptied. The vapor above the remaining liquid in the tank will expand to fill the void space at the vapor pressure of the liquid stored in the tank at storage temperature. As the tank is filled, the vapors are compressed into a smaller void space until the set pressure on the vent/relief system is reached. There are also some filling losses that are associated with the expansion of the liquid entering the tank. Fig. 11.7 provides a graphi- cal approach to estimating the filling losses as a percentage of the liquid being pumped into the tank.

Filling lasses from storage containers

Fig. 11.7-Filling losses from storage containers


OrwIly Lou, “API

Fig. 11.8-Gravity loss in degrees API vs. percent loss by

Vaporization Losses

This type of loss is characterized as the vapors generated by heat gain through the shell, bottom. and roof. The total heat input is the sum of the radiant, conductive, and convective energy forces. This type of loss is especially prevalent where light hydrocarbon liquids arc stored in full-pressure or refrigerated storage. This is less prevalent but still quite common in crude oil and finished- product storage tanks. These vapors may be recovered by the use of a vapor-recovery system.

Vapor Control and Gravity Conservation With Storage Tanks9 Crude oils and condensates are composed of many different paraffin hydrocarbons. Propane is the lightest hydrocarbon found in any measureable amount and the hydrocarbon with the greatest tendency to evaporate or vaporize from the liquid stored. When propane and other hydrocarbons pass into the vapor phase by vaporization, the volume of the liquid stored is decreased. and because these lighter hydrocarbons arc not now present in their initial amounts, the API gravity of the crude is decreased. There is a definite relationship between API gravity lost and volume lost. depending on the character of the crude (Fig. 11.8).

Factors Contributing to Vapor and Gravity Losses Several factors affect and contribute to vapor and gravi-

ty losses in storage tanks: (I) vapor pressure of the product stored, (2) temperature of the product stored. (3) surface area of the product stored. (4) agitation of the product stored, (5) pressure on the storage tanks, (6) filling loss- cs from the storage tanks, (7) breathing losses from the storage tanks, (8) size of the storage tanks, and (9) color of outside paint or coating.

Several, if not all, of these factors usually contribute to the total loss from any one tank or battery.

Vapor Pressure. The true vapor pressure (TVP) of a liquid is the actual pressure it exerts on the vapor space in a container at a given temperature. Water, for example, has a TVP of 1 psi at 100°F and a TVP of 14.7 psi at 212”F, yet it must be kept in a closed container to prevent evaporation. The same is true for crude oil if the TVP is below 14.7 psi. Crudes with a TVP of 10 psi and lower are usually relatively stable in closed-atmospheric storage.

Temperature. Temperature of crude is directly related to its vapor pressure. For example, a crude with a TVP of 8 psi at 50°F will have a TVP of 17 psi at 90°F. The vaporization loss is then approximately doubled at the 90°F temperature.

Surface Area. Directly related to the rate of evaporation is the surface area of the crude. Take, for example, two tanks with a capacity of 500 bbl each, one a high 500-bbl tank and the other a low .500-bbl tank. If both are filled one-half full, the high 500 bbl has 0.74 sq ft of surface area exposed per barrel stored, whereas the low 500 bbl has 1.46 sq ft of surface area exposed per barrel stored. The low SOO-bbl tank then has twice the evaporation rate of the high 500-bbl tank.

Agitation. Agitation of the stored product is related directly to the vapor pressure. If two crudes under the same conditions receive equal agitation, the one with the higher vapor pressure will show the greatest evaporation loss.

Tank Pressures. The higher the pressure maintained on the storage tank, the less will be the tendency for the crude to evaporate. Pressure storage, considered to be in excess of 1 psig, is required for all stored products with a TVP in excess of 14.7 psi to prevent excessive evaporation losses. High-gravity crudes and distillates or condensates usually require a higher storage pressure than the normal 1 to 4 oz. The crude purchaser often dictates allowable storage pressure.

Filling Losses. When 475 bbl are run from a 500-bbl tank, crude-oil vapors occupy the displaced oil. When the tank is filled again, these vapors are forced from the tank into the atmosphere. These expelled vapors may be equivalent to one or more barrels, depending on the type of crude.

Breathing Losses. Temperature changes between day and night cause vapors to be expelled from the tank and air to be breathed in. These reactions are similar to, but smaller in volume than, the filling and running losses.

Storage Size. A greater vapor space and longer storage time will increase evaporation losses. As an example, consider two tanks with 100 bbl of stored crude each, one a 250-bbl tank and the other a high 500-bbl tank. The 250-bbl tank has 948 cu ft of vapor space while the high 500-bbl tank has 2,457 cu ft or two and one-half times as much. This added vapor space increases the evaporation losses from the larger tank.

Preventing Evaporation and Gravity Losses

Much can be done by the producer to prevent undue losses of crude oil by evaporation. Products should be introduced into storage as cool as possible and kept that way. Some types of heat-exchange equipment should be employed between an emulsion treater, or other heating equipment, and the tanks to cool the oil before it enters storage. If fluid heat exchangers are used, a preventive maintenance program should be employed to guard against buildup of

OIL STORAGE 11-13

scale. paraffin. salt. etc.. which are common to many pro duced fluids. Most modern tank batteries are equtpped with lease automatic custody-transfer (LACT) units. The run tank should be of sufficient size to allow approximately I2 hours’ settling time. Where batteries are equipped with storage for bad oil. this storage should be kept to a minimum and the battery treating capacity should be capable of treating a certain amount of bad oil. Steel tanks should be painted with a reflective or white paint. Tests show the vapor-space temperature of a tank painted with aluminum paint to average 4%“F above atmospheric temperature. while a red-painted tank averaged 14°F above atmospheric temperature.

Tanks should be selected with smaller diameters. greater heights. and smaller capacities, all other considerations being equal. These factors will allow the stored product to have relatively smaller surface areas and vapor spaces as well as a shorter length of storage time before being sold.

Downcomer pipes prevent undue agitation in the tank. They are usually made by installing a line inside the tank from the inlet connection in the tank deck to 1 ft above the tank bottom. The downcomer must have a vacuum breaker hole at the top to allow gas to escape and thereby prevent agitation, splashing, and accumulation of static clcctricity.

All tank openings should be maintained closed and pressure on the tank should be as high as practical (at least r/z in. of water column). Tanks in a battery all should be connected together into a common vent line to keep breathing and filling losses to a minimum. Bypass thief hatches are manufactured that will do much to prevent evaporation losses when a tank is gauged through the thief hatch. These special hatches have the tank-battery vent lines connected to them and will close off or isolate all other tanks except the one being gauged. This allows all other tanks in the battery to maintain their pressure while the tank being gauged is depressured.

The producer may install one of several types ofground- level gauging and sampling devices available that will permit gauging and sampling without opening the tank. The tank remaining closed goes far toward eliminating evaporation losses.

Vapor-Recovery System

Vapor-recovery systems are of two basic types. One type connects a vacuum line to the tank and transports the tank vapors to a processing or gasoline plant. The other type consists of a small compressor located by the tank hat- tery. which compresses the tank vapors to a pressure suitable for lease use or sales.

Vacuum-Line System. The vacuum-line system usually is found only in large oil fields, where many tank batteries can be connected together into a relatively short gathering system. This system must employ well-maintained and properly functioning pressure/vacuum relieving devices and dependable control valves to prevent the tanks from collapsing or air from entering the gathering system.

Compression System. The compressor system is usually electrically driven and all components are skid mounted. Some of these systems use a vane-type compressor

and inject a refined oil by way of a lubrication system to seal the vanes against the compressor walls. In these systems an actual liquid recovery is accomplished by the sealing oil absorbing the condensed hydrocarbons from the compressed vapors and transporting them to storage with the returning sealing oil. Applications for this type of system are twofold: (I) compression of the rich stock- tank vapors for sale to a gasoline plant and (2) the recovery of liquids from the rich stock-tank vapors. Liquid hydrocarbons also can be recovered from the compressed vapor-recovery unit (VRU) vapor by either one or both of the following means.

I. A vapor cooling system (air-cooled or water-cooled) heat exchanger can be installed complete with a separator downstream of the hot compressed VRU vapor stream. Multiple vapor cooling systems may he used in between the stages of a multiple-staged compressron system.

2. A mechanical refrigeration unit may be installed downstream of the VRU for a higher-yield liquid hydrocarbon recovery.

Underground Storage Underground storage* is most advantageous when large volumes are to be stored. Underground storage is especially advantageous for high-vapor-pressure products. Solution-mined and conventionally mined caverns are not typically used for underground storage of refrigerated products. Underground storage allows most of the surface area (except for the entry wells) to be used for other purposes. This is especially beneficial in high-value, congested areas.

Type of Construction

Types of underground storage are (I ) caverns constructed in salt by solution minmg or conventional mining, (2) caverns constructed in nonporous rock by conventional mining, (3) caverns developed by conversion of depleted coal. limestone, or salt mines to storage, and (4) depleted reservoirs.

The solution-mined cavern is constructed by drilling a well or wells into the salt and circulating low-salinity water over the salt interval to dissolve the salt. Fig. 1 I .9 shows a typical solution-mined cavern.

Conventionally mined caverns can be constructed any place a nonporous rock is available at adequate depth to withstand product pressures. An engineer or geologist experienced in underground storage should evaluate any specific site for the feasibility of constructing underground storage. Most product caverns are constructed in shale, limestone, dolomite, or granite. This type cavern is opcr- ated “dry” (the product is recovered by pumping).

Operation

The cavern may be operated by brine displacement of product or pump-out methods (see Figs. 1 I .9 and 1 I. IO).

Most solution-mined caverns are operated by the brine- displacement technique (Fig. 1 I .9). A suspended displacement string of casing is installed near the bottom of the cavern. and product is injected into the annulus between the product casing (casing cemented at cavern roof) and the displacement casing, forcing brine up the displace-

II-14 PETROLEUM ENGINEERING HANDBOOK

Ertne In/Out -

Producl

L-L- Product

2L.L.

Fig. 11.9-Brine displacement cavern operation (solution-mined Fig. ll.lO-Pump-out cavern operation (fracture-connected

cavern). solution-mined cavern in bedded salt).

ment casing. The procedure is reversed for product recovery. In this type operation, a brine storage reservoir usually is provided.

Some solution-mined caverns are operated “dry” by installing a pump at cavern depth either within the cavern or in a well connected to the cavern by fracturing. Both submersible electrically driven pumps and deep-well vertical multistage pumps are used for this purpose (Fig. 11.10).

References I. API Specification 12B: Specific~ti~n~for Bolred Prnducfion Tunks,

12th edition. API Div. of Production, Dallas (Jan. 1977). 2. API Specification 12F: Specifications for Shop- Welded Tanks Ji)r

Storage of Production Liquids, eighth edltmn, Dallas (Jan. 1982). 3. Koppers Protective Coatings, Koppers Co. Inc.. Pittsburgh (Ott

1980). 4. Koppers Protective Coatings. Koppers Co. Inc., Pittsburgh (Jan.

19801. 5. Koppers Protective Coatings, Koppera Co. Inc., Pittsburgh (March

1981). 6. Design and Fabn’~d~~ ojG&anrzed Products, American Hot Dip

Galvanizers Assn., and the Zinc Inst. (Nov. 1983). 7, API Standard 2o(M: Vmtmg Atmosphenc and Low-Prrssuw Storcrge

Tanks, third edition, Dallas (Jan. L982).

8. API RP 12RI: Recommended Prwtice for SrttinR. Connwting, Main- tenance and Operation ofleuse Tanks, second edition, Dallas (Feb. 1981).

9. Vapor and Gravity Conrro( in Crude Oil Production. first edirion, Petroleum Extension Service. U. of Texas, Div of Extension. Austin (1956)

Chapter 12

Oil and Gas Separators H. Vernon Smith, Meridian Corp.

Summary This chapter is a discussion of the design, use, functions, capacities, classifications, performance, operation, and maintenance of oil and gas separators. Vertical. horizontal. and spherical separators in both two- and three-phase arrangements are discussed. Quality of effluent fluids is approximated. Equations for calculating the sizes and capacities of separators and capacity curves and tables for sizing oil and gas separators are provided. These capacity curves and tables can be used to estimate capacities of separators as well as to determine the size of separator required to handle given volumes of fluids. Sample calculations for sizing separators are included.

Introduction The term “oil and gas separator” in oilfield terminology designates a pressure vessel used for separating well fluids produced from oil and gas wells into gaseous and liquid components. A separating vessel may be referred to in the following ways:

1, Oil and gas separator. 2. Separator. 3. Stage separator. 4. Trap. 5. Knockout vessel, knockout drum. knockout trap,

water knockout, or liquid knockout. 6. Flash chamber, flash vessel, or flash trap 7. Expansion separator or expansion vessel. 8. Scrubber (gas scrubber). dry or wet type. 9. Filter (gas filter). dry or wet type.

IO. Filter/separator. The terms “oil and gas separator.” “separator,” “stage

separator,” and “trap” refer to a conventional oil and gas separator. These separating vessels are normally used on a producing lease or platform near the wellhead, manifold, or tank battery to separate fluids produced from oil and gas wells into oil and gas or liquid and gas. They must be capable of handling “slugs” or “heads” of well

fluids. Therefore, they are usually sized to handle the highest instantaneous rates of flow.

A knockout vessel, drum, or trap may be used to remove only water from the well fluid or to remove all liquid, oil plus water, from the gas. In the case of a water knockout for use near the wellhead, the gas and liquid petroleum are usually discharged together, and the free water is separated and discharged from the bottom of the vessel.

A liquid knockout is used to remove all liquid, oil plus water. from the gas. The water and liquid hydrocarbons are discharged together from the bottom of the vessel, and the gas is discharged from the top.

A flash chamber (trap or vessel) normally refers to a conventional oil and gas separator operated at low pressure, with the liquid from a higher-pressure separator being “flashed” into it. This flash chamber is quite often the second or third stage of separation, with the liquid being discharged from the flash chamber to storage.

An expansion vessel is the first-stage separator vessel on a low-temperature or cold-separation unit. This vessel may be equipped with a heating coil to melt hydrates, or a hydrate-preventive liquid (such as glycol) may be injected into the well fluid just before expansion into this vessel.

A gas scrubber may be similar to an oil and gas separator. Usually it handles fluid that contains less liquid than that produced from oil and gas wells. Gas scrubbers are normally used in gas gathering, sales, and distribution lines where they are not required to handle slugs or heads of liquid, as is often the case with oil and gas separators. The dry-type gas scrubber uses mist extractors and other internals similar to oil and gas separators. with preference shown to the coalescing-type mist extractor. The wet- type gas scrubber passes the stream of gas through a bath of oil or other liquid that washes dust and other impurities from the gas. The gas is flowed through a mist ex-

PETROLEUM ENGINEERING HANDBOOK

Fig. 12.1-Typical surface production equipment for handling oil and gas-oil and gas separators and other related equipment.

tractor where all removable liquid is separated from it. A “scrubber” can refer to a vessel used upstream from any gas-processing vessel or unit to protect the downstream vessel or unit from liquid hydrocarbons and/or water.

The “filter” (gas filter or filter/separator) refers to a dry-type gas scrubber, especially if the unit is being used primarily to remove dust from the gas stream. A filtering medium is used in the vessel to remove dust, line scale, rust, and other foreign material from the gas. Such units will normally remove liquid from the gas.

An oil and gas separator generally includes the following essential components and features.

1. A vessel that includes (a) primary separation device and/or section, (b) secondary “gravity” settling (separating) section, (c) mist extractor to remove small liquid particles from the gas, (d) gas outlet, (e) liquid settling (separating) section to remove gas or vapor from oil (on a three-phase unit, this section also separates water from oil), (f) oil outlet, and (g) water outlet (three-phase unit).

2. Adequate volumetric liquid capacity to handle liq-

uid surges (slugs) from the wells and/or flowlines. 3. Adequate vessel diameter and height or length to al-

low most of the liquid to separate from the gas so that the mist extractor will not be flooded.

4. A means of controlling an oil level in the separator, which usually includes a liquid-level controller and a diaphragm motor valve on the oil outlet. For three-phase operation, the separator must include an oil/water interface liquid-level controller and a water-discharge control valve.

5. A backpressure valve on the gas outlet to maintain a steady pressure in the vessel.

6. Pressure relief devices. In most oil and gas surface production equipment sys-

tems, the oil and gas separator is the first vessel the well fluid flows through after it leaves the producing well. However, other equipment-such as heaters and water knockouts-may be installed upstream of the separator. Fig. 12.1 shows a typical surface production equipment system for handling crude oil using an oil and gas separator along with related equipment.

OIL AND GAS SEPARATORS 12-3

Well Fluids and Their Characteristics Primary Functions of Oil and Some of the physical characteristics of well fluids han- Gas Separators dled by oil and gas separators are briefly outlined in this section.

Separation of oil from gas may begin as the fluid flows through the producing formation into the wellbore and may progressively increase through the tubing, flowlines,

Crude Oil. Crude oil is a complex mixture of hydro- and surface handling equipment. Under certain conditions,

carbons produced in liquid form. The API gravity of crude the fluid may be completely separated into liquid and gas

oil can range from 6 to 5O”API and viscosity from 5.0 before it reaches the oil and gas separator. In such cases,

to 90,000 cp at average operating conditions. Color var- the separator vessel affords only an “enlargement” to per-

ies through shades of green, yellow, brown, and black. mit gas to ascend to one outlet and liquid to descend to

Detailed characteristics of crude oils are given in another.

Chap. 21. Removal of Oil From Gas

Condensate. This is a hydrocarbon that may exist in the producing formation either as a liquid or as a condensible vapor. Liquefaction of gaseous components of the condensate usually occurs with reduction of well-fluid temperature to surface operating conditions. Gravities of the condensed liquids may range from 50 to 120”API and viscosities from 2.0 to 6.0 cp at standard conditions. Color may be water-white, light yellow, or light blue.

Difference in density of the liquid and gaseous hydrocarbons may accomplish acceptable separation in an oil and gas separator. However, in some instances, it is necessary to use mechanical devices commonly referred to as “mist extractors” to remove liquid mist from the gas before it is discharged from the separator. Also, it may be desirable or necessary to use some means to remove nonsolution gas from the oil before the oil is discharged from the separator.

Natural Gas. A gas may be defined as a substance that Removal of Gas From Oil

has no shape or volume of its own. It will completely fill any container in which it is placed and will take the shape of the container. Hydrocarbon gas associated with crude oil is referred to as natural gas and may be found as “free” gas or as “solution” gas. Specific gravity of natural gas may vary from 0.55 to 0.90 and viscosity from 0.01 1 to 0.024 cp at standard conditions.

Free Gus. Free gas is a hydrocarbon that exists in the gaseous phase at operating pressure and temperature. Free gas may refer to any gas at any pressure that is not in solution or mechanically held in the liquid hydrocarbon.

S&&ion Gas. Solution gas is homogeneously contained in oil at a given pressure and temperature. A reduction in pressure and/or an increase in temperature may cause the gas to be emitted from the oil, whereupon it assumes the characteristics of free gas.

Condensible Vapors. These hydrocarbons exist as vapor at certain pressures and temperatures and as liquid at other pressures and temperatures. In the vapor phase, they assume the general characteristics of a gas. In the vapor phase. condensible vapors vary in specific gravity from

0.55 to 4.91 (air= 1 .O), and in viscosity from 0.006 to 0.011 cp at standard conditions.

Water. Water produced with crude oil and natural gas may be in the form of vapor or liquid. The liquid water may be free or emulsified. Free water reaches the surface separated from the liquid hydrocarbon. Emulsified water is dispersed as droplets in the liquid hydrocarbon.

Impurities and Extraneous Materials. Produced well tluids may contain such gaseous impurities as nitrogen, carbon dioxide, hydrogen sulfide, and other gases that are not hydrocarbon in nature or origin. Well fluids may contain liquid or semiliquid impurities, such as water and paraffin. They may also contain solid impurities, such as drilling mud, sand, silt. and salt.

The physical and chemical characteristics of the oil and its conditions of pressure and temperature determine the amount of gas it will contain in solution. The rate at which the gas is liberated from a given oil is a function of change in pressure and temperature. The volume of gas that an oil and gas separator will remove from crude oil is dependent on (1) physical and chemical characteristics of the crude, (2) operating pressure, (3) operating temperature, (4) rate of throughput, (5) size and configuration of the separator, and (6) other factors.

Rate of throughput and liquid depth in the separator determine the “retention” or “settling” time of the oil. Retention time of 1 to 3 minutes is generally adequate to obtain satisfactory separation of crude oil and gas unless foaming oil is being handled. When foaming oil is separated, retention time should be increased to 5 to 20

minutes, dependent on the stability of the foam and on the design of the separator. Advancements in field processing systems and production procedures-such as automatic custody transfer-emphasize the need for complete removal of nonsolution gas from the oil. Agitation, heat, special baffling, coalescing packs, and filtering materials can assist in the removal of nonsolution gas that otherwise may be retained in the oil because of the viscosity and surface tension of the oil.

Separation of Water From Oil

In some instances it is preferable to separate and to remove water from the well fluid before it flows through pressure reductions, such as those caused by chokes and valves. Such water removal may prevent difficulties that could be caused downstream by the water-such as corrosion, hydrate formation, and the formation of tight emulsion that may be difficult to resolve into oil and water.

The water can be separated from the oil in a three-phase separator by use of chemicals and gravity separation. If the three-phase separator is not large enough to separate the water adequately, it can be separated in a free-water


Fig. 12.2—Horizontal skid-mounted three-phase well tester on offshore drilling platform off coast ofBrazil.

knockout vessel installed upstream or downstream of theseparators. If the water is emulsified, it may be neces-sary to use an emulsion treater to remove it. Figs. 12.2through 12.5 are illustrations of three-phase separators.

Secondary Functions of Oil andGas SeparatorsMaintain Optimum Pressure on SeparatorFor an oil and gas separator to accomplish its primaryfunctions, pressure must be maintained in the separatorso that the liquid and gas can be discharged into their re-spective processing or gathering systems. Pressure is

FLOAT NOZZLES MIST EXTRACT

NONWEIGHTED MOTOR VALVE (DMV)

WEIGHTEDFLOAT

SECTION A-A SECTION B-B

maintained on the separator by use of a gas backpressurevalve on each separator or with one master backpressurevalve that controls the pressure on a battery of two or moreseparators. Fig. 12.6 shows a typical low-pressure gasbackpressure valve, and Fig. 12.7 shows a typical high-pressure gas backpressure valve used to maintain thedesired pressure in separators.

The optimum pressure to maintain on a separator is thepressure that will result in the highest economic yield fromthe sale of the liquid and gaseous hydrocarbons. This op-timum pressure can be calculated theoretically or deter-mined by field tests.

GAS BPV-

/

INLETFLUID IN-+ 5EPARATlNG

E L E M E N T

NONWEIGHTEDF L O A T .

WEIGHTED -FLOAT

Fig. 12.3—Schematic of typical horizontal three-phaseoil/gas/water separator.

Fig. 12.4—Schematic of a typical vertical three-phaseoil/gas/water separator.


GASOUTLET

CENTRIFUGAL-TYPEINLET SEPARATING

FLUIDbI - - - - -

OIL LEVEL

NONWEIGHTED DT

Fig. 12.5—Schematic of a typical spherical three-phaseoil/gas/water separator.

Fig. 12.6—Low-pressure-gas backpressure valve.

Maintain Liquid Seal in SeparatorTo maintain pressure on a separator, a liquid seal mustbe effected in the lower portion of the vessel. This liquidseal prevents loss of gas with the oil and requires the useof a liquid-level controller and a valve similar to thoseshown in Figs. 12.8 and 12.9. A lever-operated valvesimilar to the one shown in Fig. 12.10 can be used to

Fig. 12.7—High-pressure-gas backpressure valve.

maintain the liquid seal in a separator when the valve isoperated by a float that is actuated by the oil level in theseparator. The oil discharge control valve shown in Fig.12.9 can be actuated by a float-operated pilot (not illus-trated), by a floatless liquid-level controller similar to theone shown in Fig. 12.11, or by a torque tube-type (dis-placement) liquid-level controller similar to the one shownin Fig. 12.8.


Fig. 12.8—Torque-tube (displacement)-type liquid-level con-troller.

Fig. 12.9—Diaphragm-motor-type oil-discharge control valve.

Special Problems in Oil andGas Separation

Separating Foaming Crude OilWhen pressure is reduced on certain types of crude oil,tiny spheres (bubbles) of gas are encased in a thin filmof oil when the gas comes out of solution. This may re-sult in foam, or froth, being dispersed in the oil and cre-ates what is known as “foaming” oil. In other types ofcrude oil, the viscosity and surface tension of the oil may

Fig. 12.10—Lever-type valve for controlling oil discharge fromoil and gas separators. Valve is float operated.

mechanically lock gas in the oil and can cause an effectsimilar to foam. Oil foam will not be stable or long-lastingunless a foaming agent is present in the oil. Crude oil ismore likely to foam when (1) the API gravity is less than40° API, (2) the operating temperature is less than 160°F,and (3) the crude oil is viscous, having a viscosity great-er than 5,000 SSU (about 53 cp).

Foaming greatly reduces the capacity of oil and gasseparators because a much longer retention time is re-quired to separate adequately a given quantity of foam-ing crude oil. Foaming crude oil cannot be measuredaccurately with positive-displacement meters or with con-ventional volumetric metering vessels. These problems,combined with the potential loss of oil and gas becauseof improper separation, emphasize the need for specialequipment and procedures in handling foaming crude oil.

There are many special designs of separators for han-dling foaming crude oil. The special horizontal separatorfor handling foaming oil shown in Fig. 12.12 is one ofthe simpler, more effective units available for this serv-ice. The special degassing element used on the inlet ofthis separator shown in Section CC of Fig. 12.12 gentlyagitates the well fluid and assists in removing gas fromthe oil and in breaking foam bubbles as they flow throughthe inlet element.

The defoaming plates, which extend from near the in-let end to near the outlet end of the separator, are spaced4 in. apart and are shaped with an apex at the verticalcenter of the separator. The plates that are immersed inoil assist in removing nonsolution gas from the oil andin breaking foam in the oil. The plates that are above theoil/gas interface in the gas section of the separator remove


oil mist from the gas and assist in breaking foam that mayexist in the gas section of the vessel.

The 6-in.-thick knitted-wire-mesh mist extractor (locat-ed below the gas outlet) removes the remainder of the liq-uid mist from the gas and breaks or removes the remainingfoam bubbles from the gas.

The vertical separator shown in Fig. 12.13 can be usedto handle foaming crude oil. As the oil cascades downthe plates in this unit, the foam bubbles will be distortedand broken. This design can increase the capacity of theseparator to handle foaming oil by 10 to 50%.

The main factors that assist in “breaking” foaming oilare settling, agitation (baffling), heat, chemicals, and cen-trifugal force, These factors or methods of “reducing”or “breaking” foaming oil are also used to remove en-trained gas from oil. They are discussed on Pages 12-13through 12-15. Many different designs of separators forhandling foaming crude oil have evolved. They are avail-able from various manufacturers-some as standard foam-handling units and some designed especially for a specif-ic application.

ParaffinParaffin deposition in oil and gas separators reduces theirefficiency and may render them inoperable by partiallyfilling the vessel and/or blocking the mist extractor andfluid passages. Paraffin can be effectively removed fromseparators by use of steam or solvents. However, the best

Fig. 12.11—Floatless liquid-level controller and diaphragm-motoroil-control valve on high-pressure oil and gasseparator.

LEGEND 1Fig. 12.12—Horizontal oil and gas separator with special internals for separating foaming crude oil.

12-8

GAS BPV

DEGASSING ELEMENT

-MIST EXTRACTOR

INLET SEPARATING

, AND OEGASSING ELEMENT

BAFFLES FOR - REMOVING GAS

FROM OIL

OIL OUT

Fig. 12.13-Vertical oil and gas separator with special baffling to remove gas from oil, especially beneficial in handling foaming oil. Upper left view and Section A-A show inlet separating element that assists in removing gas from oil.

solution is to prevent initial deposition in the vessel by heat or chemical treatment of the fluid upstream of the separator. Another deterrent, successful in most instances, involves the coating of all internal surfaces of the separator with a plastic for which paraffin has little or no affinity. The weight of the paraffin will cause it to slough off of the plastic-coated surface before it builds up to harmful thickness.

Sand, Silt, Mud, Salt, Etc.

If sand and other solids are continuously produced in appreciable quantities with well fluids, they should be removed before the fluids enter the pipelines. Medium- grained sand in small quantities can be removed by settling in an oversized vertical vessel with a conical bottom and by periodically draining the residue from the vessel. Salt may be removed by mixing water with the oil, and after the salt is dissolved, the water can be separated from the oil and drained from the system.

Corrosion

Produced well fluids can be very corrosive and cause early failure of equipment. The two most corrosive elements are hydrogen sulfide and carbon dioxide. These two gases may be present in the well fluids in quantities from a trace up to 40 to 50% of the gas by volume. A discussion of the problems caused by these two corrosive gases is included in Chaos. 14 and 44. .


Methods Used To Remove Oil From Gas in Separators Liquid mist can be effectively removed from the gas stream in an oil and gas separator by a well-designed mist extractor. Condensible vapors in the gas cannot be removed by mist extractors. Condensation of these vapors, caused by reduction of temperature, may occur after the gas has been discharged from the separator. Thus, existence of liquid in the effluent gas from an oil and gas separator in many instances may not necessarily reflect the efficiency of the separator. Because condensible vapors may have the characteristics of natural gas at separator temperature and pressure, condensation of these vapors may occur immediately after being discharged from the separator.

Density difference of liquid and gas may accomplish separation of liquid droplets from a gas stream where the velocity of the stream is slow enough and sufficient time is allowed to accomplish separation. Limiting the gas velocity in a separator may obtain satisfactory separation without a mist extractor. However, mist extractors are generally installed in conventional oil and gas separators to assist in separation and to minimize the amount of liquid (mist) carried out with the gas.

The methods used to remove oil from gas in oil and gas separators are density difference (gravity separation), impingement, change of flow direction, change of flow velocity, centrifugal force, coalescence, and filtering. Mist extractors used in oil and gas separators can be of many different designs using one or more of these methods. Fig. 12.14 shows a vane-type mist extractor, Fig. 12.15 a centrifugal one, and Fig. 12.16 shows a knitted-wire-mesh (coalescing)-type mist extractor.

Density Difference (Gravity Separation)

Natural gas is lighter than liquid hydrocarbon. Minute particles of liquid hydrocarbon that arc temporarily suspended in a stream of natural gas will, by density difference or force of gravity, settle out of the stream of gas if the velocity of the gas is sufficiently slow. The larger droplets of hydrocarbon will quickly settle out of the gas, but the smaller ones will take longer.

At standard conditions of pressure and temperature, the droplets of liquid hydrocarbon may have a density 400 to 1,600 times that of natural gas. However, as the operating pressure and temperature increase, the difference in density decreases. At an operating pressure of 800 psig, the liquid hydrocarbon may be only 6 to 10 times as dense as the gas. Thus, operating pressure materially affects the size of the separator and the size and type of mist extractor required to separate adequately the liquid and gas.

The fact that the liquid droplets may have a density 6 to 10 times that of the gas may indicate that droplets of liquid would quickly settle out of and separate from the gas. However, this may not occur because the particles of liquid may be so small that they tend to “float” in the gas and may not settle out of the gas stream in the short period of time the gas is in the oil and gas separator.

ly are needed to remove smaller particles from the gas.

Particles of liquid hydrocarbon with diameters of 100 pm and larger will generally settle out of the gas in most average-sized separators. However, mist extractors usual-

OIL AND GAS SEPARATORS

As the operating pressure on a separator increases, the density difference between the liquid and gas decreases. For this reason, it is desirable to operate oil and gas separators at as low a pressure as is consistent with other process variables, conditions, and requirements.

Impingement

If a flowing stream of gas containing liquid mist is im- pinged against a surface, the liquid mist may adhere to and coalesce on the surface. After the mist coalesces into larger droplets, the droplets will gravitate to the liquid section of the vessel. If the liquid content of the gas is high, or if the mist particles are extremely fine, several successive impingement surfaces may be required to effect satisfactory removal of the mist. A mist extractor that provides repeated impingement to remove fine oil mist from the gas is shown in Fig. 12.14a.

Change of Flow Direction

When the direction of flow of a gas stream containing liquid mist is changed abruptly, inertia causes the liquid to continue in the original direction of flow. Separation of liquid mist from the gas thus can be effected because the gas will more readily assume the change of flow direction and will flow away from the liquid mist particles. The liquid thus removed may coalesce on a surface or fall to the liquid section below. The mist extractor shown in Fig. 12.14a uses this method of mist extraction.

Change of Flow Velocity

Separation of liquid and gas can be effected with either a sudden increase or decrease in gas velocity. Both conditions use the difference in inertia of gas and liquid. With a decrease in velocity, the higher inertia of the liquid mist carries it forward and away from the gas. The liquid may then coalesce on some surface and gravitate to the liquid section of the separator. With an increase in gas velocity, the higher inertia of the liquid causes the gas to move away from the liquid, and the liquid may fall to the liquid section of the vessel. Fig. 12.14a shows one version of a vane-type mist extractor that uses change of flow velocity. This mist extractor is used in the typical vertical oil and gas separator shown in Fig. 12.14b.

Centrifugal Force

If a gas stream carrying liquid mist flows in a circular motion at sufficiently high velocity, centrifugal force throws the liquid mist outward against the walls of the container. Here the liquid coalesces into progressively larger droplets and finally gravitates to the liquid section below. Centrifugal force is one of the most effective methods of separating liquid mist from gas. Efficiency of this type of mist extractor increases as the velocity of the gas stream increases. Thus for a given rate of throughput, a smaller centrifugal separator will suffice.

Fig. 12.15 illustrates a horizontal, dual-tube, two-phase oil and gas separator that uses two stages of centrifugal mist extraction to remove liquid mist from the gas. The inlet impingement element is a cone with outwardly spiraling vanes that impart a swirling motion to the well fluid as it enters the separator. The larger droplets of liquid are thrown outward against the shell of the separator and gravitate down to the liquid section of the vessel. The gas

12-9

I @ CHANGE OF DIRECTION

0 CHANGE OF VELOCITY

(a)

GAS BACK-PRESSURE VALVE

VANE-TYPE MIST EXTRACTOR,

WELL FLUID INLET d

DEGASSING ELEMENT-

(1:

SAFETY HEAD

vLOT 1

IL DUMP VALVE

DRAIN

Fig. 12.14-Vertical two-phase oil and gas separator. (a) Illus- tration of impingement, change of flow direction, and change of flow velocity methods of mist extraction. (b) Separating and degassing element on inlet of vessel shown in detail in Section A-A.

flows to the secondary element, which consists of inward- spiraling vanes that accelerate the gas to around 80 ft/sec at normal capacity. This high velocity forces the particles of the liquid mist to the center of the element where they coalesce and separate from the gas when the velocity reduces to 2 to 8 ft/sec downstream of the secondary element. Oil separated by the primary centrifugal element flows from the upper shell cylinder to the lower one through the downcomer at the left. Oil removed from the gas by the secondary centrifugal element flows from the upper shell cylinder to the lower shell cylinder through the downcomer at the right. The lower cylinder of the separator is divided into two compartments with crude oil being discharged from each by use of two liquid-level controllers and two oil-discharge control valves.


Fig. 12.15-Dual-tube horizontal two-phase oil and gas separator with centrifugal primary and secondary separating elements.

4 e 4 5 5 5

1 1 1 1 Mist-laden Gas 2 Mist Eliminator. Extra Fine Mesh. 3” Thick 3 Mist Eliminator, Fme Mesh. 6” Thick 4 Liquid Drain Cylinders, 3” Diameter 5 Dry Gas 6 Liquid Dram to Vessel Sump

Fig. 12.16-Coalescing-type mist-eliminator pad with drain cylinders.

Separators and scrubbers using centrifugal force for the removal of liquid mist from the gas can handle large volumes of gas. One such unit installed near Princess, Alta., Canada, handles 3.5 X lo9 scf/D at 1,000 psig. ’ This type of gas-cleaning unit is generally used in gas gathering, transmisssion, and distribution systems.

Small-diameter oil and gas separators (below 3 or 4 ft in diameter) using centrifugal force are generally not used

as the primary separator on producing leases. This is because of the possibility that the small vessels may be in- undated with a “slug” or “head” of liquid that may allow excessive liquid to exit with the gas and excessive gas to exit with the liquid. Therefore, primary separators on oil and gas streams are usually “conventional” units (other than centrifugal) to prevent the possibility of “overloading” the separators with liquid.

Coalescence

Coalescing packs afford an effective means of separating and removing liquid mist from a stream of natural gas. One of their most appropriate uses is the removal of liquid mist from gas in transmission and distribution systems where the amount of liquid in the gas is low.

Coalescing packs can be made of Berl saddles, Raschig rings, knitted wire mesh, and other such tower-packing materials. The packs use a combination of impingement, change of direction, change of velocity, and coalescence to separate and to remove liquid mist from gas. These packs provide a large surface area for collection and

coalescence of the liquid mist. Fig. 12.17 is a schematic of a knitted wire mesh coalescing pack-type mist extractor used in some oil and gas separators and gas scrubbers.


A word of caution is appropriate concerning the use of coalescing packs in oil and gas separators for general field use. Coalescing packs may be made of frangible material that can be damaged during transit or installation if they are installed in the separator in the manufacturing shop before shipment to point of use. Knitted wire mesh may foul or plug from paraffin deposition and other foreign material and thus make a separator inoperative after a short period of service. Also, excessive pressure drop across the pack may force the pack out of place and allow channelling around or through the pack.

Even though coalescing packs are very effective in the removal of liquid mist from gas, it is usually preferred to use vane-type mist extractors for most oil and gas separators because they may be used under widely varying field conditions. Because of the “fouling” tendency of coalescing-type mist extractors, their use may appropriately be restricted to gas scrubbers used in gas gathering, transmission, and distribution systems.

Filtering

Porous filters are effective in the removal of liquid mist from gas in certain applications. In effect, the porous material strains or filters the liquid mist from the gas. The porous material may use the principles of impingement, change of flow direction, and change of velocity to assist in separation of liquid mist from gas.

Pressure drop through mist extractors used in separators should be as low as practical while maximum separating efficiency is still maintained. Generally, filter-type mist extractors will have the highest pressure drop per unit volume of capacity and the coalescing type will have the lowest. Pressure drop through the other types of mist extractors will usually range between these two extremes.

Mist Extractors Used in Oil and Gas Separators Vane-Type Mist Extractors

Vane-type mist extractors are widely used in oil and gas separators to remove the liquid mist from the gas. These mist extractors can be of many designs. One design is shown in Fig. 12.14B. An enlargement of the mist extractor is shown in Fig. 12.14A. This is a simple but effective mist extractor, consisting of four layers of steel angles placed parallel to each other with the apex of the angle pointing upward. The angles are spaced % in. apart horizontally; that is, they have a S-in. gap between the legs of the angles in the horizontal plane. There is also a %-in. space between the apex of the angle and the leg of the angles in the row above.

Gas flowing through the mist extractor follows the path illustrated in Fig. 12.14A. This flow pattern takes advantage of impingement, change of flow direction, change of flow velocity, and coalescence to separate liquid mist from the gas. Literally thousands of these mist extractors have been used, with all of them giving good performance. They are inexpensive to manufacture and usually will not plug or foul with foreign material and paraffin.

Another design of a vane-type mist extractor is shown in Fig. 12.18. As the gas enters the mist extractor, it is divided into many vertical ribbons (A). Each ribbon of gas is subjected to multiple changes of direction of flow

12-11

GAS OUTLET

f-l

COALTEySpC:NG-

MIST EXTRACTOR

LIQUID OUTLET

Fig. 12.17-Coalescing-type mist extractor with knitted wire mesh. Used in gas scrubbers and oil and gas separators.

(B) as it flows through the mist extractor. This causes mild turbulence and causes the gas to roll against the vanes, as at (C). The entrained droplets of liquid impinge against the vanes, where they adhere and coalesce (D). The liquid droplets move into the vane pockets (E). The liquid flows downward in these channels to the bottom of the mist extractor and then through the drain to the liquid reservoir in the bottom of the vessel where it can be drained from the separator. 2 The liquid drainage from this mist extractor occurs with the liquid out of the gas stream and with the movement of the liquid flow at a right angle to the direction of flow of the gas.

The separating efficiency of this mist extractor depends on the (1) number of vanes in the element, (2) distance between the vanes, (3) number of drainage channels, (4)

PRINCIPLE OF OPERATION

Fig. 12.18-Vane-type mist extractor with liquid channels.


_-_.-

t 01 01 02 04 06 0810 2 4 6 810

- Gas velocity, m/s

Fig. 12.19-Pressure drop and flooding velocity for 6-in.-thick mist-eliminator pad with and without drain cylinders.

width and depth of drainage channels, (5) distance between drainage channels, and (6) size of liquid particles to be removed from the gas.

It is claimed that this mist extractor will remove all entrained liquid droplets that are 8 to 10 pm and larger. If liquid particles smaller than 8 pm in diameter are present in the gas, an agglomerator should be installed upstream of the separator to coalesce the liquid into particles that are large enough for the vane-type mist extractor to remove. Some agglomerators are capable of achieving removal of 99.5 % of all particles 1 .O pm and larger. Pres- sure drop across this vane mist extractor is very low, varying from 2 to 3 in. of water up to 6 to 8 in. of water.*

Fibrous-Type Mist Extractors

Fibrous packing has been used to remove liquid mist from natural gas since the early 1950s. Most of these fibrous packs have been knitted wire mesh. The main use of such mist eliminators has been to remove fine droplets, 10 to 100 pm in diameter. from a stream of gas. Standard mist- eliminator pads made of knitted wire mesh have low pressure drop, high separating efficiency, relatively low initial cost, and low maintenance cost.

In the late 1960’s and early 1970’s, considerable de- velopmental work was done to improve the separating efficiency of knitted-wire-mesh mist-eliminator pads. It was

found that through use of a combination of filaments of different materials and diameters, the separating capacity of the pads could be greatly increased. It was found that a pad 9 in. thick with one 3-m-thick pad of coknitted O.CMlO8-in.-diameter fiberglass filaments and 0.01 l-in.- diameter stainless steel filaments used as the bottom portion and one 6-in.-thick pad of 0.01 l-m-diameter stainless-steel wire mesh as the top portion of the pad would give the highest separating efficiency at the lowest initial cost. In Fig. 12.16. No. 2 is the multifilament bottom portion of the pad, and No. 3 is the coarser monofila- ment top portion of the pad.

The extra-fine fiberglass filaments (0.0008 in. in diameter) coknitted with the 0.011 -in. stainless-steel wire used in the bottom portion of the pad will agglomerate mist particles of 1 to 10 pm into larger particles so that the larger-diameter wire fibers (0.011 in. in diameter) used in the upper 6 in. of the mist eliminator can remove these agglomerated particles from the gas. Even though these combination multifilament pads appreciably increased the separating efficiency, they could not be used widely because they would flood at velocities 50% below those of regular (O.Oll-in. diameter) wire pads.

U.S. Patent No. 4,022,593’ issued May 10, 1977, solved this flooding problem for fine-wire pads. This patent discloses that the use of liquid drain cylinders installed underneath the mist eliminator pad, No. 4 of Fig. 12.16, will cause liquid to drain from the mist eliminator pad as quickly as it collects so that the mist pads remain free of liquid. This reduces the pressure drop through the pads and increases their separating capacity and efficiency.

These drain cylinders are made of the same material as the bottom portion of the mist eliminator pad, are about 3 in. in diameter, and are spaced underneath the pad on 12-m centers. The drain cylinders provide a preferential “escape” route from the pad for the liquid. The preferential drain route results from the added gravity head provided by the drain cylinders. 4 The drain cylinders are generally made of the same material and mesh as the bottom portion of the mist-eliminator pad. The liquid draining from the drain cylinders is shielded from the drag friction of the upflowing gas. Small rivulets or streams of liquid flow down through and across the mist pads down through the drain cylinders. The liquid flows from the drain cylinders in large drops or small streams. Re- entrainment is minimized or eliminated by use of the drain cylinders at the same time the separating efficiency of the pad is increased. Any free liquid anywhere in the pad tends to flow into the drain cylinder and be removed from the pad.

Fig. 12.19 shows the comparative pressure drop and flooding characteristics of a 6-m-thick knitted-mesh mist- eliminator pad made of high-density polypropylene with and without drain cylinders. The fluids used in this test were air and water. The left curve is for the 6-in.-thick pad without drain cylinders. The middle curve is with the same pad but with drain cylinders installed. The straight line marked “Ref. Dry” represents the pressure drop through the same pad with no water in the pad, i.e., with the pad dry. The two points marked “Flood” indicate the air velocity that caused the pad to flood with water. The flood velocity for the pad without drain cylinders was almost 7 ftisec. For the pad with drain cylinders, the flood velocity was 11.8 ftisec.


Methods Used To Remove Gas From Oil In Separators Because of higher prices for natural gas, the widespread reliance on metering of liquid hydrocarbons, and other reasons, it is important to remove all nonsolution gas from crude oil during field processing.

Methods used to remove gas from crude oil in oil and gas separators are settling, agitation, baffling, heat, chemicals, and centrifugal force.

Settling

Gas contained in crude oil that is not in solution in the oil will usually separate from the oil if allowed to settle a sufficient length of time. An increase in retention time for a given liquid throughput requires an increase in the size of the vessel and/or an increase in the liquid depth in the separator. Increasing the depth of oil in the separator may not result in increased emission of nonsolution gas from the oil because “stacking up” of the oil may prevent the gas from emerging. Optimum removal of gas from the oil is usually obtained when the body of oil in the separator is thin-i.e., when the ratio of surface area to retained oil volume is high.

Agitation

Moderate, controlled agitation is helpful in removing nonsolution gas that may be mechanically locked in the oil by surface tension and oil viscosity. Agitation usually will cause the gas bubbles to coalesce and to separate from the oil in less time than would be required if agitation were not used. Agitation can be obtained by properly designed and placed baffling.

Baffling

An inlet degassing element similar to that shown in Fig. 12.13 can be installed on the inlet of the separator to assist in introducing the well fluid into the separator with minimum turbulence and in removing gas from the oil. This element disperses the oil in such a manner that gas can more readily escape from the oil. This type of element eliminates high-velocity impingement of fluid against the opposite wall of the separator. The baffles placed in the separator (Fig. 12.13) between the inlet and the oil level spread the oil into thin layers as it flows downward from the inlet to the oil section. The oil is rolled over and over as it cascades down the baffles, and the combination of spreading and rolling is effective in releasing entrained gas bubbles. This type of baffling is effective in handling foaming oil.

Special perforated baffles or tower packing can be used to remove nonsolution gas from crude oil. Such baffling or packing provides slight agitation. which allows the gas bubbles to break out of the oil as it flows through the baffles or packing.

Heat

Heat reduces surface tension and viscosity of the oil and thus assists in releasing gas that is hydraulically retained in the oil. The most effective method of heating crude oil is to pass it through a heated-water bath. A spreader plate that disperses the oil into small streams or rivulets increases the effectiveness of the heated-water bath. Up- ward flow of the oil through the water bath affords slight

agitation, which is helpful in coalescing and separating entrained gas from the oil. A heated-water bath is probably the most effective method nf removmg foam bubbles from foaming crude oil. A heated-water bath is not practical in most oil and gas separators, but heat can be added to the oil by direct or indirect fired heaters and/or heat exchangers, or heated free-water knockouts or emulsion treaters can be used to obtain a heated-water bath.

Chemicals

Chemicals that reduce the surface tension of crude oil will assist in freeing nonsolution gas from the oil. Such chemicals will appreciably reduce the foaming tendency of the oil and thereby increase the capacity of a separator when foaming oil is handled. In one particular case, the capacity of an oil and gas separator was increased from 3,800 to 9,600 B/D when silicone was injected into and mixed with the oil upstream of the separator with no other change made in the system. Silicone is effective in reducing the foaming tendency of crude oil when it is mixed with the oil in such small quantities as parts per million or parts per billion.

Centrifugal Force

Centrifugal force is effective in separating gas from oil. The heavier oil is thrown outward against the wall of the vortex retainer while the gas occupies the inner portion of the vortex. A properly shaped and sized vortex will allow the gas to ascend while the liquid flows downward to the bottom of the unit. The separators and scrubbers shown in Figs. 12. I5 and 12.20 through 12.22 use centrifugal force for separation. Oil from such units will usually contain less nonsolution gas than that from units that do not use centrifugal force.

Estimated Quality of Separated Fluids Crude Oil

The free (nonsolution) gas content of separated crude oil will vary widely depending on many factors, such as size and configuration of the separator, design and arrangement of the separator internals. operating pressure and temperature, rate of flow, GOR, depth of liquid in the separator, viscosity, and surface tension of the oil. Table 12.1 indicates the estimated free gas and water content of separated crude oil discharged from average oil and gas separators operating under average field conditions. The values shown in this table are only approximate and are intended as an indication of the general range of results that may be expected; they are not intended to be exact and limiting. Table 12. I indicates that appreciable quantities of free gas and water may be left in the separated crude oil; such undesirable performance may be obtained unless particular attention is given to the controlling factors indicated previously.

The water content of separated crude oil probably will be within the wide range indicated in Table 12. I. The factors listed previously, in addition to the agitation resulting from pressure reduction and flow, well-fluid water content, impurities, and degree of emulsification of the oil and water will determine the water content of the separated crude oil.

The approximate values given in Table 12. I assume that special chemicals, equipment, procedures. and techniques


Well Fluids Inlet /

Vortex Finder -

Cyclone Cone -

’ c3 i , ,

Gas-Oil Interface -l~&y&&.&&

Oil-Water Interface --IkSY

Fig. 12.20-Vertical three-phase separator with centrifugal force to obtain primary separation

GAS OUTLET

VORTEX SECTION

VORTEX SECTION LIP>

Fig. 12.21-Diverging vortex separator

Fig. 12.22--Schematic of vertical recycling separator with centrifugal force to obtain primary and secondary separation of oil and gas.


TABLE 12.1-ESTIMATED QUALITY OF SEPARATED CRUDE OIL

Estimated

Approximate Oil

Retention Time

(minutes)

1 to 2 2 to 3 3 to 4 4 to 5 5 to 6 6+

Free (Nonsolution) Gas Content of

Effluent Oil (%) l

Minimum Maximum

5.0 20.0

4.0 16.0 3.0 12.0

2.5 10.0 2.0 8.0 1.5 6.0

Estimated Range of Water

Content of Effluent Oil

Minimum Maximum

(w-4 W” 0-v-n) W”

-1.60 16,000 80,000 8.00 8,000 0.80 40,000 4.00 4,000 0.40 20,000 2.00 2,000 0.20 10,000 1 .oo 1,000 0.10 5,000 0.50

500 0.05 2,500 0.25

‘Expressed as a percent of the total 011 volume wth the 9as measured at standard pressure and temperature

-‘VoI~me basis

have not been used or applied to improve the quality of the separated crude oil. When these are applied, appreciably improved results may be obtained.

Separated Water

It is probable that the effluent water from a three-phase separator will contain oil somewhere within the range indicated in Table 12.2. The quality of the separated water discharged from a three-phase separator depends on the same factors as previously listed for controlling the water content of the effluent oil. It is assumed that special chemicals and separating methods have not been used to improve the estimated quality of the effluent water shown in Table 12.2.

If the difference in the specific gravities of the oil and water at separator operating conditions is less than 0.20, special attention is required because the small difference in the densities of the oil and water will result in limited and incomplete separation. Lower qualities of effluent oil and water may result in such cases.

Gas

The oil (liquid hydrocarbon) content of the gas discharged from an oil and gas separator probably will be in the range shown in Table 12.3. Currently, it is difficult to measure the amount of oil in the separated gas under field operating conditions. With experience and patience, it can be done with a laser liquid particle spectrometer. The range of oil content in the separated gas shown in Table 12.3 has been accepted in recent years as an approximation of the performance of standard commercially available oil and gas separators under normal or average conditions equipped with suitably designed mist extractors.

TABLE 12.2-ESTIMATED QUALITY OF SEPARATED WATER

Estimated Range of Oil Content of Effluent Water

Water Retention Time Minimum Maximum

(minutes) (pm) (Oh)* (wm) W)’

1 to 2 4,ooo0.40 TTGiii2.00 2 to 3 2,000 0.20 10,000 1 .oo 3 to 4 1,000 0.10 5,000 0.50 4 to 5 500 0.05 2,500 0.25 5 to 6 200 0.02 1,000 0.10 6+ 40 0.004 200 0.02

‘Volume basis

Gas Quality From Scrubbers

The liquid content of gas discharged from gas scrubbers is usually less than the liquid content of gas discharged from oil and gas separators. Gas scrubbers are normally installed downstream of oil and gas separators or other separating equipment. If there is a separator upstream of the scrubber, the liquid hydrocarbon content of the scrubbed gas should be less than 0.10 gal/MMscf (less than 0.01335 ppm on a volume basis).

Measuring Quality of Separated Fluids

The quality of separated fluids discharged from oil and gas separators and similar equipment can be measured by state-of-the-art instruments currently available from various manufacturers. The measurements of the quality of effluent fluids and the instruments used to make these measurements are presented in Table 12.4

The instruments indicated are delicate. Each must be carefully selected, calibrated, applied, and operated, and

TABLE 12.3-ESTIMATED QUALITY OF SEPARATED GAS

Estimated Oil

Operating Operating Content of Effluent Gas

Pressure Temperature Minimum Maximum

kWJ (“V @t-W (gal/MMscf) (ppm) (gal/MMscf)

0 to 3,000 60 to 130 0.01335 0.10’ 0.1335 1.00”

‘Eqwalent to 14 129 L/NM’ x lo6 “Eqwalent to 141 29 L/NM3 x 10’


TABLE 12.4-MEASUREMENTS OF EFFLUENT FLUIDS QUALITY

EXTRACTOR

F L UIN

Measurement

Oil in effluent gas

Instrument

Laser liquid particle spectrometerGas in effluent-oil Nucleonic DensitometerWater in effluent oilOil in effluent water

BS&W monitor (capacitance measurement unit)Ultraviolet absorption unit

Oil in effluent water Solvent extraction/infrared absorbance

NONWEIGHTEDFLOAT

Fig. 12.23-Schematic of typical vertical two-phase oil and gasseparator.

Fig. 12.24-Typical field installation of a vertical two-phase oiland gas separator.

the results must be expertly analyzed and interpreted toobtain reliable and reproducible results.

Classification of Oil and Gas SeparatorsClassification by ConfigurationOil and gas separators can have three general configura-tions: vertical, horizontal, and spherical. Vertical sepa-rators can vary in size from 10 or 12 in. in diameter and4 to 5 ft seam to seam (S to S) up to 10 or 12 ft in di-ameter and 15 to 25 ft S to S. Vertical separators areshown in Figs. 12.4, 12.13, 12.14, and 12.20 through12.24.

Horizontal oil and gas separators are manufactured withmonotube and dual-tube shells. Monotube units have onecylindrical shell, and dual-tube units have two cylindri-cal parallel shells with one above the other. Both typesof units can be used for two-phase and three-phase serv-ice. A monotube horizontal oil and gas separator is usuallypreferred over a dual-tube unit. The monotube unit hasa greater area for gas flow as well as a greater oil/gasinterface area than is usually available in a dual-tube sepa-rator of comparable price. The monotube separator willusually afford a longer retention time because the largersingle-tube vessel retains a larger volume of oil than thedual-tube separator. It is also easier to clean than the dual-tube unit.

In cold climates, freezing will likely cause less troublein the monotube unit because the liquid is usually in closecontact with the warm stream of gas flowing through theseparator. The monotube design normally has a lower sil-houette than the dual-tube unit, and it is easier to stackthem for multiple-stage separation on offshore platformswhere space is limited.

Horizontal separators may vary in size from 10 or 12in. in diameter and 4 to 5 ft S to S up to 15 to 16 ft indiameter and 60 to 70 ft S to S. Horizontal separators areshown in Figs. 12.2, 12.3, 12.12, 12.15, and 12.25through 12.27.

Spherical separators are usually available in 24 or 30in. up to 66 to 72 in. in diameter. Spherical separatorsare shown in Figs. 12.5 and 12.28.

Classification by FunctionThe three configurations of separators are available fortwo- and three-phase operation. In the two-phase units,gas is separated from the liquid with the gas and liquidbeing discharged separately. In three-phase separators,well fluid is separated into gas, oil, and water with thethree fluids being discharged separately.

Classification by Operating PressureOil and gas separators can operate at pressures rangingfrom a high vacuum to 4,000 to 5,000 psi. Most oil andgas separators operate in the pressure range of 20 to 1,500psi.


Separators may be referred to as low pressure, medi-um pressure, or high pressure. Low-pressure separatorsusually operate at pressures ranging from 10 to 20 up to180 to 225 psi. Medium-pressure separators usually oper-ate at pressures ranging from 230 to 250 up to 600 to 700psi. High-pressure separators generally operate in the widepressure range from 750 to 1,500 psi.

Classification by ApplicationOil and gas separators may be classified according to ap-plication as test separator, production separator, low-temperature separator, metering separator, elevated sepa-rator, and stage separators (first stage, second stage, etc.).

Test Separator. A test separator is used to separate andto meter the well fluids. The test separator can be referredto as a well tester or well checker. Test separators canbe vertical, horizontal, or spherical. They can be two-phase or three-phase. They can be permanently installedor portable (skid or trailer mounted). Test separators canbe equipped with various types of meters for measuringthe oil, gas, and/or water for potential tests, periodic pro-duction tests, marginal well tests, etc.

Production Separator. A production separator is usedto separate the produced well fluid from a well, groupof wells, or a lease on a daily or continuous basis. Pro-duction separators can be vertical, horizontal, or spheri-cal. They can be two phase or three phase. Productionseparators range in size from 12 in. to 15 ft in diameter,with most units ranging from 30 in. to 10 ft in diameter.They range in length from 6 to 70 ft, with most from 10to 40 ft long.

Low-Temperature Separator. A low-temperature sepa-rator is a special one in which high-pressure well fluidis jetted into the vessel through a choke or pressure-

12-17

GAS OUT

FLOAT NOZZLE\

VAJ$ TYPE MIST EXTRACTOR

NONWEIGHTED FLOAT

SECTION’A-A’

Fig. 12.25--Schematic of typical horizontal two-phase oil and gasseparator.

reducing valve so that the separator temperature is reducedappreciably below the well-fluid temperature. The tem-perature reduction is obtained by the Joule-Thompson ef-fect of expanding well fluid as it flows through thepressure-reducing choke or valve into the separator. Thelower operating temperature in the separator causes con-densation of vapors that otherwise would exit the separa-tor in the vapor state. Liquids thus recovered requirestabilization to prevent excessive evaporation in thestorage tanks.

Metering Separator. The function of separating wellfluids into oil, gas, and water and metering the liquidscan be accomplished in one vessel. These vessels are com-monly referred to as metering separators and are availa-ble for two- and three-phase operation. These units areavailable in special models that make them suitable foraccurately metering foaming and heavy viscous oil.

Fig. 12.26--Horizontal monotube two-phase oil and gas separator. Unit is skid mounted with acces-sories. Operator is checking liquid level in separator.


GASO U T L E T

2 ’ S A F E T Y P O P

Fig. 12.27—Typical horizontal dual-tube two-phase oil and gasseparator.

Fig. 12.29—Schematic of a vertical two-phase metering sepa-rator. Liquid is metered in integral metering com-partment in lower portion of vessel.

GAUGE COCK

I

M I S T E X T R A C T O R

OILO U T L E T

Fig. 12.28—Schematic of a typical spherical two-phase oil andgas separator with float-operated lever-type oil-control valve.

A two-phase metering separator separates well fluidsinto liquid and gas and measures the liquid in the lowerportion of the vessel. A typical two-phase metering sepa-rator is shown in Fig. 12.29. A three-phase metering sepa-rator separates the oil, water, and gas and measures onlythe oil or both the oil and water. Metering of the liquidis normally accomplished by accumulation, isolation, anddischarge of given volumes in a metering compartmentin the lower portion of the vessel.

Fig. 12.30 illustrates a three-phase metering separatorin which the free water is measured with a positive-displacement meter. The metering separator shown in Fig.12.31 is designed especially for separating large volumesof foaming and/or viscous oil. This unit uses hydrostatic-head liquid-level controllers to measure the oil accurate-ly on a weight basis rather than by volume. It uses pres-sure flow into and out of the dual compartments and doesnot rely on gravitational flow. The unit shown in Fig.12.31 is a two-phase vessel with two combination separat-ing and metering compartments operating in parallel onan alternate basis. It is equipped with controls and valvesthat are arranged to permit constant flow of well fluid intothe vessel. With pressure flow into and out of each of thetwo compartments, this separator can handle much larg-er volumes than separators with two compartments thatrely on gravity flow from upper to lower compartments.These units are furnished with hydrostatic-head liquid-level controls for metering foaming oil or float-operatedcontrols for nonfoaming oil.

Foam Separator. Oil and gas separators that handle foam-ing crude oil are generally referred to as foam separa-tors. For a discussion of the design and application ofseparators for handling foaming oil refer to Pages 12-6and 12 -7 .

Elevated Separators. Separators can be installed on plat-forms at or near tank batteries or on offshore platformsso that the liquid can flow from the separator to storage


Fig. 12.30-Schematic of a vertical three-phase metering separator with free water metered with a positive displacement meter.

or to downstream vessels by gravity. This permits the separator to be operated at the lowest possible pressure to capture the maximum amount of liquid and to minimize the loss of gas and vapor to the atmosphere or to the low- pressure gas system.

Stage Separators. When produced well fluid is flowed through more than one separator with the separators in series, the separators are referred to as stage separators. The first separator is referred to as the first-stage separator, the second separator is called the second-stage separator, etc. For a more detailed discussion on stage separation refer to Page 12-32.

Classification by Principle Used for Primary Separation

Separators may be classified according to the method used to accomplish primary separation in the separator. Such a classification is density difference (gravity separation), coalescence and/or impingement, and centrifugal force.

Density Difference (Gravity Separation). This classification includes all units that have no inlet element, deflector, impingement plate, or pack on the inlet to the vessel.

Fig. 12.31-Schematic of vertical two-phase metering separator for separating and metering viscous and/or foaming oil on a weight basis. Pressure flow into and out of the dual separating and metering compartments increases fluid-handling capacity.

Primary separation is obtained solely by the difference in density of the oil and gas or vapor. These units are few in number and most separators will have a mist extractor near the gas outlet to remove oil mist from the gas.

Impingement and/or Coalescence. This type of separator includes all units that use an impingement plate or device or a pack of tower packing on the inlet of the separator to accomplish initial separation of the oil and gas. An infinite number of design arrangements can be used

on the inlet of a separator, but one of the simplest and most effective arrangements is illustrated in Fig. 12.14.

Centrifugal Force. Centrifugal force can be used for both primary and secondary separation of oil and gas in a separator. The centrifugal force can be obtained with either a properly sized tangential inlet in the separator (see Figs. 12.20 through 12.22) or a properly sized internal spiral or involute element with the top and bottom of the element open or partially open. These centriFuga1 elements cause cyclonic flow of the incoming fluid at velocities high enough to separate the fluid into an outer layer or cylinder of liquid and an inner cone or cylinder of gas or vapor. The velocity required for centrifugal separation will vary


from about 40 to about 300 ftisec. The most common operating velocity range is between about 80 and 120 ftisec.

Most centrifugal separators are vertical. However, a centrifugal separating element can be used on the inlet of horizontal separators to accomplish the initial separation of oil and gas. A second centrifugal element can be installed in the vessel to remove liquid mist from the exiting gas.

Centrifugal Oil and Gas Separators and Gas Scrubbers Increased use of oil and gas separators on offshore platforms for handling larger volumes of well fluid has increased efforts to develop more compact separators to reduce space and weight on offshore platforms. Positive results have been achieved from these efforts, resulting in some separators that use centrifugal force to accomplish both initial and final separation of the oil and gas. Three centrifugal-force separators are illustrated in Figs. 12.20 through 12.22.

Centrifugal Separators

The vertical centrifugal oil and gas separator shown in Fig. 12.20 operates as described here. 5 Well fluid enters the separator through the adjustable tangential slot at high velocity, inducing a cyclone within the vessel. The cyclone, stabilized by the vortex finder, moves down the cyclone cone. High cyclonic velocity ensures that a stable, thin film of liquid is maintained. The cyclone cone provides a long path for well fluid, enabling free gas to break out, a factor that is very effective with foaming oil. Residence time is not critical in gas/liquid separation. When the gas/oil interface is reached, the gas cyclone is reversed with the gas flowing upward and exiting from Nozzle A. The cyclone cone provides a smooth transi- tion for the liquid to flow to the sump, preventing re- entrainment and assisting oil/water separation.

Oil/water separation is accomplished by gravity with the required residence time dependent on the well-fluid properties. Oil and water are drawn off from Nozzles C and D, respectively. Gas Vent B is provided to equalize the pressure and remove any gas that separates in the sump. Gas Vent B is normally connected to Gas Outlet A.

Diverging Vortex Separator

Fig. 12.21 shows a centrifugal gas/liquid separator that has been designed and patented recently. ‘*? This unit uses centrifugal force to separate the gas and liquids. If the liquid is separated into oil and water, the separation is done by gravity.

The diverging vortex separator (DVS) is a bottom-entry, high-performance cyclone separator. Performance ranges from 99% to 99.99+ % over a flow rate range of 10 to 1. Mist particles are typically removed to below 5 pm and, depending on design specifications, additional condensate can be obtained because of the centrifugal force field. There are no moving parts and no change in gas flow direction. Pressure losses are minimal, ranging from inches of water to a few psi.

The oil-laden gas tangentially enters the bottom of the DVS vortex section, Fig. 12.21. Both the separated oil and gas corotationally spiral outward and up in a constant

vortex flow field. At the top of the vortex section, the oil circumferentially flows over the vortex section lip (Coanda effect), around and down the shroud to the vessel bottom, and out the liquid outlet. The gas continues an upward spiral to the gas outlet. This flow regime minimizes the oil-to-gas relative velocity, thereby minimizing re-entrainment and maximizing vortex-section surface area for coalescing and gathering oil. This surface is flow-wise continuous, allowing the oil to form a nearly uniform film. This film is stable. and the oil movement is consistent with inlet velocity and centrifugal force. Separation performance is independent of inlet liquid loadings up to 10 lbm of liquid per pound of gas.

The shroud extends below the oil level in the annular liquid section. The large annular liquid volume allows the separator to accommodate flow surges and liquid slugs. Performance is not affected by oil level within control range.

The oil and gas equilibrium is affected by centrifugal force in the vortex section. The equilibrium shift favors condensate formation. thereby effecting a dewpoint depression in the outlet gas. This dewpoint depression correlates directly with vortex section centrifugal force and is thus a function of inlet velocity. In-situ condensate formation is greatest at the vortex section inlet. This aids in the coalescing of entrained oil in the inlet stream, which improves separator performance. The outlet gas dewpoint depression means both a drier gas and a slightly higher liquid vapor pressure. If the inlet gas contains water vapor, moisture recovery will be consistent with the dewpoint- depression characteristic. This moisture will collect with the oil and condensate, allowing their final separation in the annulus formed by the vortex wall and the shell of the separator vessel.

Centrifugal Gas Scrubbers

Several different designs of centrifugal gas scrubbers are available from several suppliers. One of the most popular and effective centrifugal gas scrubbers is illustrated in Fig. 12.22. Operation of this unit is as follows.8 The centripetal-flow-type recycling separator has two effective stages of separation. The term centripetal flow denotes gas flow converging toward the center of the vessel, as in a whirlpool. In the first stage, all the free liquid and most of the entrained liquid are spun out of the gas by centrifugal force. In the second stage, the small amount of entrained liquid remaining in the gas is spun out under the influence of greatly increased centrifugal force and is collected by a recycling circuit as indicated in Fig. 12.22.

The well fluid enters the separator through the tangential inlet nozzle, which causes the stream to whirl around the inlet chamber. The spinning stream then moves downward between the smoothing sleeve and the separator shell into the vortex chamber. Liquid in the spinning stream is thrown outward by centrifugal force to the wall of the vortex chamber and runs down past the baffle plate into the liquid chamber, from which it is discharged. The gas, still spinning, converges toward the center of the vortex chamber, increasing in velocity, and enters the vortex tinder tube. In the vortex finder tube, entrained liquid remaining in the rapidly spinning gas collects on the vortex finder wall and is swept upward by the gas toward the gas outlet. This liquid, together with a sidestream of about


5% of the total gas, is then sucked through a gap in the tube wall, down the recycling line, and through the central hole of the baffle plate into the vortex chamber. The low-pressure area along the axis of the vortex provides the necessary suction. Recycled liquid and sidestream gas thus entering the vortex chamber through the hole in the baffle plate mix with the rapidly spinning gas in the core of the vortex, and the liquid is thrown out to the wall and runs down with the rest of the liquid into the liquid chamber. The mainstream gas, now clean, continues up the vortex finder past the gap to the gas outlet.

Comparison of Oil and Gas Separators Table 12.5 compares the advantages and disadvantages of two- and three-phase horizontal, vertical, and spherical oil and gas separators. This table is not intended as an “absolute” guide but affords a relative comparison of the various characteristics or features of the different separators over the range of types, sizes, and working pressures. The comparison of oil and gas separators in Table 12.5 assumes that the horizontal oil and gas separators are monotube vessels.

The liquid chamber in these separators contains baffles for liquid stilling or for isolation of the level-control float as needed. The liquid chamber may be made oversized to handle very large liquid flows and may be constructed for gravity separation of the oil and water.

Illustrations of Oil and Gas Separators Typical oil and gas separators are illustrated in Figs. 12.2 through 12.5 and 12.23 through 12.28. Fig. 12.25 is a schematic of a typical horizontal two-phase oil and gas separator. Fig. 12.26 is a photograph of a horizontal monotube two-phase oil and gas separator. Fig. 12.2 shows a horizontal skid-mounted three-phase oil/gas/water well tester on an offshore drilling platform off the coast of Brazil. The well tester is a separator with oil, gas, and water meters piped on it so that the well fluid can be separated into oil, gas, and water and each of the fluids metered before they are recombined or discharged separately.

Fig. 12.23 is a schematic of a typical vertical two-phase oil and gas separator. Fig. 12.24 is a photograph of a typical field installation of a vertical two-phase oil and gas separator. Fig. 12.28 depicts a typical two-phase spherical oil and gas separator with float-operated lever-type oil-control valve.

Estimating the Sizes and Capacities of Oil and Gas Separators The oil and gas capacities of oil and gas separators will vary as the following factors vary.

1. Size (diameter and length) of separator. 2. Design and arrangement of separator internals. 3. Number of stages of separation. 4. Operating pressure and temperature of separator. 5. Physical and chemical characteristics of well fluid

(gravity, viscosity, phase equilibrium, etc.) 6. Varying gas/liquid ratio. 7. Size and distribution of liquid particles in gas in

the separator upstream of mist extractor. 8. Liquid level maintained in separator. 9. Well-fluid pattern, whether steady or surging.

10. Foreign-material content of well fluid. 11. Foaming tendency of oil. 12. Physical condition of separator and its components. 13. Other factors.

Fig. 12.3 is a schematic of a typical horizontal three- phase oil/gas/water separator. Fig. 12.4 shows a typical vertical three-phase oil/gas/water separator, and Fig. 12.5 illustrates a typical spherical three-phase oil/gas/water separator. Fig. 12.27 is a photograph of a horizontal dual- tube two-phase oil and gas separator.

Items 5 and 7 are generally not known with sufficient detail and accuracy to permit accurate calculation of the size or performance of a separator. However, such calculations can be based on empirical data and assumptions for comparative and budgetary purposes. When separators are being sized for maximum performance and when the performance must be guaranteed, Items 5 and 7 be-

come very important and must be available to the designer. In a vertical separator, the liquid particles to be removed

from the gas must settle downward against the upflowing column of gas. Conversely, in a horizontal separator, the

TABLE 12.5-COMPARISON OF ADVANTAGES AND DISADVANTAGES OF HORIZONTAL, VERTICAL, AND SPHERICAL OIL AND GAS SEPARATORS, TWO- AND THREE-PHASE

SDherical

Considerations Horizontal

(Monotube)*

Efficiency of separation 1 Stabilization of separated fluids 1 Adapatability to varying

conditions 1 Flexibility of operation 2 Capacity (same diameter) 1 Cost per unit capacity 1 Ability to handle foreign material 3 Ability to handle foaming oil 1 Adaptability to portable use 1 Space required for installation

Vertical plane 1 Horizontal plane 3

Ease of installation 2 Ease of inspection and

maintenance 1

Vertical (Monotube)*

2 2

2

2 2

3

3

‘(One Compartment)’

3 3

2

‘Rattngs (1) mosl favorable: (2) Intermed~ale; (3) least favorable


path of a liquid particle to be separated from the gas as it flows through the vessel resembles the trajectory of a bullet fired from a gun. This difference in the flow pattern of the separated liquid particles indicates that a horizontal separator of a given diameter and length will separate a larger volume of well fluid than a vertical vessel of the same size. This is generally true, but the liquid level that must be carried in a monotube horizontal separator subtracts from this advantage and may cancel it out completely if a high liquid level is maintained in the horizontal separator,

The maximum gas velocity in an oil and gas separator that will allow separation of liquid mist from the gas can be calculated with the following form of Stokes’ law:

where

VR = maximum allowable gas velocity, ft/sec, F,., = configuration and operating factor

(empirical) (see Fig. 12.32 for values), PL = density of liquid at operating conditions,

Ibm/cu ft, and

PR = density of gas at operating conditions, lbm/cu ft.

The value of F,.,) in Eq. 1 is an empirical independent variable; it includes all factors that affect separation of liquid from gas in an oil and gas separator except (1) the compressibility factor of the gas, (2) base and operating pressure, (3) base and operating temperature, and (4) density of the fluids to be separated. F,., does include and varies with the length/diameter (L/D) ratio of the separator, the configuration of the separator vessel, the design of the internals of the separator, the liquid depth in the separator, foaming tendency of the oil, steady or pulsating flow of gas, heading or steady flow of liquid, gasiliq- uid ratio, presence of foreign materials and impurities,

and the degree of separation required. F, varies in direct proportion to the L/D ratio. Of the variables listed above on which F,, is dependent, the LID ratio of the separator vessel is the most dominant. The use of straightening vanes or cylinders, baffling, and special inlet degassing elements can increase the value of F,., and the separating capacity of a separator.

The original values of F,, used in Eq. 1 were determined by the assignment of values to F, that would result in a velocity in the separator, expressed in feet per second and calculated by Eq. 1, that would provide the desired degree of separation. Thus the original values of F,, were for the customary system of units. To change the velocity calculated by Eq. 1 from customary to SI units, a multiplier of 30.48 must be used. That is, the velocity in feet per second must be multiplied by 30.48 to change it to centimeters per second.

The maximum allowable gas velocity vR of Eq. 1 is the maximum velocity at which the gas can flow in the separator and still obtain the desired quality of gas/liquid separation. Only the open area of the separator available for gas flow is considered in calculating its capacity. The gas separating capacity of an oil and gas separator can be stated as

4x =Ag”g, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (2)

where

qK = volume of gas flowing through separator, cu ftlsec,

A, = cross-sectional area of separator for gas flow, sq ft, and

“R = gas velocity, ft/sec, from Eq. 1.

The pX of Eq. 1 is calculated from Eq. 3 as follows:

PM, pg=- z,RT3 “‘.‘..............__..___,,__ (3)

12.32-Configuration and operation factor F,, for oil and gas separators and gas scrubbers (see Eqs. 1 and 4 through 6).


where p = separator operating pressure, psia,

M, = molecular weight of gas, 7 ‘.S = compressibility factor of gas, R = gas constant, 10.732, and T = operating temperature, “R.

Substituting in Eq. 2 for vK and simplifying yields

. .

Equation for Gas Capacity of Oil and Gas Separator

When Eq. 4 is corrected for compressibility, pressure, and temperature and when units are changed to cubic feet per day, the equation becomes

Y,q =

where ph = base pressure, psia, and Tl, = base temperature, “R.

Equation for Sizing an Oil and Gas Separator for Gas Capacity

Eq. 5 can be rewritten to solve for A,.

*,s = q, 86,40OF,.,,

. . . . . . . . . . . . . . . . . . (6)

Eq. 5 can be used to calculate the volume of gas a separator of a given size will handle under given operating conditions. Eq. 6 can be used to calculate the size of separator required to handle a given volume of gas under certain operating conditions.

When Eqs. 5 and 6 are used for horizontal separators, A,Y is the area for gas flow. The area occupied by liquid must be excluded. In a vertical separator, A,q is usually the entire internal cross-sectional area of the vessel.

The range of values of F,, normally used in Eqs. 5 and 6 is as follows. For vertical separators, the range is 0.10 to 0.167, and for horizontal separators, the range is 0.35 to 0.707. Values of F,,, can be obtained from Fig. 12.32. The values of F,, shown in Fig. 12.32 are in customary units. Multiply these values by 30.48 for SI Units. For the L/D ratio of any given separator, any value of F,.,, indicated in Fig. 12.32 can be used. Experience aids in the selection of the optimum value of F,, The nonideal (conservative) values of F,,, will result in lower gas velocities, i.e., larger vessels. Use of F,,, values for ideal (liberal) conditions will result in higher gas velocities, i.e., smaller vessels. Where the oil and gas tend to separate clean and dry and ideal operating conditions prevail, the ideal (liberal) values of F,.,, can be used. Where condi-

tions are less than ideal-such as slugging flow, high operating pressure and temperature, and excessive platform vibration or movement-the nonideal (conservative) values of F,., should be used. For estimating purposes for unknown operating conditions, the nonideal (conservative) values of F,, should be used. As experience is gained in a particular oil field or producing zone or area, it may be possible and practical to use higher values of F,, until the optimum balance between cost and performance is reached.

The values of F,, for horizontal separators are higher than for vertical separators. A typical value of F, for vertical separators is 0.167; for horizontal separators a typical value of F,, is 0.500.

If the proper F,., value is used in Eq. 5, all liquid particles larger than 100 pm should be removed by gravity separation upstream of the mist extractor. If a properly designed and sized mist extractor is used, all liquid part- cles larger than 10 pm should be removed by the mist extractor.

The value of F,, in Eqs. 1 and 4 through 6 varies as the L/D ratio of the separator vessel varies. With a given diameter separator, as the length of the separator increases, the value of F,., increases. With a given length separator, as the diameter of the separator increases, the value of F,., decreases. This relationship is more pronounced in horizontal separators than it is in vertical separators; in vertical separators, if the L/D is greater than about 2.0, the value of F,., will change little, if at all, regardless of how much the L/D is increased. Refer to Fig. 12.32 for an indication of the relationship between LID and F,, for both vertical and horizontal separators.

There is a range of length-to-diameter (L/D) ratios for oil and gas separators that will adequately meet each separator capacity requirement. This range of L/D ratios is minimum at about 1.0 to 2.0 and maximum at about 8.0 to 9.0. There is not just one diameter and length of separator that will satisfy a given capacity requirement; rather, there is a series of sizes (L/D ratios) that can be used for each application.

In a vertical separator, the well-fluid inlet is located about one-third of the length of the shell below the top head-shell weld seam, and the gas flows from the inlet up through the vessel to the gas outlet at the top of the vessel. The oil (liquid) flows downward from the inlet to the bottom of the vessel. Thus either the gas volume (above the inlet) or the liquid oil volume (below the inlet) can determine the required separator diameter.

For vertical separators, if the gas volume determines the size of the vessel, the L/D ratio of the vessel should be from about 2.0 to 3.0; if the liquid volume determines the separator size, the L/D ratio should be from about 2.0 to 6.0.

For horizontal separators, the L/D ratio of the vessel should be from about 2.0 to 6.0. The gas and liquid added together determine the size of a horizontal separator because the two fluids flow concurrently through the vessel from inlet end to outlet end.

For vertical separators similar to the configuration shown in Fig. 12.33, LID or SLID can be used to determine the value of F,., from Fig. 12.32. From the standpoint of determining the gas separating capacity of a vertical separator, the only pertinent part of the separator is that from the well-fluid inlet to the gas outlet. This


GAS OlJTLEl TABLE 12.6-VALUES FOR EXAMPLE PROBLEM 1

Difference in Values of F...

MIST EXT

ELEMENT -I I-

WELL FLUID INLET

GAS

:

OIL

Fig.

UD=10/2=5.0 L’/D’ =7.0/1.33=5.26 w

F,, = 0.725 F,, = 0.747 (ideal conditions) (ideal conditions)

F,, = 0.636 F,, = 0.655 (nonideal conditions) (nonideal conditions)

+3.03

+2.99

can include the mist extractor because the mist extractor should be as effective as density-difference separation would be in the space that it occupies. Fig. 12.32 shows both LID and SLID for vertical separators.

Above an LID ratio of 2.0 and a SLID ratio of 0.67, additional shell length does not increase the gas capacity of vertical separators. Additional shell length can increase the liquid capacity of the separator when the additional shell length is located between the inlet and the oil outlet.

For configurations of horizontal separators similar to the one shown in Fig. 12.34, L’lD’ instead of LID should be used to determine the value of F,, . This usually will result in a smaller separator for a given volume of gas, as indicated in the following three examples.

Example Problem 1. Assume a 24-in.-diameter x lo-ft (S to S) horizontal separator operating with a liquid depth of 8 in. and with dimensions A= 1.5 ft and B= 1.5 ft (Fig. 12.34). The values of F, are from Fig. 12.32. Refer to

H !&LET Table 12.6 for further data.

Example Problem 2. Assume a 48in.-diameter X 16 ft S to S horizontal separator operating with a liquid depth of 24 in. and with dimensions A=2 ft and B=2 ft (Fig. 12.34). Refer to Table 12.7 for more data.

Example Problem 3. Assume a 96-in-diameter X 30 ft S to S horizontal separator operating with a liquid depth of 32 in. and with dimensions A=3 ft and B=4 ft (Fig. 12.34). Refer to Table 12.8 for more data.

For a given horizontal separator, the value of F, ob- tamed by use of L’ID’ will vary directly with the liquid depth.

12.33-One-third l/D vs. UD for vertical oil and gas separators.

Fig, 12.34-L’/D’ vs. UD for horizontal oil and gas separators


TABLE 12.8-VALUES FOR EXAMPLE PROBLEM 3

Difference in Values of F_.

TABLE 12.7-VALUES FOR EXAMPLE PROBLEM 2 Difference in Values of F,,

L/D= 16/4=4.00 L'/D'=12/2=6.00 W)

F,, =0.640 F,, = 0.608 (ideal conditions) (ideal conditions) + 26.25

f co = 0.560 F,, =0.716 (nonideal conditions) (nonideal conditions) + 15.60

L/D=3018=3.75 L'/D'=23/5.33=4.32 (%) Lu

F,, =0.618 F, = 0.670 (ideal conditions) (ideal conditions)

F,, = 0.541 F,, =0.562 (nonideal conditions) (nonideal conditions)

+a.41

+4.10

The use of straightening tubes, cylinders, plates, vanes, and Dixon plates will generally increase the value of F,,, and increase the capacity of a given separator. The increase in capacity to be obtained by such devices and other design features can be estimated from empirical data that must be verified by field tests.

Computer Sizing of Oil and Gas Separators

The LID for the vertical separator was indicated by the computer to vary from 2.0 to 2.8. The L/D for the horizontal separator was indicated by the computer to vary from 2.0 to 6.0. Fig. 12.32 shows the relation between L/D and the values of F,., . The values of F,., to be used in Eqs. 5 and 6 can be obtained from Fig. 12.32. Table 12.9 is an example of computer sizing of a vertical oil and gas separator. Table 12.10 is an example of computer sizing of a horizontal separator.

Tables 12.9 and 12.10 are printouts of a computer program’ for sizing vertical and horizontal separators.

The liquid separating capacity of most oil and gas separators is controlled by retention or residence time, the time

The format of the computer printout has been modified. liquid is retained in the separator. This time usually var-

TABLE lP.Q-COMPUTER DESIGN OF TWO-PHASE VERTICAL OIL AND GAS SEPARATOR

lnout Data

50 30,000

0.70 0.934

115 185 250

2

Output Data

GOR, scflbbl Density of gas, lbmlcu ft Density of oil, lbmlcu ft Compressibility factor Inlet nozzle size, in. Gas outlet size, in. Oil outlet size, in. Relief valve

Inlet body size, in. Orifice area; sq in.

1,666.7 0.68

58.30 0.970

14 14 6

8.06;

Gas rate, MMscflD Oil rate, B/D Specific gravity of gas (air = 1) Specific gravity of oil (water = 1) Operating temperature, OF Operating pressure, psig Maximum design pressure, psig Oil retention time. minutes

Output Data

Distribution inlet Inside Mist-Extractor Height of Height of Total Liquid Connection From Cross-Sectional Cross-Sectional

Length to Number Diameter Ratio l:)

Water Level Oil Level Height Bottom Seam Area Area

WI (ft) vu (fu (=I ft) 6s f0

2.8 6.15 0.00 7.94 7.94 9.92 29.67 14.16 2.7 6.25 0.00 7.68 7.60 9.68 30.68 14.16 2.6 6.50 0.00 7.10 7.10 9.15 33.18 14.16 2.4 6.75 0.00 6.58 6.58 8.69 35.78 14.16 2.3 7.00 0.00 6.12 6.12 8.28 38.48 14.16 2.2 7.25 0.00 5.71 5.71 7.92 41.28 14.16 2.1 7.50 0.00 5.33 5.33 7.60 44.18 14.16 2.1 7.75 0.00 4.99 4.99 7.32 47.17 14.16 2.0 8.00 0.00 4.69 4.69 7.07 50.27 14.16

1 2 3 4 5 6 7 a 9

Output Data

Length of Minimum Shell Plus Minimum Shell Minimum Head

Shell Height Heads Thickness Thickness Approximate Weight Approximate Vessel of Head Pius Shell Weight

(IW W-N

9,299 10,693 9,591 11,029

10,331 11,881 11,127 12,797 11,982 13,780 12,097 14,832 13,075 15,956 14,917 17,154 16,025 18.429

Actual Velocity of Gas (ftlsec)

1.54 1.49 1.38 1.28 1.19 1.11 1.04 0.97 0.91

(ft) - (W (in,)

17.06 20.56 0.53 16.93 20.40 0.54 16.65 20.33 0.56 16.44 20.25 0.58 16.28 20.22 0.61 16.17 20.24 0.63 16.10 20.29 0.65 16.07 20.39 0.67

16.07 20.52 0.69

‘Based on gas volume only wfh maximum backpressure of 10%

(in.)

0.53 0.54 0.56 0.58 0.60 0.62 0.64 0.67 0.69


TABLE 12.10-COMPUTER DESIGN OF TWO-PHASE HORIZONTAL OIL AND GAS SEPARATOR

Input Data

Gas rate, MMscflD 100

Oil rate, B/D 50,000 Specific gravity of gas (air = 1) 0.80 Specific gravity of oil (water = 1) 0.85

Operating temperature, OF 110 Operating pressure, psig 800 Maximum design pressure, psig 1,000 yjbretention time, minutes 1.5

2.0, 3.0, 4.0, 5.0, 6.0

‘Based on gas volume only with mawmum backpressure 01 10%

Output Data

Minimum Shell Minimum Shell Minimum Head Mist-Extractor Vessel Minimum ID Length Thickness Thickness Area Area

Number UD (ft) (fV (in.) (in.) (=I ft) (sq ft)

---2.0 1 6.97 13.93 2.47 2.40 14.95 38.12 2 3.0 6.24 18.72 2.22 2.15 14.95 30.60 3 4.0 5.67 22.68 2.01 1.96 14.95 25.26 4 5.0 5.26 26.32 1.87 1.82 14.95 21.77 5 6.0 4.99 29.92 1.77 1.72 14.95 19.53

Output Data

Output Data

GOR, scflbbl Compressibility factor of gas Density of gas, lbmlcu ft Density of oil, Ibm/cut ft inlet nozzle size, in. Gas outlet size, in. Oil outlet size, in. Relief valve

Inlet body size, in. Orifice area,’ sq in.

Area Assigned to

Oil Dead Space Gas

(sq ft) (%) (sq ft) (%) (sq ft) (%)

20.9755.0 -12.1 4.63 12.5232.8

Actual Velocity of Gas (ftlsec)

1.53

15.60 51.0 4.19 13.7 10.80 35.3 1.77 12.88 51.0 3.88 15.3 8.50 33.7 2.25

11.10 51.0 2.77 12.7 7.89 36.3 2.42 9.77 50.0 2.36 12.1 7.40 37.9 2.58

Output Data

Height of Height of Oil Level Dead Space

Vessel Diameter Vessel Diameter

W) Feet co/o) Feet

53.94 3.76 9.71 0.68 50.79 3.17 10.86 0.68 50.79 2.88 12.20 0.69 50.79 2.67 10.10 0.53 50.00 2.49 9.57 0.48

Output Data

Heighl of Gas

Vessel diameter

(W ft

36.36 2.53 38.36 2.39 37.01 2.10 39.12 2.06 40.43 2.02

Weight of Shell and Heads

(IW

Approximate Weight of Separator

VW

44,292.8 50,936.7 43,324.0 49,822.6 41,100.g 47,266.0 39,760.2 45,724.2 39.636.4 45.581.9

2,ooo.oo 0.834

3.70 53.06

14 IO

8

4 4.480

Any one of the hve separators indicated will satisfy the design capacity requ!rements


60 {k,f

Fig. 12.35-Gas capacity of vertical oil and gas separators.

ies from 20 seconds to 1 to 2 hours, depending on many factors. For retention time to result in separation of water from oil, the liquid must be relatively quiet and free from agitation. Typical retention time for two-phase separation is 30 seconds to 2 minutes. Typical retention time for three-phase separation is from 2 to 10 minutes, with 2 to 4 minutes the most common.

Separators can be sized for liquid capacity in two ways. One way is to base the sizing on test data on the well fluid to be separated. The other way is to base the sizing on experience in separating fluid from that producing zone or area from neighboring wells or leases. If it is known that a certain retention time will be required to accomplish separation, sizing the separator for liquid becomes a simple volume calculation.

Capacity Curves for Vertical and Horizontal Oil and Gas Separators The gas and liquid capacities of vertical and horizontal oil and gas separators can be estimated from the curves shown in Figs. 12.35 through 12.38. These capacity curves are based on calculations made with a computer

program developed by George 0. Ellis’ and the author. Sample printouts of this computer program for vertical and horizontal separators listing the input and output data

for each is shown in Tables 12.9 and 12.10. The oil and gas capacity curves in Figs. 12.35 through

12.38 can be used reversibly to determine (1) the size separator required to separate a given volume of fluid un-

der given operating conditions and (2) the volume of fluid a given separator will handle under given operating conditions.

The gas capacity of vertical separators is shown in Fig. 12.35. The capacities shown are based on the assumptions listed at the lower left side of Fig. 12.35. Condi- tions other than these will result in different capacities. These capacities are suitable for preliminary sizing and estimating purposes. If accurate sizing or performance data are required, calculations should be made with pertinent data. The gas capacity of a vertical separator does not vary directly with a change in shell length. If a standard length of 10 ft is assumed, an increase in shell length of 100%) to 20 ft, will result in an increase in gas capacity of about 23%.


- . . It

I 1s *..

[

i : :

LIQUID CAPACITY, BID

Fig. 12.36-Liquid capacity of vertical oil and gas Separators.


Fig. 12.38-Liquid capacity of horizontal oil and gas separators.

The liquid (oil) capacity of vertical separators is shown in Fig. 12.36. The liquid capacity of a vertical separator is controlled primarily by the volume of liquid retained in the accumulation (settling) section of the separator. Normal practice is to maintain a liquid depth above the oil outlet connection of from one to three diameters of the vessel. The optimum liquid depth depends on the design of the separator, the rate of throughput, and the characteristics of the fluid being separated.

The gas capacity of horizontal separators is shown in Fig. 12.37. The gas capacity of a horizontal separator is directly proportional to the cross-sectional area of the vessel available for gas flow. Thus, the diameter of a horizontal separator and the depth of liquid maintained in the vessel determine its gas capacity under given conditions. The gas separating capacity of a horizontal separator is proportional to the length but not directly proportional. For instance, if the length of the separator is increased from 10 to 20 ft, the gas separating capacity will increase 46%. Refer to the lower left corner of Fig. 12.37 for the relationship between shell length and gas capacity of a horizontal separator. The indicated multipliers assume that no special internals are used in the separators.

The liquid (oil) capacity of horizontal separators is shown in Fig. 12.38. The liquid capacity of a horizontal separator depends on the volumetric liquid-settling capacity of the accumulation (retention) section of the separator. This volumetric capacity is determined by inside shell diameter, shell length, and liquid depth.

The liquid depth in a horizontal separator for two-phase operation is normally assumed to be one-third of the diameter of the vessel. However, it can vary from 3 to 4 inches up to 60 to 70% of the cross-sectional area of the

separator. The liquid depth in a horizontal separator for three-phase operation is normally assumed to be at the horizontal centerline of the vessel (one-half full of liquid). However, it can vary from 8 to 10 inches up to 80 to 90% of the cross-sectional area of the vessel.

The gas capacities shown in the graphs in Figs. 12.35 and 12.37 are only approximate. These graphs should be used only for approximating the sizes and performance of separators. Calculations are recommended for more precise sizing and performance requirements especially where performance must be guaranteed.

Vertical Separator Sizing

Example Problem 4. See Table 12.11 for the given data.

From Fig. 12.32, F,=0.167; M,=M,i,y,=28.97~ 0.7=20.28.

Gas Sizing. Substituting in IZq. 3 yields

(199.7)(20.28)

PK = (0.97)( 10.73)(575) =0.68 lbm/cu ft

TABLE 12.11-GIVEN DATA FOR SIZING VERTICAL SEPARATORS-EXAMPLE PROBLEM 4

Maximum gas flow rate, MMscf/D 50.0 Specific gravity of gas 0.70 Maximum oil rate, BID 30,000 Specific gravity of oil 0.934 Operating temperature, OF 115 Operating pressure, psig 185 Design pressure, psig 250 Oil retention time, minutes 2.0 Gas compressibility factor 0.97 Two-phase operation No water

12-30

TABLE 12.12-GIVEN DATA FOR SIZING HORIZONTAL SEPARATORS-EXAMPLE PROBLEM 5

Maximum gas flow rate, MMscflD 100.0

Specific gravity of gas 0.80 Maximum oil rate, B/D 50,000 Specific gravity of oil 0.85 Operating temperature, OF 110

Operating pressure, psig 800 Design pressure, psig 1,000 Oil retention time, minutes 1.5 Gas compressibility factor 0.834 Two-phase operation No water

Substituting in Eq. 6 gives

A,=

50,000.000


Gas Sizing. Substituting in Eq. 3 yields

(814.7)(23.18)

pg = (0.834)(10.73)(570) =3.70 Ibmku ft.

Substituting in Eq. 6 gives

A,=

lcKl,ooo.ooo

(86,400)(0.707) (&) (=J (Z) (~~~o;IJo170> ’

A, =7.40 sq ft for gas.

Oil Sizingfor the Same Separator. The oil volume required in the vessel is

(86,400)(O.l67)(&) (z) (g) (58~~~80’68~ ’ V,,=50’ooox5.615 x1.5=292.4 cu ft. II

24x60

A, = 29.76 sq fi for gas flow, and D = (29.7610.7854)” =6.15 ft ID.

A 72-in.-ID vessel can be used if the volume of 50.0 MhlscfiD includes a small safety margin. If the separator must handle 50.0 MMscfiD with a written guarantee for performance, then a 78-in.-OD or -ID vessel can be used.

Oil Sizing for the Same Separator. Volume required in the separator for oil, V,, is

30,000 V,,=- =41.67 bbl-233.98 cu ft.

144012

The height of oil, h ,~, in the vertical separator is

233.98 h,, = -=8.28 ft in a 72-in.-ID vessel.

28.27

where the cross-sectional area of the 72-in.-ID vessel is 28.27 sq ft. If the separator has a 78kr. ID, the h,, will be

233.98 h,, = - =7.05 ft,

33.18

where the cross-sectional area of the 78-in. -ID vessel is 33.18 sq ft.

The length required for this separator shell will be about 16 to 18 ft.

Horizontal Separator Sizing

Example Problem 5. See Table 12.12 for the given data. From Fig. 12.32, F,,,=0.707; M,=M,i,~,~=28.97~ 0.80=23.18.

Select a vessel length of 30 ft. The cross-sectional area of oil is

292.4 A,>=-

30 =9.74 sq ft.

For vessel sizing, Area for gas = 7.40 sq ft Area for oil = 9.74 sq ft Dead space = 2.00 sq ft (about 25% of area

for gas)

Total area 19.14 sq ft

Note that dead space is considered to be “reserve” space between oil and gas and is usually assumed to be about 10 to 30% of the gas space for reserve capacity.

The vessel will be 9.74/19.14=51% full of liquid. The area of the vessel’s ID= 19.14 sq ft.

=4.94 ft=59.2 in.

Use a 60-in.-ID or 66-in.-OD x 30-ft S to S separator, depending on reserve capacity desired.

The volume of the two heads on horizontal separators is normally not considered in separator-sizing calculations. This volume will compensate for the internals and for other variables.

Capacities of Spherical Separators

Spherical oil and gas separators use the same principles of separation used in horizontal and vertical separators. A spherical separator can be considered as a truncated vertical separator. When the fluid-handling capacity of spherical separators is considered, allowance must be made for the reduced height available for separation above the fluid inlet. The same consideration must be applied to spherical separators that is applied to the trays in a frac-


tionating column: the smaller the tray spacing, the lower the allowable capacity. Fig. 12.28 is a schematic of a typical two-phase spherical oil and gas separator. A typical three-phase spherical separator is shown in Fig. 12.5.

The chief advantage of spherical separators is their relatively low silhouette, which allows all component parts to be readily accessible to operating personnel. However, the horizontal separator offers this same advantage. The spherical separator may be easier and less expensive to install; on smaller units a gin-pole truck or crane may not be required to unload and place it on location.

The oil capacities of spherical oil and gas separators are shown in Table 12.13. The gas capacities of spherical oil and gas separators are shown in Fig. 12.39. The table and curves were originated by Vondy. lo The gas capacities shown in Fig. 12.39 for spherical separators are conservative and assume that the spherical vessels contain no internals. If a properly designed separating element is used on the inlet of the spherical separators, and if an effective mist extractor is used, the gas capacities shown in Fig. 12.39 can be increased by a multiplier that ranges from 1 .O to 3.0, as shown in Table 12.14.

The amount of increase in spherical separator capacity that can be obtained by vessel design is dependent on several factors: (1) location of inlet connection in the ves- se1 with respect to the liquid level in the separator; (2) size, configuration, and location of the inlet separating and spreading element; (3) vertical distance between the inlet separating and spreading element and the mist extractor; (4) size, design, and location of the mist extractor; (5) physical and chemical characteristics of the well fluid being separated; (6) operating pressure and temperature of the separator; (7) flow pattern into the separator (heading or steady); and (8) other factors.

80 60

1.0 IO 20 30 50K)100 200xK)500 I.000 2,000 5,000

OPERATING PAESSURE,PSlA

Flg. 12.39-Gas capacity of spherical oil and gas separators.

The oil (liquid hydrocarbon) capacity of spherical oil and gas separators is shown in Table 12.13. The liquid capacity for each size unit is shown for two different liquid depths in the vessel-with the liquid depth equal to one-third and one-half the ID of the sphere. The first condition is appropriate for two-phase (oil/gas) separation; the second condition is a propriate for three-phase (oil/gas/water) separation. NY

Spherical separators are more appropriately used for two-phase separation than for three-phase separation. This is especially true of sizes smaller than 36 in. in diameter. Field tests should be made on spherical separators to determine and/or to confirm their capacity because, of the three shapes of separator vessels available, they are the most difficult to rate properly for oil and gas capacities.

TABLE 12.13-LIQUID CAPACITIES OF SPHERICAL OIL AND GAS SEPARATORS

ID of Sphere

(in.)

16 (1)'

(2)’ * 20 (3)

(4) 24 (5)

(6) 30 (7)

(8) 36 (9)

(10) 42 (11)

(12) 48 (13)

(14) 60 (15)

(16) 72 (17)

(18) 04 (19)

(20) 96 (21)

(22) 100 (23)

(24)

Assumed Liquid

Depth in

Liquid Capacity, BID, 42 gallbbl, Steady Flow

Liquid-Retention Time

Sphere fin.) 20 seconds 30seconds 60 seconds 2.5 minutes 5.0 minutes 10 minutes 20 mrnutes 30 minutes -

5.33 246 165 I33 33 17 0 4 3 8.00 447 318 159 64 32 16 a 5 6.66 463 322 161 64 32 16 a 5

10.00 932 622 311 124 62 31 16 10 8.00 836 557 278 111 56 28 14 9

12.00 1,611 1,074 537 215 107 54 27 19 10.00 1,632 1,088 544 216 109 54 27 18 1500 3,147 2,098 1,049 420 210 105 52 35 12.00 2,820 1,860 940 376 188 94 47 31 18.00 5,438 3,625 1.813 725 363 181 91 60 14.00 4,470 2,985 1,493 597 299 149 75 50 21.00 8,636 5.757 2,879 1,151 576 288 144 96 16.00 6,684 4,456 2,226 691 446 223 111 74 24.00 12.890 8,594 4,297 1,719 859 430 215 143 2000 13,056 8.704 4,352 1,741 870 435 216 145 30.00 25,177 16,785 8,392 3,357 1,678 839 420 280 24.00 22,560 15,040 7,520 3,008 1,502 752 375 250 36.00 43,505 29,004 14,502 5,601 2,900 1,450 725 483 28.00 35,625 23,880 11,942 4,777 2,388 1,194 597 398 42.00 69,085 46,057 23,028 9,211 4,606 2,303 1,151 766 32.00 53,477 35,652 17,826 7,130 3,565 1,783 891 594 48.00 103,125 68,750 34,375 13,750 6,675 3,437 1,719 1,146 33.33 60,444 40,296 20,148 8,059 4,030 2,015 1,007 672 50.00 116,560 77,707 38,853 15,541 7,771 3,885 1,943 1,295

‘Odd-numbered conditons recommended for Iwo-phase operaban “Evewnumbered condttlons recommended for three-phase o!xraf~on


TABLE 12.14-MULTIPLIERS FOR GAS CAPACITIES SHOWN IN FIG. 12.39

Type of Separator Service

Foamina crude oil Nonfoaming crude oil Nonfoaming crude oil

Multipliers for Gas

Design of Vessel Capacities Shown internals in Fig. 12.39

Averaae 0.25 to 0.75 Average Superior Average Superior

1.0 to 1.5 1.5 to 2.0

Distillate or-condensate Distillate or condensate

2.0 to 2.5 2.5 to 3.0

Practical Considerations in Sizing Oil and Gas Separators To ensure acceptable separation at all times, an oil and gas separator should be sized so that it will never operate above its maximum rated capacity. A separator must be sized for the maximum instantaneous flow rate to which it will be subjected rather than for the total daily production rate. Many wells produce by “heads” or “slugs” as a result of natural causes or intermittent gas lift. Such a well may produce a total of only 200 bbl of liquid in 24 hours. However, if that well “heads” or “intermits” only once each hour, it may produce x24 of its total daily production in a matter of 2 or 3 minutes, which would

result in an instantaneous flow rate of about 4,000 to 6,000 B/D. The separator should be sized to handle the maximum instantaneous rate of fluid produced during these short intervals, or it must be of sufficient size to store a portion of these slugs while it separates and discharges the balance.

In addition to serving as a means for separating the well fluid into gas and liquids, the separator vessel, in some instances, must also serve as an accumulation or storage vessel, particularly when the wells flow by “heads,” when intermittent gas lift is used, when the tubing string and/or flowlines may unload liquid into the separator at high instantaneous flow rates and when the liquid is transferred from the separator by pump and the separator must serve as the accumulator/reservoir vessel for the pumping operation. In these instances the separator vessel must be large enough to store the extra volume of liquid in addition to performing the function of separating. This consideration of storage may often dictate larger vessels than otherwise would be required if the flow of fluid into and out of the separator were steady and continuous.

It is extravagant to install grossly oversized separators where their excess capacities will never be used. Most pumping wells, continuous-flow gas-lift wells, and some naturally flowing wells always produce at uniform rates. For such wells, separator sizes may be selected on the basis of maximum total daily production.

Field tests should be made on oil and gas separators to determine their oil- and gas-handling capacities under

TABLE 12.15-NUMBER OF STAGES VS. DIFFERENTIAL SEPARATION

Approximate Percent

Number of Stages Approach to Differential of Separation Vaporization

2 0 3 75 4 90 5 96 6 90’/2

actual conditions. Manufacturers’ rated capacities for separators are intended for “general” or “average” con-

ditions. The only way to determine the exact capacity of a particular separator under a given set of conditions is to actually test the separator under those conditions.

Stage Separation of Oil and Gas Theoretical Considerations of Stage Separation

Stage separation of oil and gas is accomplished with a series of separators operating at sequentially reduced pressures. Liquid is discharged from a higher-pressure separator into the next-lower-pressure separator. The purpose of stage separation is to obtain maximum recovery of liquid hydrocarbons from the well fluid and to provide maximum stabilization of both the liquid and gas effluent.

Two processes liberate gas (vapor) from liquid hydrocarbon under pressure. They are flash separation (vaporization) and differential separation (vaporization). Flash separation is accomplished when pressure is reduced on the system with the liquid and gas (vapor) remaining in contact; i.e., the gas (vapor) is not removed from contact with the liquid as reduction in pressure allows the gas to come out of solution. This process yields the most gas (vapor) and the least liquid. Differential separation is accomplished when the gas (vapor) is removed from contact with the liquid as reduction in pressure allows the gas (vapor) to come out of solution. This process yields the most liquid and least gas (vapor).

In a multiple-stage-separator installation, both processes of gas liberation are obtained. When the well fluid flows through the formation, tubing, chokes, reducing regulators, and surface lines, pressure reduction occurs with the gas in contact with the liquid. This is flash separation. When the fluid passes through a separator, pressure is reduced; also, the oil and gas are separated and discharged separately. This is differential separation. The more nearly the separation system approaches true differential separation from producing formation to storage, the higher the yield of liquid will be.

An ideal oil and gas separator, from the standpoint of maximum liquid recovery, is one constructed so that it reduces the pressure of the well fluid from the wellhead at the entrance of the separator vessel to, or near, atmospheric pressure at the discharge from the separator. The gas and/or vapor is removed from the separator continuously as soon as it is separated from the liquid. This special application of differential vaporization or separation is not practical and is never used.

Some of the benefits of an ideal separator may be obtained by use of multiple-stage separation. The number of stages does not have to be large to obtain an appreciable benefit, as can be seen from Table 12.15. ” Econom- ics usually limits the number of stages of separation to


three or four, but five or six will pay out under favorable (4) gas and vapors from second-stage separator, conditions. Seven stages have been used on large volumes (5) liquid from second-stage separator, of oil, but such installations are rare. (6) gas and vapors from storage tank (third stage), and

Ratios of operating pressures between stages in (7) liquid in storage tank (stock-tank oil) (third stage). multiple-stage separation can be approximated from the The first-stage separator operates at 850 psia and 90°F; following equations ’ ’ : the second-stage separator operates at 250 psia and 76°F;

the third stage (storage tank) is maintained at atmospher-

Al- ” PI

ic pressure (14.7 psia) and is assumed to be at 100°F F= -, (7) (summer conditions).

PS Flash calculations were made on this system with a computer program available from Simulation Sciences, Inc. I2

,,=$=p,,F”l-‘, . _. .@I The &suits are shown in Table 12.16.

and Economic Considerations of Stage Separation

The extent of application of stage separation will depend

p3=~=p,,,+, . . (9) on two principal considerations: the terms of the gas sales contract and the price structure for the gaseous and liquid hydrocarbons.

where F = stage pressure ratio p1Ip2 =pzIp:,

=, .P,ilP.$> ni = number of interstages (number of

If gas is sold on volume only, it will usually be desirable to remove most condensible vapors from it. If, on the other hand, gas is sold on liquid content, it may be desirable to permit condensible vapors to remain in the gas, depending on conditions, facilities, and location.

stages - l), p I = first-stage separator pressure, psia, p2 = second-stage separator pressure, psia. px = third-stage separator pressure, psia, and p,Y = storage-tank pressure, psia.

Equilibrium flash calculations should be made for several assumed conditions of pressures and temperatures to determine the conditions that will yield the most stock- tank liquid. However, the above equations will give a practical approximation that can be used when no other information is available. Fig. 12.40 schematically shows typical two-, three-, and four-stage separation systems.

Other considerations in the application of stage separation are (1) physical and chemical characteristics of the well fluid, (2) flowing wellhead pressure and temperature. (3) operating pressures of available gas-gathering systems, (4) conservation features of liquid-storage facilities, and (5) facilities for transporting liquids.

Two-stage separation is normally considered to be obtained when one oil and gas separator is used in conjunction with a storage tank. Three-stage separation is obtained when two separators are used in a series at different pressures in conjunction with a storage tank. The storage tank is considered a separation stage because of the gas contained in solution in the oil at a pressure below the operating pressure of the low-pressure separator. Some of this gas will not separate from oil immediately upon entry into the tank but will “weather” from the oil for a period of a few minutes up to a few days. Such gas or vapors can be captured with a vapor recovery system (see Chap. 14).

The point of diminishing returns in stage separation is reached when the cost of additional stages of separation is not justified by increased economic gains. The optimum number of stages of separation can be determined by field testing and/or by equilibrium calculations based on laboratory tests of the well fluid. Flash calculations can be made for various numbers of stages of separation to determine the optimum number of stages of separation for each installation.

Flash Calculations for Oil and Gas Separators

Flash (equilibrium) calculations can be made to determine accurately the gas and liquid analysis and yield from oil and gas separators if the composition of the well fluid is known. As an example. fluid produced from seven wells is gathered and separated in three stages in such a system as that depicted in Fig. 12.41.

Stabilization of Separated Liquid Hydrocarbons

If an oil and gas separator is operated under a vacuum and/or at a temperature higher than well-stream temperature, the liquid hydrocarbons flowing through the separator will have more gas and/or vapors removed than otherwise would be removed. This tends to stabilize the liquid and results in loss of less gas and condensible vapors from the storage tanks. By the use of a stabilization unit similar to that shown in Fig. 12.42, the yield of stock- tank liquid has been increased by 10 to 15% over that obtainable with standard two-stage separation.

In Fig. 12.41, Fl represents the first-stage separator, When a stabilizer of this type is used, a separator in- F2 represents the second-stage separator, and F3 repre- stalled upstream of the stabilizer removes gas from the sents the third stage, the storage tank. The fluid streams liquid, and the liquid is discharged to the stabilizer. The are represented by numbers as follows: liquid discharged from the stabilizer to the tanks is com-

If the liquid hydrocarbon is sold on the basis of volume and API gravity, it may be desirable to remove the condensible vapors from the gas and add them to the liquid to increase its sales price. If, on the other hand, the liquid is sold on the basis of volume only, it may be desirable to leave the condensible vapors in the gas.

(1) well tluid inlet to first-stage separator, (2) gas and vapors from first-stage separator, (3) liquid from first-stage separator,

pletely stabilized and has a Reid vapor pressure of 11 to 13 psi, which is less than atmospheric pressure. There- fore, there will be no loss of gas or vapor from the storage


TWO STAGE SEPARATION

LOW GAS-OIL RATIO LOW FLOWING PRESSURE

FLUIDS

-3

r 104

n

10 lOOIl

THREE STAGE SEPARATION

LOW TO INTERMEDIATE GRAVITY OIL LOW TO INTERMEDIATE GRAVITY OIL

1 OO# 1ou 1ou INTERMEOIATE TO HIGH GAS OIL RATIO INTERMEOIATE TO HIGH GAS OIL RATIO

*to *to to INTERMEOIATE WELL HO INTERMEOIATE WELL HO

5oou 7% 7% FLOWING PRESS FLOWING PRESS

FOUR STAGE SEPARATION

4

5004 1oo)I r) to to

-r

HIGH GRAVITY OIL HIGH GAS-OIL RATIO

1OP HIGH FLOWING PRESSURE

10 USE OF HIGH PRESSURE

ISOO! 5OOP 750 GAS FOR MARKET OR

,,_i~

PRESSURE MAINTENANC E

rl

Fig. 12.40-Flow diagrams for two-, three-, and four-stage separation

H i2iAS -$B L.F$AS OUT OUT

HIGH-

i

LOW- - PRESSURE

SEPARATOR

(6) VAPOAS

STORAGE TANK

SECOND THIRD STAGE STAGE

F2 F3

111 WELL FLU10

STAGE

Fl

Fig. 12.41-Three-stage separator installation with two separators and storage tank(s). Refer to equl- librium flash calculations on Page 12-33 and Table 12.16.


tanks. In some installations the initial cost of a stabilizer may be less than the initial cost of multiple-stage oil and gas separators. Use of a stabilization unit has resulted in liquid recovery comparable to that from four to six stages of separation. Each new installation should be studied to determine whether a stabilizer should be used. If a highly volatile liquid is being handled, the use of stabilizers may result in increased profit.

Selection and Application of Separators and Scrubbers Oil and gas separators are manufactured in three basic configurations: vertical, horizontal, and spherical. Gas scrubbers are manufactured in two basic shapes: vertical and horizontal. Each of these units has specific advantages and uses. Selection of the particular unit to use for each application is usually based on which will obtain the desired results at the lowest equipment, installation, and maintenance costs. Table 12.5 gives a comparison of the advantages and disadvantages of vertical. horizontal, and spherical separators.

The following summary indicates the general recommended uses of vertical, horizontal, and spherical oil and gas separators and gas scrubbers.

Vertical Oil and Gas Separators

Applications for vertical oil and gas separators include the following.

1, Well fluids having a high liquid/gas ratio. 2. Well fluids containing appreciable quantities of

sand, mud, and similar finely divided solids. 3. Installations with horizontal space limitations but

with little or no vertical height limitations, such as crowd- ed tank batteries and/or offshore production platforms.

4. Well fluids where liquid volume may vary widely and instantaneously, such as slugging wells and/or intermittent gas lift wells.

5. Downstream of other production equipment that allows or causes liquid condensation or coalescence.

6. Upstream of other field process equipment that will not perform properly with entrained liquid in the gas.

7. Where economics favors the vertical separator.

Horizontal Oil and Gas Separators

Applications for horizontal oil and gas separators include these situations.

1. Liquid/liquid separation in three-phase separator installations to obtain more efficient oil/water separation.

2. Separating foaming crude oil where the larger liquid/gas contact area of the horizontal vessel will allow and/or cause faster foam breakdown and more efficient gas/liquid separation.

3. Installations where vertical height limitations indicate the use of a horizontal vessel because of its lower silhouette.

4. Well fluids with a high GOR. 5. Well with relatively constant flow rate and with lit-

tle or no liquid heading or surging. 6. Installations where the operator and/or conditions

require or indicate internal water-weir and oil-bucket construction to eliminate the use of oil/water interface liquid level controller.

7. Where portable units (either skid or trailer mounted) are required for either test or production use.

CONDENSER

RECTIFICATION w--w-- SECTION --., -----

----- --- --

LIQUID ----- FROM __)I SEPARATOR

----- ----- -----

REFLUX ACCUMULATOR

STRIPPING SECTION

- --1 -

LIOUID LIQUID

TO- *

STABILIZER AEBOILER

Fig. 12.42--Schematic of typical stabilization unit used for stabilizing and increasing the yield of liquid hydrocarbons at field separation stations.

8. Where multiple units can be stacked to conserve floor space.

9. Upstream of other field process equipment that will not perform satisfactorily with entrained liquid in the gas.

10. Downstream of other production equipment that allows or causes liquid condensation or coalescence.

11. Where economics favors the horizontal separator.

Spherical Oil and Gas Separators

The following is a list of applications for spherical oil and gas separators.

1. Well fluids with high GOR’s, constant flow rates, and no liquid slugging or heading.

2. Installations where both vertical and horizontal space and height limitations exist.

3. Downstream of process units-such as glycol dehydrators and gas sweeteners-to scrub and to salvage the expensive process fluids, such as amine and glycol.

4. Installations where economics favors the spherical separator.

5. Installations requiring a small separator where one man can transport the spherical separator to location and install it.

6. Scrubber for fuel and process gas for field and/or plant use.

Gas Scrubbers

Most vertical and horizontal gas scrubbers are used where

the gas previously has been separated, cleaned, transported and/or processed with other equipment. That is, most of the impurities-such as entrained liquids, silt, line scale, and dust-have been removed from the gas by other equipment, and the gas scrubber is being used to “polish” the gas. Gas scrubbers generally are used to ensure that the gas contains no materials that will be detrimental to the equipment that the scrubber is installed to protect-such as compressors, dehydrators, sweeteners, meters. and regulators.


TABLE 12.16--SEPARATOR FLASH CALCULATIONS

Unrt ldentrfrcation Fl F2 F3

Type Flash Flash Flash Feeds (1) (3) (5) Products (2) W)

$“c.b’d 145; I:; ;‘1’,:;

Temperature, OF 75.82 100.00 Pressure, psia 850.00 250.00 14.70 Fraction liquid 0.062 0.748 0.597

Duty, MMBtulD 0.000 0.000 5,694

Stream Component Flow Rates, Ibm mol/D

Stream identification (1) (2) Phase Well Fluid Vaoor

1 Cl 26,081.7315 25,585.5952 2 c2 1,783.4190 1,663.0260 3 c3 1,098.4380 925.2687 4 IC4 296.2080 217.4148

5 NC4 398.0295 268.3454 6 IC5 185.1300 94.2998

7 NC5 129.5910 59.1854 8 NC6 249.9255 65.1976 9 NC7 316.2638 40.4824

10 NC8 316.2638 18.6834

Totals 30,855.OOOO 28.937.4988

(3) (4) Liquid Vapor

496.1363 385.1366 120.3930 48.3420 173.1693 29.1554

78.7932 6.3136 129.6841 7.4733

90.8302 2.2505 70.4056 1.3462

I 84.7277 1.1809 275.7814 0.5803 297.5804 0.2113

1.917.5012 481.9902

Temperature, “F 90.0000 90.0000 90.0000 75.8247 Pressure, psia 850.0000 850.0000 850.0000 250.0000 Mole fraction liquid 0.0621 0.0000 1 .oooo 0.0000

Stream Identification Phase

(5) Liquid

110.9998 72.0510

144.0139 72.4796

122.2108 88.5797 69.0594

183.5467 275.2010 297.3691

1.43551 IO

(6) Vapor

110.1207 69.2178

126.8146 55.0069 84.3322 42.4336 28.6450 34.6597 19.5618

7.4581 578.2504

(7) Liquid

0.8791 2.8332

17.1993 17.4727 37.8786 46.1461 40.4144

148.8870 255.6392 289.91 to 857.2606

Cl 2 c2 3 c3 4 IC4 5 NC4 6 IC5 7 NC5 8 NC6 9 NC7

10 NC8 Totals

Temperature, OF Pressure, psia Mole fraction liqurd


75.8247 250.0000

1 .oooo

100.0000 14.7000

0 .oooo

100.0000 14.7000

1 .oooo

Stream Molal Compositions

(11 (2) 13). .C4) Well Fluid

1 Cl 2 c2 3 c3 4 IC4 5 NC4 6 IC5 7 NC5 8 NC6 9 NC7

IO NC8 Totals

8.4530 x 10 -’ 5.7800x10-* 3.5600 x IO -’ 9.6000 x 10 -s 1.2900x10-’ 6.0000 x 10 -3 4.2000 x 10 -3 a.fooo x lo -s 1.0250xlo~2 1.0250x10~’

30,855.0000

Vapor Lrqurd

8.8417x 10-l 2.5874x10-l 5.7470x 10-Z 6.2786x IO-’ 3.1975x lo-’ 9.0310x 10 -2 7.5133x 10 -3 4.1092x IO -2 9.2733x 10 -3 6.7632x 10 -* 3.2587x 10 -s 4.7369x IO -’ 2.0453x 10 -3 3.6717x10-’ 2.2531 x 10 -3 9.6338 x 10 -’ 1.3990x lo-3 1.4382x to-1

6.4565~10-~ 1.5519x 10-l 28,937 4988 1,917.5012

Vapor

7.9905 x IO ’ l.oo3oxlo~’ 6.0490 x IO -’ 1.2099x10-’ 1.55o5x1o-2 4.6692 x IO -3 2.7930 x 10 m3 2.4501 x IO m3 1.2040~10-~ 4.3836 x IO -4

481.9902 -

Temperature, OF 90.0000 90 0000 90.0000 75.8247 Pressure, psia 850.0000 850.0000 850.0000 250.0000 Mole fraction liquid 0.0621 0.0000 1 .oooo 0.0000


TABLE 12.16-SEPARATOR FLASH CALCULATIONS (continued)


1 Cl 2 c2 3 c3 4 IC4 5 NC4 6 IC5 7 NC5 8 NC6 9 NC7

10 NC8 Totals

Temperature, OF Pressure, psia Mole fraction liquid


Ibm mol/D Temperature, OF Pressure, psia H, Btullbm Mole fraction liquid

(1) Well Fluid

30,855.OOO 90.000

850.000 38.422

0.06214

M IbmlD 672.911 Molecular weight 21.809 Specific gravity 0.3586

M IbmlD 552.396 552.396 0.000 10.163 Standard M cu ft/D 10,981.491 10,981.491 0.000 182.910 Actual M cu ft/D 174.741 174.741 0.000 10.455 Z 0.87018 0.87018 0.00000 0.94366

Liauid

M IbmlD Standard liquid

cu ft/D Actual gallons per

minute cu ft/D


Ibm mol/D Temperature, OF Pressure, psia H, Btullbm Mole fraction liquid

M IbmlD Molecular weight Specific Gravity

Vapor

M IbmlD Standard M cu ft/D Actual M cu ft/D Z

Liquid

M IbmlD Standard liquid

cu ft/D Actual gallons per

minute

(5) (6) (7) Liqurd Vapor Liauid

7.7324x 10 -2 1.9044x10-’ 1.0254~10-~ 5.0192x 10-Z 1.1970x10-’ 3.3050x10-3 1.0032x 10-l 2.1931 x 10 -’ 2.0063x 10 -* 5.0490x 10-z 9.5126x 10 -’ 2.0382x10-2 8.5134x 10-Z 1.4584x10-’ 4.4186~10~’ 6.1706x lo-* 7.3383 x 10 -’ 5.3830x10-2 4.8108~10~~ 4.9537x 10 -2 4.7144x10-2 1.2786x IO-’ 5.9939xlo-2 1.7368~10-~ 1.9171 x10-1 3.3829x 10 -* 2.9820 x 10 -’ 2.0715x IO-’ 1.2898x10-’ 3.3818x10-’

1,435.5110 578.2504 857.2606

75.8247 100.0000 250.0000 14.7000

1 .oooo 0.0000

Stream Summarv

(2) (3) Vapor Liquid

28,937.499 1,917.501

120.515

3.268.282

19.3504 3.724.960

90.000 850.000

41.548 0.00000

552.396 19.089 0.3302

0.000

0.000

0.0000 0.000

Stream Summary

(5) (6) Liquid Vapor

1,435.511 578.250 75.825 100.000

250.000 14.700 19.306 179.133

1 .ooooo 0.00000

110.352 28.467 76.873 49.229 0.6330 0.5315

0.000 28.467 0.000 0.000 219.440 0.000 0.000 232.310 0.000 0.00000 0.98332 0.00000

110.352 0.000

2.798.155 0.000

16.3754 0.0000

100.0000 14.7000

1 .oooo

90.000 850.000

24.098 1 .ooooo

120.515 62.850 0.5919

120.515

3,268.282

19.3504 3.724.960

(7) Liquid

857.261 100.000

14.700 33.287

1 .ooooo

81.885 95.520 0.6781

81.885

1,938.415

11.7773

- (4)

Vapor

481.990 75.825

250.000 76.102

0.00000

10.163 21.085 0.3470

0.000

0.000

0.0000 0.000

cu RID 3,152.270 0.000 2.267.129


Some of the uses for gas scrubbers are to clean gas (1) Spherical shells. for fuel for heaters, boilers. steam generators. and engines: (2) for control gas for processing plants and equipment; (3) upstream of compressors; (4) upstream of h,n =

p.,.d-i

dehydrators and sweeteners; (5) downstream of dehydra- 20E-0.2p,,

tors and sweeteners to conserve processing fluids; (6) up- and stream of gas distribution systems; (7) upstream of and in gas transmission lines to remove liquid, dust, rust. and

2uEh \., PA, =

scale; (8) upstream and/or downstream of pressure regu- ri +0.2h,,

lation stations; and (9) downstream of gas-transmission- line compressor stations to remove lubricating oil from Ellipsoidul heads. the line.

Both vertical and horizontal scrubbers can be effective and efficient in the applications listed above. The selec- h Ph =

Puhd,

tion of a vertical or horizontal scrubber may be a matter 2aE-0.2p,l,

of personal preference, space limitations, cost consider- and ation, and/or availability.

Gas scrubbers normally are limited to applications where there is little or no liquid to be removed and where liquid slugging seldom or never occurs. A large number of different types, designs, and sizes are available from many manufacturers.

2aEh (>/,

“‘= di +0.2h,h ’

Torispherical heads.

Construction Codes for Oil and h rh

= 0.8W,hrc. Gas Separators UE-0. lp,,l ASME Code for Unfired Pressure Vessels and

Most oil and gas separators furnished for field use are designed, constructed, pressure tested, and labelled in accordance with the ASME Code for Unfired Pressure Ves- sels. I3 Outside the U.S., other similar codes may be used (Table 12.17). Sec. VIII, Div. I of the ASME code is used for most unfired pressure vessels. Div. II of Sec. VIII may be used for offshore installations because of weight reduction obtained by its use. Div. I design equations are based on a safety factor of 4.0, while Div. II equations are based on a safety factor of 3 .O. Quality control is more stringent under Div. 11 than Div. I.

Use of the ASME or some equivalent code for construction of pressure vessels ensures the purchaser of receiving vessels that are designed, constructed, and pressure-tested in accordance with established standards, inspected by a disinterested party, and certified safe for use at specified design pressures and temperatures. All vessels labeled with ASME code stamp must be constructed in accordance with the ASME code requirements, and written reports verifying this information must be furnished to the purchaser if requested.

The ASME Code for Unfired Pressure Vessels, Sec. VIII, Div. I, is generally preferred in the U.S. Informa- tion about and copies of the ASME Code may be obtained from Ref. 13. Information on other codes for construction of pressure vessels can be obtained from the government authorities shown in Table 12.17. I4

Design equations for shell and head thicknesses and working pressures according to ASME Code for Unfired Pressure Vessels, Sec. VIII. Div. I” are as follows.

Cylindrical shells

aEh,h P/h =

0.885r,. +O. lh,,,

Hemispherical heads.

hhh= Phhrc,

2&--0.2Phh

and

2aEh h,, Phh=

rc. +0.2hj,h

where h = minimum required thickness (exclusive of

corrosion allowance), in.. p = design pressure, psi, ri = inside radius of shell course under consid-

eration before corrosion allowance is added, in.,

u = maximum allowable stress value, psi (use one-quarter of tensile strength listed in Table 12.19 for a safety factor of 4.0),

E = joint efficiency for appropriate joint in cylindrical shells and any joint in spherical shells or the efficiency of ligaments between openings, whichever is less (see Table 12.18).

di = inside diameter, in., and r(. = inside spherical or crown radius, in.

h,., = Pc.sri Joint efficiency, E, for the above ASME design equa-

UEE-0.6p,, tions is shown in Table 12.18.

and aEh L:T

Materials of Construction for Separator

PC.7 = r, +0.6h,,

The most common steels used in the shells. heads, nozzles, and flanges in pressure vessels built under the ASME


TABLE 12.17-CODE AUTHORITIES FOR VARIOUS COUNTRIES

Country

Au&alla

Austria

Canada

People’s Republic of China

France

West Germany

lndonesla

Italy

Japan

Mexico

Government Authority and Address’

SAA, Standards Assn of Australia 8-86 Arthur Street North Sydney, NSW 2060, Australta ONORM, Oesterrelchlsches Normungstnstltut Leopoldsgasse 4A-1021 Weln 2. Austria

CSA. Canadian Standards Assn 178 Rexdale Blvd. Rexdale 603, Ont., Canada

China Assn. for Standardization P.0 Box 820 Beijing, Chma

AFNOR. Assoclallon Franqalse de Normalizatton Tour Europe, Cedex 7. 92080 Pans, La Defense, France

DNA, DIN. Deutsches NotmenausschuB 4-7 Burggrafenstrafle Postfach 1107 Berlin 30. West Germany

Badan Kerjasama Standardisas LIpI-Ydn! Jln. Teuku Chlk Dltiro 43 P 0 Box 250 Jakarta, lndonesla

UNI, Ente Nazlonale Italian0 de Unlflcazione Piazza Armando Dlaz 2, 120123 Milano, Italy

JISC. Japanese lndustnal Standards CommIttee Agency of Industrial Science and Technology Mlmstry of lnternallonal Trade and Industry 1-3-1 Kasumlgasekl Chlyoda-Ku Tokyo 100, Japan

DGN, Direction General de Normas Calle Puente de Tecamachalco No. 6 Lomas de Tecamachalco Secclon Fuentes Naucalpan de Juarez 53-950 Mexico DF, Mexico

Country

The Netherlands

Norway

Saud1 Arabia

Singapore

Sweden

Switzerland

United Kingdom

USSR

Venezuela

Government Authonty and Address’

NNI. Nederlands Normalisatle lnstltuut Kalfjeslaan 2 P 0 Box 5059 2600 GB Delft, The Netherlands

SSF. Norges Standardlserlngsforbund Haakon VII’s Gate 2 Oslo 1. Norway

SASO, Saud1 Arabian Standards Organlzatlon P.O. Box 3437 Rlyadh, Saudi Arabia

SIRU. Smgapore lnshlute of Standards and Industrial Research

Maxwell Road P.O. Box 2611 Singapore. Singapore

SIS. Standard!seringskommlsslonen Tegnergatan 11 P 0 Box 3295 S-103-66 Stockholm, Sweden

SNV. Assoclatlon Sulsse de Normallsatlon Kirchenweg 4 Zurich, Switzerland

BSI, Brltlsh Standards lnstltute 2 Park Street London WlA-28s. England

USSR State CommIttee for Standards Lenlnsky Prospekt 9 Moskva 117049, USSR

COVENIN. Comlslon Venezolana de Normas lndustriales

Avenlda Andres Bello Edificio Torre Fondo Comun PISO 11 Caracas 1050. Venezuela

‘The names and addresses of Ihe government agences for other countr~s can be obtaIned from Ref 14

Code for Unfired Pressure Vessels, Sec. VIII, Div. I, are listed in Table 12.19, along with the most important information concerning these materials. For more detailed information on these steels refer to the corresponding ASTM specifications indicated in Table 12.19.

Controls, Valves, Accessories, and Safety Features for Oil and Gas Separators Controls

The controls required for oil and gas separators are liquid- level controllers for oil and oil/water interface (three- phase operation) and gas back-pressure control valve with pressure controller.

Valves

The valves required for oil and gas separators are oil- discharge control valve, water-discharge control valve (three-phase operation), drain valves, block valves. pressure relief valve, and valves for sight glasses.

Accessories

The accessories required for oil and gas separators are pressure gauges, thermometers, pressure-reducing regulators (for control gas), level sight glasses, safety head with rupture disk, piping, and tubing.

Safety Features for Oil and Gas Separators

Oil and gas separators should be installed at a safe distance from other lease equipment. Where they are installed on offshore platforms or in close proximity to other equipment, precautions should be taken to prevent injury to personnel and damage to surrounding equipment in case the separator or its controls or accessories fail.

The following safety features are recommended for most oil and gas separators.

High- and Low-Liquid-Level Controls. High- and low- liquid-level controls normally are float-operated pilots that actuate a valve on the inlet to the separator, open a bypass around the separator, sound a warning alarm, or perform some other pertinent function to prevent damage that might result from high or low liquid levels in the separator.

High- and Low-Pressure Controls. High- and low- pressure controls are installed on separators to prevent excessively high or low pressures from interfering with normal operations. These high- and low-pressure controls can be mechanical, pneumatic, or electric and can sound a warning, actuate a shut-in valve, open a bypass, or perform other pertinent functions to protect personnel, the separator, and surrounding equipment.


TABLE 12.18-JOINT EFFICIENCY, ASME SEC. VIII, DIV. I

Single-welded Butt Joints Double-Welded (back-up strip left in place) Butt Joints

tom (O/o Radiograph) (O/O)

90 100 100

:05 spot 85 None 70

ing fluids. A valve should not be used between the safety head and the separator because it may inadvertently be closed. Water should not be allowed to accumulate on top of the rupture disk because ice could form and alter the rupture characteristics of the disk. Operation of an oil and gas separator without a properly sized and installed safety head (rupture disk) is not recommended.

Pressure relief valves may corrode and leak or may

High- and Low-Temperature Controls. Temperature controls may be installed on separators to shut in the unit, to open or to close a bypass to a heater, or to sound a warning should the temperature in the separator become too high or too low. Such temperature controls are not normally used on separators, but they may be appropriate in special cases.

Safety Relief Valves. A spring-loaded safety relief valve is usually installed on all oil and gas separators. These valves normally are set at the design pressure of the vessel. Safety relief valves serve primarily as a warning, and in most instances are too small to handle the full rated fluid capacity of the separator. Full-capacity safety relief valves can be used and are particularly recommended when no safety head (rupture disk) is used on the separator.

Safety Heads or Rupture Disks. A safety head or rupture disk is a device contammg a thin metal membrane that is designed to rupture when the pressure in the separator exceeds a predetermined value. This is usually from 1 1/4 to 1% times the design pressure of the separator vessel. The safety head disk is usually selected so that it will not rupture until the safety relief valve has opened and is incapable of preventing excessive pressure buildup in the separator.

Operation and Maintenance Considerations for Oil and Gas Separators Periodic Inspection

In refineries and processing plants, it is normal practice to inspect all pressure vessels and piping periodically for corrosion and erosion. In the oil fields, this practice is not generally followed, and equipment is replaced only after actual failure. This policy may create hazardous conditions for operating personnel and surrounding equipment. It is recommended that periodic inspection schedules for all pressure equipment be established and followed to protect against undue failures.

Installation of Safety Devices

All safety relief devices should be installed as close to the vessel as possible and in such manner that the reaction force from exhausting fluids will not break off, un- screw, or otherwise dislodge the safety device. The discharge from safety devices should not endanger personnel or other equipment.

Safety Heads (Rupture Disks)

The discharge from a safety head should be open and without restriction. The discharge line from a safety device should be parallel to a vertical separator and perpendicular to a horizontal one; otherwise the separator may be blown over by the reaction force from exhaust-

“freeze” in the closed position. They should be checked periodically and replaced if not in good working condition. Discharge lines, especially those on full-capacity relief valves, should be such that reaction force from discharge will not move the separator. Safety relief valves with “try” handles are recommended for general use.

Some operators use pilot-operated relief valves where frequent testing of the relief valves is required.

Mist Extractors

Some mist extractors in oil and gas separators require a drain or liquid downcomer to conduct liquid from the mist extractor to the liquid section of the separator. This drain will be a source of trouble when pressure drop through the mist extractor becomes excessive. If the pressure drop across the mist extractor, measured in inches of oil, exceeds the distance from the oil level in the separator to the mist extractor, the oil will flow from the bottom of the separator up through the mist-extractor drain and out

with the gas. This condition may be aggravated by partial plugging of the mist extractor with paraffin or other foreign material. This explains why some separators have definite fixed capacities that cannot be exceeded without liquid carryover through the gas outlet, and it also explains why the capacities of some separators may be lowered with use. In recent years, separators of advanced design have used mist extractors that do not require drains or downcomers. These designs eliminate this source of trouble (see Fig. 12.14).

Low Temperatures

Separators should be operated above hydrate-formation temperatures. Otherwise hydrates may form in the vessel and partially or completely plug it, thereby reducing the capacity of the separator and, in some instances when the liquid or gas outlet is plugged or restricted, causing the safety valve to open or the safety head to rupture. Steam coils can be installed in the liquid section of oil and gas separators to melt hydrates that may form there. This is especially appropriate on low-temperature separators.

Corrosive Fluids

A separator handling corrosive fluid should be checked periodically to determine whether remedial work is required. Extreme cases of corrosion may require a reduction in the rated working pressure of the vessel. Periodic hydrostatic testing is recommended, especially if the fluids being handled are corrosive. Expendable anodes can be used in separators to protect them against electrolytic corrosion. Some operators determine separator shell and head thickness with ultrasonic thickness indicators and calcu-

late the maximum allowable working pressure from the remaining metal thickness. This should be done yearly offshore and every two to four years onshore.


TABLE 12.19-PROPERTIES OF MATERIALS OF CONSTRUCTION FOR PRESSURE VESSELS

Properties of Materials Carbon and Low Allov Steel’

Nominal Form Composition

Plate C

C

C-Si

C-Si

C-Si

C-Si

C-Si

C-Si

C-Mn-Si

C-Mn-Si

Flange and

Fitting

C-Mn-Si

C-Si

C-Mn

Pipe

C-Mn-Si

C-Mn

C-Mn

C-Mn

Bolting ICr-115 MO.

Specification

Grade Number

SA-283 C

SA-285 C

SA-515 55

SA-515 60

SA-515 65

SA-515 70

SA-516 55

SA-516 60

SA-516 65

SA-516 70

SA-105

SA-181

SA-350

I

LFl

SA-53

SA- 106

LF2

B

B

SA-333 6

SA-193 87

SA-194 2H

SA-307 B

Application

Structural quality. For pressure vessel may be used with limitations. See Note I.

Boilers for stationary service and other pressure vessels

Primarily for intermediate- and high-temperature service




For moderate- and lower- temperature servrce

For moderate- and lower- temperature service

For moderate- and lower- temperature service

For moderate- and lower- temperature service*’

For high-temperature service

For general service

For low-temperature service

For general service

For high-temperature service

Low-temperature and sour-gas service

For high-temperature service 2Win. diameter or less

For high-temperature service nut

Machine bolt for general use

Specification

Number

SA-283

SA-285

SA-515

SA-515

SA-515

SA-515

SA-516

SA-516

SA-516

SA-516

SA-105

SA-181

SA-350

SA-350

SA-53

SA-106

SA-333

SA-193

SA-194

SA-307

Grade

C

C

55

60

65

70

55

60

65

70

-

I

LFl

LF2

B

B

6

87

2H

B

Tensile Yield P Strength Point

Number (1,000 psi) (1,000 psi) Notes

1 55.0 30.0 1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

-

-

-

55.0 30.0 2.6

55.0 30.0 3

60.0 32.0 3

65.0 35.0 3

70.0 38.0 3

55.0 30.0 3,8

60.0 32.0 33

65.0 35.0 3,8

70.0 38.0 3,8,10

70.0 36.0 23

60.0 30.0 23 60.0 30.0 -

70.0 36.0

60.0 35.0 GM7 60.0 35.0 3

60.0 35.0 9

125.0 105.0 - <21/z-in. diameter

55.0 - -

55.0 5 -

‘Data of the most frequently used materials from ASME Code Set II and VIII ‘^Wlth charpy tests. this steel 1s used for low-temperature and sour-gas service

1 SA-283 C plate may be used for pressure parts I” pressure vessels provided all of the followlng requirements are met (I) The vessels are not used to contam lethal substances, elther liquid or gaseous, (2) The maternl IS not used in the constructlo” of unfwd steam hollers. (3) The design temperature at which the material IS used 1s between - 20nF and 65OOF; (4) For shells, heads, and nozzles only. the thickness of plates on which strength welding IS applied does not exceed 5/ m ; and (5) The steel IS manufactured by the electnc furnace, open-hearth process, or the basic oxygen process.

2 For servw temperatures above 850°F. It IS recommended that hilled Steels cantalnlng not less than 0.10% residual s~llcon be used Kllled steels that have been deoxldlzed wth large amounts of aluminum and rimmed steels may have creep and stress-rupture prapertles m the temperature range above 850°F. which are somewhat less than those on whach the values I” the table are based

3 Upon prolonged exposure to temperatures above about 800°F, the carblde phase of carbon steel may be converted to graphite 4 Only ktlled sleel shall be used above 850°F 5 Not permItted above 45O’F. allowable stress value 7.000 psi. 6 The maler~al shall not be used I” thicknesses above 2 I”. 7 For welded p!pe the maximum allowable stress values are 15% less No !ncrease in these stress values shall be allowed for the performance of radiography. 8 The stress values fo be used for temperatures below - 2o’F when steels are made to conform wth Supplement (5) SA-20 shall be those that are given m the column for - 20 lo

65O=‘F. 9. SA-333 pape IS SA-106-S pape wth charpy tests made at the mill by the manufacturer of lhe pipe

10 Far low-temperature and sour-gas serwce. charpy tests must be made on this steel before fabrlcatmn of the vessel . Most 01 the mformatlon for this table was taken from Ref. 14


Paraffin

A separator handling paraffin-based oil may need to be steamed periodically to prevent plugging and a resultant decrease in capacity. This reduction in capacity often results in liquid carryover in the gas or discharge of excessive gas with the liquid.

High-Capacity Operation

Where separators are operating near or at their maximum rated capacity, they should be checked carefully and periodically to determine whether acceptable separation is being accomplished.

Pressure Shock Loads

Wells should be switched in and out of the separator slowly. Fast opening and closing of valves causes damaging shock loads on the vessel, its components, and piping.

Throttling Discharge of Liquid

Throttling discharge of small volumes of liquid from separators normally should be avoided. Throttling may cause erosion or wire drawing of the inner valves and seats of the liquid-dump valves and may erode the dump-valve bodies to the extent that they may burst at or below rated working pressures.

However, throttling discharge may be necessary because processing units, such as lower-pressure separators or stabilization units, downstream of the separator may require relatively steady flow. Liquid-discharge control valves on separators should be sized for the volume of liquid the separator must handle. Such valves usually should be smaller than the lines in which they are installed. Reduced inner valves can be used to size the valve properly to minimize wire drawing during throttling service.

Pressure Gauges

Pressure gauges and other mechanical devices on separators should be tested for accuracy at regular intervals. Isolating valves should be used so that pressure gauges can be easily removed for testing, cleaning, repairs, or replacement.

Gauge Cocks and Glasses

Gauge cocks and gauge glasses should be kept clean so that the liquid level observed in the sight glass indicates at all times the true liquid level in the separator. Periodic flushing of the gauge glass or cleaning with special solvents and swabs is recommended.

Cleaning of Vessels

It is recommended that all separator vessels be equipped with manways, cleanout openings, and/or washout connections so that the vessels can be cleaned periodically. Larger vessels can be equipped with manways to facilitate their cleaning. Smaller vessels can be equipped with handholes and/or washout connections so that they can be easily cleaned or washed out periodically.

SI Metric Unit Conversions Most conversions from customary units to Sl metric units in Chap. 12 can be made directly by using the conversion factors presented in the tables in the publication en- titled “The SI Metric System of Units and SPE Metric Standard.” The exceptions are listed below.

The values of F,, in Fig. 12.32 should be multiplied by 30.48 to change them for use in the SI metric units. F,, is used in Eqs. 1, 4, 5, and 6.

In Eq. 3, the gas constant R in customary units has the value 10.732. In SI metric units, the value of R is 0.083 145 where the gas volume is in cubic meters, pressure in bars, temperature in degrees Kelvin, and n in kmol. I6

Nomenclature

A, = cross-sectional area of separator for gas flow, sq ft

A, = cross-sectional area of separator for oil flow, sq ft

di = ID, in. D = diameter of separator, in. (see Fig. 12.34

for D’) E = joint efficiency for appropriate joint in

cylindrical shells and any joint in spherical shells or the efficiency of ligaments between openings, whichever is less (Table 12.18)

F = stage pressure ratio p, lpz =p2/p3

=.. .PnJP., F,,, = configuration and operation factor

(empirical) (see Fig. 12.32 for values) h = thickness (exclusive of corrosion

allowance), in. h, = height (or depth) of oil in separator, ft

L = length of separator, ft (see Fig. 12.34 for

L ‘1 M, = molecular weight of gas

n; = number of interstages (number of stages - 1)

p = design pressure or separator operating pressure, psia

pb = base pressure, psi ps = storage-tank pressure, psia p 1 = first-stage separator pressure, psia p2 = second-stage separator pressure, psia p.1 = third-stage separator pressure, psia

q‘q = volume of gas flowing through separator, cu ftisec

r(. = inside spherical or crown radius, in. T; = inside radius of shell course under consid-

eration before corrosion allowance is added, in.

R = gas constant T = operating temperature, “R

Tb = base temperature, “R

vg = gas velocity, fttsec V, = volume required in separator for oil, cu ft

ZK = gas compressibility factor

-fx = specific gravity of gas

p.s = density of gas at operating conditions, Ibm/cu ft

pL = density of liquid at operating conditions, lbmicu ft


g = maximum allowable stress value, psi (use l/4 of tensile strength listed in Table 12.19 for a safety factor of 4.0)

Acknowledgments Appreciation is expressed to George 0. Ellis of Ellis En- gineering Co., Houston, for allowing the usage of the Comsign Computer Program to calculate the oil and gas capacities for vertical and horizontal separators shown in Figs. 12.35 through 12.38 and for the sample calculations in Tables 12.9 and 12.10.

References I. 2.

3.

4.

5.

6.

7.

Sales Bullerin. Porta-Test Systems Ltd. Edmonton. Alta. (1982) 4. “Peerless Vane-Type Line Separators.” Bull. 11000. Peerless Man- ufacturing Co., Dallas (1965). Lerner, B.J.: “Mist Elimination,” U.S. Patent No. 4,022.593 (1977). “Mist Eliminator Allows 100% Greater Throughput in Vessels,” Chemical Engineen’n~. ACS Industries Inc. (Aug. 20. 1984) 53, 54. S&s Bulk-fin. Plenty Metro1 Ltd., Newhury. Berkshire. England (1980). Schilling, J.R.: “Diverging Vortex Separator,” U.S. Patent No. “Gulf of Mexico Region Production Safety Systems.” OCS Order No.

4.394.138 (1982). 5, U.S. Dept. of the Interior, Metairie, LA (1980).

“Gas-Liquid Seperation Systems.” Vortec Inc., Woodside, CA (1984).

K.

9. IO.

“People. Products. Porta-Test,” Porta-Test Systems Ltd., Edmon- ton, Alta. (1979).

11.

12.

Comsign Computer Program. Ellis Engineering Inc., Houston. Vondy. D.: “Spherical Process Vessels.“ O,c a& Gus J. (Oct. 15, 1456) 148:(Nov. 12. 1956) 213: (Dec. 3. 1956) 13.5; (Jan. 7. 1957) 130; (Feb. 4. 1957) 136: (Feb. 11. 1957) 130; (March 11, 1957) 189: (April 8, 1957) 121-122. Klmmell. G.O.: “Stage Separation.” paper 48.PET-15 presented at the ASME Annual Meeting. Oklahoma City. Oct. 1949. “Separator Flash Calculations, Process Version 0882.” Simula- tion Science5 Inc.. Houston.

13. 14.

15.

ASME Boiler und Prrs~ure Ve.ur/ Code, Sec. VIII. Div. 1 (1986). KWIC Index of Irmmutionrrl Srmzdurd~. Intl. Orgamzation for Stan- dardlzatton. Geneva. Megye\y. E.F.: Pwxww I’~~,wl Hmdbook. Pressure Vessel Hand- hook Publishing Inc., Tulsa.

General References “Advancing Technology Gives New Production Equipment for Gaa-

011 Separation.” Bull. MTI-//79 Plenty-Metro1 Ltd.. Ncwbury, En-

gland (1979).

Alien. S.N rr ul.: “Uliyanie Fizicheskikh Svoistv Zhidkosti Na Rabotu Gdzovykh Separatorov,” Nrji K/w; (Jan. 1982) 40-43.

Barrett. R: “Unique High Efficiency Cyclone Separator,” Europe und OiL (1970).

Barry, A.F. and Parks, A.S.: “Low Temperature Separation Increases Recovery ot Condensate from Natural Gas.” W&r/ Oil, 7. 203.

Broussard. W.F. and Gravix, C.K.: “Three-Phase Separators.” l+‘wld Oil (April 1960) 127-32.

Campbell, J. M. : Gus Cortdiiionmg u/x/ Procr.wi,z~. Campbell Pctrolc- urn Series, Norman. OK (1979) 1. 119-14.

Chepkasov. V.M., Ovchinnikov, A.A., and Nlkolaev, N.A.: “Raschct Gidravlicheskogo Soprotivleniya Vikhrevykh Separatorov S Aksial- Nymi Zavikhntelyaml.” I~~’ vv.w/z Uchebn ZmwIk~ f.31: (May 1981) 43-48.

12-43

Cooper, F.E. et ul.: “Field Separatmn of Liquids.” Oil and GOJ J. (Feb. 9, 1959) 91-99.

Davies, E.E. and Watson, P.. “Miniaturized Separators Provide High Performance,” World Oil (April 1980).

Dixon. P.C.: “Method of and Means for Separating Liquid and Gas or Gaseous Fluid,” U.S. Patent No. 2,349,944 (1944).

Dottenveich, F.H.: “Mechanical Removal of Entrained Materials From Natural Gas,” Per. Eng. (May 1949) 546-49.

Field Handling of Nufural Gas, third ed., Petroleum Extension Serv- ice. U. of Texas, Austin (1974).

Fekete, L.A.: “Vortex Tube Separator May Solve Weight/Space Limi- tations,” World Oil (July 1986) 40-44.

Gaskell, T.F. and Wood. P.M.: “Production Improved in New Do- mains,” Oil and Gas Intl. (June 1963) 102-04.

Gravis, C.K.: “The Oil and Gas Separator,” World Oil (Jan. 1960) 84-92.

Guerrero, E.T.: “How To Make Stage-Separation Calculations,” Oii und Gas J. (Oct. 3, 1966).

Hwfropn Sulphide Corrosion in Oil und Gas Production: A Compila- rion of Classic Pqm, Natl. Assn. of Corrosion Engineers. Houston (1981).

Katz. D.L.: “Overview of Phase Behavior in Oil and Gas Production.” J. Pet. Tec,h. (June 1983) 1205- 14.

Mapes. G.J.: “The Low Temperature Separatton Unit,” World Oi/(Jan. 1960) 93-99.

Morris, J.K. and Smith. R.S.: “Crude Stabdlzer Can Save Money Off- shore,” Oil und Guu J. (May 7, 1984) 112-16.

Pemck. D.P. and Thrasher, W.B.: “Challenges Associated With the Design of Oil-Gas Separation Systems for North Sea Platforms. Pmt.. Ninth Annual Offshore Technology Conference. May 2-5, 1977.

Penick, D.P. and Thrasher, W.B.: “Mobil’s Design Considerations for North Sea OlllGas Separation Facilities.” Pet. Eng. (Oct. 1977) 22. 24, 26.

Peters. B.A.: “Get Better Separation With Proper Water-Weir Heights.” Oii and Gus J. (Nov. 19, 1973) 73, 74. 77.

Pollak, A. and Work, L.T.: “The Separation of Liquid From Vapor Usmg Cyclones.” Trans., ASME (Jan. 1942) 31-41.

“Pressure Vessel inspection Code-Maintenance, Inspection. Rating, Repair, and Alteration,” API Std. 5/O. third ed. (1983).

“Recommended Practice for Analysis Design, Installation. and Test- mg of Basic Surface Safety Systems on Offshore Production Plat- rorms,” API RP /4-C, third ed. (April 1984)

“Recommended Practice for Design and Installatmn of Offshore Pro- duction Platform Piping Systems,” API RI’ 14-E. founh cd. (April 1984).

Robinson. J.: “Fundamentals of Oil and Gas Separation.” Pmt., 1979 Gas Cond. Conference, March 5-7.

“Separator Loses Weight, Bulk-But Boosts Production,” Offshore Services and Technology, Newbury. England (May 1980).

“Separators-Technique.” Oil and Gu\ S~parcrrron ‘%ory, Applirarion and Lkipz. AIME (1977) 5.


Sinaiskii, E.G.: “0 Razdelenii Gazozhidkostnykh Smesei V Warner, B.J. and Scauzillo, F.: “The Design of Fibrous Filters for Mist Separatorakh,” I:v ~~~r.rh Udwhin Znvrd Neff GIX (March 1980) Elimination,” paper presented at the 1963 Gas Technology Confer- 46.54. ence, Norman, OK.

“Specification for Oil and Gas Separators,” API Spec. /Z-J, fifth ed. Whinery, K.F. and Campbell, J.M.: “A Method for Determining Op-

(Jan. 1982). timum Second Stage Pressure in Three Stage Separation,” J. Per. Tech. (April 1958) 53-54.

“Standard for Welding Pipelines and Related Facilities,” APISld. 1104. Worley, S.M. and Lawrence, L.L.: “Oil and Gas Separation is a 16th ed. (1983). Science,” J. Pet. Tech. (April 1957) 11-16.

Chapter 13

Gas Measurement and Regulation John M. Campbell Sr.. The Campbell Cornpaws*

Introduction

Measurement and regulation are an integral part of the production of gas. As the value of gas has increased, the need for more accurate measurement has become apparent. Proper regulation is necessary to prevent unnecessary flaring or production curtailment.

Today the engineer has a choice from among many instruments and controls, some of which are rather sophisticated. The advent of electronic and digital equipment offers many alternatives. Remote metering and control now are more commonly employed to minimize personnel overhead. The net result is a more complex production system than the traditional one.

Many seemingly superior instruments and controls are not suitable for some production systems. They may not be packaged to withstand exposure to the weather and elements in the environment-rain, salt spray, sand. wind, lightning, etc. Some devices are so sensitive that mere vibration may be a problem. Calibration and other like necessities may be inconvenient (or maybe impractical) in remote locations. Also, properly trained personnel may not be available for maintenance.

If transfer of custody is involved, contractural obliga- tion may limit the choice of system. Other factors like spare parts availability and vendor service all influence the proper selection of equipment and vendor.

Gas Measurement The choice of meter depends on the absolute volume involved and the rangeability-the ratio of maximum to minimum flow rate that a given size meter can accommodate. Reliability also is a factor. Both accuracy and precision must be cdnsidered. Accuracy is the difference between true and measured rate. Precision is the repeatability of a measurement, however wrong it may be.

There are a series of volumetric meters that measure volume directly by use of bellows, reciprocating pistons, rotating vanes and cams, etc. ’ These are used primarily for measuring small volumes of low-pressure gas and have little application in production operations.

Velocity Meters

A velocity meter is any device that measures flow by impact kinetic energy or by using the change in pressure that accompanies a controlled change in velocity. However this change is induced, meter performance is governed by an energy balance.

The Energy Balance. Fig. 13. I illustrates a general flow system on which a general flow equation may be based. Points 1 and 2 represent any two points in a flow system between which an energy balance may be written

The fluid entering at Point 1 has an internal energy, U 1 , which is the result of previous energy obtained. In addition, a certain amount of pV work must be done to get the fluid past Point I-i.e., the eiiergy necessary for it to flow. This fluid also possesses potential energy , X 1 , above the datum plane, and kinetic energy vI ‘i2g. The fluid leaving at Point 2 possesses the same forms of energy but the amount depends on the work W or heat Q gained or lost by the system between the two points. The general equation may then be written

2

lJ,+p,V,+Z,+~+Q-W 21:

=u2+p2v1+z?+$; . . . . . . . . . . . . . (1)


L----------------------J I DATUM X=0 2

Fig. 13.1 -General flow system.

where U = internal energy, which includes all energy

such as heat, electrical, chemical, and

surface, p = pressure,

V = volume or specific volume,

Z = height above an arbitraty datum plane,

v = velocity in conduit,

Q = heat gained or lost by system (plus if

gained, minus if lost), and

W = work done by system (plus if done by

system, minus if done on system).

The many ramifications of Eq. 1 are developed in any standard text on thermodynamics.

Eq. 1 may be applied to any of the systems shown in Fig. 13.2. When writing it between Points 1 and 2 of any of the three systems shown, it may be reduced to

2 2

p,Y,-p~v*=~-~. . . . . . . . . . . . . (2)

Eq. 2 develops on the normally correct premise that Q, W, and AU are substantially zero between the two points. These two points are, furthermore, so close together that any change in potential energy is negligible in comparison with the other changes involved. Eq. 2 is the basic equation governing velocity or head meters and expresses the general fact that changes in velocity in

’ such a system must be accompanied by a corresponding change in static pressure.

Forms of Meter. Fig. 13.2 shows several common ways in which a velocity change may be imposed on the system. In each instance, a constriction is supplied for increasing the velocity (decreasing static pressure), following which the velocity returns to normal. The amount of permnerzr pressure loss across the device is small and depends largely on the amount of turbulence and friction loss involved.

The Venturi tube shown in Part a is designed so that a high-pressure differential can be induced and still minimize the permanent pressure loss due to turbulence and eddy currents. In general, it will restore at least 90% of the pressure drop at the throat. The sections involved,

1 though, are expensive to manufacture and bulky to handle, especially at elevated pressures. Consequently, their

Fig. 13.2-Diagrammatic representation of fluid flow through a flat-plate orifice showmg relative positlons of the pressure taps in common use and the change in static pressures (flow is from left to right). (a) Venturi tube, cross section; (b) flow nozzle, cross section; (c) location of taps; (d) schematic view of pitot tube.

use is usually limited to low-pressure applications where high-pressure recovery is essential. The flow nozzle is essentially a cross between the three devices shown and represents an attempt to incorporate the best features of each. General usage is limited.

The flat-plate orifice is used in the majority of all gas- production and transportation facilities. It is simple, inexpensive, handy to change and stock, and gives reproducible results. Although the pressure recovery is seldom higher than 65%, this is not a critical fault, for most systems are designed to operate with a low differential pressure.

Derivation of an Orifice Equation. Eq. 2 may be simplified by assuming that V, = V? = V, an assumption that is approximately ttue and is later corrected by a term called the expansion factor. Then

VI 2

-v2 2 =2gV(p* -p, )=2gh. . . . . (3)

The term h is the differential head loss between Points 1 and 2 expressed as “feet of fluid” (that fluid flowing in the system).

The velocities may be expressed in terms of volume rate of flow qR and diameters d; and d, (internal diameter of pipe and orifice opening, respectively). This substitution yields an intermediate equation

‘R =A2 &i Jql, . (4)

where A2 =area of orifice plate opening, and Fd =d,ldi. The second term on the right side of Eq. 4 is often referred to as the approach factor.

GAS MEASUREMENT AND REGULATION

Further modification of our theoretical equation is necessary. Part c of Fig. 13.2 shows that the diameter of the flowing stream becomes smaller than that of the orifice opening. This point, known as the vena contrac- ta, is really the limiting diameter, and some correction must be made when using the orifice-opening diameter in the equation. There is also an energy loss due to turbulence and friction not yet accounted for. Both of these considerations are easily handled by use of an efficiency factor, K,, that must be determined experimentally.

For convenience we furthermore like to express h in terms of inches of water, h ,I.. If all these corrections and conversions are superimposed on Eq. 4 the equation reads

where

Yx - gas-flow rate, std cu ft/hr measured at I)),

and Th, di = internal diameter, in.,

d,, = orifice-opening diameter. in.,

Pf = flowing pressure, psia,

P \C’ = base pressure, psia,

Tf = flowing temperature, “R.

T,,. = base temperature, “R,

h ,,. = differential across orifice, in. H20,

Yr: = specific gravity of gas (air= 1 .O), and

K, = efficiency factor, which includes the

approach factor.

Eq. 5 may be modified to a more standard form by (1) combining terms, (2) writing it for a gas whose specific gravity= 1.0, and (3) assuming that p,,( = 14.73 psia and T,,. = Tf=520”R.

This yields the two basic orifice equations

qs=C’Jh,pf . . . . . . . . . . . . . (6)

and

C’=FhF,,I,F,i,F,~FdF,YF,,.F,,,F,,F,. _. (7)

The value C’ is known as the orifice constant. It is determined primarily from F,, the basic orifice flow factor, which is equal to 338.17 K,d,,‘. The other multipliers shown in Eq. 7 are correction factors to correct for the basic assumptions necessary to arrive at Eq. 6. Only the first five are usually used in routine lease operations. in measuring gas for proration records and the like. The last six correction factors are relatively small and are used to obtain accurate records in the sale or purchase of gas. The last two apply only to meters with mercury manometers.

Table 13.1 summarizes these constants except for F,,,. , which is calculated from compressibility factor, and F,,. calculated from the expansibility of metals used for orifice construction. 2.x

13-3

The comparable metric standard for the orifice equations and constants is ANSI/API 2530. * At the time this material was being prepared, the conversion effort to SI units had not been completed. Until it is, calculations may be made in English units and converted.

q,(m3)=(0.028 33)(cu ft),

where qR =m3 measured at 101.315 kPa and 15°C and cu ft=volume measured at 14.73 psia and 60°F.

Orifice Constants

The value of the constants in Eq. 7 depends in many cases on the points between which the differential head loss, h H,, is measured. Two standards are provided in gas measurement-flange taps and pipe taps. With the former, the flange or orifice holder is so tapped that the center of the upstream and downstream taps is 1 in. from the respective orifice-plate surfaces. For standard pipe taps, the upstream tap is located two and one-half pipe diameters upstream and eight pipe diameters downstream. The location of Sec. 1 (Tables 13. la, g, i, and 1) of the orifice tables is forflange taps and Sec. 3 (Tables 13. lb, h, j, and m) is for pipe taps. Sec. 2 (the remaining parts of Table 13.1) shows the fluid-condition factors common to both types.

Basic Orifice Factor Fb. This factor, as noted above, is based on those assumptions necessary to go from Eq. 5 to 6. These are T,,,. =520”R, yR = 1 .OO, and Tf=520”R. It is a function of the experimental constant K,,. which means that it depends on the location of the differential pressure taps and the internal pipe diameter, in addition to the orifice diameter.

The value of Fb may be found from tables of Fh shown in Table 13.1 a for flange taps and Table 13.1 b for pipe taps if the meter run is of standard internal diameter. For gas-measurement work involving sale and purchase by a gas-transmission or distribution company, corrections are provided for values of F,, not fitting the standard tables. These are not shown here, for they are not normally used in production operations.

Pressure-Base Factor Ft,b. Values of this factor are shown in Sec. 2 (Table 13.1~) of the charts. It corrects the value of Fh for the case where the pressure base used is not 14.73 psia. It my be determined by the equation F,,f, = 14.73 psc.

Temperature-Base Factor Frb. This is shown in Table 13. Id and corrects for any contract wherein the base temperature is not 520”R (60°F). This factor may be computed by the equation F,/, = T,,,.i520.

Specific-Gravity Factor Fg. This factor, shown in Table 13. le, is to correct the basic orifice equation for those cases where the specific gravity of the gas is other than 1.00. The equation is F, =(l/y,)“.s.

Flowing-Temperature Factor Fu. This factor corrects for those cases where the flowing temperature of the is other than 60°F. The equation is F,f=(520/Tf-) 8

as 5.

Values are given in Table 13. If.

(contmued on Page 8)

13-4

Orifice Diameter,

;K)

0.250 0.375 0.500 0.625 0.750

0.875 1.000 1.125 1.250 1.375

1.500 1.625 1.750 1.875 2.000

2.125 2.250 2.375 2.500 2.625

2.750 2.875 3.000 3.125 3.250

3.375 3.500 3.625 3.750 3.875

4.000 4.250 4.500 4.750 5.000

5.250 5.500 5.750 6.000 6.250

6.500 6.750 7.000 7.250 7.500

7.750 8.000 8.250 8.500 8.750

9.000 9.250 9.500 9.750

10.000

10.250 10.500 10.750 11.000 11.250

PETROLEUMENGINEERINGHANDBOOK

TABLE 13.la-FLANGE TAPS: BASIC ORIFICE FACTORS, F,

lnternalDiameterofPipe,d,, in

1.689

12.695 28.474 50.777 80.090 117.09

162.95 219.77 290.99 385.78 -

-

-

- - - -

-

-

-

- - - - -

-

-

- - - - -

- - -

-

- - - -

- - - - -

- - - - -

- 2

1.939 2.067

12.707 28.439 50.587 79.509

115.62

159.56 212.47 276.20 353.58 448.57

-

- - - - -

-

-

-

- - - - -

-

-

- - - -

- - - - -

- - - - -

- - - - -

- - - - -

12.711 28.428 50.521 79.311

115.14

158.47 210.22 271.70 345.13 433.50

542.26 -

-

- - - - -

-

-

-

- - - -

- - - - -

- - - - -

- - - - -

- - - - -

-

- - -

- - - -

3

2.300

12.714 28.411 50.435 79.052

114.52

157.12 207.44 266.35 335.12 415.75

510.86 623.91

-

- - - - -

-

- - -

- - - - -

- - - - -

- - - - -

- - - - -

- - - - -

- - - - -

- - - -

-

2.626 2.900 3.068

12.712 28.393 50.356 78.818

113.99

156.00 205.18 262.06 327.39 402.18

487.98 586.82 701.27 834.88

- - - - -

- - -

-

- - - - -

- - - - -

- - - - -

- - - - -

-

- - -

- - - - -

-

-

12.708 28.382 50.313 78.686

113.70

155.41 204.04 259.95 323.63 395.80

477.36 569.65 674.44 793.88 930.65

1091.2 - - - -

- - - - -

- - - - -

- - - - -

- - - - -

- - - - -

-

-

-

- - - - -

-

-

12.705 28.376 50.292 78.625

113.56

155.14 203.54 259.04 322.03 393.09

472.96 562.58 663.42 777.18 906.01

1052.5 1223.2

-

-

- - - - -

- - - - -

- - - - -

- - - - -

- - - - -

-

-

-

- - - - -

- -

-

GASMEASUREMENTANDREGULATION 13-5

TABLE 13.la-FLANGE TAPS: BASIC ORIFICE FACTORS, F, (continued)

Orifice Diameter,

d 0’ (in.)

0.250 0.375 0.500 0.625 0.750

0.875 1.000 1.125 1.250 1.375

1.500 1.625 1.750 1.875 2.000

2.125 2.250 2.375 2.500 2.625

2.750 2.875 3.000 3.125 3.250

3.375 3.500 3.625 3.750 3.875

4.000 4.250 4.500 4.750 5.000

5.250 5.500 5.750 6.000 6.250

6.500 6.750 7.000 7.250 7.500

7.750 8.000 8.250 8.500 8.750

9.000 9.250 9.500 9.750

10.000

10.250 10.500 10.750 11.000 11.250

3.152

12.703 28.373 50.284 78.598 113.50

155.03 203.33 258.65 321.37 391.97

471.14 559.72 658.96 770.44 896.06

1038.1 1499.9

- -

- - - - -

- - - - -

- - - - -

- - - - -

- - - - -

-

-

- - - - -

-

-

-

4

- 3.438

12.697 28.364 50.258 78.523 113.33

154.71 202.75 257.63 319.61 389.03

466.39 552.31 647.54 753.17 870.59

1001.4 1147.7 1311.7 1498.4

-

- - - - -

- - - - -

- - - - -

- - - - -

- - - - -

-

-

-

- - -

-

-

-

Internal Diameterof Pipe,d,, in

3.826 4.026

12.687- 28.353 50.234 78.450 113.15

154.40 202.20 256.69 318.03 386.45

462.27 545.89 637.84 738.75 849.41

970.95 1104.7 1252.1 1415.0 1595.6

1797.1 - - - -

- - - - -

- - - -

12.683 28.348 50.224 78.421

113.08

154.27 201.99 256.33 317.45 385.51

460.79 543.61 634.39 733.68 842.12

960.48 1089.9 1231.7 1387.2 1558.2

1746.7 1955.5 2194.9

- - - - -

- - - - -

-

-

- - - - -

-

-

-

- -

-

-

- - -

- - - - -

- - - - -

-

-

- - -

-

-

-

-

6

4.897

- -

50.197 78.338

112.87

153.88 201.34 255.31 315.83 382.99

456.93 537.77 625.73 721.03 823.99

934.97 1054.4 i 182.9 1320.9 1469.2

1628.9 1801.0 1986.6 2187.2 2404.2

2639.5 2895.5 3180.8

- -

- - - - -

-

-

- - - - -

- -

-

- - - - -

-

-

-

5.189 - -

50.191 78.321 112.82

153.78 201.19 255.08 315.48 382.47

456.16 536.64 624.09 718.69 820.68

930.35 1048.1 1174.2 1309.3 1453.9

1608.7 1774.5 1952.4 2143.4 2348.8

2569.8 2808.1 3065.3 3345.5 3657.7

- - - - -

-

-

-

- - - - -

- - - - -

- -

-

- - - - -

5.761 -

50.182 78.296

112.75

153.63 200.96 254.72 314.95 381.70

455.03 535.03 621.79 715.44 816.13

924.07 1039.5 1162.6 1293.8 1433.5

1582.1 1740.0 1907.8 2086.4 2276.5

2479.1 2695.1 2925.7 3172.1 3435.7

3718.2 4354.8

- - -

-

- - - -

- - - - -

- - - - -

- - - - -

6.065 - -

50.178 78.287 112.72

153.56 200.85 254.56 314.72 381.37

454.57 534.38 620.88 714.19 814.41

921.71 1036.3 1158.3 1288.2 1426.0

1572.3 1727.5 1891.9 2066.1 2250.8

2446.8 2654.9 2876.0 3111.2 3361.5

3628.2 4216.6 4900.9

- -

- - - - -

- -

-

- - -

-

- -

- -

- - - - -



Orifice Internal Diameter of Pipe, d,, in. Diameter,

d 8 "\ (in.) 7.625 7.981 8.071 9.564 10.020 10.136

0.250 0.375 0.500 0.625 0.750

0.875 1.000 1.125 1.250 1.375

1.500 1.625 1.750 1.875 2.000

2.125 2.250 2.375 2.500 2.625

2.750 2.875 3.000 3.125 3.250

3.375 3.500 3.625 3.750 3.875

4.000 4.250 4.500 4.750 5.000

5.250 5.500 5.750 6.000 6.250

6.500 6.750 7.000 7.250 7.500

7.750 8.000 8.250 8.500 8.750

9.000 9.250 9.500 9.750 10.000

10.250 10.500 10.750 11.000 11.250

- - - - - - - - - -

153.34 153.31 200.46 200.39 253.99 253.89 313.91 313.78 380.25 380.06

453.02 452.78 532.27 531.95 618.02 617.60 710.32 709.77 809.22 808.50

914.79 913.86 1027.1 1025.9 1146.2 1144.7 1272.3 1270.3 1405.4 1402.9

1545.7 1542.5 1693.4 1689.3 1848.8 1843.5 2011.6 2005.2 2182.6 2174.6

2361.8 2352.0 2549.7 2537.7 2746.5 2731.8 2952.6 2934.8 3168.3 3146.9

3394.3 3368.5 3879.4 3842.3 4412.8 4360.5 5000.7 4928.1 5650.0 5551.1

6369.3 6236.4 7170.9 6992.0

- 7830.0 - - - -

- - - - -

- - - - -

-

- -

- - - - -

- - - - -

- - - - -

- -

- -

- - - - -

- - -

-

153.31 200.38 253.87 313.74 380.02

452.72 531.87 617.50 709.84 808.34

913.64 1025.6 1144.3 1269.8 1402.3

1541.8 1688.4 1842.3 2003.8 2172.9

2349.9 2535.0 2728.6 2930.8 3142.1

3362.9 3834.2 4349.0 4912.2 5529.5

6207.3 6953.6 7777.8 8706.9

-

- - - -

- - - - -

- - - -

- - - - -

10

- - - - - -

200.20 253.55 313.31 379.44

451.95 530.87 616.21 707.99 805.23

910.97 1022.2 1140.1 1264.5 1395.6

1533.4 1678.0 1829.4 1987.8 2153.2

2325.7 2505.6 2692.8 2887.6 3090.1

3300.6 3746.1 4226.0 4742.7 5298.6

5897.4 6543.1 7240.0 7993.3 8808.9

9693.3 10654 11711

- -

- - - - -

- - - -

- - - - -

- - -

- - -

- - -

- - - -

253.48 253.47 313.20 313.18 379.29 379.26

451.76 451.72 530.63 530.57 615.90 615.83 707.61 707.51 805.76 805.65

910.38 910.24 1021.5 1021.3 1139.2 1139.0 1263.4 1263.1 1394.2 1393.9

1531.7 1531.3 1675.9 1675.4 1826.9 1826.3 1984.7 1984.0 2149.5 2148.6

2321.2 2320.2 2500.1 2498.9 2686.2 2684.7 2079.7 2877.9 3080.7 3078.5

3289.3 3286.8 3730.2 3726.7 4204.1 4199.2 4712.8 4706.2 5258.5 5249.6

5843.6 5831.8 6471.9 6456.3 7146.9 7126.5 7873.0 7846.6 8654.8 8621.1

9498.1 9455.3 10409 10355 11394 11327 12467 12381 13656 13541

- - - - -

- - - -

- - - - -

- - - -

-

- -

- - - - -

GASMEASUREMENTAND REGULATION 13-7


Orifice Internal Diameter of Pipe, d,. in. Diameter,

d "3 (in.) 11.376

0.250 -

12

11.938 12.090 14.688 -

0.375 0.500 0.625 0.750

0.875 1.000 1.125 1.250 1.375

1.500 1.625 1.750 1.875 2.000

2.125 2.250 2.375 2.500 2.625

2.750 2.875 3.000 3.125 3.250

3.375 3.500 3.625 3.750 3.875

4.000 4.250 4.500 4.750 5.000

5.250 5.500 5.750 6.000 6.250

6.500 6.750 7.000 7.250 7.500

7.750 8.000 8.250 8.500 8.750

9.000 9.250 9.500 9.750

10.000

10.250 10.500 10.750 11.000 11.250

- -

- -

312.94 378.94

451.30 530.04 615.16 706.68 804.61

908.98 1019.8 1137.1 1260.8 1391.1

1528.0 1671.4 1821.4 1978.1 2141.5

2311.7 2488.7 2672.6 2863.5 3061.4

32664 3698.4 4160.4 4653.4 5179.0

5730.5 6333.8 6966.9 7640.4 8357.3

9121.0 9935.2 10804 11732 12725

13787 14927 16158 17505

- -

- - - - -

- - -

-

-

- - -

312.85 378.82

451.14 529.83 614.90 706.36 804.23

908.51 1019.2 1136.4 1260.0 1390.1

1526.8 1670.0 1819.7 1976.1 2139.2

2308.9 2485.4 2668.7 2858.8 3055.9

3260.0 3689.6 4148.4 4637.2 5157.4

5710.0 6296.6 6919.0 7579.0 8278.9

9021.7 9810.5 10649 11540 12489

13500 14578 15730 16962 18296

-

- - - - - -

312.83 378.79

451.10 529.78 614.84 706.28 804.13

908.39 1019.1 1136.2 1259.8 1389.9

1526.5 16696 18193 1975.6 2138.6

2308.2 2484.6 2667.7 2857.7 3054.6

32585 3687.5 4145.5 4633.4 51523

5703.3 6287.9 6907.8 7564.7 8260.7

8998.7 9781.6 10613 11496 12434

13433 14498 15633 16845 18148

19565 - - - -

- - - - -

-

- - -

- - - -

450.53 529.06 613.94 705.18 802.78

906.77 1017.1 1133.9 1257.1 1386.7

1522.7 1665.2 1814.1 1969.6 2131.5

2299.9 2474.9 2656.4 2844.6 3039.4

3240.8 3663.8 4113.9 4591.5 5097.2

5631.4 6194.8 6788.1 7412.3 8068.4

8757.3 9480.4 10239 11035 11869

12745 13664 14628 15642 16706

17826 19004 20245 21552 22930

24385 25924 27567 29331

-

- -

- - - - -

- - - - - - - - - -

450.48 -

528.99 528.94 613.85 613.78 705.07 704.99 802.65 802.55

906.61 906.49 1017.0 1016.8 1133.7 1133.5 1256.8 1256.6 1386.4 1386.1

1522.4 1522.1 1664.8 1664.5 1813.7 1813.3 1969.0 1968.6 2130.9 2130.4

2299.2 2298.7 2474.1 2473.5 2655.5 2654.8 2843.5 2842.7 3038.1 3037.2

3239.4 3238.3 3661.9 3660.5 4111.5 4109.7 4588.4 4586.0 5093.1 5090.1

5626.1 5622.3 6188.1 6183.1 6779.6 6773.3 7401.5 7393.6 8054.8 8044.8

8740.3 8727.9 9459.4 9444.0 10213 10194 11003 10980 11831 11803

12698 12664 13607 13566 14560 14511 15560 15501 16609 16539

17711 17628 18868 18770 20085 19969 21365 21230 22712 22555

24132 23948 25628 25416 27210 26962 28899 28600 30710 30348

16

15.000 15.250


TABLE 13.lb-PIPE TAPS-BASIC ORIFICE FACTORS, F,

Orifice Diameter,

d Al

Internal Diameter of Pipe, d,, in.

2 3 4

2.626 2.900 3.068 3.152 3.438 (in.) 1.689 1.939 2.067 2.300

0.250 12.850 12.813 12.800 12.782 0.375 29.359 29.097 29.005 28.882 0.500 53.703 52.816 52.481 52.019 0.625 87.212 84.919 84.063 82.922 0.750 132.23 126.86 124.99 122.45

0.875 192.74 181.02 177.08 171.92 1.000 275.45 251.10 243.27 233.30 1.125 391.93 342.98 327.98 30943 1.250 - 465.99 437.99 404.52 1.375 - - 583.96 524.68

1.500 - 679.10 1.625 1.750 - - - -

1.875 - - - -

2.000 - - -

2.125 - - -

2.250 - - - -

2.375 - - -

12.765 28.771 51.591 81.795 120.06

167.23 224.56 293.79 377.36 478.68

602.45 755.34 946.99

12.753 26.710

12.748 28.682 51.243 80.835 118.00

163.31 217.52 281.66 357.12 445.74

549.94 672.95 819.05

- -

12.745 28.669 51.196 80.703 117.70

162.76 216.55 280.02 354.45 44148

543.31 662.81 803.77 971.19 1171.8

1415.0 -

51.353 61.142

118.67

164.58 219.76 285.48 363.41 455.82

565.79 697.43 856.37 1050.4 1290.7

- -

- -

993.98 1205.6

1465.1

- -

12.737 28.634 51.064 80.332 116.86

161.17 213.79 275.42 347.03 429.83

525.40 635.76 763.51 911.98 1085.5

1289.7 1532.0 1822.8

Reynolds-Number Factor F,. This takes into account the variation of the discharge coefficient with Reynolds number. In gas measurement, the variation is slight and is often ignored in production operations. Values are shown in Tables 13. lg and 13. lh. It has been assumed in preparation of these tables that gas viscosity is substantially constant. The constant b shown in the tables is then primarily a function of pipe diameter, orifice diameter, and the location of the differential-pressure taps.

Expansion Factor Y. This factor accounts for the change in gas density as the pressure changes across the orifice. Inasmuch as the differential involved is usually small, this correction is small and often ignored. The value used depends on which one of the differential- pressure taps is used to measure static pressure and the location of the tap. The additional primary variables involved are (1) Fd, (2) ratio of differential pressure to absolute pressure, and (3) the specific-heat ratio C,,/C,, . In the standard table the last variable is taken as constant and equal to 1.3. This factor is shown in Tables 13. li through 13. lm.

Supercompressibility Factor F,,” . The variation from the ideal-gas laws by an actual gas is the function of this factor. The factor may be measured experimentally or determined by detailed methods outlined in AGA Com- mittee Report 3. 2 The correction is usually small and is often ignored. It may be estimated from the equation Fpr =( I/zJ~.~, where z is equal to the compressibility factor obtained from standard correlations.

Manometer Factor F,. This is used only with mercury-type meters to correct for the slight error in measurement caused by having different heads of gas above the two legs of the manometer. For all practical purposes it is insignificant. Table 13. In lists this factor vs. gas specific gravity.

Gauge Location Factor, Ft. This corrects the manometer factor, F, , for elevation and latitudes other than sea level and 45” latitude (see Table 13. lo).

Thermal Expansion Factor, F,. This corrects for expansion or contraction of the orifice opening when operating at temperatures substantially different from that at which the orifice was made. Normally, it is applied only at temperatures above 120°F or below 0°F.

Example Problem 1. A 0.7-specific-gravity natural gas is measured in an orifice meter having flange taps. Determine the flow rate in scfihr at 14.4 psia and 60°F if the following data apply:

h,,, = 40 in. H,O,

pf = 143 psig (measured

downstream), avg flowing temp = 84”F,

line size = 4.026 in.. and orifice-plate opening = 1.50 in.

Solution.

The respective factors are

Fb = 460.19, Fpb = 1.0229,

Ftb = 1.0000,

F, = 1.1952, Fd = 0.9777, F, = 1.0004,

Y = 1.0016, and

C’ = 551.89 (the product of the above numbers).

Then, qR =551.89 [(40)(143+14.73)]“-5=43,810 scf/hr.

As a practical matter, the accuracy of the meter device itself does not justify the number of significant figures shown in the charts. Consequently, in most production operations and engineering calculations the values of the coefficients may be rounded off and sometimes ignored with no practical loss of accuracy.

(continued on Page 36)


TABLE 13.1 b-PIPE TAPS-BASIC ORIFICE FACTORS, F, (continued)

Orifice Diameter,

;i.\

0.250 0.375 0.500 0.625 0.750

0.875 1.000 1.125 1.250 1.375

1.500 1.625 1.750 1.875 2.000

2.125 2.250 2.375 2.500 2.625

2.750 2.875 3.000 3.125 3.250

3.375 3.500 3.625 3.750 3.875

4.000 4.250 4.500 4.750 5.000

5.250 5.500

4

Internal Diameter of Pipe, dj, in.

6 8 3.826 4.026

12.727 12.722 28.598 28.584 50.936 50.886 79.974 79.835 116.05 115.73

159.57 158.94 211.03 209.91 270.90 269.10 339.87 337.05 418.79 414.51

508.76 502.38 611.11 601.80 727.54 714.16 860.17 841.19

1011.7 985.04

1185.3 1148.4 1385.4 1334.4 1617.2 1547.3 1887.6 1792.3 2206.0 2075.9

- 2407.0 - - - - - - - -

- - - - -

- - - - -

- -

- - - - -

- - - - -

- -

4.897 5.189 5.761 6.065 7.625 7.981 6.071 - - - - - - - -

50.739 79.436 114.81

157.11 206.62 263.71 328.73 402.06

484.20 575.73 677.38 789.99 914.57

1052.3 1204.7 1373.4 1560.5 1768.3

1999.8 2258.5 2548.6 2875.2 3244.8

3665.6 -

- - -

- - -

- - - - -

- -

50.705 79.349 114.61

156.71 205.91 262.51 326.85 399.30

480.23 570.14 669.63 779.40 900.28

1033.2 1179.4 1340.2 1517.2 1712.3

1927.6 2165.9 2430.2 2724.4 3052.8

3420.9 3835.7 4305.7

-

50.652 50.628 79.217 79.162 114.32 114.20

156.13 155.89 204.84 204.41 260.71 259.98 324.02 322.86 395.08 393.33

474.20 471.69 561.73 558.24 658.08 653.33 763.77 757.39 879.38 870.93

1005.6 994.52 1143.2 1128.8 1293.1 1274.6 1456.4 1432.7 1634.3 1604.3

1828.3 1790.3 2039.9 1992.2 2271.2 2211.6 2524.3 2450.1 2801.8 2709.9

3106.9 2993.3 3443.0 3303.0 3814.4 3842.3 4226.3 4014.8 4684.9 4425.1

5197.7 4878.4

- - - - -

- - -

-

- - - - -

- -

- - - -

- -

- - - -

- -

155.10 203.00 257.82 319.10 387.82

463.39 546.61 637.51 736.34 843.34

958.78 1083.0 1216.3 1359.2 1512.0

1675.4 1849.9 2036.0 2234.7 2446.5

2872.5 2913.7 3171.1 3446.0 3739.9

4054.2 4751.4 5554.7 6485.3 7571.4

8850.3 -

154.99 154.96 202.80 202.75 257.28 257.20 318.56 318.44 386.81 386.62

462.19 461.92 544.92 544.53 635.19 634.65 733.23 732.52 839.29 838.35

953.58 952.38 1076.4 1074.9 1208.0 1206.1 1348.8 1348.5 1499.2 1496.3

1659.7 1656.1 1830.6 1826.3 2012.7 2007.3 2206.4 2199.9 2412.4 2404.7

2631.6 2622.3 2864.7 2853.7 3112.7 3099.6 3376.6 3381.0 3657.6 3639.2

3957.0 3935.2 4616.6 4586.6 5369.0 5327.9 6231.1 6175.2 7224.3 7148.7

8376.3 8274.0 9723.8 9585.1


Orlflce

Diameler.

fli.;

1.000 1.125 1.250 1.375

1500

1.625

1.750 1.675 2.000 2.125

2.250 2.375 2.500 2 625 2.750

2.875 3.000 3125 3.250 3.375

3.500 3.625 3750 3.875 4000

4250 4500 4 750 5 000 5.250

5500 5.750 6.000 6.250

6.500

6.750

7 000 7.250 7.500 7.750

8.000 8.250 8.500 8 750

9 000

9250

9 500 9750

10000 10250

10500

TABLE 13.lb-PIPE TAPS-BASIC ORIFICE FACTORS, F, (continued)

lnlernal Diameter of Pipe. d,. in.

9.564

202.16 256.22

316.90 384 29

45852

539.72 628.03 723.61

826.63 937.28

1055.7 1182.2 1316.9 14600 1611.8

1772.5 1942.5 2122.1 2311.6 25115

2722.3 2944.3 3178.1 3424.3 36835

4243.6

4865.1 5554.9 6322.2 7177.7

a1341 9207 0

10415

11783 13340

-

10

10.020 10.136

12

11.376 11.938 12090

16

14.688 15.000 15.250

- 255 96

316 49 383.66

457 63

538 45 626.29 721 27

62354 93327

1050.6 1175.8 1309.0 1450.3 1600.1

17584 19256 2102.0 2287.0 24834

2689.1 2905.5 3132.7 3371.5 3622.1

4161.6

47561 5411 5

6134.9 69344

7820.0 8803.1 98978

11121 12492

14038 15790

-

- - - 256.01

31656 383 79

45779

538 69 626.61 721.70

82412 934.02

1051.6 1177.0 1310.5 1452.1 1602.3

1761.0 19288 2105.7 2292.2 2486.6

2695.3 2912.7 3141.2

33613 3633.5

4176.8

4776 2 5437.9

61692 6978.9

7077.2

6876.3 9991 2

11240

12644

14230 16035

- 315.84 38266 456.16

536 38 623.44 71743

61648 926.72

1042.3 11653 1295.9 1434.3 15807

1735.1 18978 2069.0 2248.9 24377

2635.6

2843.0 3060.2 3287.4

3524.9

4032.8 4587.1 5191.5

58506 65694

73541 8211 4 91495

10178

11307

12550 13923 15442 17131

19017

315.57 31551 362.30 382 22 455.64 455.52

535.66 535.46 62245 622.20 716.10 715.78

816 73 816.30 924 44 923.88

1039.4 1038.7 1161.6 1160.7 1291.4 1290.2

1428.7 1427.4 1573.9 1572.2

1726.9 1724.9 18881 1885.7 2057.5 2054.7 2235.4 2232.1 2421.8 2418.0

2617.2 2612.6 2821.6 2816.3 3035.3 3029.3 3258.7 3251.7

3492.0 3483.9

3989.5 3979.0 4530.8 45172 5119.0 5101 5

5757.8 5735.4 6451.5 6423.2

72051 7169.5 8024.2 7979.6 8915.4 8859.8

9886.1 9817.2

10945 10860

12103 11996 13371 13242 14762 14604 16294 16101 17986 17750

19861 19572 21947 21593

- - - - -

- - - -

-

- -

-

- -

-

- -

- - -

-

- - -

- - -

- -

- -

-

- -

-

-

- - -

453.92 533.27 619.18 711.73

810.99 917.01

1029.9 1149.7 1276.5 1410.5 1551 7

17001 18561 2019.5 2190 7 2369.6

2556.5 2751.4 2954.5

3165.9 3385.6

3851.6 43534 4892.9 5471.9

60925

6757.0 7468.0 8228.5 9041.6 9911.2

10841 11837 12902 14044

15268

16583 17996 19517 21156

22926

24841

26917 29172 31629 34315

-

45378

53307 61892 71139

81053 916.43

1092.6 1148.8 1275.4 1409.1 15499

1698.1 1853.6 2016.6 2187.2 2365.5

2551.7

2745.9 2948.1 3158.6

3377.5

3840.9 4339.8 4875.8

5450.5 6065.9

67241 7427.6 81792 6981 7 9838.7

10754 11732 12777 13894 15090

16371 17746

19221 20807

22515

24356

26346 28501 30839 33383 36160

532.93 618.73 711 13

810.19 915.99

1028.6 1148.1 1274.5

1408.0 1548.6

1696.5 1851.7 2014.3

2184.5 23624

2548.1 2741.7 2943.3 3153.1

3371.2

3832.8 4329.6 4862.9

5434.3 6045 9

66994 73974 81423

89370 97847

10689 11654 12684 13783 14959

16216 17561

19003 20551

22214

24003 25932 26014 30268 32713

35372

GAS MEASUREMENT AND REGULATION 13-11

TABLE 13.1 b-PIPE TAPS-BASIC ORIFICE FACTORS, F, (continued)

Internal Dtameier of Pipe. d,, in Orifice

Diameter, d (ii

2.000 2.125

2.250 2.375

2500

2.625 2.750 2.675 3.000 3.125

3.250 3.375 3.500

3625 3750

3.875 4.000 4.250

4.500 4.750

5.000 5.250 5.500 5.750 6 000

6.250 6.500 6.750

7.000 7.250

7.500 7.750 8.000 8.250 8.500

8.750

9 000 9.250 9.500 9.750

10.000 10.250 10.500 10.750

11.000

11.250

11.500 11 750 12 000 12.500

13.000 13 500 14.000 14.500

15.000

15500

16.000 16 500 17.000 17.500

18000 18 500 19000 19.500 20 000

20

19.000 19.250 18.814

806.71 911.51

1022.9 11410

1265.7

1397.2

1535.5 1680.7 1832.7 1991.8

21580 2331.3 25119

2699 7 2895 0

30977 3308.0 3751.6 42268 4734.1

5274.6 5849 0 64566 7104.4 77879

8510.4

92734 10079

10926 11823

12767

13762 14810 15914

17073

18305

19598 20963 22402 23923

25529 27227 29023 30925

32940

35078

37348 39761 42330 47991

54463

-

-

-

806.57 806.40 911.35 911.13

1022.7 1022.4 11407 1140.4

12654 1265.0

13968 13963 1535.0 1534.4

16801 1679.3

18321 1831.2 1991.0 1990.0

2157.0 2155.8 2330.2 23287 2510.6 2508.8 2698.2 2696 2 2893.2 2890 9

3095.7 3093.0 3305.7 33027 3748.7 3744.8

42230 42181 4729.4 4723.3

5268.7 5261.1 5841.9 5832.6 6449.9 6438.7

7094.0 7080.4

7775.4 7759.1

8495.4 84760

9255.6 9232.5 10058 10030

10903 10871 11794 11756

12733 12689 13722 13670 14763 14703

15860 15791

17015 16935

18232 18139

19515 19408 20866 20743 22292 22151

23796 23634

25384 25198 27061 26649

28834 28592 30709 30434

32694 32381

34798 34443

37030 36626

39400 38941 41920 41399 47461 46790

53778 52914

22.626

24 30

23.000 23.250 28.628 29.000 29.250

1136.8

1260.6

13909 1527.9 1671.5

1821.9 1978.9

2142.8 23135 2491 2

26758 28674

30660 32718 3705.0 41676 46600

5163.0 5737.1 6322.9 6941.3 7592.8

8278.3 8998.8 9755.0

10548 11379

12249 13160 14113

15109 16150

17237

18373 19560

20800 22094

23447 24859 26335 27878 29490

31175

32938 34783 36714 40855

45406 50420 55959 62099

68929

76562

- -

11365

1260.2

1390.5 1527.3

1670.9 1821.1 1978.0

2141.7 2312.3 2489.7 2674.0 2865.4

3063.8 3269.2 3701.7 4163.4 46548

5176.4 5729.1 6313.2 6929.7 7579.0

82620 8979 5 97324

10522 11348

12214 13119 14065

15054 16087

17166 16292 19468 20695

21976

23312 24708 26164 27685 29273

30932

32666 34478 36373 40429

44877 49763

55147 61094

67607

75025

83231

-

- -

11363 12599

13902 1527.0 1670.4

1820.6 1977.4

2141.0 2311.5 24888 2673.0 2864.1

3062.3 32676 3699.6 4160.7 46514

5172.3

5723.9 63070 6922 2 7570 1

8251 5 8967.1 9717.9

10505 11329

12191 13093 14035 15020 16048

17121 18241 19409

20628 21900

23227 24612

26056 27563 29136

30779

32494 34285 36158 40161

44544

49352 54638 60468

66915

74074

82055

- - - -

1663.8 1812.7 1968.1

2130.2 2298.8 2474.1

2656.0 2844.6

3040.0 32422 3666.9 4119.3 45996

51082 56454 6211 8 68077

74336

BOB99 87772 9496 0

10247 11030

11847 12697 13582 14501 15457

16450 17480 18548

19656 20805

21995 23228 24505 25827 27196

28613

30080 31598 33169 36478

40024 43823

47890 52271

56967

62017

67453 73314 79641 86485

93900 101950 110720

120300 130780

-

-

-

1812.3 19677

2129.6 2298.2 2473.3

2655.2 2843.7

3038 9 3240.9 3665.3 4117.3 45971

5105.0 5641.6 6207.2 6802.1 7426.9

8082.0 8768.0 9485.2

10234 11016

11830 12678 13560 14477

15429

16418 17444 18508 19611 20754

21938 23165 24434 25749 27109

28516 29972 31479 33038 36318

39829 43589 47615 51932

56562

61583

66878 72630 78831 85525

92765 100610

109130 118420 128560

- - -

-

- 1812.0 1967.4

21293 2297.7 2472.9

26546 2843.0

30382 32401 3664.3 41160 45954

5103.0 56391 62042 67985 74226

80769 87619 9478 1

10226 11006

11819 12665 13546 14461

15411

16397 17421 18482 19582 20721

21901 23124

24388 25698 27052

28453

29903 31402 32953 36215

39704 43437 47433 51714

56301

61223

66509 72193 78313 84913

92042 99758

108130

117230 127150


TABLE 13.lc-PRESSURE-BASE FACTORS, Fp,

F,, = 14.73 + base pressure, psia

Factor Pressure base, psia F, Pressure base, psia

14.4 1.0229 15.2 (8 oz. above 14.7). 14.65 (4 oz. above 14.4). 1.0055 15.325 (IO oz. above 14.7). 14.73.. . 1 .oooo 15.4 (1 psi above 14.4). 14.9 (8 oz. above 14.4). 0.9886 15.7 (1 psi above 14.7) 14.95 (4 oz. above 14.7). 0.9853 16.4 (2 psi above 14.4), 15.025 (10 oz. above 14.4). 0.9804 16.7 (2 psi above 14.7).

Factor

F, 0.9691 0.9612 0.9565 0 9382 0.8982 0.8820

TABLE 13.lci-TEMPERATURE-BASE FACTORS, F,

‘=, = 460 + temperature base OF

520

Temoerature Factor TemDerature Factor TemDerature Factor bake, OF F

Ih bake, OF F

--LL bake, OF F

rb 45 0.9712 65 1.0096 a5 1.0481 50 0.9808 70 1.0192 90 1.0577

55 0.9904 75 1.0288 95 1.0673

60 1 .oooo 80 1.0385 100 1.0769

GASMEASUREMENTANDREGULATION

TABLE 13.1e--SPECIFIC-GRAVITY FACTOR&f,

13-13

Specific gravity

YQ

0.500 0.505 0.510 0.515 0.520

0.525 0.530 0.535 0.540 0.545

0.550 0.555 0.560 0.565 0.570

0.575 0.580 0.585 0.590 0.595

0.600 0.605 0.610 0.615 0.620

0.625 0.630 0.635 0.640 0.645

0.650 0.655 0.660 0.665 0.670

11 12 13 14 15

16 17 18 19 20

Factor OF

1.0621 1.0609 1.0598 1.0586 1.0575

1.0564 1.0552 1.0541 1.0530 1.0518

1.0507 1.0496 1.0485 1.0474 1.0463

1.0452 1.0441 1.0430 1.0419 1.0408

21 22 23 24 25

26 27 28 29 30

31 32 33 34 35

36 37 38 39 40

Factor f

D 1.4142 1.4072 1.4003 1.3935 1.3868

1.3801 1.3736 1.3672 1.3608 1.3546

1.3484 1.3423 1.3363 1.3304 1.3245

1.3188 1.3131 1.3074 1.3019 1.2964

1.2910 1.2856 1.2804 1.2752 1.2700

1.2649 1.2599 1.2549 1.2500 1.2451

1.2403 1.2356 1.2309 1.2263 1.2217

Specific gravity

YQ

0.675 0.680 0.685 0.690 0.695

0.700 0.705 0.710 0.715 0.720

0.725 0.730 0.735 0.740 0.745

0.750 0.755 0.760 0.765 0.770

0.775 0.780 0.785 0.790 0.795

0.800 0.805 0.810 0.815 0.820

0.825 0.830 0.835 0.840 0.845

d 1.0000 f,= - YQ

Factor F

L 1.2172 1.2127 1.2082 1.2039 1.1995

1.1952 1.1910 1.1868 1.1826 1.1785

1.1744 1.1704 1.1664 1.1625 1.1586

1.1547 1.1509 1.1471 1.1433 1.1396

1.1359 1.1323 1.1287 1.1251 1.1215

1.1180 1.1146 1.1111 1.1077 1.1043

1.1010 1.0976 1.0944 1.0911 1.0879

Specific gravity

.--AL 0.850 0.855 0.860 0.865 0.870

0.875 0.880 0.885 0.890 0.895

0.900 0.905 0.910 0.915 0.920

0.925 0.930 0.935 0.940 0.945

0.950 0.955 0.960 0.965 0.970

0.975 0.980 0.985 0.990 0.995

1.00 1.01 1.02 1.03 1.04

Factor F

9 1.0847 1.0815 1.0783 1.0752 1.0721

1.0690 1.0660 1.0630 1.0600 1.0570

1.0541 1.0512 1.0483 1.0454 1.0426

1.0398 1.0370 1.0342 1.0314 1.0287

1.0260 1.0233 1.0206 1.0180 1.0153

1.0127 1.0102 1.0076 1.0050 1.0025

1.0000 0.9950 0.9901 0.9853 0.9806

Specific gravity

Yg 1.05 1.06 1.07 1.08 1.09

1.10 1.11 1.12 1.13 1.14

1.15 1.16 1.17 1.18 1.19

1.20 1.21 1.22 1.23 1.24

I.25 1.26 1.27 1.28 1.29

1.30 1.31 1.32 1.33 1.34

1.35 1.36 1.37 1.38 1.39

TABLEl3.lf-FLOWING-TEMPERATUREFACTORS,F,

520 F,=

460 +actual flowing temperature

Factor OF Factor "F Factor OF __~---- 1.0398 41 1.0188 61 0.9990 81 1.0387 42 1.0178 62 0.9981 1.0376 43 1.0168 63 0.9971 1.0365 44 1.0157 64 0.9962 1.0355 45 1.0147 65 0.9952

1.0344 46 1.0137 66 0.9943 1.0333 47 1.0127 67 0.9933 1.0323 48 1.0117 68 0.9924 1.0312 49 1.0107 69 0.9915 1.0302 50 1.0098 70 0.9905

1.0291 51 1.0088 71 0.9896 1.0281 52 1.0078 72 0.9887 1.0270 53 1.0068 73 0.9877 1.0260 54 1.0058 74 0.9868 1.0249 55 1.0048 75 0.9859

1.0239 56 1.0039 76 0.9850 1.0229 57 1.0029 77 0.9840 1.0219 58 1.0019 78 0.9831 1.0208 59 1.0010 79 0.9822 1.0198 60 1.0000 80 0.9813

82 83 84 85

86 87 88

il:

91 92 93 94 95

t ;

2 100

Factor

0.9804 0.9795 0.9786 0.9777 0.9768

0.9759 0.9750 0.9741 0.9732 0.9723

0.9715 0.9706 0.9697 0.9688 0.9680

0.9671 0.9662 0.9653 0.9645 0.9636

Factor F

9 0.9759 0.9713 0.9667 0.9623 0.9578

0.9535 0.9492 0.9449 0.9407 0.9366

0.9325 0.9285 0.9245 0.9206 0.9167

0.9129 0.9091 0.9054 0.9017 0.8980

0.8944 0.8909 0.8874 0.8839 0.8805

0.8771 0.8737 0.8704 0.8671 0.8639

0.8607 0.8575 0.8544 0.8513 0.8482

OF Factor

E-O.9551 120 0.9469 130 0.9388 140 0.9309 150 0.9233

160 0.9158 170 0.9085 180 0.9014 190 0.8944 200 0.8876

210 0.8810 220 0.8745 230 0.8681 240 0.8619 250 0.8558

260 0.8498 270 0.8440 280 0.8383 290 0.8327 300 0.8272

13-14

Orifice Diameter,

d (inD.i

0.250 0.375 0.500 0.625 0.750

0.875 1.000 1.125 1.250 1.375

1.500 1.625 1.750 1.875 2.000

2.125 2.250 2.375 2.500 2.625

2.750 2.875 3.000 3.125 3.250

3.375 3.500 3.625 3.750 3.875

4.000 4.250 4.500 4.750 5.000

5.250 5.500 5.750 6.000 6.250

6.500 6.750 7.000 7.250 7.500

7.750 8.000 8.250 8.500 8.750

9.000 9.250 9.500 9.750 10.000

10.250 10.500 10.750 11.000 11.250


TABLE 13.19-b VALUES FOR REYNOLDS NUMBER FACTOR, f,, FLANGE TAPS

b

Fr=‘+ Jh,p,

Internal Diameter of Pipe, d,, in

1.689

0.0879 0.0677 0.0562 0.0520 0.0536

0.0595 0.0677 0.0762 0.0824

-

-

-

-

- -

-

-

-

- - -

-

- - - - -

- - - - -

- - - - -

- - - - -

- - - - -

- - - - -

2

1.939

0.0911 0.0709 0.0576 0.0505 0.0485

0.0506 0.0559 0.0630 0.0707 0.0772

-

-

-

- - - - -

-

-

- - - - -

- - - - -

- - - - -

- - - - -

- -

- -

- - - - -

- - - - -

- 3 4

2.067

0.0926 0.0726 0.0588 0.0506 0.0471

0.0478 0.0515 0.0574 0.0646 0.0715

0.0773

-

-

- - - - -

-

-

- - - - -

- - -

-

- - - - -

- - - - -

- - - - -

- - - - -

- - - - -

2.300

0.0950 0.0755 0.0612 0.0516 0.0462

0.0445 0.0458 0.0495 0.0550 0.0614

0.0679 0.0735

- - -

- - - - -

- - - - -

- - - - -

- - - - -

- - - - -

- - - - -

- - - - -

- - - - -

- - - - -

2.626 -

0.0792 0.0648 0.0541 0.0470

0.0429 0.0416 0.0427 0.0456 0.0501

0.0554 0.0613 0.0669 0.0717

-

- - - - -

- - - - -

- - -

-

- - - - -

- - - - -

- - - - -

- - - - -

- - - - -

- - - - -

2.900

0.0820 0.0677 0.0566 0.0466 0.0433

0.0403 0.0396 0.0408 0.0435 0.0474

0.0522 0.0575 0.0628 0.0676 0.0715

- - - - -

- - - - -

- - -

-

- - - - -

- - - - -

- - - - -

- - - - -

- - - - -

- - - - -

3.068

0.0836 0.0695 0.0583 0.0498 0.0438

0.0402 0.0386 0.0388 0.0406 0.0436

0.0477 0.0524 0.0574 0.0624 0.0669

0.0706 - - - -

- - - -

- - - - -

- - - - -

- - - - -

- - - - -

-

- -

- - - - -

- - - - -

3.152 3.438

0.0844 -

0.0703 0.0867 0.0591 0.0728 0.0504 0.0618 0.0442 0.0528

0.0403 0.0460 0.0383 0.0411 0.0381 0.0380 0.0394 0.0365 0.0420 0.0365

0.0457 0.0378 0.0500 0.0402 0.0549 0.0434 0.0598 0.0473 0.0642 0.0517

0.0685 0.0563 - 0.0607 - 0.0648 - 0.0683 - -

- - - - - - - - - -

- - - - -

- - - - -

- - - - -

- - - - -

- -

- -

- - - - -

- - -

-

-

-

- - - - -

-

-

-

- - - - -

- - - -

- - - - -

- - - - -

3.826

- -

0.0763 0.0653 0.0561

0.0487 0.0430 0.0388 0.0361 0.0347

0.0345 0.0354 0.0372 0.0398 0.0430

0.0467 0.0507 0.0548 0.0589 0.0626

0.0659 -

- -

- -

-

- - - - -

- - - - -

- - - - -

- - - - -

- - - - -

- ( -

- - -

4.026

0.0779 0.0670 0.0578 0.0502 0.0442

0.0396 0.0364 0.0344 0.0336 0.0338

0.0350 0.0370 0.0395 0.0427 0.0462

0.0501 0.0540 0.0579 0.0615 0.0647

- - - - -

- - - - -

- - - - -

- - - - -

- - - - -

- - - - -

- - - - -

- - - - -

13-15 GASMEASUREMENTANDREGULATION

TABLE 13.19-b VALUES FOR REYNOLDS NUMBER FACTOR, F,, FLANGE TAPS (continued)

b F,=l+-

Jh,p,

Orifice Diameter,

d “3 6


8 10

(in.) 4.897 5.189 5.761 6.065 7.625 7.981 8.071 9.564 10.020 10.136 - - - - - -

0.0753 0.0785 0.0665 0.0701

0.0587 0.0625 0.0520 0.0557 0.0462 0.0498 0.0414 0.0447 0.0375 0.0403

0.0344 0.0367 0.0322 0.0337 0.0306 0.0314 0.0298 0.0298 0.0296 0.0287

0.0300 0.0281 0.0310 0.0281 0.0324 0.0286 0.0342 0.0295 0.0365 0.0308

0.0391 0.0324 0.0418 0.0343 0.0448 0.0366 0.0479 0.0389 0.0510 0.0416

0.0541 0.0443 0.0569 0.0472 0.0597 0.0500 0.0621 0.0527 0.0640 0.0553

- 0.0578 - 0.0620 - - - - - -

-

-

-

- - - - -

- - - -

- - - - -

- - - - -

-

-

-

- - - - -

- - - -

- - - - -

- - - - -

0.250 0.375 0.500 0.625 0.750

0.875 1.000 1.125 1.250 1.375

1.500 1.625 1.750 1.875 2.000

2.125 2.250 2.375 2.500 2.625

2.750 2.875 3.000 3.125 3.250

3.375 3.500 3.625 3.750 3.875

4.000 4.250 4.500 4.750 5.000

5.250 5.500 5.750 6.000 6.250

6.500 6.750 7.000 7.250 7.500

7.750 8.000 0.250 8.500 8.750

9.000 9.250 9.500 9.750 10.000

10.250 10.500 10.750 11.000 11.250

- -

0.0836 0.0734 0.0645

0.0567 0.0500 0.0444 0.0399 0.0363

0.0336 0.0318 0.0307 0.0305 0.0308

0.0318 0.0334 0.0354 0.0378 0.0406

0.0436 0.0468 0.0500 0.0533 0.0564

0.0594 0.0620 0.0643

-

- - - - -

-

-

-

- - - - -

- - - - -

- - - - -

- - - - -

- - -

0.0801 0.0718

0.0643 0.0576 0.0517 0.0464 0.0419

0.0381 0.0348 0.0322 0.0303 0.0288

0.0278 0.0274 0.0274 0.0279 0.0287

0.0300 0.0314 0.0332 0.0353 0.0375

0.0400 0.0426 0.0452 0.0479 0.0505

0.0531 0.0579 0.0618

- -

-

-

-

- - -

-

- -

-

- - - - -

- - -

-

- - - - -

0.0723 0.0660 0.0602 0.0549 0.0501

0.0457 0.0418 0.0383 0.0353 0.0327

0.0304 0.0286 0.0271 0.0259 0.0251

0.0246 0.0244 0.0245 0.0248 0.0254

0.0263 0.0273 0.0286 0.0300 0.0316

0.0334 0.0372 0.0414 0.0457 0.0500

0.0539 0.0574

-

-

- -

- - - - - - - - - -

0.0738 0.0676 0.0619 0.0566 0.0518

0.0474 0.0435 0.0399 0.0366 0.0340

0.0315 0.0295 0.0278 0.0264 0.0253

0.0245 0.0240 0.0238 0.0239 0.0242

0.0248 0.0255 0.0265 0.0274 0.0289

0.0304 0.0338 0.0386 0.0418 0.0457

0.0497 0.0535 0.0569

- - - - - -

0.0742 0.0680 0.0623 0.0571 0.0523

0.0479 0.0439 0.0403 0.0371 0.0343

0.0318 0.0297 0.0280 0.0265 0.0254

0.0245 0.0240 0.0237 0.0237 0.0240

0.0244 0.0251 0.0260 0.0271 0.0283

0.0297 0.0330 0.0366 0.0405 0.0446

0.0487 0.0524 0.0559 0.0588

-

-

0.0738 0.0685 0.0635 0.0588

0.0545 0.0504 0.0467 0.0433 0.0401

0.0372 0.0346 0.0322 0.0302 0.0283

0.0267 0.0254 0.0243 0.0234 0.0226

0.0221 0.0219 0.0218 0.0218 0.0221

0.0225 0.0238 0.0256 0.0279 0.0307

0.0337 0.0370 0.0404 0.0438 0.0473

0.0505 0.0536 0.0562

-

-

- - 0.0701 0.0705 0.0652 0.0656 0.0606 0.0610

0.0563 0.0568 0.0523 0.0527 0.0485 0.0490 0.0451 0.0455 0.0419 0.0414

0.0389 0.0383 0.0362 0.0356 0.0337 0.0330 0.0315 0.0308 0.0296 0.0287

0.0278 0.0269 0.0263 0.0253 0.0250 0.0252 0.0239 0.0241 0.0230 0.0231

0.0223 0.0224 0.0218 0.0218 0.0214 0.0214 0.0213 0.0212 0.0213 0.0211

0.0214 0.0212 0.0222 0.0219 0.0236 0.0231 0.0254 0.0249 0.0277 0.0270

0.0303 0.0295 0.0332 0.0323 0.0363 0.0354 0.0396 0.0386 0.0437 0.0418

0.0462 0.0451 0.0493 0.0483 0.0523 0.0513 0.0550 0.0540 0.0572 0.0564

- -

- - - - -

- - - - - - - - - - - - - - - - -

- - - - - - - - - -

- - - -

- - - - -

- - - - - - -

- - - - - - -

- - - - - -

- -

13-16 PETROLEUM ENGINEERINGHANDBOOK

TABLE13.lg-bVALUESFORREYNOLDSNUMBER FACTOR,F,, FLANGETAPS(continued)

Internal Diameter of Pipe, d,

12

11.938 12.090 14.688

- -

- - -

, in.

16

15.000

-

-

d (in0.i

0.250 0.375 0.500 0.625 0.750

0.875 1.000 1.125 1.250 1.375

1.500 1.625 1.750 1.875 2.000

2.125 2.250 2.375 2.500 2.625

2.750 2.875 3.000 3.125 3.250

3.375 3.500 3.625 3.750 3.875

4.000 4.250 4.500 4.750 5.000

5.250 5.500 5.750 6.000 6.250

6.500 6.750 7.000 7.250 7.500

7.750 8.000 8.250 8.500 8.750

9.000 9.250 9.500 9.750 10.000

10.250 10.500 10.750 11.000

11.376 15.250

-

- -

0.0698 0.0654

0.0612 0.0573 0.0536 0.0501 0.0469

0.0438 0.0410 0.0383 0.0359 0.0336

0.0316 0.0297 0.0278 0.0264 0.0251

0.0239 0.0229 0.0221 0.0214 0.0208

0.0204 0.0200 0.0201 0.0207 0.0217

0.0231 0.0249 0.0270 0.0294 0.0320

0.0347 0.0376 0.0406 0.0435 0.0463

0.0491 0.0517 0.0540 0.0560

-

-

- - - - -

0.0714 0.0671

0.0631 0.0592 0.0555 0.0521 0.0488

0.0458 0.0429 0.0402 0.0377 0.0354

0.0332 0.0312 0.0294 0.0278 0.0263

0.0250 0.0238 0.0228 0.0219 0.0212

0.0206 0.0198 0.0195 0.0196 0.0202

0.0212 0.0226 0.0243 0.0263 0.0285

0.0309 0.0335 0.0362 0.0390 0.0418

0.0446 0.0473 0.0498 0.0522 0.0543

- - - -

- 0.0718 0.0676

0.0635 0.0597 0.0560 0.0526 0.0492

0.0463 0.0434 0.0407 0.0382 0.0358

0.0336 0.0317 0.0298 0.0282 0.0266

0.0253 0.0241 0.0230 0.0221 0.0213

0.0207 0.0198 0.0194 0.0194 0.0199

0.0208 0.0221 0.0237 0.0255 0.0277

0.0300 0.0325 0.0351 0.0379 0.0407

0.0434 0.0461 0.0487 0.0511 0.0534

0.0553 - -

- - - - - - -

- - - - - -

-

- -

- -

- -

- - -

0.0706 0.0670 0.0636 0.0604 0.0572

0.0542 0.0514 0.0487 0.0461 0.0436

0.0413 0.0391 0.0370 0.0350 0.0331

0.0314 0.0298 0.0282 0.0268 0.0255

0.0243 0.0223 0.0206 0.0193 0.0184

0.0178 0.0176 0.0176 0.0180 0.0186

0.0195 0.0206 0.0220 0.0235 0.0252

0.0271 0.0291 0.0312 0.0334 0.0357

0.0380 0.0402 0.0425 0.0447 0.0469

0.0489 0.0508 0.0526 0.0541

0.0713 0.0678 0.0644 0.0612 0.0581

0.0551 0.0523 0.0496 0.0470 0.0445

0.0422 0.0399 0.0378 0.0358 0.0339

0.0321 0.0305 0.0290 0.0275 0.0262

0.0249 0.0228 0.0210 0.0196 0.0185

0.0178 0.0174 0.0174 0.0176 0.0180

0.01'88 0.0198 0.0210 0.0224 0.0240

0.0257 0.0276 0.0296 0.0317 0.0338

0.0361 0.0383 0.0406 0.0428 0.0449

0.0470 0.0490 0.0509 0.0526

0.0684 0.0650 0.0618 0.0587

0.0558 0.0529 0.0502 0.0476 0.0452

0.0428 0.0406 0.0385 0.0365 0.0346

0.0328 0.0311 0.0295 0.0281 0.0267

0.0254 0.0232 0.0213 0.0198 0.0187

0.0179 0.0174 0.0172 0.0173 0.0177

0.0183 0.0192 0.0202 0.0216 0.0230

0.0246 0.0264 0.0283 0.0303 0.0324

0.0346 0.0368 0.0390 0.0412 0.0434

0.0455 0.0475 0.0495 0.0513

11.250 - - - - 0.0541 0.0528


TABLE 13.lg-b VALUES FOR REYNOLDS NUMBER FACTOR, F,, FLANGE TAPS (continued)

13-17

F, b

=‘+=

Diameter of Pipe

24

22626 23000

Internal !, d,, in. OhX Diameler

-g-j

0 250 0375 0 500 0625 0 750

0875 1 000 1125 1250 1375

1 500 1625 1750 1875 2000

2 125 2.250 2 375 2 500 2625

2 750 2875 3000 3125 3 250

3 375 3 500 3 625 3 750 3 875

4000 4 250 4 500 4 750 5000

5 250 5 500 5 750 6 000 6 250

6 500 6 750 7 000 7250 7 500

7 750 a 000 a 250 8 500 8750

9 000 9 250 9500 9750 10000

10250 10500 10750 11000 11 250

11 500 11 750 12000 12500 13000

13500 14000 14500 15000 15500

16000 16500 17000 17500 18000

18500 19000 19500 20000 20500

21 000 21 500

20

19000

30

29 000 29 250 23250 28628 16814

-

- -

- - -

- -

-

00667

00640 00614 00586 00563 00540

00517 0 0494 00473 00452 00433

00414 0 0395 00378 00361 00345

0 0329 00301 00275 0 0252 00232

00214 0 0199 00186 00176 00167

00161 00157 00155 00155 00157

0 0160 00166 00172 00180 0 0190

00201 00213 00226 00240 00256

00271 0 0288 00305 00322 00340

00358 00376 0 0394 0 0429 00463

0 0494 00520

19250

- -

- -

- - - -

- - - -

.- -

- - - -

- - - -

- -

- -

- - - - -

- - - -

0 0671

00644 00618 0 0592 00568 00544

00521 0 0499 00477 00457 00437

00418 0 0399 00382 00365 0 0349

0.0333 00304 0 0279 00256 00235

00217 00201 00188 00177 00168

00162 00157 00155 00154 00155

00158 00163 00169 00177 00186

00196 00208 00220 00234 0 0249

00264 00280 0 0297 00314 0 0332

0 0349 00367 00385 00420 00454

00485 00512

- 00676

- - -

00649 00622 0 0597 00573 00549

- - - - - -

0 0659 00665 0 0669 00636 0 0642 00646

- - - - -

00662 00644 00626 00608

0 0590

- - - - -

- - - -

0 0649 00652 00631 00634 0.0613 00616

0 0596 0 0599 0 0579 00582 00562 00566 00546 00550 00530 00534

00614 00620 00624

0 0592 0 0599 00603 00571 00578 00582 00551 00557 00562 00531 00538 00542 00511 00520 00523

0 0493 00500 00504 00474 00481 00486

00526 00504 00483 00462 00442

00423 00405 00387 00370 00354

00574 00557 00541

00457 00464 00468 00440 00447 00451 00423 00430

00407 00414 00376 00384 00348 00355 00322 00328 0 0297 00304

00275 00281 00254 00260 00236 00241 0 0219 00224 0 0204 00208

00191 00195 00179 00183 00169 00172 00161 00163 00154 00156

0 0148 00150 0 0144 00145 0 0142 00142 00141 00140 00141 00140

00143 00141 0 0146 00143 0 0150 00146 00155 00150 00161 00155

00168 0 0162 00176 00169 00185 00176 00194 00186 00205 00196

00216 00207 0 0228 00218 00241 00230 00267 00255 0 0296 00282

00326 00311 00356 00341 00386 00370 00415 00400 00443 00428

00470 00455 0 0494 00480

00503 -

00435

00419 00388 00360 00333 00308

00285 00264 00245 00228 00212

00198 00185 00174 00165 00157

00151 00146 00142 00140 00139

00140 00141 00144 0.0147 00152

00525

0 0509 0 0479 00450 00423 0 0397

00373 0 0349

00515 00518 0 0485 00488

0 0339 00310 00283 00260 0 0239

00220 00204 00191 00179 00170

00456 00460 0 0429 00433 00403 00407

00378 00382 00355 0 0359

00327 0 0333 00337 00306 00312 00316 00287 0 0292 0 0296

0 0269 00274 00277 00252 00257 00260 00236 00240 00244

00163 00157 00154 00153 00154

00156 00160 00165 00172 00180

00190 00201 00213 00226

00221 00208

0 0195 00184 00174 00164 00156

00149 00143 00138 00133 00130

00158 00128 00164 00126

00226 0 0229 00212 00215

00199 00202 00187 00190 00177 00179 00168 00170 0 0159 00161

00152 00153 00145 00146 00139 00141 00135 00136 00131 00132

00128 00128 00126 00126 00125 00125 00124 00124 00125 00124

00126 00125 00128 00127 00131 00129 00138 00136 00148 00145

00125 00125 00126

00128 00130 00134 00142 00153

00166 00160 00157 00182 00175 00171

00240

00255 00270 00286 00303 00320

00338 00355 00373 00408 00442

0 0474 00502

-

00172 00181 0 0190

00200 00211 00223 00248 00274

00302 00331 00360 0 0390 00418

00446 00471 0 0494

-

00199 00218 0 0239

00260 00283 00307

00192 00187 0 0209 00204 00230 00224

0 0250 00244 00273 00266 0 0296 002.38 00319 00312 00343 00335

00366 00358 00390 0 0382 00412 00404 00434 00426 00455 00448

00331 00355

0 0379 00402 00424 00446 00466

00485 00475 00467 .- - ~ 00492 00485


TABLE 13.lg--b VALUES FOR REYNOLDS NUMBER FACTOR, F,, FLANGE TAPS (continued)

b F,=l+-

Jh,p,

InteN rnal Diameter of Pipe, d,, in.

2 4

F, 0.10 0.11 0.12 0.13 0.14

0.15 0.16 0.17 0.18 0.19

0.20 0.21 0.22 0.23 0.24

0.25

1.689 1.939 2.067 2.300 3.152

0.1027 0.1012 0.0987 0.0985 0.0970 0.0945 0.0945 0.0930 0.0905 0.0907 0.0892 0.0867 0.0871 0.0856 0.0831

2.626 2.900 - 0.0958 0.0928 0.0916 0.0896 0.0876 0.0856 0.0638 0.0818 0.0802 0.0782

3.068

0.0927 0.0885 0.0845 0.0807 0.0771

3.438 3.826

0.0905 0.0886 0.0863 0.0844 0.0823 0.0804 0.0785 0.0766 0.0749 0.0730

4.026

0.1062 0.1020 0.0981 0.0943 0.0906

0.0872 0.0840

0.0921 0.0880 0.0840 0.0802 0.0765

0.0731 0.0699 0.0868 0.0639 0.0612

0.0587 0.0562 0.0540 0.0520 0.0501

0.0483

0.0877 0.0835 0.0795 0.0757 0.0721

0.0687 0.0854

0.0837 0.0805 0.0774 0.0745

0.0822 0 0797 0.0768 0.0748 0.0789 0.0764 0.0736 0.0715

0.0736 0.0704 0.0673 0.0645

0.0715 0.0696 0.0682 0.0663 0.0652 0.0633 0.0623 0.0604

0.0809 0.0780

0.0759 0.0730 0.0703

0.0677 0.0652 0.0630 0.0610 0.0591

0.0574

0.0734 0.0705 0.0685 0.0705 0.0676 0.0656 0.0678 0.0649 0.0629

0.0653 0.0624 0.0603 0.0628 0.0599 0.0579 0.0606 0.0577 0.0556 0.0585 0.0556 0.0536 0.0566 0.0538 0.0517

0.0549 0.0520 0.0500 0.0532 0.0504 0.0483 0.0518 0.0489 0.0469 0.0505 0.0476 0.0456 0.0494 0.0465 0.0444

0.0484 0.0455 0.0434 0.0474 0.0445 0.0425 0.0467 0.0438 0.0417 0.0460 0.0431 0.0411 0.0455 0.0426 0.0406

0.0450 0.0422 0.0401 0.0447 0.0418 0.0398 0.0446 0.0417 0.0397 0.0445 0.0416 0.0396 0.0445 0.0417 0.0397

0.0446 0.0418 0.0398 0.0449 0.0420 0.0400 0.0452 0.0423 0.0403 0.0456 0.0427 0.0407 0.0461 0.0432 0.0412

0.0466 0.0438 0.0418 0.0473 0.0445 0.0425 0.0479 0.0451 0.0431 0.0488 0.0459 0.0440 0.0496 0.0468 0.0448

0.0505 0.0477 0.0457 0.0514 0.0486 0.0467 0.0525 0.0497 0.0478 0.0535 0.0508 0.0488 0.0546 0.0518 0.0498

0.0557 0.0529 0.0510 0.0568 0.0540 0.0521 0.0580 0.0553 0.0534 0.0592 0.0565 0.0545 0.0605 0.0578 0.0558

0.0617 0.0590 0.0571 0.0628 0.0602 0.0583 0.0641 0.0614 0.0595 0.0653 0.0626 0.0608 0.0665 0.0638 0.0620

0.0677 0.0651 0.0632

0.0624 0.0595

0.0753 0.0718

0.0728 0.0693 0.0703 0.0668 0.0681 0.0646 0.0660 0.0625 0.0642 0.0606

0.0624 0.0589

0.0617

0.0592 0.0567 0.0545 0.0525 0.0506

0.0489 0.0472 0.0458 0.0445 0.0433

0.0423 0.0414 0.0406 0.0399 0.0395

0.0390 0.0387 0.0386 0.0385 0.0386

0.0387 0.0389 0.0392 0.0396 0.0401

0.0407 0.0414 0.0421 0.0429 0.0437

0.0446 0.0456 0.0467 0.0477 0.0488

0.0499 0.0511 0.0523 0.0535 0.0548

0.0560 0.0572 0.0585 0.0597 0.0610

0.0622

0.0596 0.0577 0.0568

0.0570 0.0551 0.0542 0.0546 0.0527 0.0518 0.0524 0.0505 0.0496 0.0503 0.0484 0.0475 0.0484 0.0465 0.0457

0.0467 0.0448 0.0439 0.26 0.0607 0.0572 0.27 0.0593 0.0558 0.28 0.0580 0.0545 0.29 0.0569 0.0534

0.0557 0.0543

0.0467 0.0452

0.0451 0.0431 0.0423 0.0436 0.0417 0.0408 0.0423 0.0404 0.0395 0.0412 0.0393 0.0384

0.0530 0.0518

0.0508 0.0499 0.0491

0.0440 0.0428

0.0418 0.0409 0.0401

0.30 0.31 0.32 0.33 0.34

0.35 0.36

0.0559 0.0524 0.0549 0.0514 0.0541 0.0506 0.0534 0.0499 0.0529 0.0495

0.0402 0.0383 0.0392 0.0373 0.0385 0.0366 0.0378 0.0359 0.0373 0.0354

00369 0.0350 0.0366 0.0347 0.0364 0.0345 0.0383 0.0344 0.0364 0.0345

0.0365 0.0346 0.0368 0.0349 0.0371 0.0352 0.0375 0.0356 0.0380 0.0361

0.0386 0.0367

0.0374 0.0364 0.0357 0.0350 0.0345

0.0341 0.0338 0.0337 0.0338 0.0337

0.0338 0.0340 0.0343 0.0348 0.0353

0.0359 0.0365 0.0372 0.0380 0.0389

0.0484 0.0479

0.0475 0.0471

0.0394 0.0389

0.0525 0.0490 0.0521 0.0487 0.0520 0.0485 0.0519 0.0484

0.0385 0.0382

0.37 0.38

0.0470 0.0469 0.0470

0.0471 0.0473 0.0476 0.0480 0.0485

0.0490 0.0497 0.0504 0.0512 0.0520

0.0529 0.0538 0.0549 0.0559 0.0569

0.0580 0.0591 0.0603 0.0615 0.0628

0.0640 0.0651 0.0664 0.0676 0.0687

0.0699

0.0380 0.0380 0.0380

0.0382 0.0384 0.0387 0.0391 0.0396

0.0402

0.39 0.0520

0.40 0.0521 0.41 0.0523 0.42 0.0526 0.43 0.0530 0.44 0.0534

0.45 0.0540 0.46 0.0546 0.47 0.0553 0.48 0.0561 0.49 0.0569

0.50 0.0578 0.51 0.0587

0.0485

0.0486 0.0488 0.0491 0.0495 0.0500

0.0506 0.0512 0.0519 0.0527 0.0535

0.0544 0.0553 0.0564 0.0574 0.0584

0.0595 0.0606 0.0618 0.0630 0.0642

0.0654 0.0666 0.0678 0.0690 0.0701

0.0713 0.0724 0.0735 0.0745 0.0755

0.0765 0.0773 0.0781 0.0789 0.0795 0.0800

0.0409 0.0393 0.0374 0.0415 0.0400 0.0381 0.0424 0.0432

0.0441 0.0451 0.0462 0.0472 0.0483

0.0494 0.0506 0.0518 0.0530 0.0543

0.0556 0.0567 0.0580 0.0593 0.0605

0.0617

0.0408 0.0389 0.0416 0.0397

0.0426 0.0407 0.0435 0.0417 0.0446 0.0427 0.0457 0.0438 0.0467 0.0449

0.0479 0.0460 0.0490 0.0472 0.0503 0.0484 0.0515 0.0496 0.0528 0.0510

0.0540 0.0522 0.0552 0.0534 0.0565 0.0548 0.0578 0.0560 0.0590 0.0572

0.0602 0.0585 0.0614 0.0597 0.0626 0.0608 00637 0.0620 0.0648 0.0631

0.0658 0.0641 0.0667 0.0651 0.0676 0.0660 0.0685 0.0669 0.0693 0.0677 0.0699 0.0683

0.0398 0.0408 0.0419 0.0430

0.52 0.0598 0.53 0.0608 0.54

0.55

0.0617

0.0629 0.0639 0.0651 0.0663 0.0675

0.0687 0.0698 0.0711 0.0722 0.0733

0.0745 0.0756 0.0766

0.0440

0.0452 0.0464 0.0476 0.0488 0.0501

0.0514 0.0526 0.0539 0.0552 0.0564

0.56 0.57 0.58 0.59

0.60 0.61 0.62 0.63 0.64

0.65 0.66 0.67

0.0577 0.0589 0.0601 0.0612 0.0623

0.0710 0.0688 0.0662 0.0643 0.0633 0.0721 0.0699 0.0673 0.0655 0.0645 0.0732 0.0710 0.0684 0.0666 0.0656 0.0741 0.0720 0.0694 0.0676 0.0667

0.0629 0.0640 0.0652 0.0662

0.0672 0.0681

0.68 0.69

0.70 0.71

0.0777 0.0786

0.0795 0.0803 0.0811 0.0818 0.0825 0.0829

0.0751 0.0730 0.0704 0.0687 0.0677 0.0759 0.0738 0.0713 0.0695 0.0686 0.0768 0.0747 0.0722 0.0704 0.0695 0.0776 0.0755 0.0730 0.0713 0.0704 0.0782 0.0762 0.0738 0.0721 0.0711 0.0788 0.0767 0.0743 0.0726 0.0718

0.0634 0.0643 0.0652 0.0662 0.0670 0.0677

0.72 0.73 0.74 0.75

0.0690 0.0699 0.0707 0.0713


TABLE 13.19-b VALUES FOR REYNOLDS NUMBER FACTOR, F,, FLANGE TAPS (continued)

F,=l+b

JhwP, Internal Diameter of Pipe, di, in,

6 8 10 12

-A-- F 4.897 5.189 5.761 6.065 __ 7.625 ~ 7981 8.071 9.564 10.020 10.136 11.376 .l1.938 12.090

0.10 0.0859 0.0846 0.0825 0.0815 0.0777 0.0771 0.0769 00747 0.0742 0.0741- 0.0729 0.0724 0.0725

0.11 0.0818 0.0805 0.0783 0.0773 0.0736 0.0729 00727 0.0706 0.0700 0.0699 0.0687 0.0682 0.0681

0.12 0.0777 0.0764 0.0742 0.0732 0.0695 0.0688 00687 00665 0.0660 0.0658 0.0646 0.0642 0.0640

0.13 0.0738 0.0725 0.0703 0.0694 0.0656 0.0649 00648 0.0626 0.0621 0.0619 0.0607 0.0603 0.0602

0.14 0.0701 0.0688 0.0666 0.0657 0.0619 0.0612 0.0611 0.0589 0.0584 0.0583 0.0571 0.0566 0.0565

0.15 0.0666 0.0653 0.0631 0.0622 0 0584 0.0578 00576 0.0554 0.0549 0.0548 0.0536 0.0531 0.0530

0.16 0.0633 0.0620 0.0598 0.0589 0.0551 0.0545 0.0543 0.0522 0.0516 0.0515 0.0503 0.0498 0.0497

0.17 0.0601 0.0589 0.0567 0.0558 0.0520 0.0514 0.0512 0.0491 0.0485 0.0484 0.0472 0.0467 0.0466

0.18 0.0571 0.0558 0.0537 00527 0.0490 0.0484 0.0482 0.0461 0.0455 0.0454 0.0442 0.0438 0.0436

0.19 0.0544 0.0531 0.0510 0.0500 0.0463 0.0456 00455 00433 0.0428 0.0427 0.0415 0.0410 0.0409

0.20 0.0517 0.0504 0.0483 0.0474 0.0436 0.0430 00428 0.0407 0.0402 0.0401 0.0389 0.0384 0.0383

0.21 0.0492 0.0480 0.0459 0.0449 0.0412 0.0405 00404 0.0383 0.0377 0.0376 0.0364 0.0360 0.0359

0.22

0.23

0.24

0.25

0.26

0.27

0.28

0.29

0.30

0.31

0.32

0.33

0.34

0.35

0.36

0.37

0.0469 0.0457

0.0448 0.0435

0.0428 0.0416

0.0436

0.0414

0.0395

0.0377

0.0360

0.0345

0.0332

0.0319

0.0308

0.0298

0.0290

0.0426

0.0405

0.0385

00367

0.0350

0.0336

0.0389 0.0383 0.0381 0.0360 0.0355

0.0366 0.0362 00360 0.0339 0.0334

0 0349 0.0342 0.0341 0.0320 0.0314

00331 0.0324 00323 0.0302 0.0297

0.0314 0.0308 00306 0.0285 0.0280

0.0299 0.0293 0 0292 0.0271 0.0266

00286 0.0280 0.0278 0.0258 0.0252

0.0274 0.0268 0.0266 0.0245 0.0240

0.0263 0.0257 00255 0.0235 0.0230

0.0254 0.0247 00246 0.0225 0.0220

0.0245 0.0239 0.0237 0.0217 0.0212

0.0353

0.0333

0.0313

0.0296

0.0279

0.0264

0.0337

0.0316

0.0297

0.0279

0.0263

0.0248

0.0336

0.0321

0.0302

0.0284

0.0267

0.0253

0.0315

0.0296

0.0278

0.0262

0.0247

0.0410

0.0393

0.0398 0.0361

0.0378 0.0366

0.0365 0.0352

0.0352 0.0340

0.0341 0.0328

0.0331 0.0319

0.0322 0.0310

0.0322

0.0310

0.0251 0.0240 0.0235 0.0234

0.0239 0.0228 0.0223 0.0222

0.0299 0.0289 0.0280

0.0228 00217 0.0213

0.0219 0.0208 0.0203

0.0211 00199 0.0195

0.0211

0.0202

0.0194

0.0187

0.0182

0.0177

0.0174

0.0314 0.0302 0.0282 0.0273 0.0238 0.0232 0.0230 0.0210 0.0205 0.0204 0.0192 0.0188

0.0308 0.0296 0.0276 00267 0.0232 0.0226 00225 0.0204 0.0199 0.0198 0.0187 0.0183

0.0303 0.0291 0.0272

0.0299 0.0287 0.0268

0.0296 0.0284 0.0265

0.0294 0.0282 0.0263

0.0293 0.0281 0.0262

0.0292 0.0281 0.0261

0.0292 0.0281 0.0262

00263 0.0228 0.0221 00220 0.0200 0.0259 0.0224 0.0218 0.0216 0.0196

0.0256 0.0221 0.0215 00214 0.0194

00254 00220 0.0214 00212 0.0193

0.0253 0.0219 0.0213 0.0211 0.0192

0.0253 0.0219 0.0213 0.0212 0.0192

0.0253 0.0220 0.0214 00213 0.0193 00255 00222 00216 00214 0.0195

0.0257 0.0224 0.0218 0.0217 0.0198

0.0260 0.0227 0.0221 0.0220 0.0201

0.0263 0.0231 0.0225 0.0224 0.0205 0.0268 0.0235 0.0230 0.0228 0.0210

0.0272 0.0240 0.0235 0.0233 0.0215

0.0195

0.0192

0.0189

0.0188

0.0187

0.0187

0.0189

0.0191

0.0193

0.0197

0.0201

0.0194

0.0190

0.0188

0.0187

0.0186

0.0186

0.0187

0.0189

0.0192

0.0196

0.0200

0.0204

0.0209

0.0183 0.0178

00179 0.0175

0.0177 0.0173 0.0172

00176 0.0171 0.0170 0.38

0.39

0.40

0.0175 0.0171

0.0175 0.0171

0.0177 0.0173

00179 0.0175

0.0182 0.0178

0.0185 0.0181

0.0189 0.0185

00194 0.0190

0.0199 0.0195

0.0170

0.0170

0.41 0.42

0.0172

0.0174 0.0294 0.0282 0.0263

0.0296 0.0285 0.0266 0.43

0.44

0 45

0.46

0.47

0.0177

0.0180

0.0184

0.0189

0.0194

0.0298 0.0287 0.0268

0.0301 0.0290 0.0272

0.0305 0.0294 0.0276

0.0309 0.0298 0.0280

0.0205

0.0210

0.48 0.0314 0.0303 0.0285 0.0277 0.0245 0.0240 0.0238 0.0220 0.0216 0.0215 0.0205 0.0201 0.0200

0.49 0.0319 0.0308 0.0290 0.0282 0.0251 0.0245 0.0244 0.0226 0.0222 0.0221 0.0211 0.0207 0.0206

0.50 0.0324 0.0314 0.0296 0.0288 0.0257 0.0252 0.0250 0.0233 0.0228 0.0227 0.0217 0.0213 0.0212

0.51 0.0330 0.0319 0.0302 0.0294 0.0263 0.0258 0.0257 0.0239 0.0235 0.0234 0.0224 0.0220 0.0219

0.52 0.0335 0.0325 0.0308 0.0300 0.0270 0.0265 0.0263 0.0246 0.0242 0.0241 0.0231 0.0227 0.0226

0.53 0.0342 0.0332 0.0315 0.0307 0.0277 0.0272 0.0270 0.0253 0.0249 0.0248 0.0238 0.0235 0.0234

0.54 0.0348 0.0338 0.0321 0.0314 0.0284 0.0279 0.0278 0.0261 0.0256 0.0255 0.0246 0.0242 0.0241

0.55 0.0355 0.0345 0.0328 0.0321 0.0291 0.0286 0.0285 0.0268 0.0264 0.0263 0.0254 0.0250 0.0249

0.56 0.0361 0.0351 0.0335 0.0327 0.0298 0.0293 0.0292 0.0276 0.0272 0.0271 0.0261 0.0258 0.0257

0.57

0.58

0.59

0.60

0.61

0.62

0.63

0.64

0.65

0.66

067

0.68

0.69

0.70

071 0.72

0.0368

0.0375

0.0381

0.0388

0.0393

0.0399

0.0358

0.0365

0.0372

0.0378

0.0384

0.0390

0.0396

0.0402

0.0407

0.0412

0.0417

0.0422

0.0425

0.0428 -

0.0342

0.0349

00356

0.0363

0.0369

0.0375

0.0381

0.0387

0.0393

0.0398

0.0403

0.0408

0.0411

00415

0.0335 0.0306 0.0301

0.0313 0.0309

0.0321 00316

0.0328 0.0323

0.0335 0.0330

0.0342 0.0337

0.0348 0.0344

0.0355 0.0350

00361 0.0357

0.0366 0.0362

0.0372 0.0368

0.0377 0.0373

0.0381 0.0377

0.0386 0.0382 - -

0.0300 0.0283 0.0279 0.0307 0.0291 0.0287

00315 0.0299 0.0295

0.0322 0.0307 0.0303

0.0329 0.0314 0.0310

0.0336 0.0321 0.0317

0.0343 0.0328 0.0324

0.0349 0.0334 0.0331

0.0355 0.0341 0.0337

0.0361 0.0347 0.0343

0.0367 0.0352 0.0349

0.0372 0.0358 0.0355

0.0376 0.0363 0.0359

0.0381 0.0367 0.0364

0.0278 0.0269 0.0266 0.0265

0.0286 0.0277 0.0274 0.0273

0.0294 0.0285 0.0282 0.0281

0.0302 0.0293 0.0290 0.0289 0.0309 0.0300 0.0297 0.0296

0.0316 0.0307 0.0304 0.0303

0.0323 0.0315 0.0311 0.0311

0.0330 0.0322 0.0318 0.0318

0.0336 0.0328 0.0325 0.0324

0.0342 0.0334 0.0331 0.0331

0.0349 0.0341 0.0338 0.0337

0.0354 0.0346 0.0344 0.0343

0.0359 0.0351 0.0348 0.0347

0.0363 0.0356 0.0353 0.0352 -

0.0342

0.0349

0.0356 0.0362

0.0368

0.0375

0.0381

0.0386

0.0391

0.0397

0.0402

0.0405

0.0409 -

0.0405

0.0411

0.0416

0.0421

0.0425

0.0430

0.0433

0.0436

0.74 - - - - - - -

0.75 - - - - - - - -


TABLE 13.lh-b VALUES FOR REYNOLDS NUMBER FACTOR, f,, PIPE TAPS

b F,=l+-

KE

Orifice Diameter,


2 3 4

1.689 1.939 2.067 2.300 2.626 2.900 3.068

0.1105 01091 0.1087 0.1081 - - -

3.152 3.438

0.250 0.375 0.500 0.625 0.750

0.875 1.000 1.125 1.250 1.375

1.500 1.625 1.750 1.875 2.000

2.125 2.250 2.375

-

0.0908 0.0918 0.0763 0.0778 0.0646 0.0662 0.0555 0.0568

0.0489 0.0496 0.0443 0.0443 0.0417 0.0407 0.0408 0.0387 0.0408 0.0379

0.0418 0.0382 0.0435 0.0392 0.0456 0.0408 0.0477 0.0427 0.0495 0.0448

0.0509 0.0467 00483

- 0.0494

0.0890 0.0758 0.0893 0.0675

0.0684 0.0702 0.0708

-

0.0878 00734 0.0647 0.0608

0.0602 0.0614 0.0635 0.0650

-

0.0877 0.0729 0.0635 0.0586

0.0570 0.0576 0.0595 0.0616 0.0629

-

0.0879 0.0728 0.0624 0.0559

0.0528 0.0522 0.0532 0.0552 0.0574

0.0590 -

0.0888 0.0737 0.0624 0.0546

0.0497 0.0473 0.0469 0.0478 0.0496

0.0518 0.0539 0.0553

-

0.0898 0.0750 0.0634 0.0548

0.0488 0.0452 0.0435 0.0434 0.0443

0.0460 0.0482 0.0504 0.0521 0.0532

- - -

0.0905 0.0758 0.0642 0.0552

0.0488 0.0445 0.0422 0.0414 0.0418

0.0431 0.0450 0.0471 0.0492 0.0508

0.0519 - -

- -

- - - - - - - - - -

- - - - - - -

- -

Internal Diameter of Pipe, d,, in. Orifice Diameter,

$.j

0.500 0.625 0.750 0.875 1.000

1.125 1.250 1.375 1.500 1.625 1.750 1.875 2.000

2.125 2.250 2.375 2.500 2.625 2.750 2.875 3.000

3.125 3.250 3.375 3.500 3.625 3.750 3.875 4.000

4.250 4.500 4.750 5.000

5.250 5.500

6

5.761 6.065

- -

0.0789 0.0802 0.0703 0.0718 0.0625 0.0842 0.0556 0.0573

0.0495 0.0512 0.0442 0.0458 0.0397 0.0412 0.0360 0.0372 0.0329 0.0339 0.0304 0.0311 0.0285 0.0290 0.0273 0.0273

0.0265 0.0262 0.0261 0.0258 0.0262 0.0253 0.0267 0.0254 0.0274 0.0258 0.0284 0.0265 0.0295 0.0274 0.0308 0.0285

0.0323 0.0297 0.0338 0.0311 0.0353 0.0325 0.0367 0.0339 0.0381 0.0354 0.0393 0.0367 0.0404 0.0380 0.0413 0.0391 -

-

- -

-

- -

4

4.026

0.0810 0.0697 0.0602 0.0524 0.0461

0.0412 0.0377 0.0353 0.0340 0.0336 0.0340 0.0349 0.0363

0.0380 0.0398 0.0417 0.0435 0.0450 0.0462

-

- -

- - - - -

-

-

- -

8

7.625 7.981 8.071 5.189 -

0.0762 0.0672 0.0592 0.0523

0.0464 0.0413 0.0373 0.0340 0.0315 0.0298 0.0287 0.0281

0.0281 0.0285 0.0293 0.0304 0.0317 0.0331 0.0347 0.0364

0.0380 0.0394 0.0408 0.0419 0.0428

3.826

0.0799 0.0685 0.0590 0.0513 0.0453

0.0408 0.0376 0.0358 0.0350 0.0351 0.0358 0.0371 0.0388

0.0407 0.0427 0.0445 0.0460 0.0472

-

- - - - - - - -

-

-

- -

4.897

0.0850 0.0747 0.0655 0.0575 0.0506

0.0448 0.0401 0.0363 0.0334 0.0313 0.0300 0.0293 0.0292

0.0297 0.0305 0.0316 0.0330 0.0345 0.0362 0.0379 0.0395

0.0410 0.0422 0.0432

- -

- - -

0.0716 0.0652

0.0592 0.0538 0.0489 0.0445 0.0404 0.0389 0.0338 0.0311

0.0288 0.0268 0.0252 0.0239 0.0230 0.0224 0.0220 0.0219

0.0220 0.0223 0.0228 0.0235 0.0243 0.0252 0.0262 0.0273

0.0296 0.0321 0.0344 0.0364

0.0381 -

-

- 0.0733 0.0662 0.0613 0.0560 0.0510 0.0466 0.0425 0.0388 0.0355 0.0327

0.0301 0.0280 0.0261 0.0246 0.0233 0.0224 0.0218 0.0213

0.0211 0.0212 0.0214 0.0218 0.0224 0.0230 0.0238 0.0246

0.0268 0.0290 0.0314 0.0336

0.0356 0.0372

- -

0.0730 0.0668

0.0609 0.0555 0.0506 0.0462 0.0421 0.0384 0.0352 0.0323

0.0298 0.0277 0.0259 0.0244 0.0232 0.0224 0.0218 0.0214

0.0213 0.0214 0.0216 0.0221 0.0227 0.0234 0.0243 0.0252

0.0273 0.0296 0.0320 0.0342

0.0361 0.0377


TABLE IXlh--b VALUES FOR REYNOLDS NUMBER FACTOR, F,, PIPE TAPS (continued)

b F,=1+-

JGI

Internal Diameter of Pipe, di, in. Orifice

Diameter,

;ri.;

1.000 1.125 1.250 1.375

1.500 1.625 1.750 1875 2.000

2.125

2.250 2.375 2.500 2625 2.750

2.875 3.000

3.125

3250 3.375 3500 3625 3.750 3 a75 4000

4.250

4.500 4.750 5.000

5.250 5500

5.750 6.000

6.250 6500 6.750 7.000

7.250 7.500 7.750

8.000

8.250 8500 8750 9 000

9 250 9.500 9 750

10000

10250 10.500

9.564

0.0728 0.0674

0.0624 0.0576

0.0532 0 0490 00452 00417 00385

0.0355 0 0329 00305 00263

00265 00248 00234 00222

0.0212

0.0204

0.0199 00195

00193 00192 00193 00195

0.0203 00215 0.0230 0.0248

0.0267 00287 00307 0.0326

0.0343

00358 -

10.020 10.136

- -

0.0690 0.0694

0.0641 0.0646 0.0594 0 0599

00550 00555 0.0509 0.0514 0.0471 0.0476 0.0436 0.0440

0.0403 0.0407

0.0372 0.0377 0.0345 0.0349 0.0320 0.0324

0.0298 0.0301

0.0277 0.0281 0.0260 0.0262 0.0244 0.0246

0.0230 0.0232

0.0218 0.0220

0.0209 00210

0.0201 0.0202 0.0195 0.0196 00191 0.0191 0.0188 0.0188 0.0187 0.0186 0 0187 00186

0.0192 0.0189

0.0200 0.0197

0.0212 0.0206 0.0228 00223

00244 0.0239 0.0263 0.0257 0.0282 0.0276 0.0302 0.0295

0.0320 0.0316

0.0336 0.0331 0.0351 0.0346 0.0363 0.0359

11.376

-

o.osa7 0.0643 0.0601 0.0561 0.0523 0.0488

0.0454

00423 0.0394 0.0367 0.0342

0.0319 0 0298 0.0279 0.0262

0.0244

0.0232 0.0220 0.0210 0.0200 0.0193 0.0187 0.0182

00176 00175 0.0178 0.0185

0.0195 0.0207 00221 00231

00253 00270 00288 00304

00320 00334

00347

-

-

- -

12

11.938 12.090 - - -

00704 0.0661

0.0620 00560 00543 00508

00475

00443 0.0414 0.0387 0.0361 0.0337 00316 0.0295 0.0277

00260 00245 00232 0.0220 0 0209 00200 0.0192 00185

0.0176 0.0172 0.0171 00174

0.0181 0.0190 00202 00215

00230 00246 00262 0 0279

0 0295 00310 00325 00338

0 0349

-

0.0708 0.0666

0.0625 0.0585 0.0548 0.0513 0.0480

0.0449 0.0419 0.0392 0.0366 0.0342 0.0320 0.0300 0.0281

0.0264 0.0249 0.0235 0.0222 0.0212 0.0202 0.0194

0.0187

0.0177 0.0171

0.0170 0.0173

0.0178 0.0186 00197 0.0210

0.0224 0.0239 00256 0.0272

00288 00304

0.0318 0.0332

0.0344

16

14.686 15.000 15.250

- - - - -

0.0697 0.0662 0.0628 0.0594 0.0563

0.0532 0.0503 0.0475 0 0449 00424 00400 0.0378

00356

00336

00317 0.0300 00283 00268 0.0254 00240 0.0228

00207 00190 00176 00166

00160 00156

0.0155 0.0157

0.0161 0.0167 0.0174 0.0184

0.0195 0.0206

0.0219 0.0232

0.0246

0.0259 0.0273

0.0286

0.0299

0.0311 0.0322 00332

00341 -

0.0705 0.0670 0.0636 0.0603

0.0572

0.0541 0.0512 0.0484

0.0458 0.0433 0.0409 0.0387 0 0365

0.0345 0.0326 00308 0.0291 0.0275 0.0261 0.0247

00235

0.0213

0.0194 0.0180 0.0168

0.0161 0.0156 0.0154 0.0154

0.0157 0.0162 00169 0.0177

0.0187 0.0198

0 0209 0.0222

0.0235

0.0248 0.0262

0.0276

0 0288

0.0300 0.0312 0.0323

0.0333 0.0341

0.0676 0.0642 0.0610

0.0578

0.0548 0.0519 0 0492 0.0466 0.0440 0.0417 0 0394 0.0372

00352 0.0332

0.0314 0 0297 00281 00267 00253 00240

00217 00198 00182 0.0170

0.0162 00156 0.0153

00153

00154 00159 00164 00172

00181 00191

00202 00214

0.0227

00240 00253

00267

00280

0 0292 0.0304 00315

0.0326 0.0335

13-22

Diameter,

g.;

2.000 2.125 2.250 2.375

2.500 2.625 2.750 2.875 3.000

3.125

3.250 3.375 3.500 3.625 3.750 3.875 4.000

4.250 4.500

4.750 5.000

5.250 5.500 5.750

6.000

6.250

6.500 6.750 7.000

7.250 7.500

7.750 8000

8 250 8500 8 750

9.000

9.250 9 500

9.750 10000

10.250 10500 10750

11000

11.250

11.5oa 11750 12000

12.500

13.000

13500 14000

14500 15000

15500 16000

16500 17000

17500

la000

18500 19000

19500 20000


TABLE 13.lh-b VALUES FOR REYNOLDS NUMBER FACTOR, F,, PIPE TAPS (continued)

20

19.250 19.000

0.0667 0.0639

0.0613 00588

0 0562 0.0539 0.0515 0.0492 0.0470

0.0449 0.0429 0.0410 0.0391 0.0373

0.0356 0.0340

0.0324

0.0295

0.0269 0.0246 0.0225

0.0206 00190 0.0177 0.0165

0.0156 0 0149 00144 0.0141

00140 0.0140 0.0141 0.0144

0.0148 0.0154 0.0160

0.0168

0.0176 0.0185

0.0194 0.0204

0.0214 00225 0.0236

0.0247

0.0261 0.0268 0.0278 0.0288

0.0307 0.0323

0.0672 - - 00644

0.0618 0.0593

0.0568 0.0544 00520 0.0498 00476

0.0455 0.0435 0.0416 0 0397

0.0379 0.0362 0.0346 0.0330

0.0301 0.0274

0.0250 0.0229

00210 00194 0.0180

0.0168

0.0158

0.0150 0.0145 0.0141

0.0139 0.0139

00140 0.0142

0.0146 00151 0.0157

0.0163

0.0171

0.0180 0.0189 0.0198

0.0208 00219 0 0229 0.0240

0.0251

0.0262 0.0272 0.0282

0.0301 0.0318

18.814

0.0663 0.0635 0.0609 0.0583

0.0558 0.0534 0.0510

0.0488 0.0466

00445 0.0425 0.0406 0.0387

0 0369 00352 0.0336 0.0320

0.0291 00265

00242 0.0221

0.0203 0.0188

0.0175 0.0164

0.0155 0.0146 0.0143 0.0141

0.0140 00140 0.0142

0.0146

0.0151 0.0156 00163

0.0171

00180

00189 0.0198 0.0209

0.0219 0.0230 0.0241

0.0252

0.0263

0.0273 0.0284 0.0293

0.0312 0.0327

-

-

-

- -

-

-

-

- -

b

Fr=‘+ -&&


24

22.626 23.000 23.250

- 0.0658

0.0635 0.0613 0.0591 0.0570

0.0549

0.0529 0.0509 0.0490 0.0471

0.0454 0.0436 0.0419 0.0403

0.0372 0.0343 00316 0.0292

0 0269 00248 00230 0.0212

0.0197 0.0164 0.0172 0.0162

0.0153 0.0146

0.0140 00136

00133 0.0132 0.0131 0.0131

0.0133 0.0136 0.0139 0.0143

0.0148 00154 0.0160 0.0168

0.0175 0.0183 0.0191 0.0200

0.0218 00236

0.0254

00272

0 0289 0.0304

0.0318 -

-

0.0665 0.0642 0.0620 0 0598 0.0577

0.0556

0.0536 0.0516 0.0497 0.0479 0.0461 0.0444 00427 0.0411

0.0380 0.0351 00324 0 0299

00276 00255 00236

00218

0.0202

0.0189 0.0176 00166

00156 00148

00142 00138

00134 00132 0.0130

0.0130

0.0131 0.0133 0.0136 0.0140

0.0144 0.0150 0.0155 0.0162

0.0169 0.0176 0.0184 0.0192

0.0210 0.0228

0.0246

0.0264 0.0260

0.0296

0.0311 00323

-

-

0.0669

0.0646 0.0624 0.0603 0.0582

0.0561

0.0541 0.0521 0.0502 0.0484

0.0466 0.0449 0.0432

0.0416

0.0385 0.0356

0.0328 0.0303

0.0280 0 0259 0.0240

0.0222

0.0206 0.0192 0.0179 0.0168

0 0158 0.0150 0.0144

0.0138

0.0132 0.0130 0.0130

0.0130

0.0130 0.0132 0.0134 0.0138

0.0142 0.0147

0.0152 0.0158

0.0165 0.0172 0.0180 0.0190

0.0204 0.0222

0.0240

0.0258

0.0275 0.0291

0.0306 00318

30

28.626 29.000 29.250 - -

-

- -

00667 0 0649

0 0630 0.0613 0 0595 0.0578

0.0561 0.0545 00528 0.0513

0.0482

00453 0.0425 0 0399

00374 00350

00328 00307

00287 0 0269 00252 00236

00221 00207

0.0195 0.0183

0.0173 0.0164 0.0155

0.0148

0.0142

0.0136

0.0132 00128

00125 00123 0.0122 0.0121

00122

00122 0.0124

00126

00132 0.0140

00150 0.0161

0.0173 00186

0.0200 00215

0.0230 0.0244

0.0259 0.0272

0.0286 0.0298

0.0309 0.0318

- - - - - - -

- -

- - 0.0654 0.0657

0.0636 0.0639 0.0616 0.0622 0.0601 0.0604 0.0584 0.0587 0.0567 0.0571 0.0550 0.0554 0.0534 0.0538 0.0518 0.0522

0.0488 0.0492 0.0459 0.0463 0.0431 0.0435 0.0405 0 0409

0.0380 0.0384 0.0356 0.0360 0.0334 00338

0.0313 0.0317

0 0293 0 0297 0.0274 00278 0.0257 00260 0.0241 00244

0.0226 0 0229 0.0212 00215

0.0199 0.0202 00187 0.0190

0.0177 00179 0.0167 00169 00158 0.0161

0.0151 0.0153

0 0144 0.0146

0.0138 0.0140 0.0133 00134 0.0129 0.0130

0.0126 0.0127 00124 0.0124 00122 0.0122 0.0121 0.0121

0.0121 0.0121 0.0121 0.0122 0.0123 0.0122 0.0124 0.0123

0.0130 0.0128 0.0137 0.0135

0.0146 0.0143 0.0156 0.0153

0.0168 0.0165 00181 0.0177

0.0194 0.0190 0 0209 00204

0.0223 0.0219 0.0238 0.0233

0.0252 00248 0.0266 0.0261

0.0279 0.0275 0.0292 0.0288

00303 0.0299 00313 00310


TABLE 13.1 h-6 VALUES FOR REYNOLDS NUMBER FACTOR, F,, PIPE TAPS (continued) b

F,=l+-

JKE


F, 0.10 0.11 0.12 0.13 0.14

0.15 0.16 0.17 0.18 0.19

0.20 0.21 0.22 0.23 0.24

0.25 0.26 0.27 0.28 0.29

0.30 0.31 0.32 0.33 0.34

0.35 0.36 0.37 0.38 0.39

0.40 0.41 042 0.43 0.44

0.45 0.46 0.47 0.48 0.49

0.50 0.51 0.52 0.53 0.54

0.55 0.56 0.57 0.58 0.59

0.60 0.61 0.62 0.63 0.64

0.65 0.66 0.67 0.68 0.69 0.70

1.689

0.1295 0.1253 0.1212 0.1172 0.1134

0.1098 0.1065 0.1033 0.1001 0.0973

0.0945 0.0919 0.0894 0.0672 0.0851

0.0831 0.0812 0.0796 0.0781 0.0766

0.0753 0.0741 0.0730 0.0720 0.0712

0.0705 0.0699 0.0693 0.0689 0.0684

0.0681 0.0678 0.0677 0.0676 0.0675

0.0675 0.0676 0.0676 0.0677 0.0678

0.0680 0.0682 0.0684 0.0687 0.0689

0.0692 0.0694 0.0696 0.0699 0.0701

0.0704 0.0705 0.0706 0.0707 0.0708

0.0709 0.0708 0.0708 0.0708 0.0705 0.0704

1.939 2.067

0.1209 0.1173 0.1167 0.1132 0.1126 0.1090 0.1086 0.1051 0.1049 0.1013

0.1013 0.0978 0.0980 0.0944 0.0948 0.0912 0.0916 0.0881 0.0888 0.0853

0.0861 0.0825 0.0835 0.0800 0.0811 0.0776 0.0788 0.0753 0.0767 0.0733

0.0748 0.0714 0.0730 0.0695 0.0713 0.0679 0.0699 0.0665 0.0685 0.0651

0.0672 0.0638 0.0661 0.0627 0.0650 0.0616 0.0640 0.0606 0.0632 0.0599

0.0626 0.0593 0.0620 0.0587 0.0615 0.0582 0.0611 0.0578 0.0607 0.0575

0.0604 0.0572 0.0603 0.0571 0.0602 0.0570 0.0601 0.0570 0.0601 0.0570

0.0602 0.0571 0.0603 0.0572 0.0604 0.0574 0.0606 0.0576 0.0608 0.0578

0.0611 0.0581 0.0613 0.0584 0.0615 0.0587 0.0619 0.0590 0.0622 0.0594

0.0625 0.0598 0.0628 0.0601 0.0632 0.0605 0.0635 0.0608 0.0638 0.0612

0.0641 0.0615 0.0643 0.0618 0.0645 0.0620 0.0648 0.0623 0.0649 0.0625

0.0651 0.0627 0.0651 0.0628 0.0653 0.0629 0.0653 0.0630 0.0652 0.0629 0.0651 0.0629

2.300 2.626 2.900 3.068 3.152 3.438 3.826 4.026

0.1118 0.1058 0.1077 0.1016 0.1035 0.0975 0.0996 0.0935 0.0958 0.0898

0.0923 0.0863 0.0889 0.0829 0.0856 0.0798 0.0827 0.0767 0.0799 0.0739

0.0771 0.0712 0.0746 0.0687 0.0722 0.0663 0.0700 0.0641 0.0679 0.0621

0.0660 0.0602 0.0642 0.0584 0.0626 0.0568 0.0612 0.0554 0.0598 0.0541

0.0586 0.0529 0.0575 0.0518 0.0564 0.0508 0.0555 0.0499 0.0548 0.0492

0.0542 0.0486 0.0537 0.0481 0.0532 0.0477 0.0529 0.0474 0.0525 0.0471

00523 0.0469 0.0522 0.0468 00521 0.0468 0.0522 0.0469 0.0522 0.0470

0.0524 0.0472 0.0526 0.0474 0.0528 0.0476 0.0530 0.0479 0.0532 0.0482

0.0536 0.0486 0.0539 0.0490 0.0543 0.0494 00547 0.0499 0.0551 0.0503

0.0555 0.0508 0.0559 0.0513 0.0563 0.0517 0.0567 0.0522 0.0571 0.0527

0.0575 0.0532 0.0578 0.0535 0.0581 0.0539 0.0585 0.0543 0.0587 0.0546

0.0590 0.0549 0.0591 0.0551 0.0594 0.0554 0.0595 0.0556 0.0595 0.0557 0.0595 0.0558

0.1017 0.0996 0.0986 0.0957 0.0976 0.0934 0.0895 0.0858

0.0823 0.0789 0.0758 0.0727 0.0699

0.0672 0.0647 0.0623 0.0602 0.0581

0.0563 0.0545 0.0530 0.0516 0.0502

0.0490 0.0480 0.0470 0.0461 0.0455

0.0449 0.0444 0.0440 0.0437 0.0435

0.0433 0.0432 0.0433 0.0434 0.0435

0.0437 0.0440 0.0442 0.0446 0.0449

0.0453 0.0458 0.0462 0.0467 0.0472

0.0477 0.0482 0.0487 0.0492 0.0497

0.0502 0.0506 0.0510 0.0515 0.0518

0.0522 0.0525 0.0528 0.0531 0.0532 0.0533

0.0954 0.0945 0.0913 0.0903 0.0874 0.0864 0.0837 0.0827

0.0801 0.0792 0.0768 0.0758 0.0737 0.0727 0.0706 0.0696 0.0678 0.0669

0.0651 0.0626 0.0603 0.0581 0.0561

0.0542 0.0524 0.0509 0.0495 0.0482

0.0470 0.0460 0.0450 0.0441 0.0435

0.0429 0.0424 0.0421 0.0418 0.0415

0.0414 0.0414 0.0414 0.0415 0.0417

0.0419 0.0422 0.0424 0.0428 0.0431

0.0436 0.0440 0.0445 0.0450 0.0455

0.0460 0.0466 0.0471 0.0476 0.0482

0.0487 0.0491 0.0495 0.0500 0.0504

0.0508 0.0511 0.0514 0.0517 0.0518 0.0520

0.0642 0.0617 0.0593 0.0571 0.0551

0.0533 0.0515 0.0500 0.0486 0.0473

0.0461 0.0451 0.0441 0.0432 0.0426

0.0915 0.0874 0.0835 0.0797

0.0762 0.0729 0.0698 0.0667 0.0639

0.0613 0.0588 0.0564 0.0543 0.0523

0.0504 0.0487 0.0471 0.0458 0.0445

0.0433 0.0423 0.0413

0.0420 0.0416 0.0412 0.0409 0.0407

0.0405 0.0405 0.0405 0.0402 0.0408

0.0410 0.0413 0.0416 0.0420 0.0423

0.0428 0.0433 0.0437 0.0442 0.0448

0.0453 0.0458 0.0464 0.0469 0.0474

0.0480

0.0405 0.0398

0.0393 0.0388 0.0385 0.0382 0.0380

0.0379 0.0379 0.0379 0.0381 0.0382

0.0385 0.0388 0.0391 0.0395 0.0399 0.0404 0.0409 0.0413 0.0419 0.0424

0.0430 0.0435 0.0441 0.0447 0.0453

0.0458 0.0484 0.0463 0.0489 0.0468 0.0493 0.0473 0.0497 0.0477

0.0501 0.0481 0.0504 0.0485 0.0508 0.0489 0.0511 0.0492 0.0512 0.0494 0.0514 0.0496

0.0924 0.0909 0.0882 0.0867 0.0841 0.0826 0.0802 0.0787 0.0765 0.0750

0.0729 0.0715 0.0696 0.0682 0.0685 0.0650 0.0634 0.0620 0.0607 0.0592

0.0580 0.0566 0.0555 0.0541 0.0532 0.0518 0.0510 0.0496 0.0490 0.0476

0.0472 0.0458 0.0455 0.0441 0.0440 0.0426 0.0426 0.0412 0.0413 0.0399

0.0401 0.0388 0.0391 0.0378 0.0382 0.0368 0.0374 0.0360 0.0368 0.0354

0.0362 0.0349 0.0358 0.0345 0.0355 0.0341 0.0352 0.0339 0.0350 0.0337

0.0349 0.0336 0.0349 0.0336 0.0350 0.0337 0.0352 0.0339 0.0354 0.0341

0.0357 0.0344 0.0360 0.0347 0.0363 0.0351 0.0367 0.0355 0.0372 0.0359

0.0377 0.0365 0.0382 0.0370 0.0387 0.0375 0.0393 0.0381 0.0398 0.0387

0.0404 0.0393 0.0410 0.0399 0.0416 0.0405 0.0422 0.0412 0.0428 0.0418

0.0434 0.0424 0.0439 0.0429 0.0444 0.0434 0.0450 0.0440 0.0454 0.0444

0.0459 0.0449 0.0463 0.0453 0.0467 0.0458 0.0471 0.0462 0.0473 0.0464 0.0476 0.0467

3 4


TABLE 13.1 h-b VALUES FOR REYNOLDS NUMBER FACTOR, F,, PIPE TAPS (continued) b

Fr=l+ Jh,p,

Internal Diameter of Pipe, di, in.

6 a 10 12

Fd 4.897 5.189 5.761 6.065 7.625 7.981 8.071 9.564 10020 10.136 11.376 11.938 12.090

0.10- 0.0845 0.0836 0.0821 0.0814 0.0784 00778 0.0777 0.0757 0.0752 0.0751 0 0739 0.0734 0.0733 0.11 0.0803 0.0795 0.0779 0.0772 0.0742 0.0736 0.0735 0.0716 0.0710 0.0709 0 0697 0.0692 0.0691 0.12 0.0763 0.0755 0.0739 00732 0.0702 0.0696 0.0695 0.0676 0.0670 0.0669 0.0657 0.0652 00651 0.13 0.0725 0.0717 0.0701 0.0694 0.0664 0.0658 0.0657 0.0638 0.0632 0.0631 00619 0.0614 0.0613 0.14 0.0689 0.0680 0.0665 0.0658 0.0628 0.0622 0.0621 0.0601 0.0596 0.0595 0.0583 0.0576 0.0577

0.15 0.0655 0.0646 0.0631 0.0624 0.0594 00588 0.0587 0.0567 0.0562 0.0561 0.0549 0.0544 0.0543

0.16 0.0623 0.0614 0.0599 0.0591 0.0561 0 0556 0.0554 0.0535 0.0530 0.0528 0.0516 0.0512 0.0510 0.17 0.0592 0.0583 0.0568 0.0561 0.0531 0.0525 0.0524 0.0504 0.0499 0.0498 0.0486 0.0481 0.0480 0.18 0.0563 0.0554 0.0539 0.0532 0.0502 0.0496 0.0495 0.0475 0.0470 0.0469 0.0457 0.0452 0.0451

0.19 0.0536 0.0527 0.0512 0.0505 0.0475 0.0469 0.0468 0.0446 0.0443 0.0442 0.0430 0.0425 0.0424

0.20 0.0511 0.0502 0.0487 0.0479 0.0449 0.0444 0.0442 0.0423 0.0418 0.0416 0.0404 0.0400 0.0398

0.21 0.0486 0.0477 0.0462 0.0455 0.0425 0.0419 0.0418 0.0398 0.0393 0.0392 0.0380 0.0375 0.0374

0.22 0.0464 0.0455 0.0440 0.0433 0.0403 0.0397 0.0396 0.0376 0.0371 0.0370 0.0358 0.0353 0.0352 0.23 0.0444 0.0435 00420 0.0412 0.0382 0.0377 0.0375 0.0356 0.0351 0.0350 0.0338 0.0333 0.0332

0.24 0.0425 0.0416 0.0401 0.0393 0.0364 0.0358 0.0357 0.0337 0.0332 0.0331 0.0319 0.0314 0.0313

0.25 0.0407 0.0399 0.0383 0.0376 0.0346 0.0341 0.0339 0.0320 0.0315 0.0313 0.0301 0.0297 0.0295

0.26 0.0391 0.0382 0.0367 0.0360 0.0330 0.0324 0.0323 0.0303 0.0298 0.0297 0.0265 0.0280 0.0279

0.27 0.0377 0.0368 0.0353 0.0345 00315 0.0310 0 0309 0.0289 0.0264 0.0283 0.0271 0.0266 0.0265 0.28 0.0364 00355 0.0340 0.0332 a.0303 0.0297 0.0296 0.0276 0.0271 0.0270 0.0258 0.0253 0.0252

0 29 0.0352 00343 0.0328 0.0321 0.0291 0.0286 0.0264 0.0265 0.0260 0.0258 0.0246 0.0242 0.0240

0.30 0.0342 0.0333 0.0318 0.0311 0.0281 0.0275 0.0274 0.0255 0.0249 0.0248 0.0236 0.0231 0.0230

031 0.0333 0.0324 0.0309 0.0302 0.0272 00266 0.0265 0.0245 0.0240 0.0239 0.0227 0.0222 0.0221

0.32 0.0325 0.0317 0.0301 0.0294 0.0264 0.0259 0.0257 0.0238 0.0233 0.0232 0.0220 0.0215 0.0214

0.33 0.0319 0.0310 0.0295 0.0288 0.0258 0.0252 0.0251 0.0231 0.0226 0.0225 0.0213 0.0208 0.0207

0.34 0.0314 0.0305 0.0290 0.0283 0.0253 00247 0.0246 0.0226 0.0221 0.0220 0.0208 0.0203 0.0202

0.35 00310 0.0301 0 0286 0.0278 0.0294 0.0243 0.0242 0.0222 0.0217 0.0216 0.0204 0.0199 0.0198

0.36 0.0306 0.0298 0.0283 0.0275 0.0246 0.0240 0.0239 0.0219 0.0214 0.0213 0.0201 0.0196 0.0195

0 37 0.0305 0.0296 0 0281 0.0274 00244 0.0239 0.0237 0.0218 0.0213 0.0212 0.0200 00195 0.0194

0 38 0.0304 0.0285 0 0280 0.0273 0.0244 0.0238 0.0237 0.0217 0.0212 0.0211 0.0199 00195 0.0193

0.39 00305 0.0296 00281 0.0274 0.0244 0.0239 0.0238 0.0218 0.0213 0.0212 0.0200 0.0195 00194

0.40 00306 0.0296 0.0283 0.0276 0.0246 0.0240 0.0239 0.0220 0.0215 0.0213 0.0202 0.0197 0.0196

0.41 0.0309 0.0300 0.0285 0.0276 0.0248 0.0243 0.0241 0.0222 00217 0.0216 0.0204 0.0199 0.0198

0.42 0.0312 0.0303 0 0288 0.0281 0.0252 0.0246 0.0245 00226 0.0221 0.0219 0.0206 0.0203 0.0202 0.43 0.0316 0.0308 0 0293 00286 0.0256 0.0251 0 0249 0.0230 0.0225 0.0224 0.0212 0.0207 0.0206

0.44 00321 0.0313 0.0298 0.0291 0.0261 0.0256 0.0254 0.0235 0.0230 0.0229 0.0217 0.0213 00211

0.45 0.0327 0.0319 0.0304 00297 00267 0.0262 0.0261 0.0241 0.0236 0.0235 0.0223 0.0219 0.0217

0 46 0.0334 0.0326 0.0311 00304 0.0274 0.0269 0.0267 0.0248 0.0243 0.0242 0.0230 0.0226 0.0224

0.47 0.0341 0.0333 0.0318 00311 0.0262 0.0276 0.0275 0.0256 0.0251 0.0249 0.0238 0.0233 0.0232 0 48 0.0350 00341 0.0326 0.0319 0.0290 00284 0.0283 0.0264 0.0259 0.0258 0.0246 0.0242 0.0240

0.49 0.0358 0.0349 00335 0.0328 0.0299 0.0293 0.0292 0.0273 0.0268 0.0267 0.0255 0.0250 0.0249

0.50 0.0368 0.0359 0.0344 0.0337 0.0308 00303 0.0302 0.0283 0.0278 0.0276 0.0265 0.0260 0.0259

0.51 0.0378 0.0369 0 0354 0.0347 0.0318 00313 0.0312 0.0293 0.0288 0.0287 0.0275 0.0270 0.0269

0.52 0.0388 00380 00365 0.0358 0.0329 0.0324 0.0323 0.0304 0.0299 0.029B 0.0286 0.0281 0.0280 0.53 0 0399 0 0391 00376 0.0369 00340 00335 0.0334 0.0315 0.0310 0.0309 0.0297 0.0293 0.0291

0.54 0.0410 0.0402 0 0387 0.0380 0.0351 0.0346 0.0345 00326 0.0321 0.0320 0.0308 0.0304 0.0303

055 0.0422 0.0413 0 0399 0.0392 00363 00358 0.0357 0.0338 0.0333 0.0332 0.0321 0.0316 00315

0.56 0.0433 0.0425 00411 00404 00375 00370 0.0369 0.0350 0.0345 0.0344 0.0333 0.0328 00327

0.57 0 0446 0.0438 0.0423 00416 00368 0.0383 0.0361 0.0363 0.0358 0.0357 0.0346 00341 0.0340 0 58 00458 0.0450 00436 0 0429 00401 0.0395 0.0394 0.0376 0.0371 0.0370 0.0358 00354 00353

0 59 0.0472 0.0463 0 0449 00442 00414 0 0409 0.0408 0.0389 0.0364 0.0363 0.0372 0.0367 0.0366

0 60 0.0484 0.0476 0.0462 0.0455 0.0427 00422 00421 0.0402 0.0398 0.0397 0.0385 00381 00360

0 61 0.0497 0.0489 0.0474 0 0468 0.0440 00435 00433 0.0415 0.0411 0.0409 0.0398 0 0394 0 0393

0.62 0.0510 0.0502 0.0486 00481 0.0402 00448 0.0447 0.0429 0.0424 0.0423 0.0412 00407 00406

0.63 0.0523 0.0515 0.0501 0.0494 00466 00461 0.0460 0.0442 0.0437 0.0436 0.0425 0.0421 00420

0.64 0 0535 00527 0.0513 00507 0 0479 00474 0.0473 0.0455 0.0451 0.0449 0.0438 0.0434 00433

0 65 0.0548 0.0540 0.0526 0.0520 0.0492 00487 0.0486 0.0468 0.0464 0.0463 0.0452 00447 00446

0.66 0.0560 0.0552 0.0538 0.0532 00505 00500 0 0499 0.0481 0.0476 0.0475 0.0465 00480 0 0459

0.67 0.0572 0.0564 0.0551 0.0544 00517 00512 0.0511 0.0494 0.0489 0.0488 00477 00473 00472

0.68 0.0584 0.0576 0.0563 00556 00530 00525 0.0523 0.0506 0.0502 0.0500 0.0490 00486 00484

0.69 0.0595 00587 0.0574 00567 0.0541 0.0536 00535 0.0518 0.0513 0.0512 0.0502 0 0497 0 0496

0.70 0.0606 0 0598 0.0585 0 0579 0.0552 0.0548 00546 0.0529 0.0525 0.0524 00513 0 0509 00508

0.71 0.0616 0.0606 0.0595 0.0589 0.0563 0.0558 0.0557 0.0540 0.0535 0.0534 00524 0.0520 00519

0.72 0.0625 00616 00605 0.0599 0.0573 0.0566 00567 00550 0.0546 0.0545 00535 00530 0 0529

0 73 0.0635 0.0627 00614 0.0608 0.0583 0.0578 00577 0.0560 0.0556 0.0555 0.0545 00541 00540

0 74 0.0643 0.0636 0.0623 0.0617 0 0592 0.0587 0.0586 0.0570 0.0565 0.0564 0.0554 00550 0 0549

0 75 0.0650 0.0643 0.0630 0.0624 0.0599 0 0594 0 0594 0.0577 0.0573 00572 0.0562 00558 0.0557


TABLE 13.1 h-b VALUES FOR REYNOLDS NUMBER FACTOR, F,, PIPE TAPS (continued) tl

F,=1+=


16 20 24 30

F, 14688 15.000 15250 18.814 19.000 19250 22626 23 000 23.250 28.628 29.000

0.0669 0.0627

0.0586 0.0548 0.0511

0.0476 0.0443

0.0413 0.0383 0.0356

0.0330 0.0306

0.0283 0.0262 0.0243

0.0226 0.0210 0.0195

0.0182 0.0171

0.0160 0.0151

00143

0.0137 0.0131

0.0127 00124 0.0122

0.0121 0.0121

0.0122 0.0124

0.0126

0.0129 0.0133

00136 00143

00149

0.0155 0.0161

0.0168 0.0176 0.0183

0.0191 0.0199

0.0207 0.0216

0.0224

0.0233 0.0241

0.0249

0.0257 00265

0.0273 0.0281

0.0288 0.0295

0.0302 0.0308 0.0314 0.0319

29.250

0.0668 0.0627

0.10 0.0706 0.0705 0.0704 0.0690 0.0689 0.0688 0.0680 0.11 0.0665 0.0663 0.0662 0.0648 0.0647 0.0647 0.0638 0.12 0.0624 0.0622 0.0621 0.0607 0.0607 0.0606 0.0597 0.13 0.0565 0.0584 0.0582 0.0569 0.0568 0.0567 0.0558 0.14 0.0549 0.0548 0.0546 0.0532 00531 0.0530 0.0522

0.15 0.0514 0.0512 0.0511 0.0497 0.0497 0.0496 00487

0.16 0.0481 0.0479 0.0478 0.0464 0.0464 0.0463 00454 0.17 0.0450 0.0448 0.0447 0.0433 0.0433 0.0432 00423 0.18 0.0420 0.0419 0.0417 0.0404 0.0403 0.0402 0 0394 0.19 0.0393 0.0391 0.0390 0.0376 0.0376 0.0375 00366

0.20 0.0367 0.0365 0.0364 0.0350 00350 0 0349 0.0340 0.21 0.0343 0.0341 0.0340 0.0326 0.0326 0.0325 00316 0.22 0.0320 0.0318 00317 0.0304 0.0303 00302 0 0294 0.23 0.0299 0.0298 0 0296 0.0263 0.0282 0.0281 00273 0.24 00280 0.0278 00277 0.0264 0.0263 0.0262 00254

0.25 0.0262 0.0261 0.0260 0.0246 0.0246 00245 00236 0.26 0.0246 0.0244 0.0243 0.0230 0.0229 0.0228 0 0220 0 27 0.0231 0.0230 0.0229 0.0215 00215 0.0214 00206

0.28 0.0218 0.0217 00216 0.0202 0.0202 0.0201 00193 0.29 0.0206 0.0205 0.0204 0.0191 0.0190 0.0189 00181

0 30 0.0196 0.0194 0.0193 0.0180 0.0179 00179 00170

0.31 0.0187 0.0185 00184 0.0171 0.0170 0.0170 00161 0.32 0.0179 0.0177 0.0176 0.0163 0.0162 0.0162 00153

0.33 0.0172 0.0170 0.0169 0.0156 0.0156 0.0155 00147 0.34 0.0166 0.0165 0.0164 0.0151 00150 0.0150 00142

0.35 0.0162 0.0161 00160 0.0147 0.0146 0.0145 0.0137 0.36 0.0159 0.0157 0.0156 0.0144 0.0143 0.0142 00134 0.37 0.0157 0.0155 0.0154 0.0141 0.0141 0.0140 00132 0.38 0.0155 0.0154 0.0153 0.0140 0.0140 0.0139 00131 0.39 0.0155 0.0154 0.0153 0.0140 0.0139 0.0139 00131

0.40 0.0156 0.0154 0.0153 0.0141 0.0140 00139 0.0132

0.41 0.0157 0.0155 0.0154 0.0142 0.0142 0.0141 00133

0.42 0.0159 0.0158 0.0157 0.0144 0.0144 0.0143 0.0136 0.43 0.0162 0.0161 0.0160 0.0148 0.0147 0.0146 0.0139 0.44 0.0166 0.0164 0.0163 00151 0.0151 0.0150 00143

0.45 0.0170 0.0169 0.0168 0.0156 0.0155 0.0155 0.0147 0.46 0.0175 0.0174 0.0173 0.0161 0.0160 0.0160 0.0152 0.47 0.0180 0.0179 0.0178 0.0186 00166 0.0165 0.0158 0.48 0.0186 0.0185 0.0184 0.0172 00172 0.0171 0.0164 0.49 0.0192 0.0191 0 0190 0.0178 0.0178 0.0177 0.0170

0.50 0.0199 0.0198 00197 0.0185 0.0185 00184 0.0177

0.51 0.0206 0.0205 00204 0.0193 0.0192 00191 0.0184 0.52 0.0213 0.0212 00211 0.0200 0.0199 0.0199 0.0192 0.53 0.0221 0.0220 0.0219 0.0206 0.0207 0.0207 0.0200 0.54 0.0229 0.0227 0.0226 0.0215 0.0215 0.0214 0.0208

0.55 0.0237 0.0235 00234 0.0224 0.0223 0.0223 0.0216 0.56 0.0244 0.0243 00242 0.0232 00231 0.0231 0.0224

0.57 0.0253 0.0251 00250 0.0240 0.0239 0 0239 0.0232 0.58 0.0261 0.0260 0 0259 0.0248 00248 00247 0.0241 0.59 0.0269 0.0268 00267 0.0256 0.0256 00255 0.0249

0 60 0.0277 0.0276 00275 0.0265 00264 0.0264 0.0257

0 61 0.0284 0.0283 0.0282 0.0272 0.0272 0.0271 0.0265 0.62 0.0292 0.0291 0.0290 0.0280 0.0280 0 0279 00273

0.63 0.0299 0.0298 0.0297 0.0266 0.0287 0.0287 0.0281 0.64 0.0306 0.0305 0.0304 0.0295 0.0295 0.0294 0.0288

065 0.0313 00312 00312 00302 0.0302 00301 0.0295 0.66 0.0320 0.0319 00318 0.0309 0.0308 0.0308 0.0302

0.67 0.0326 0.0325 00324 0.0315 0.0315 00314 0.0309

0.68 0.0332 0.0331 0.0330 0.0322 0.0321 0.0321 0.0315 0.69 0.0337 0.0336 00335 0.0327 00326 0.0326 0.0320 0.70 0.0342 0.0341 00340 0.0332 00332 0.0331 0.0326

0.0879 0.0637

0.0598

0.0678 0.0669 0.0637 0.0628 0.0596 0.0567 0.0557 0.0548 0.0520 0.0511

0.0486 0.0477 0.0453 0.0444 0.0422 0.0413 0.0392 0.0383 0.0365 0.0356

0.0339 0.0330 0.0315 0.0306 0.0293 0.0264 0.0272 0.0263 0.0253 0.0244

0.0235 0.0226 0.0219 0.0210 0.0204 0.0196 0.0191 0.0183 0.0180 0.0171

0.0169 0.0161 0.0160 0.0152

0.0152 0.0144

0.0146 0.0137 0.0140 0.0132

0.0136 0.0128 0.0133 00125 0.0131 0.0123 0.0130 0.0122 0.0130 0.0122

0.0130 0.0122 0.0132 0.0124 0.0134 0.0126 0.0138 0.0130 0.0141 0.0134

0.0146 0.0138 0.0151 0.0143

00157 0.0149

0.0163 0.0155 0.0169 0.0162

0.0176 0.0169 0 0183 0.0176 0.0191 0.0183 0.0199 0.0191 0.0207 0.0199

0.0215 0.0208 0.0223 0.0216

0.0231 0.0224 0.0240 0.0233 0.0248 0.0241

0.0256 0.0250

0.0264 0.0258 0.0272 0.0266

0.0280 0.0273 0.0287 0.0281

0.0294 0.0288 0.0301 0.0295 0.0308 0.0302 0.0314 0.0308 0.0320 0.0314 0.0325 0.0319

0.0586 0.0547

0.0511

0.0476 00443 0.0412

0.0383 0.0355

0.0329 00305

0.0283 0.0262 00243

0.0226 0.0209 0.0195

00182 00170

00160 00151 00143

00136 00131

0.0127 00124

00122

00121 00121

00122 00123

00126

00129 00133

00137 00143

00148

00154 00161

0.0168 0.0175 0.0183

00191 0.0199

0.0207 0.0215

0.0224

0.0232 0.0241

0.0249

0.0257 0.0265

0.0273 0.0280

00288 0.0295 0.0302

0.0308 00313 0.0319

0.0558 0.0521

0.0486 0.0453 0.0423

0.0393 0.0366

0.0340 0.0316 0.0293

0.0272 0.0253

0.0236 0.0219

0.0205 0.0192 0.0180

0.0170 0.0161

0.0153 0.0146 0.0141

0.0137 0.0134

0.0132

0.0130 0.0130

0.0131 0.0132

0.0135

0.0138 0.0142

0.0146 0.0151

0.0157

0.0163 0.0169

0.0176 0.0184 0.0191

0.0199 0.0207

0.0215 0.0223

0.0232 0.0240 0.0248

0.0257

0.0265 0.0272

0.0280 0.0288

0.0295 0.0301

0.0308 0.0315 0.0320 0.0325


TABLE 13.1 i-EXPANSION FACTORS: FLANGE TAPS, Y, ; STATIC PRESSURE TAKEN FROM UPSTREAM TAPS

F,= d.

h, Pfl Ratio

00

01

02

03

0.4

0.5

0.6

07 08

0.9

10

1.1 12

I3

14

1.5

1.6 17

1.8

19

2.0

21 2.2

2.3

2.4

25

26 2.7

28

29

3.0

3.1

32 3.3

3.4

35

3.6 3.7

3.8

3.9 4.0

01

1.0000

0.9989

0.9977

0 9966

0.9954 0.9943

0.9932

0.9920 0.9909

0.9898 0.9886

0.9875

0.9863

0.9852

0.9841 0.9829

0.9618

045 0.60 0.61

Tici- 1.0000

0.2 0.3

1.0000 1.0000

09989 0 9989

0.9977 0.9977

0 9966 0.9966

0.9954 0.9954 0.9943 0 9943

0.9932 09931

0.9920 0 9920 0.9909 0.9908

0.9897 0.9897 0.9886 0.9885

0.9875 0.9874

09863 0.9862

0.9852 0.9851

0.9840 0.9840 0.9829 0.9828

0.9818 0.9817

0.9806 0.9805

0.4 0 50 0.52

0.9568

1.0000

0.9565

1.0000

09988 0.9988

0.9976 0.9976 0 9964 0 9964

0.9952

0.9556

0.9952

0.9553

0.9940 0.9940

0.9928 0 9927

0.9916 0 9915 0.9904 0.9903

0.9892 0.9891 0.9880 0.9879

0.9868 0 9867

0.9856 0 9855

0 9844 0.9843

0.9832 0.9831 0.9820 0.9819

0.9808 0.9806

0.9796 0.9794

0.9784 0.9782 0.9772 0.9770

0 9760 0.9758

0.9748 0.9746

0.9736 0.9734

0.9724 0.9722 0.9712 0.9710 0.9700 0.9698

0.9688 0.9686

0.9676 0.9673 0.9664 0.9661

0.9652 0.9649

0.9640 0.9637

0.9628 0.9625

0.9616 0.9613 0.9604 0.9601

0.9592 0.9589

0.9580 0.9577

0.54 0 56

1.0000

0.9988

0.58

1 .oooo

0.9988

0.9975

0.9963

0.9950 0.9938

0.9925

0.9913 0.9900

062

1.0000

0.9987

0 9974

0.9962

0.9949 0 9936

0 9923

0.9910 09897

09885 0 9872

0 9859 0 9846

09833

0.9821 09808

0 9795

09782

0.9769 0 9756 0.9744

0.9731

0.9718

0.9705

0.9692

0.9680

0.9581

1 0000

09988

0.9977

0 9965

0 9953 0 9942

0 9930

09919 0 9907

0.9570

0.9895 0 9884

09872 09860

0.9849

0.9837 09826

09814

09802

09791

0 9779

0.9767

0 9756

0 9744

0 9732 09721

0 9709

09698

0 9686 0 9674

0.9663

0.9651

0.9639

0.9628 0.9616

0.9604

0 9593

1.0000

0.9988

0.9976

0 9965

0.9953 0.9941

0.9929

0.9918 0.9906

0.9894 0.9882

0.9870 0 9859

0.9847

0.9835 0.9823

0.9811 0.9800

0.9788

0.9776

0.9764

0.9753

0.9741

0.9729 0.9717

0.9705

0.9694

0.9682

0 9670

1.0000

0.9988 0.9987 0.9987

0.9975 0.9975

0.9962 0.9962

0.9949 0.9949 0.9937 0.9936

0.9924 0.9924

0.9912 0.9911

0.9899 0.9898

0.9886 0.9885 0 9874 0.9873

0.9838

0.9861

0.9888

0.9836

0.9825

0.9848

0.9875

0.9823 0.9813

0.9863

0.9810

0.9800

0.9551

0.9850

0.9798

0.9545

0.9788 0.9785

0.9775 0.9772

0.9763 0.9760 0.9750

0.9538

0.9747

0.9532

0.9738 0.9734

0.9725 0.9722

0.9713 0.9709 0.9700 0.9897 0.9688 0.9684

0 9675 0.9671

0.9663 0.9659

0.9650 0.9646

0.9838 0.9633 0.9626 0.9621

0.9613 0.9608

0.9601 0.9595 0.9588 0.9583

0.9576 0.9570 0.9563 0.9558

0.9835

0.9860

0.9822

0.9809

0.9847

0.9796

0.9784

0.9771

0.9758 0 9745 0 9733 0.9720 0.9707 0.9694 0.9682

0.9669

0.9656 0 9644

09831 0.9618

0.9605

0.9593

0.9580

0.9567 0 9554

0 9542

0.9529

09516

0.9504 0.9491

0.9976

0.9951 0.9939

0.9963

0.9927

0.9915

0.9560

0.9902

0.9890 0.9878

0.9866

0.9853

0.9841

0.9829 0.9817

0.9548

0.9805 0.9792

0.9780 0 9768

0.9756

0.9744

0.9731

0.9719

0.9707 0.9695

0.9683

0.9670

0 9658

0.9646 0.9634

0.9622

0.9609

0.9597

0.9585

0 9573

0.9951

0.9975

0.9938

0.9926

0.9963

0.9914 0.9901

0.9889

0.9840

0.9827

0 9877

0.9815

0.9803

0.9790

0.9864

0.9778

0.9556

0.9766

0 9852

0.9753

0.9741

0.9729

0.9716

0.9704 0.9692

0.9679

0.9543

0.9667

0.9654

0.9642 0.9630 0.9617

0.9605 0.9593

0.9580 0.9568

0.9795 0.9795 0.9794 09784 09783 09782 0.9772 0.9772 09771

09761

0.9750

0.9738 0.9727 0.9715

0.9704

0.9693

0.9681

0.9670 0.9658

0.9847

0.9635

0.9624

0.9613 0.9602

0.9761

0.9749

0.9738

0.9726 0.9715

0.9704

0 9759 09748

0.9736

0 9725

09713

0.9702

0.9691

0.9679

0.9668

0.9656

0.9645

0.9633 0.9622

0.9610

0.9599

0.9587

0.9576

0.9564

0.9553 0.9542

0.9667

0.9654 0.9692 0.9681 0.9641

09628 0.9615

0.9603

0.9590 0 9577

0.9669

0.9658

0.9647

0 9635

0.9624

0.9612 0.9601

0.9658

0.9647

0.9576

0.9635

0.9623 0.9611

0.9599

0 9588

0.9564

0.9564 09551

0.9590 0.9590 0.9579 0.9578

0.9538

0.9526

0.9513

0.9500 09487

0.9567 0.9567 0.9556 0.9555 0.9545 0.9544

0.9558 0.9552 0.9544 0.9540 0.9536 0.9531 0.9526 0.9546 0.9540 0.9532 0.9528 0.9524 0.9519 0.9513 0.9535 0.9529 0.9520 0.9516 0.9512 0.9506 0.9501

0.9520 0.9507 0.9494

GAS MEASUREMENTAND REGULATION 13-27

TABLE 13.li-EXPANSION FACTORS: FLANGE TAPS, Y,; STATIC PRESSURE TAKEN FROM UPSTREAM TAPS (continued)

h, PO

Ratto

0.0 0.1 0.2

03

0.4

0.5

0.6 0.7

0.8

0.9

1.0

11 1.2

1.3

1.4

15

16 1.7

1.8 1.9

2.0

2.1

2.2

2.3

2.4

25

2.6 2.7

2.8

2.9

3.0

3.1 3.2

3.3

3.4

35

3.6 3.7

3.8

39

4.0

1.0000

0.9987

0.9974

09961

0.9948 0.9934

09921

0.9908 0.9895

0.9882 0.9869

0.9856 0.9843 0.9829

0.9816 0.9803

0.9790 0.9777

0.9764 09751

0.9738

0.9725

0.9711

0.9698

0.9685

0.9672

0 9659

0 9646

0.9633

0.9620 0.9606

0.9593

0 9580

0.9567

0.9554

09541

0.9528

0.9515

0.9502

0.9488 0.9475

0.65 0.66 0.67 0.68 0 69 0.70

1.0000

0.9986

0.9973

0.9959

0.9945 0.9931

0.9918

0.9904 0.9890

0.9877

0.9863

0.9849

0.9835

0.9822

0.9808 0.9794

0.9781 0.9767

0.9753

0.9739

0.9726

0.9712

0.9698

0.9685

0.9671

0.9657

0.9643

1.0000

0.9987

0.9974

0.9960

0.9947 0.9934

0.9921

0.9907 0.9894

0.9881 0.9868

0.9854 0.9841

0.9828

0.9815 0.9802

0.9788

0.9775

0.9762

0 9749

0.9735

0.9722

0.9709

0.9696

0.9683

0.9669

0.9656

0.9643 0.9630 0.9616

0 9603

0.9590

0.9577 0.9564

0.9550 0.9537 0.9524 0.9511

0.9497

0.9484 0.9471

1.0000

0.9987

0.9973

0 9960

0.9947 0 9933

0.9920

0 9907 0.9893

0.9880 0 9867

0 9853

0 9840

0.9827

0.9813 0.9800

0.9787 0.9773

0 9760

0 9747

0 9733

0 9720

0.9706

0.9693

0.9680

0.9666

0 9653

0.9640

0.9626

0.9613

0.9600

0.9586

0.9573

0.9560

0.9546 0 9533

0.9520 0.9506 0.9493 0.9480 0.9466

10000 1 .oooo

0.9987 0.9986

0 9973 0.9973

0 9960 0 9959

0 9946 0.9946 0.9933 0.9932

0.9919 0.9918

0.9906 0.9905 0.9892 0.9891

0.9879 0.9878

0.9865 0.9864

0.9852 0.9851

0 9838 09837

0.9825 0.9823

0.9812 0.9810 0 9798 0.9796

0.9785 0.9783

0.9771 0.9769

0.9758 0.9755 0 9744 0 9742

0.9731 0.9728

09717 0.9715

09704 0.9701

09690 0.9688

0 9677 0.9674

0.9663 0.9660

0 9650 0.9647

0 9637 0.9633 0 9623 0.9620

09610 0 9606

0 9596 0.9592

09583 0.9579

0 9569 0.9565 0 9556 0.9552

0.9542 0.9538 0 9529 0.9524

0.63

1.0000

0.9987 0.9974

0.9961

0.9948

0.9935

0.9923 0.9910

0.9897

0.64

1.0000

0.9987

0.9974

0.9961

0.9948 0.9935

0.9922

0.9909 0.9896

0.9883 0.9870

0.9857

0.9844

0.9831

0.9818 0.9805

0.9792

0.9779

0.9766

0.9753 0.9740

0.9727

0.9714

0.9701

0.9688 0.9675

0.9662

0.9649

0.9636

0.9623 0.9610

0.9597

0.9584

0.9571

0.9558 0.9545

0.9532

0.9518

0.9505

0.9492 0.9479

0.71 0.72

1.0000 1.0000

0.9986 0.9986

0.9972 0.9972

0.9958 0 9958

0.9945 0.9944 0.9931 0.9930

0.9917 0.9916

0.9903 0.9902 0 9869 0.9888

0.9875 0.9874 0.9861 0.9860

0.9848 0.9846 0 9834 0.9832

0 9820 0.9818

0.9806 0.9804 0.9792 0.9790

0.9778 0.9776 0.9764 0.9762

0.9751 0.9748 0.9737 0.9734

0.9723 0.9720

0.9709 0.9706

0.9695 0.9692

0.9681 0.9678

0.9668 0.9664

0.9654 0.9650

0.9640 0.9636

0.9626 0.9622 0.9612 0.9608

0.9598 0.9594 0.9584 0.9580

0.9571 0.9566

0.9557 0.9552

0.9543 0.9538

0.9529 0.9524 0.9515 0.9510

0.9501 0.9496

0.9487 0.9482 0.9474 0.9468

0.9460 0.9454 0.9446 0.9440

0.73

1.0000

0.9986

0 74

10000

0.9986

0.9971

0.9957

0.75

1.0000

0.9986 0.9971

0.9957

0 9972 0.9958 0.9943 0.9929 0.9915

0.9901

0.9887

0.9873 0.9859

0.9844

0 9830

0.9816

0.9802 0 9788

0.9774

0.9760

0.9745 09731

0.9717

0.9703

0.9689

0.9675 0.9661

0.9646

0.9632 0.9618

0 9604

0.9590 0.9576

0.9562 0.9547

0.9943 0.9929

0.9914

0.9900 0.9886

0.9871 0.9857

0.9843

0.9828

0.9814

0.9800 0.9786

0.9771

0.9757

0.9743

0.9728 0.9714

0.9700

0.9685

0.9671

0.9657

0.9643

0.9628 0.9614

0 9600

0.9585 0.9571

0.9557

0.9542

0.9528 0.9514 0 9500

0.9485 0 9471

0.9457

0 9442 0 9428

0 9942 0.9928

09913

0.9899 0 9884

0.9870 0 9855

0.9884 0.9871

0.9858 0.9845

0.9841

0.9826

0 9832 0.9819

0.9806

0.9793 0.9780 0 9768 0 9755 0.9742

0.9729 09716

0.9703

0.9690

0.9677

0.9664 0.9651

0.9638

0.9625

0.9613

0.9600 0.9587 0.9574

0.9561

0.9548

0.9535 0.9522

0.9509

0.9496 0.9483

09812

0.9798 0.9783

0.9769 0.9754

0 9740

0 9725 0.9711

0.9696

0.9682

0.9667

0.9653 0.9639

0.9624 0.9610

0.9595

0.9581

0.9566

0.9552

0.9537 0.9523

0.9508 0.9494

0.9480

0.9465

0.9451

,0.9436 0.9422

0.9630 0.9616

0.9602 0.9588 0.9575 0.9561 0.9547

0.9534 0.9520

0.9533 0.9519 0 9505

09515 0.9511 0.9506 0 9502 0.9497 0.9492

0.9491

0.9477

0.9463

0.9448 0.9434

0 9488 0.9484 0.9479 0 9475 0.9470 0.9465 0.9462 0.9457 0.9451


TABLE 13.lj-EXPANSION FACTORS: PIPE TAPS, Y, ; STATIC PRESSURE TAKEN FROM UPSTREAM TAPS

h, Pi1 Ratio 0.45

1.0000

0.50

1 .oooo

0.9982 0.9964 0.9946

0.52

1.0000

0.9981 0.9962 0.9944 0.9925 0.9906

0.54

1.0000

0.56

1 .cmoo

0.9979 0.9959 0.9938 0.9917 0.9897

0.58 0.60

1.0000 1.0000

0.9978 0.9977 0.9957 0.9954

0.2

1.0000

0.3

1.0000

0.9988 0.9976

0.4

1.0000

0.1

1.0000 0.0

0.1 0.2 0.3 0.4 0.5

0.6 0.7 0.8 0.9 1.0

1.1 1.2 1.3 1.4 1.5

1.6 17 1.8 1.9 2.0

2.1 2.2 2.3 2.4 2.5

2.6 2.7 2.8 2.9 3.0

3.1 3.2 3.3 3.4 3.5

3.6 3.7 3.8 3.9 4.0

0.9990 0.9989 0.9981 0.9979 0.9971 0.9968 0.9962 0.9958

0.9985 0.9984 0.9971 0.9968 0.9956 0.9952 0.9942 0.9936

0.9980 0.9961 0.9941 0.9921

0.9964 0.9951

0.9935 0.9931 0.9913 0.9908 0.9928

0.9910

0.9892 0.9874 0.9857 0.9839

0.9952 0.9947 0.9939 0.9943 0.9937 0.9927 0.9933 0.9926 0.9915 0.9923 0.9916 0.9903 0.9914 0.9905 0.9891 0.9904 0.9895 0.9878

0.9895 0.9884 0.9866 0.9885 0.9874 0.9854 0.9876 0.9863 0.9842 0.9866 0.9853 0.9830 0.9857 0.9842 0.9818

0.9847 0.9832 0.9805 0.9837 0.9821 0.9793 0.9828 0.9811 0.9781

0.9927 0.9913 0.9898

0.9919

0.9903 0.9887 0.9871 0.9855 0.9839

0.9823 0.9807 0.9791 0.9775 0.9758

0.9742

0.9902 0.9891 0.9885 0.9862 0.9840 0.9817 0.9794

0.9887

0.9812

0.9794 0.9775 0.9756

0.9869

0.9737 0.9719

0.9700 0.9681

0.9850

0.9662 0.9643 0.9625

0.9606

0.9831

0.9587 0.9568 0.9550 0.9531

0.9512 0.9493 0.9475 0.9456 0.9437

0.9418 0.9400 0.9381 0.9362 0.9343

0.9324 0.9306 0.9287

0.9803

0.9882

0.9794

0.9876

0.9783

0.9784

0.9870

0.9773

0.9862

0.9761 0.9764

0.9856

0.9752

0.9848

0.9739

0.9843

0.9744

0.9835

0.9732

0.9826

0.9718 0.9725

0.9823

0.9711

0.9814

0.9696 0.9705

0.9805

0.9690 0.9674

0.9685 0.9670 0.9652

0.9883 0.9869 0.9854 0.9840 0.9825 0.9811 0.9796 0.9782

0.9767

0.9821

0.9803 0.9785 0.9767 0.9749 0.9731

0.9713

0.9771

0.9748 0.9725 0.9702 0.9679 0.9656 0.9633 0.9610 0.9587

0.9752 0.9738 0.9723 0.9709

0.9726 0.9895 0.9710 0.9677

0.9666 0.9646

0.9649 0.9628 0.9608 0.9587

0.9631 0.9609

0.9818 0.9809

0.9799 0.9790

0.9800 0.9769 0.9790 0.9757

0.9694 0.9678 0.9662 0.9646 0.9630 0.9613 0.9597

0.9581 0.9565 0.9549 0.9533 0.9517

0.9501

0.9659 0.9641

0.9623 0.9605 0.9587 0.9570 0.9552

0.9534 0.9516 0.9498 0.9480 0.9462

0.9444 0.9426 0.9408 0.9390 0.9372

0.9354 0.9336 0.9318

0.9626 0.9607 0.9587 0.9567 0.9548 0.9528 0.9508

0.9489 0.9469 0.9449 0.9430 0.9410

0.9390

0.9587 0.9566

0.9544 0.9522 0.9500 0.9479 0.9457

0.9435 0.9414 0.9392 0.9370 0.9348

0.9327

0.9565 0.9542

0.9779 0.9745 0.9694 0.9768 0.9732 0.9680 0.9758 0.9720 0.9665 0.9747 0.9708 0.9650 0.9737 0.9696 0.9636 0.9726 0.9684 0.9621 0.9716 0.9672 0.9607 0.9705 0.9659 0.9592 0.9695 0.9647 0.9578 0.9684 0.9635 0.9563

0.9674 0.9623 0.9549

0.9566 0.9546 0.9525 0.9505 0.9484

0.9463 0.9443 0.9422 0.9401 0.9381

0.9360

0.9519 0.9496 0.9473 0.9450 0.9427

0.9404 0.9381 0.9358 0.9335 0.9312

0.9290

0.9780 0.9770 0.9761

0.9751 0.9742 0.9732 0.9723 0.9713

0.9704 0.9694 0.9684

0.9663 0.9611 0.9534 0.9485 0.9653 0.9599 0.9519 0.9469 0.9642 0.9587 0.9505 0.9452 0.9632 0.9574 0.9490 0.9436

0.9675 0.9665 0.9656 0.9621 0.9562 0.9476 0.9646 0.9611 0.9550 0.9461 0.9637 0.9600 0.9538 0.9447 0.9627 0.9590 0.9526 0.9432 0.9617 0.9579 0.9514 0.9417

0.9420 0.9404 0.9388 0 9372 0.9301 0.9268 0.9233 0.9356 0.9283 0.9249 0.9213

0.9371 0.9351

0.9339 0.9319

0.9305 0.9267 0.9283 0.9244 0.9261 0.9221 0.9240 0.9198

0.9331 0.9312

0.9292 0.9272 0.9253

0.9298 0.9277 0.9257 0.9236 0.9216 0.9195 0.9174

0.9218 0.9175 0.9196 0.9152 0.9175 0.9129 0.9153 0.9106 0.9131 0.9083


TABLE 13.1j-EXPANSION FACTORS: PIPE TAPS, Y,; STATIC PRESSURE TAKEN FROM UPSTREAM TAPS (conth

h 2 Pfl Ratio

0.0

0.1 0.2 0.3 0.4 0.5

0.6 0.7 0.8 0.9 1.0

1.1 1.2 1.3 1.4 1.5

1.6 1.7 1.8 1.9 2.0

2.1 2.2 2.3 2.4 2.5

2.6 2.7 2.8 2.9 3.0

3.1 3.2 3.3 3.4 3.5

3.6 3.7 3.8 3.9 4.0

0.61 0.62 0.63 0.64 0.65

1.0000 1.0000 1.0000 1.0000 1.0000

0.9976 0.9976 0.9975 0.9974 0.9973 0.9953 0.9951 0.9950 0.9948 0.9947 0.9929 0.9927 0.9925 0.9923 0.9920

0.66 0.67 0.68 0.69 0.70 -- 1.0000 1.0000 1.0000 1.0000 1.0000

0.9972 0.9971 0.9970 0.9969 0.9945 0.9943 0.9941 0.9938 0.9917 0.9914 0.9911 0.9907

0.9968 0.9935 0.9903 0.9871 0.9839

0.9806 0.9774

0.9906 0.9903 0.9900 0.9897 0.9893 0.9890 0.9886 0.9881 0.9876 0.9882 0.9879 0.9875 0.9871 0.9867 0.9862 0.9857 0.9851 0.9845 0.9859 0.9854 0.9850 0.9845 0.9840 0.9834 0.9828 0.9822 0.9835 0.9830 0.9825 0.9819 0.9813 0.9807 0.9800 0.9792 0.9811 0.9806 0.9800 0.9794 0.9787 0.9779 0.9771 0.9762

0.9814 0.9784 0.9753 0.9722 0.9691

0.9660 0.9629

0.9742 0.9710 0.9677

0.9645 0.9613 0.9581 0.9548 0.9516

0.9484

0.9788 0.9764 0.9741 0.9717 0.9694 0.9670

0.9782 0.9775 0.9757 0.9750 0.9733 0.9725 0.9709 0.9700

0.9768 0.9742

0.9760 0.9733 0.9707 0.9680

0.9752 0.9724

0.9696 0.9669 0.9641 0.9614

0.9742 0.9714

0.9685 0.9657

0.9733 0.9703

0.9673 0.9643 0.9614 0.9584

0.9716 0.9690 0.9664 0.9639 0.9613

0.9587 0.9561 0.9535 0.9510 0.9484

0.9458

0.9685 0.9675 0.9660 0.9650 0.9636 0.9825

0.9612 0.9600 0.9587 0.9575 0.9563 0.9550 0.9539 0.9525 0.9515 0.9500 0.9490 0.9475

0.9653 0.9627 0.9600

0.9573 0.9547 0.9520 0.9493 0.9467

0.9440

0.9628 0.9599 0.9571

0.9542 0.9514 0.9485 0.9456 0.9428

0.9399 0.9371 0.9342

0.9598 0.9567

0.9646 0.9623 0.9599 0.9576 0.9552 0.9529

0.9505

0.9586

0.9558 0.9531 0.9503 0.9476 0.9448 0.9420

0.9554 0.9536

0.9525 0.9505 0.9495 0.9474 0.9465 0.9443 0.9435 0.9412 0.9406 0.9381

0.9376 0.9351 0.9346 0 9320 0.9317 0.9289

0.9452 0.9419 0.9387 0.9355

0.9323 0.9290 0.9258 0.9226 0.9194

0.9161 0.9129

0.9258 0.9227

0.9196 0.9165 0.9134 0.9097 0.9103 0.9064

0.9481 0.9458 0.9434 0.9411

0.9387 0.9364 0.9340 0.9316 0.9293

0.9269 0.9246 0.9222 0.9199 0.9175

0.9151

0.9466 0.9450 0.9432 0.9442 0.9425 0.9406 0.9418 0.9400 0.9381 0.9393 0.9375 0.9355

0.9369 0.9350 0.9329 0.9345 0.9325 0.9303 0.9321 0.9300 0.9277 0.9296 0.9275 0.9252 0.9272 0.9250 0.9226 0.9248 0.9225 0.9200 0.9223 0.9200 0.9174 0.9199 0.9175 0.9148 0.9175 0.9150 0.9122 0.9151 0.9125 0.9097

0.9126 0.9100 0.9071 0.9102 0.9075 0.9045 0.9078 0.9050 0.9019 0.9054 0.9025 0.8993 0.9029 0.9000 0.8968

0.9413 0.9393 0.9387 0.9365 0.9360 0.9338 0.9333 0.9310

0.9313 0.9285

0.9287 0.9257 0.9227 0.9198

0.9307 0.9280 0.9253 0.9227 0.9200 0.9173 0.9147 0.9120 0.9093 0.9067

0.9040 0.9013 0.8987 0.8960 0.8933

0.9282 0.9255 0.9227 0.9200 0.9172

0.9144 0.9117 0.9089 0.9062 0.9034

0.9006

0.9256 0.9227 0.9199 0.9170 0.9142

0.9113 0.9084 0.9056 0.9027 0.8999

0.8970 0.8941 0.8913 0.8884 0.8856

0.9168 0.9138 0.9108

0.9079 0.9049 0.9019 0.8990 0.8960

0.8930

0.9072 0.9041 0.9010 0.8979 0.8948 0.8918

0.8887 0.8856 0.8825 0.8794 0.8763

0.9032

0.9000 0.8968 0.8935 0.8903 0.8871

0.8839 0.8806 0.8774 0.8742 0.8710

0.9128 0.9104 0.9081 0.9057

0.8979 0.8951 0.8924 0.8896

0.8900 0.8871 0.8841 0.8811


TABLE 13.lk-EXPANSION FACTORS: FLANGE TAPS, Y,; STATIC PRESSURE TAKEN FROM DOWNSTREAM TAPS

h, PI2 Ratlo

0.0

0.1 0.2

0.3

0.4

0.5

0.6 0.7

08

0.9

1.0

1.1 1.2

1.3

1.4

1.5

1.6 17

1.8

1.9

2.0

2.1 2.2

2.3

2.4

2.5

2.6 2.7

28

2.9

30

3.1 3.2

33

3.4

3.5

3.6

3.7 3.8

3.9 4.0

0.45

1.0000

1.0006

1.0013

1.0019

1.0026

1.0031

1.0038

1.0044

1.0050

1.0057 1.0063

1.0069

1.0075

1.0082

1.0088

1.0094

1.0101

1.0108

1.0114

1.0120 1.0126

1.0133

1.0139

1.0146

1.0151

1.0158

1.0164

1.0172

1.0176

1.0183

1.0189

1.0195

1.0201

1.0208

1.0047

10053

0.2

1.0060

1.0000

1.0066

1.0073

10007

1.0060

1.0087

1.0013

10093 1.0100

1.0020

1.0107 10114

1.0027

1.0120

1.0127

10033

1.0133

1.0140

1.0040

1.0147

1.0154

10160 1.0167

1.0173

1.0182

10186

10194

1.0200

1.0206

1.0213

10220

10227

1.0233

1.0239

1.0246 1.0252

1.0259 10265

0.3

1.0000

1.0007

1.0013

1.0020 1.0027

1.0033

1.0040

1.0047

1.0053

1.0060

1.0066

1.0073

1.0079

1.0086

1.0093

1.0099

0.4

1.0000

10006

1.0013

1.0020

1.0026 1.0032

1.0039

1.0045

1.0051

1.0058 1.0064

1.0071

1.0077

1.0084

1.0090 1.0097

1.0104

1.0110

1.0117

1.0123 1.0129

1.0136

1.0142

1.0149

0.50

1.0000

1.0006

1.0012

1.0018

1.0025 1.0030

1.0036

1.0043

1.0049

1.0055 1.0061

1.0067

1.0073

1.0080

1.0086 1.0091

0.52

1 .oooo

1.0006

1.0012

1.0018 1.0024

1.0029

1.0036

1.0042

1.0048

1.0054

1.0060

1.0066

1.0072

1.0078

1.0084 1.0090

1.0097

1.0103

1.0108

1.0115 1.0121

1.0127

1.0133

1.0139

0.54 0.56 0.58 0.60 0.61 0.62

1.0000 1.0000 1.0000 1 .oooo 1.0000 1.0000

1.0006 1.0006 1.0006 1.0005 1.0005 10005

1.0012 1.0012 1.0011 1.0011 1.0011 1.0011

1.0018 1.0017 1.0017 1.0016 1.0016 1.0016 1.0024 1.0023 1.0023 1.0022 1.0022 1.0021 1.0029 1.0028 1.0026 1.0027 1.0027 1.0027

1.0035 1.0034 1.0033 1.0032 1.0032 1.0031

1.0047

0.1

10053

1.0000

1.0060 1.0067

10007

1.0074

1.0080

1.0013

1.0087

1.0094

1.0020

1.0100

1.0106

1.0027

10114

1.0121

1.0033

10126 1.0134

1.0040

1.0140

1.0147

10154

10160 1.0167

10174

10183

1.0187

1.0194

1.0200

1.0207

10213

1.0220

1.0227

10233

10240

10246

10252

1.0041

1.0047

1.0053 1.0059

1.0065

1.0071

1.0077

1.0083

1.0088

1.0095 10101

1.0106

1.0112 1.0118

1.0124

1.0130

1.0136 1.0141

1.0148

10154

1.0161

1.0165 1.0172

1.0177

1.0183

1.0189

1.0195

1.0201

1.0207

1.0212

1.0217 1.0223

1.0040 1.0046

1.0052

1.0058

1.0063

1.0069

1.0075

1.0081

1.0086

1.0093

1.0099

1.0104

1.0110 1.0116

1.0122

1.0128

1.0133

1.0138

1.0145

1.0151

1.0158

1.0162

1.0168

1.0173

1.0179

1.0185

1.0191

1.0197

1.0202

1.0208

1.0213 1.0219

1.0225 1.0231

.0039 1.0036

.0045 1.0044

.0050 1.0049

.0056 1.0055

.0061 1.0060

.0067 1.0066

.0073 1.0071

.0079 1.0077

.0054 1.0082

1.0090 1.0096

1.0102

1.0107 1.0113

1.0119

1.0125

1.0130

1.0135

1.0141

1.0147

1.0154

1.0158

1.0164

1.0169

1.0175

1.0180

1.0186

1.0192

1.0198

1.0203

1.0208 1.0214

1.0220 1.0225

1.0038

10043

1.0046 1.0054

1.0059

1.0065

1.0070

1.0076 1.0081

1.0066

10092

1.0097

1.0103 10108

10114

1.0120

1.0125

10130 1035

10141

1.0148

1.0152

10157 1.0162

1.0168

1.0173

1.0178

1.0184

1.0190

1.0195

1.0200

1.0206

1.0211 1.0217

1.0037

10043

1.0048 1.0053

1.0056

1.0064

1.0069

1.0074 1.0080

10085 10091

1.0096

1.0101

1.0106

1.0111

1.0118

1.0123

10128

1.0133

1.0139

1.0146

1.0149

1.1055

1.0160

1.0166

1.0171

1.0176

1.0182

1.0187

1.0192

1.0197

1.0203

1.0208 1.0213

10106 1.0113

1.0120

1.0126 1.0132

1.0139

1.0146

1.0153

10159

1.0166

1.0172

10181

1.0185

1.0192 1.0198

1.0205

1.0097 1.0104

1.0110

1.0088

10094

10100

1.0104 1.0110

1.0115 1.0121

1.0127

10133

1.0137

1.0143

1.0150

1.0154

1.0160

1.0165

1.0116 1.0122

1.0129

1.0135

1.0141

1.0154

1.0162

1.0146 1.0144

1.0153 1.0150

1.0168

1.0176

1.0180 1.0187

1.0193

1.0200

1.0206

1.0213

1.0219

1.0225

1.0232

1.0238 1.0244

1.0159

1.0167

1.0171

1.0177

1.0183

1.0156

1.0164

1.0168 1.0175

1.0180

1.0186 1.0189

1.0195

1.0202

1.0171

1.0176 1.0211

1.0218

1.0192

1.0199 1.0181

1.0187

1.0192

1.0198

1.0203

1.0209

10214 10220

10225

1.0231

1.0237

1.0244 1.0250

1.0257 1.0263

1.0214 1.0208 1.0205

1.0220 1.0214 1.0210

1.0227 1.0220 1.0216

1.0232 1.0225 1.0221 1.0238 1.0231 1.0227

1.0250 1.0245 1.0237 1.0234 1.0229 10256 10251 10243 1.0240 1.0235

10259 10265


TABLE 13.lk-EXPANSION FACTORS: FLANGE TAPS, Y,; STATIC PRESSURE TAKEN FROM DOWNSTREAM TAPS (continued)

h, P12 Ratio

0.0

0.1

0.2 0.3

0.4

0.5

0.6 0.7

0.8

0.9

1.0

1.1 1.2

13 1.4

1.5

1.6 1.7

1.8

1.9

20

2.1 2.2

23

24

25

2.6 2.7

2.8

2.9

3.0

3.1 3.2

3.3

3.4

3.5

36

3.7

38

3.9

4.0

0.68 0.69

1.0000- 1.0000

0.70 0.72 0.73 0.74 0.75 0.63

10000

10005

1.0010 10016

1.0021

10026

1.0031 10036

10042

10047

10052

1.0057 10062

10067

0.64

1.0000

1.0005

1.0010

1.0015

1.0021 1.0026

1.0030

1.0036

1.0041

1.0046 1.0051

1.0056 1.0061

1.0066 1.0072 1.0077

10081

1.0087

1.0092

1.0097 1.0102

1.0108

1.0114 1.0119

10124

1.0128

1.0134

1.0141 1.0144

1.0150 10155

0.65

1.0000

1.0005

1.0010

1.0015

1.0020 1.0025

1.0030

0.67

1.0000

1.0005

1.0010 1.0014

1.0019 1.0024

1.0028

1.0033

1.0038

1.0042

1.0048

1.0053

1.0057

10062

10067 1.0072

10076

1.0082

1.0087

10091 1.0096

1.0101

1.0107

0.71

1.0000

1.0004

1.0008

1.0013

1.0017

1.0022

10025

1.0030

1.0034

1.0039

1.0043

1.0047 1.0051

1.0056

1.0060

1.0065

10069

10074

1.0078

1.0082 10087

1.0091

1.0095 10100

1.0105

1.0109

10113

1.0119

1.0122

1.0127

10131

1.0000

1.0004

1.0008

1.0012

1.0017 1.0021

1.0024

10029

1.0033

1.0037 1.0041

1.0046

10050

10054

10058 10063

10066

10071

1.0075

1.0079 10084

10088

1.0094 10097

10102

10106

10110

10115

1.0118

1.0000

1.0004

1.0006

1.0012

1.0016

1.0020

1.0023 1.0028

1.0032

1.0036 1.0040

1.0044

1.0048

1.0052

10056

1.0060

1.0064

1.0069

1.0073

1.0077

1.0081

1.0085

1.0035

1.0040

1.0000

1.0005

1.0010

1.0015

1.0020

1.0025

1.0029

1.0034

1.0039

1.0043

1.0049

1.0054

1.0059

1.0064

1.0069 1.0074

1.0045 1.0050

1.0055

1.0060

1.0065

1.0070

1.0075

1.0080 10078

10086 1.0084

1.0000 1.0000 1.0000

1.0004

1.0007 1.0011

1.0015 1.0019

1.0005 1.0004

1.0009 1.0009

1.0004

1.0009

1.0004

1.0008 1.0011

1.0015 1.0019

1.0022

10026

10031

1.0035 1.0039

1.0042 10046

10050

10054

1.0058

10062

1.0066

1.0070 1.0074

10078

10082

1.0014 1.0014 1.0013

1.0019 1.0018 1.0018 1.0024 1.0023 1.0022

1.0027 1.0027 1.0026

1.0032 1.0031 1.0031

1.0037 1.0036 1.0035 1.0041 1.0040 1.0039

1.0046 1.0045 1.0044

1.0051 1.0050 1.0049

1.0056 1.0054 1.0053

1.0061 1.0059 1.0058

1.0065 1.0063 1.0062 1.0070 1.0068 1.0067

1.0022

10025

1.0029

1.0033 10037

1.0040 1.0044

1.0048

1.0052

1.0056

1.0059

1.0064

1.0067 1.0071

1.0075

1.0079

1.0083

1.0086

1.0091

1.0094

1.0099

10103

1.0106

1.0110 1.0114

1.0118

10122

1.0125

10129 1.0133

10073

10078

10083 10089

10094

10099

10104

10110 10116

10121

10126

1.0131

10136 10143

1.0147

10152

10157

1.0075 10073 10071

1.0060 1.0078 1.0076

1.0084 1.0082 1.0080 1.0089 1.0087 1.0084 1.0094 1.0092 1.0089

1.0099 1.0096 1.0094

1.0104 1.0101 1.0098 1.0109 1.0106 1.0103

1.0114 1.0111 1.0108 1.0118 10115 1.0112

1.0123 1.0120 1.0116

1.0129 1.0126 1.0123 1.0132 1.0129 1.0126

1.0137 1.0134 10130 1.0142 1.0139 1.0135

1.0091

1.0095 1.0100

1.0105

1.0111

10116

1.0121

1.0126

1.0089

1.0093 1.0098

1.0103

1.0109 1.0114

1.0119

1.0123

1.0128

l.Oi35

1.0139

1.0144 1.0149

1.0089 1.0086 1.0093 10090 1.0111

10116 1.0098 1.0095 1.0102 10098

1.0106 1.0102

10111 1.0107 1.0114 10110

1.0119 10123

1.0120

1.0125

10132

1.0136

1.0141

1.0146

1.0150

10155

1.0159

1.0165 1.0169

1.0174

1.0179

1.0185

1.0189 1.0194

1.0131

1.0138 1.0141

1.0147 I.0152

1.0123 10127

10114 10119

10163 1.0160 1.0157 1.0153

10168 1.0165 1.0162 1.0158

1.0147 1.0143 1.0139 1.0135 1.0152 1.0148 1.0144 1.0140

1.0131

10136

1.0139

1.0144 10148

1.0153

1.0157

1.0162

1.0166 1.0170

1.0127 10123 1.0131 10127

1.0135 10130

1.0139 1.0135 10143 1.0139

1.0173

1.0179

1.0183

1.0188

1.0193

1.0199

1.0205

1.0210

1.0170 1.0166

1.0175 1.0172 1.0180 1.0177

1.0185 1.0182

1.0190 1.0187

1.0196 1.0192

1.0201 1.0197 1.0206 1.0202

1.0163

1.0169 1.0173

1.0178

1.0183

1.0156 1.0152

1.0161 1.0157

1.0165 1.0161

1.0170 1.0166

1.0175 1.0171

1.0160 1.0176

1.0185 1.0180

1.0190 1.0185

1.0146

1.0153 1.0157

1.0162

1.0144

1.0149 1.0153

1.0158

1.0162

1.0167

1.0171

1.0175

1.0148 10143 1.0138

1.0152 1.0146 1.0141 1.0166

1.0171

1.0176 1.0189

1.0193 1.0198

1.0156 1.0151 1.0145

1.0160 1.0155 1.0149 1.0164 1.0159 1.0153 1.0180


TABLE 13.1I-EXPANSION FACTORS: PIPE TAPS, Y,; STATIC PRESSURE TAKEN FROM DOWNSTREAM TAPS

h, Pt2 Ratio 0.2

1 .oooo

1.0007 1.0015 1.0023 1.0031 1.0037

1.0045 1.0053 1.0060 1.0068 1.0075

1.0082 1.0090 1.0098 1.0106 1.0113

1.0121 1.0128 1.0136

0.3

1 .oooo

1.0006 1.0012 1.0018 1.0024 1.0029

1.0035 1.0042

0.4

1.0000

1.0003 1.0007 1.0010 1.0014 1.0018

0.45

1 .oooo

1.0002 1.0004 1.0006 1.0008 1 .OOlO

1.0012 1.0014 1.0016 1.0018 1.0021

1.0023 1.0025 1.0027 1.0030 1.0032

1.0034 1.0036 1.0038

0.50

1.0000

1.0000 1.0000 1 .oooo 1.0000 1.0001

1.0001 1.0001

0.52

1 .oooo

0.9999 0.9998 0.9998 0.9997 0.9997 0.9996 0.9996

0.54 0.56 0.58 0.60

1.0000 1.0000 1.0000 1.0000

0.1

1.0000

1.0008 1.0017

0.0

0.1 0.2 0.3 0.4 0.5

0.6 0.7 0.8 0.9 1.0

1.1 1.2 1.3 1.4 1.5

1.6 1.7 1.8 1.9 2.0

2.1 2.2 2.3 2.4 2.5

2.6 2.7 2.8 2.9 3.0

3.1 3.2 3.3 3.4 3.5

3.6 3.7 3.8 3.9 4.0

0.9998 0.9997

1.9997 0.9996 0.9995 0.9995 0.9993 0.9990 0.9992 0.9989 0.9985 0.9989 0.9985 0.9980 0.9987 0.9982 0.9975 0.9985 0.9979 0.9972 0.9983 0.9976 0.9968 0.9981 0.9972 0.9963 0.9978 0.9969 0.9958 0.9976 0.9965 0.9954

0.9974 0.9962 0.9949 0.9972 0.9959 0.9945 0.9969 0.9956 0.9941 0.9967 0.9953 0.9937 0.9965 0.9950 0.9932

0.9964 0.9947 0.9928 0.9962 0.9944 0.9924 0.9960 0.9942 0.9920

1.0025 1.0033 1.0042

1.0051 1.0059 1.0068 1.0076 1.0084

1.0093 1.0101 1.0110 1.0119 1.0127

1.0136 1.0143 1.0152

0.9995 0.9994 0.9993

10021 1.0024 1.0028 1.0032 1.0036

1.0039 1.0043 1.0047 1.0050 1.0054

1.0058 1.0062 1.0065 1.0069 1.0073

1.0077 1.0081 1.0084 1.0089 1.0092

1.0096 1.0101 1.0104 1.0107 1 ,011 1

1.0115 1.0119 1.0122 1.0126 1.0130

1.0134 1.0137 1.0141 10145 1.0149

0.9991 0.9989 0.9988 0.9987

1.0048 1.0053 1.0059

1.0065 1.0071 1.0077 1.0083 1.0089

1.0095 1.0101 1.0107 1.0113 1.0119

1.0125 1.0131 1.0137 1.0142 1.0148

1.0154 1.0162 1.0166 1.0173 1.0179

1.0185 1.0190 1.0196 1.0202 1.0208

1.0214 1.0219 1.0225 1.0231 1.0237

1.0002 1.0003

0.9995 0.9995

1.0003 0.9994 0.9986

1.0004 0.9994 0.9984 1.0004 0.9994 0.9984 1.0004 0.9994 0.9982 1.0005 0.9993 0.9981 1.0005 0.9993 0.9980 1.0006 0.9993 0.9979 1.0006 0.9993 0.9978 1.0007 0.9992 0.9977

1.0161 1.0169

1.0177 1.0185 1.0194 1.0202

1.0143 1.0150

1.0041

1.0046

1.0008

1.0008 1.0048 1.0009

1.0044

1.0050 1.0010

1.0008

1.0053 1.0011 1.0056 1.0012

1.0058 1.0013 1.0061 1.0014 1.0063 1.0014 1.0065 1.0015 1.0067 1.0017

1.0070 1.0018 1.0072 1.0018 1.0075 1.0019 1.0077 1.0020 1.0080 1.0021

1.0083 1.0022 1.0085 1.0023 1.0088 1.0024 1.0091 1.0025 1.0093 1.0025

0.9992 0.9974

0 9992

0.9992

0.9976

0.9973 0.9992 0.9972 0.9992

0.9992

0.9971

0.9975

0.9992 0.9971

0.9992 0.9970 0.9992 0.9969 0.9992 0.9969 0.9992 0.9968 0.9992 0.9967

0.9993 0.9966

0.9958 0.9956 0.9954 0.9953

0.9938 09916 0.9935 0.9912

0.9933 0.9908 0.9930 0.9905 0.9928 0.9902 0.9924 0.9897

1.0158 1.0165 1.0173 1.0180 1.0188

1.0195 1.0205 1.0210 1.0217 1.0224

1.0232

0.9951 0.9949 0.9948

0.9946 0.9944 0.9943 0.9941 0.9939

0.9938

1.0210

1.0219 1.0230 1.0236 1.0244 1.0251

1.0259

0.9922 0.9894 0.9919 0.9890 0.9916 0.9885 0.9914 0.9882 0.9912 0.9879 0.9910 0.9875

0.9907 0.9872 1.0267 1.0239 1.0276 1.0247 1.0284 1.0253 1.0291 1.0261

0.9993 0.9993

0.9966 0.9936 0.9966 0.9935 0.9965 0.9934 0.9964 0.9933

0.9905 0.9869 0.9903 0.9866 0.9901 0.9863 0.9898 0.9860 0.9896 0.9857 0.9894 0.9854

0.9994 0.9994 0.9994 0.9995 0.9995 0.9995 0.9995

1.0301 1.0268 1.0309 1.0276 1.0316 1.0287 1.0324 1.0290 1.0332 1.0297

0.9964 0.9931 0.9964 0.9930 0.9963 0.9929 0.9963 0.9928 0.9963 0.9927

0.9892 0.9850 0.9889 0.9847 0.9887 0.9844


TABLE 13.lm-EXPANSION FACTORS: FLANGE TAPS, Y,; STATIC PRESSURE, MEAN OF UPSTREAM AND DOWNSTREAM

h, PI, Rail0

00

01 02

03

04

05

06 07

08

09

10

11 12

13 14

15

16 17

18

19

20

21 2.2

23

24

25

2.6 2.7

2.6

2.9

3.0

31 32

33

3.4

3.5

3.6 3.7

3.8

3.9

40

04 0.50 0.52 0 54

TiE- 1.0000

0 60 061 062 01

1.0000

0.9998 0 9995

0.9993

0.2

10000

0 9998

0 9995

0 9993

09991 0 9988

0 9986

0 9984 09981

03

1.0000

0 9998

0 9995

0 9993

0 9990 0 9988

0 9985

0 9983 09981

045

1.0000 10000

0 9997 0 9997

0 9994 0 9994

0 9992 0 9991

0.9989 0 9988 0.9986 0 9985

0 9984 0 9982

0.9981 0 9979 0 9978 0.9976

0 9975 0 9973

0 9973 09971

0.56

1.0000

0 9997

0 9993

0 9990

0 9987 0 9983

0 9980

0 9977 0 9974

09971 0 9968

0 9965

09961

0 9958

0 9955

0 9952

0 9948

0 9946

0 9943

0 9940

0 9935

0 9933

0.9931

0.9928

0.9924

0 9921

0.9919

0.9916

0.9913

0.9909

0 9907

0.9904

09901

0.9898

0.9895 0 9893

0.9890

0.9887

0.9884

0.9881 0 9879

0.58

10000 10000

0 9996

0 9993

0 9989

0 9985 09981

0 9978

0 9974

0 9970

0 9967

0 9964

0 9960

0 9957

0 9953

0 9949

0 9946

0 9943

0 9939

0 9936

0 9933 0 9927

0 9925

0 9922

09919

09915 09912

0 9909

0 9905

0 9902

0 9899 0 9895

10000

0 9996 0 9992 0 9989 0 9985 0 9981

1.0000

0.9997

10000

0 9997 0 9994 0 9991

0 9988 0 9985

0 9982

0.9979

0.9976

0 9973

0 9970

0 9967

0 9964

09961 0 9958

0 9955

0 9952

0 9950

0 9947

0 9944

0 9940

0 9938

0 9936

0 9933

0 9930

0 9927

0.9997 0 9994

0.9997 0.9993

0 9996 0 9993 0 9995

0.9992 0.9990 0.9987 0.9984

0.9981

0 9978

0.9975

0 9972 0.9969

0.9966

0.9963

0.9960

0.9957

0.9954

0.9950

0.9948 0 9945

0 9942

0 9938

0 9936

0 9934

0 9931

0 9927

0 9924

0.9990 0.9986 0.9983

0.9979 0.9976 0.9972 0 9969 0.9967 0.9963 0.9960 0.9956 0 9953 0 9950 0 9946

0 9989 0 9986 0 9982 0 9978 0 9975

0.9991

0.9988

0 9986 0 9984

09981

0 9990 0.9987 0 9984 0.9982 0 9979 0 9977 0 9975

0 9977 0 9973 0 9970 0 9967 0 9963 0 9959 0 9955 0 9952 0 9948 0 9945 09941

0 9938

0 9934

09931 0 9925

0 9923

0 9920 09917

09913

0 9909

0 9907

0 9903

0 9900

0 9896 0 9892

0 9890 0 9886 0 9883 0.9879 0 9877 0.9873

09870

0 9866

0.9863

09861

0 9971 0 9968 0 9965 09961

0 9958

0 9954 09951

0 9948

0 9944

0 9979 0 9979 0 9978 0.9977 0 9977 0 9976 0 9975 0.9973 09971

0 9968

0 9966

0 9964 0.9962

0 9959

0 9957

0 9955

0 9953 09951

0 9949

0.9947

0 9945

0 9974 0 9973 0 9970 0 9968 0 9966 0 9964 0 9962 0 9959 0 9957 0 9954 0 9953 09951

0 9949

0 9947 0 9944

0 9942

09941

0 9938

0 9936 0 9934

0 9932 09931

0 9928

0 9926 0.9924

0.9923

0 9920

0 9918 09917 0.9915

0 9974 0 9972 0 9970 0 9967 0 9965 0 9962 0 9960 0 9958 0 9956 0 9953 0 9952 0 9950 0 9947 0 9945 0 9943

0 9972 09971 0 9968 0 9970 0 9968 0 9966 0 9967 0.9965 0 9963 0 9965 0.9963 0 9960 0 9962 0.9960 0 9957 0 9960 0.9957 0 9954 0 9958 0.9955 0 9952 0 9955 0 9953 0 9949 0 9953 0.9950 0 9948 0 9950 0 9946 0 9942 0 9948 0.9945 0 9940 0 9947 0.9943 0 9938 0 9944 0.9940 0 9936 09941 0.9938 0.9933 0 9938 0.9935 0 9930

0 9943 0 9940

09941 0 9937

0 9937 0 9932 0 9930 0 9927 0 9924

0 9934 0 9929 0 9927 0 9924 09921

09921 09917 09918 09914

0 9943 0.9941

09941 0.9936 0.9933 0.9928 0 9925 0 9922 0 9939 0 9934 0.9930 0.9925 0.9922 0.9919

0 9915 09911 09912 0 9907

0 9909 0.9904

0 9905 0.9901 0 9903 0 9898

0 9900 0.9895

0 9897 0 9892

0 9894 0.9889 0 9890 0.9885 0 9888 0 9883

0.9885 0.9878

0.9882 0.9876

0.9879 0 9873 0 9876 0.9870

0.9873 0.9867

0.9939 0.9937 0.9935

0 9937 0 9932 0.9928 0.9922 09919 09916

0 9934 0 9929 0.9925 0.9919 09916 0 9913 0 9932 0.9927 0.9923 0.9917 09914 0 9911

0 9930 0.9925 0.9920 0.9914 0.9911 0.9908 0 9929 0.9923 0.9919 0.9912 0 9909 0.9905

0 9926 0.9921 0.9916 0.9909 0.9906 0.9902

0 9924 0.9918 0.9913 0.9907 0.9903 0.9899 0.9922 09916 0.9911 0.9905 0 9901 0.9897

0 9920 0.9914 0.9909 0.9902 0 9898 0 9894

09918 0 9912 0.9907 0.9899 0 9895 0 9891

09916 0.9910 0.9905 0.9897 0 9893 0 9889

09915 0.9908 0.9902 0.9894 0.9890 0.9886 09913 0.9906 0.9900 0.9892 0.9888 0.9884

0 9932 09931

0.9929

0 9927

0 9892 0 9889 0 9886 0 9882 0 9880 0 9876 0 9873 0 9870 0 9867 0 9864

0 9925 0.9923 09921

0.9919 0.9918

0.9916


TABLE 13.lm-EXPANSION FACTORS: FLANGE TAPS, Y,, STATIC PRESSURE, MEAN OF UPSTREAM AND DOWNSTREAM (co1 7tinued)

h,

Ph Ratio

00

0.1 02

0.3

04

05

06 0.7

08

09

10

11 1.2

13 14

15

16 1.7

1.8

1.9

2.0

2.1 22

23

24

25

26 27

28

29

30

3.1

32

33

34

35

36

37

38

39

40

067 066

10000 10000

0 9996 0 9996

09991 09991

0 9987 0 9987

0 9983 0 9982 0 9978 0 9978

0 9974 0.9973

0 9970 0 9969

0 9965 0 9965

0 9962 09961

0 9958 0.9957

0 9953 0.9952

0 9949 0 9948

0.9945 0.9943 09941 0.9939

0.9937 0.9935

0 9933 0.9931

0.9929 0.9927

0 9925 0 9923

09921 09919

09915 0.9912

09913 0.9910

0 9908 0.9906

0 9905 0.9902

0.9901 0.9898 0.9897 0 9894

0 69

10000

0 9995

09991

0 9986

0 9982 0 9977

0 9973

0 9968

0 9964

0 9960

0.9955

0 63

1.0000

0 9996

0 9992

0.9988

0 9984 0.9981

0.9977

0 9973

0 9969

0 9966 0 9962

0.9958

0.9954

0.9951

0 9947

0.9944

0.9940

0 64

1.0000

0 71 0.72 0 73 0.74 0 75

Tzz- 1.0000 10000 1.0000

0.9995

0 9990

0.9985

0 9980 0.9975

0.9970

0.9965

0.9961

0 9956 0.9951

0 9946

0.9941

0.9936

0.9932

0 9927

0 9922

09918 09916 09913 09910

09913 0.9911 0 9908 0 9905

0.9908 0 9906 0 9900 09901 0 9899 0 9892 0.9899 0.9896 0 9890 0 9894 0.9892 0 9884 0.9891 0.9887

0.9995 0.9995 0 9995 0 9990 0 9989 0.9989 0 9985 0.9984 0 9984

0 9980 0.9979 0 9978 0 9974 0.9974 0 9973

0 9969 0.9968 0 9967

0 9964 0.9963 0 9962

0 9960 0.9959 0 9957 0 9955 0.9953 0 9952 0 9950 0 9948 0 9947

0 9945 0 9943 09941

0 9940 0 9938 0 9936

0.9934 0.9933 09931 0.9930 0.9928 0 9926

0.9925 0 9923 0 9920

0.9920 0.9918 09915

0.9903 0 9896 0 9893 0.9888 09884

0 9878 0 9874

0 9880 0 9875 0 9870 0.9865 0.9860

0.70

10000

0 9995

09991

0 9986

09981 0.9977

0.9972

0.9967

0.9963

0 65

10000

0 9996

0 9992

0 9988

0 9984 0 9980

0 9975

09971

0 9967

0 9964

0 9960

0 9956

0 9952

0 9948 0 9944

09941

0 9937

0.66

10000

0 9996

0 9992

0 9987

0 9983 0 9979

0.9975

0.9971

0.9967

10000

0.9995

0 9990

0 9986

09981 0 9976

09971

0 9966 0 9962

0.9996

0.9992

0.9988

0.9984 0.9980

0.9976

0 9972

0.9968

0 9965 09961

0 9957

0 9953

0 9949

0 9945 0 9942

0 9938

0 9963 0 9959 0 9955 09951

0 9946

0 9942

0 9939

0 9935

09931 0 9927

0 9923

0.9917

0.9915

0.9911

0 9908

0.9904 0.9900

0.9959 0 9957 0 9954 0 9953

09951

0.9946

0 9949 0 9948

0 9945 0 9943

0 9940 0 9938

0.9936 0.9934

0 9932 0 9929

0 9927 0 9925

0 9923 09921 09918 0 9916

09914 09911

0 9907 0 9904

0 9905 0.9902

0 9900 0 9897

0 9897 0.9894

0 9892 0.9889 0 9888 0.9885

0.9942 0.9938

0.9933

0.9929

0.9925

0 9920

0.9916

0.9910

0 9908

0 9903

0 9900

0.9895

09891

0.9936 0 9934 0 9933 0.9932 0 9930 0 9928 0.9929 0 9923 09921

0 9918

09915

09911 0 9907

0 9904 0 9900 0 9897 0 9893 0 9890 0 9887 0 9883 0 9880 0 9876 0 9873 0 9869 0 9866 0 9863 0 9860 0 9857

0 9927 0 9925 0.9921 09919

09919 09917

09916 09913

09913 09910

0 9909 0 9906 0 9905 0 9902 0 9902 0 9899 0 9898 0 9895

0 9886 0.9882 09881 0 9878 0 9876 0.9873 0.9872 0.9869

0 9896 0.9893 0.9890 09887 0 9884 0.9880 0.9892 0.9889 0.9886 09883 09880 0.9876

0 9869 0 9864

0 9895 0.9891 09887

0 9884

0 9880

0 9877

0 9873 0 9870

0 9866

0.9863

0 9859 0 9856 0 9853

0 9892 0 9889 0.9886 0.9882 0.9879 0 9888 0 9885 0.9882 0 9878 0 9874 09884 0.9881 09877 0 9874 09870

09881 0 9877 0 9874 0 9870 0 9867 0 9877 0.9873 09870 0 9866 0 9862

0 9875 0.9871 0.9867 0.9864 0 9859 0 9855 09871 0.9867 0 9863 0.9859 0 9855 0 9850 0 9867 0 9863 0.9859 0 9854 0 9850 09845

0 9863 0.9859 0.9854 0.9850 0 9845 0 9840 0 9858 0 9854 0 9850 0.9845 0 9840 0.9835

0 9854 0 9850 0.9845 09841 0 9836 09831

0 9850 0.9845 0.9841 0 9836 09831 0 9825

0 9846 09841 0.9837 0 9832 0 9826 09821

0.9842 0 9837 0.9833 0.9827 0 9822 0.9816

0.9838 0 9833 0.9828 0 9822 0.9817 09811

0.9834 0 9829 0.9824 0 9818 09813 0 9807 0.9830 0 9825 0.9820 09814 0 9808 0 9802

0 9826 0 9820 0.9815 0 9809 0 9803 0 9797

09873 0 9870 0 9866 0 9863 0 9859 0 9869 0.9866 0 9862 0 9858 0 9854 0 9866 0.9863 0 9859 0 9855 0 9850

0.9862 0.9859 09855 09851 0 9846

0 9859 0.9855 09851 0 9847 09842

0 9855 0.9851 0 9847 0 9843 0 9838

09852 0.9848 09844 0 9839 0 9835

0.9849 0.9845 0 9840 0 9836 09831


TABLE13.ln--MANOMETER FACTORS(MERCURY METERS),F,

Specific Gravity,

y!J 500

0.55 0.60 0.65 0.70 0.75

0

1 .oooo 1 .oooo 1 .oooo 1.0000 1.0000

0.9989 0.9988 0.9987 0.9985

0.55 1 .oooo 0.9990 0.60 1.0000 0.9989 0.65 1 .oooo 0.9988 0.70 1.0000 0.9987 0.75 1.0000 0.9986

0.55 1.0000 0.9991 0.60 1 .oooo 0.9990 0.65 1.0000 0.9989 0.70 1.0000 0.9988 0.75 1 .oooo 0.9987

0.55 1 .oooo 0.9992 0.60 1.0000 0.9991 0.65 1.0000 0.9990 0.70 1.0000 0.9989 0.75 1 .oooo 0.9988

Flowing pressure, psig

1000 1500 2000

AmbientTemperature=O°F

0.9976 0.9960 0.9943 0.9972 0.9952 0.9932 0.9967 0.9941 0.9920 0.9961 0.9927 0.9907

- - -

AmbientTemperature=40°F

0.9979 0.9967 0.9954 0.9976 0.9962 0.9946 0.9973 0.9955 0.9937 0.9970 0.9947 0.9926 0.9965 0.9937 0.9915

Ambient Temperature=80°F

0.9981 0.9971 0.9960 0.9979 0.9967 0.9955 0.9977 0.9963 0.9948 0.9974 0.9958 0.9940 0.9971 0.9951 0.9931

AmbientTemperature=120°F

0.9983 0.9974 0.9965 0.9981 0.9971 0.9960 0.9979 0.9967 0.9955 0.9977 0.9963 0.9950 0.9975 0.9959 0.9943

2500 3000

0.9930 0.9919 0.9908 0.9896 -

0.9921 0.9910 0.9900 0.9890

0.9942 0.9932 0.9933 0.9923 0.9923 0.9913 0.9912 0.9903 0.9902 0.9893

0.9950 0.9941 0.9943 0.9933 0.9935 0.9925 0.9926 0.9915 0.9916 0.9906

0.9956 0.9948 0.9950 0.9941 0.9944 0.9934 0.9937 0.9926 0.9929 0.9918

TABLEl3.10-GAUGE LOCATION FACTOR-Fp

Gauge Elevation Above Sea Level, ft

Degrees Sea Latitude Level 2,000 4,000 6,000 8,000

0 (Equator) 0.9967 0.9986 0.9985 0.9984 0.9983 5 0.9987 0.9986 0.9985 0.9984 0.9983 IO 0.9968 0.9987 0.9986 0.9985 0.9984 15 0.9989 0.9988 0.9987 0.9986 0.9985 20 0.9990 0.9989 0.9988 0.9987 0.9986

25 0.9991 0.9990 0.9989 0.9988 0.9987 30 0.9993 0.9992 0.9991 0.9990 0.9989 35 0.9995 0.9994 0.9993 0.9992 0.9991 40 0.9998 0.9997 0.9996 0.9996 0.9994 45 1.0000 0.9999 0.9998 0.9997 0.9996

50 1.0002 1.0001 1.0000 0.9999 0.9998 55 1.0004 1.0003 1.0002 1.0001 1.0000 60 1.0007 1.0006 1.0005 1.0004 1.0003 65 1.0008 1.0007 1.0006 1.0005 1.0004 70 1.0010 1.0009 1.0008 1.0007 1.0006

75 1.0011 1.0010 1.0009 1.0006 1.0007 80 1.0012 1.0011 1.0010 1.0009 1.0008 85 1.0013 1.0012 1.0011 1.0010 1.0009 90 (Pole) 1.0013 1.0012 1.0011 1.0010 1.0009

10.000

0.9982 0.9982 0.9983 0.9984 0.9985

0.9986 0.9968 0.9990 0.9993 0.9996

0.9997 0.9999 1.0002 1.0003 1.0005

1.0006 1.0007 1.0006 1.0008

13-35

13-36

- L

Fig. 13.3-Proper installation of a meter run for an orifice meter.

Physical Setup of System

Inasmuch as standard tables are used, careful attention must be paid to the physical setup of the metering system. Otherwise the results would vary with the installation.

Straightening Vanes. The purpose of these vanes is to minimize the effect of swirls, eddy currents, or irregular velocity distribution on meter accuracy. These vanes are built into a nipple and consist of a bundle of small pipe or tubing as shown in Fig. 13.3. The dimension of each tube d should not exceed one-fourth of the inside pipe diameter d, . The length should be at least 10d.

Unless necessary, vanes should not be used because they are susceptible to erosion, introduce additional pressure loss, and clog easily.

Orifice Location. Fig. 13.3 shows the minimum distance the orifice should be from valves and fittings in order that proper metering might result.

Size of Orifice and Meter Run. The meter run should be sized so that the anticipated maximum and minimum flow rates may be handled within the satisfactory ratio of orifice to pipe diameter. In doing this, it should be kept in mind that (1) the differential h MI should not exceed the static pressure p,, (based on numerical values only, not units); (2) the meter run ID should be at least one-third larger than the orifice opening; and (3) both the differential- and static-pressure pens should preferably operate within the middle 60% of the recording-chart range.


As a practical matter, the meter run ID should not be less than 3 in. in nominal diameter regardless of the small quantity of gas flowing. As a first approximation, Eq. 6 should be solved for C’ at the flow rate h, andpf desired. Then

do= OS. _ . .

If the maximum anticipated flow rate is used to find C’, multiplying the orifice size found in Eq. 8 by 1.5 gives the approximate minimum pipe diameter needed. Once this has been established, it is a relatively simple matter to change orifice plates as the flow varies to keep the values of h w, and pf in the desired range. If flow is small and fairly constant (e.g. fuel meter) and/or a Y-in. orifice is used, a 2-in. run should be used to avoid low Fd (below 0.10).

Standard sharp-edged orifice plates should be used, the thickness of which is at least L/6 in. For pipes larger than 4 in., thickness is at least L/8 in. The thickness should not exceed one-eighth the orifice opening.

Recorder. The normal orifice meter is equipped with a two-pen recorder for measuring both static and differential pressure. The differential pen is normally actuated through a mechanical system using either a mercury manometer or a bellows. Both types are shown in Fig. 13.4.

With the former, a slight change in flow rate changes the level in the mercury manometer. A large float resting on the mercury changes level correspondingly and transmits this movement to the chart pen through a system of levers.

The bellows-type or mercuryless meter consists of two bellows filled with fluid such as glycol. As the pressure differential changes, this fluid moves between the bellows through a damping device, causing the bellows to expand or contract. The pen is actuated by a center rod, which is connected to the free ends of the bellows. The small liquid-filled bellows located on the high- pressure side serves as an expansion device to correct for changes in ambient temperature.

Both types of instruments are equipped with check valves to protect them from differential pressures that exceed the range of the instrument. This is more of a problem with the mercury meter, for excessive differential pressure can blow mercury out of the meter, which destroys its calibration. In the bellows meter, soft-seated check valves prevent the flow of fluid between the bellows. The bellows are then not likely to rupture for they are supported internally by the fluid contained within them.

The bellows meter has gained increasing popularity, particularly in field operations. Even though the initial cost is higher, many operators feel that this is compensated for by decreased maintenance. It also offers the advantage of operating either properly or not at all, with little range in between. Eliminated are the problems of mercury loss, worrying about the change in calibration with the amount of mercury, and the like. Until it fails, experience has shown that this meter needs little routine calibration.


With both types, the static pressure is read from a pen actuated by a bourdon tube. The lead for this may come

off either the upstream or downstream tap. In most installations. it is good practice to order a bourdon tube having about twice the maximum pressure anticipated. This minimizes the problem of distortion, which destroys the original calibration of the tube. These tubes arc available in many materials, the type depending on the service and the pressure rating desired.

Most meters are hooked up using a standard five-valve manifold of the type shown just above the bellows meter in Fig. 13.4. One valve is placed on each of the leads to the pressure taps, two valves on the bypass, plus a vent valve. When the meter is placed in operation the bypass valves would be open, as would the two main valves, with the vent valve closed. The meter is then placed in service by slowly closing the bypass valves, followed by opening of the vent valve. This procedure prevents momentary pressure surges that might damage the meter or upset the calibration. The reverse procedure would be used in taking the meter out of service. When the leads are bypassed, the vent valve is an excellent place to obtain a gas sample. A slightly more complex manifold is shown above the mercury meter, with drip pots for removing any liquid in the gas.

Standard calibration procedure on the differential pen calls for connecting one side of a water manometer into the high-pressure side of the meter. The low-pressure side of the meter is then opened to the atmosphere. This gives two manometers in parallel. By superimposing pressure on the high-pressure cell with a hand pump or similar device, one may compare the meter reading with that of the manometer. If they are different, the meter must be adjusted accordingly.

The bourdon tube may be calibrated simply by placing a dead-weight tester (preferred) or a calibrated pressure gauge in the lead line to the bourdon tube. By proper manipulation of the valves, the vent valve may be used to make this connection.

The circular charts used usually cover a time period of 24 hours or 7 days. Many meters have clocks that may be easily adjusted for either time period. In most production operations, 7-day periods are preferred, because this minimizes the cost of changing charts and the number of charts to be handled.

Other Forms of Metering Systems

For well-test purposes, other forms of velocity meters are used-orifice well testers, critical-flow provers, pitot tubes, and side-static methods. These methods are not so accurate as the meters discussed previously but are convenient and often yield results accurate enough for the purpose.

Orifice Well Tester. This consists of a 2-in. nipple with provision for attaching different sharp-edged orifice plates on the end. The static pressure just upstream from this plate may then be measured. Tables 13.2a and b show various applicable charts for measuring flow. a

The accuracy of this device is limited but it is suitable for measuring the amount of gas being produced where the pressures are relatively low and the production is to the atmosphere.

Fig. 1X4-Views showing construction of meters.

Critical-Flow Prover. The critical-flow prover is a similar device that exhausts the gas to the atmosphere. It is also a special pipe nipple with a flange for holding special plates to the end. It is based on the principle that the velocity of sound represents the maximum speed at which a pressure effect may be propagated through a gas-i.e., once this velocity is reached, further increase in the pressure differential will not increase the pressure at the throat. This means that the mass rate of flow would not increase as the pressure differential p2/pI was decreased below this critical value. With ideal gases, the critical ratio is 0.49 for monatomic gases and 0.53 for diatomic gases, with slightly higher values for complex gases. Saturated steam, for example, shows a critical pressure ratio of 0.55. The range of these values is the source of the common rule of thumb that, once the pressure reduction is twofold, critical-flow phenomena limit the mass rate of flow.

When metering gas under such conditions, the volume rate of flow will be a function of upstream pressure, gas gravity, and gas temperature, since it is a compressible fluid.

(continuedon Page 45)

13-38

Pressure, in of water 0.60 0.70

1.0 528 1.2 578 1.4 625 1.6 669 1.8 709

2.0 747 2.2 784 2.4 818 2.6 852 2.8 884

3.0 915 3.2 945 3.4 975 3.6 1000 3.8 1030

4.0 1060 4.5 1120 5.0 1180 5.5 1240 6.0 1290

6.5 1340 7.0 1390 8.0 1500 9.0 1590 10.0 1670

11.0 1750 12.0 1830 13.0 1910 14.0 1980 15.0 2050

1 .o 1.2 1.4 1.6 1.8

2.0 2.2 2.4 2.6 2.8

3.0 3.2 3.4 3.6 3.8

4.0 4.5 5.0 5.5 6.0

6.5 7.0 8.0 9.0 10.0

11.0 12.0 13.0 14.0 15.0

1680 1850 1990 2130 2260

2370 2490 2610 2710 2810

2920 3010 3110 3190 3280

3370 3580 3710

4120

4300 4450 4760 5050 5320

5600 5860 6100 6330 6550 6070 5670 5350 5080 4840 4640 4450 4140


TABLE 13.2a-CAPACITIES OF ORIFICE WELL TESTERS*

489 536 579 619 657

692 726 758 789 819

848 875 902 928 954

979 1040 1090 1150 1200

1250 1290 1380 1470 1550

1620 1700 1770 1830 1900

1560 1710 1840 1970 2090

2200 2310 2410 2510 2610

2700 2790 2870 2960 3040

3120 3310 3490

3820

3970 4120 4410

4930

5180 5420 5650 5860

Specific gravity

0.80 0.90 1 .oo 1.10 ~ - - ~ %in. Orifice in l/a-in. Plate

458 432 409 390 501 473 448 427 541 511 484 462 579 546 518 494 614 579 549 524

647 610 579 552 679 640 607 579 709 669 634 605 738 696 660 629 766 722 685 653

793 747 709 676 819 772 732 698 844 796 755 720 868 819 777 740 892 841 798 761

915 863 819 781 971 915 868 823 1020 965 915 873 1070 1010 960 915 1120 1060 1000 956

1170 1100 1040 995 1210 1140 1080 1030 1290 1220 1160 1100 1370 1290 1230 1170 1450 1360 1290 1230

1520 1430 1360 1300 1590 1500 1420 1350 1650 1560 1480 1410 1720 1620 1540 1460 1780 1680 1590 1520

X-in. Orifice in l/s-in. Plate

1460 1370 1300 1240 1600 1510 1430 1360 1730 1630 1540 1470 1840 1740 1650 1570 1960 1840 1750 1670

2060 1940 1840 1760 2160 2040 1930 1840 2260 2130 2020 1930 2350 2220 2100 2000 2440 2300 2180 2080

2530 2380 2260 2150 2610 2460 2330 2220 2690 2540 2410 2290 2770 2610 2470 2360 2840 2680 2540 2420

2920 2750 2610 2490 3090 2920 2770 2640 3260 3070 2920 2780 3420 3220 3060 2920 3570 3370 3190 3040

3720 3500 3330 3170 3860 3640 3450 3290 4120 3890 3690 3520 4370 4120 3910 3730 4610 4350 4120 3930

4850 4570 4340 4140 5070 4780 4540 4325 5280 4980 4730 4510 5480 5170 4900 4670

1.20 1.30 1.50

374 359 334 409 393 366 442 425 395 473 454 423 502 482 449

528 508 473 554 532 496 579 556 518 603 579 539 625 601 559

647 622 579 669 642 598 689 662 616 709 681 634 728 700 651

747 718 669 793 762 709 836 803 747 876 842 784 915 879 819

953 916 852 989 950 885 1060 1020 945 1120 1080 1000 1180 1140 1080

1240 1190 1110 1300 1250 1160 1350 1300 1200 1400 1350 1250 1450 1390 1300

1190 1300 1410 1510 1660

1680 1770 1840 1920 1990

2060 2130 2200 2260 2320

2380 2530 2660 2790 2910

3040 3150 3370 3570 3760

3960 4150 4320 4480

1140 1250 1350 1450 1530

1620 1700 1770 1840 1910

1980

2110 2170 2230

2290 2430 2560 2680 2800

1060 1170 1260 1350 1430

1510 1580 1650 1720 1780

1840 1900 1960

3030 3230 3430 3620

3810 3980 4140 4300

2070

2130 2260 2380 2500 2610

2710 2820 3010 3190 3370

3540 3700 3860 4000

GASMEASUREMENTAND REGULATION

TABLE13.2a-CAPACITIESOFORlFlCEWELLTESTERS'(continued)

13-39

Specific gravity Pressure, in. of water 1.20 1.30

2430 2660 2870 3070

1.50 0.60 0.70 0.80 0.90 1.00 1.10 - - %-in. Orifice in %-in. Plate

3560 3910 4220 4520 4790

5050

3300 3620 3910 4180 4440

4670

3090 3390

5520 5760 5980

6180 6390 6580 6780 6960

7140 7580 7980 8370 8740

9100 8450

10100 10700 11300

11900 12400 12900 12400 13900

5120 5330 5530

5720 5910 6090 6270 6440

6610 7010 7390 7750

3910 4150

4370 4580 4790

5170

5350 5530 5700

2910 3190 3450 3690 3910

4120 4320 4510 4700 4880

5050 5210 5370 5530 5680

5830

2760 3030 3270 3500 3710

3910 4100 4280 4460 4630

4790 4950 5100 5250 5390

5530 5870 6180 6480 6770

7050 7320 7820 8300 8740

2640 2890 3120 3340 3540

3780 3910 4080 4250 4410

4570 4720 4860

1.0 1.2 1.4 1.6 1.8

2.0 2.2 2.4 2.6 2.8

3.0 3.2 3.4 3.6 3.8

4.0 4.5 5.0 5.5 6.0

6.5 7.0 8.0 9.0

10.0

11.0 12.0 13.0 14.0 15.0

1.0 1.2 1.4 1.6 I.8

2.0 2.2 2.4 2.6 2.8

3.0 3.2 3.4 3.6 3.8

4.0 4.5 5.0 5.5 6.0

6.5 7.0 8.0 9.0 10.0

11.0 12.0 13.0 14.0 15.0

2520 2760

6180 6520 6830 7140

7430 7710 8240 8740 9200

9700 10100 10600 11000 11300 10800

9200 9620 10000 10400

2470 2670 2860 3030

3190 3350 3500 3640 3780

3910 4040 4160

3190 3390

3570 3740 3910 4070 4220

4370 4520 4650 4790 4920

5050

3430 3600 3760 3910 4060

4200 4340 4470 4600 4730

4850 5140 5420 5690 5940

6180 6420 6860 7280 7670

8070 8440 8790 9120 9440

6020

6180 6560 6910 7250 7570

7880 8180 8740 9270 9770

10300 10800 11200 11600 12000

5140

5270 5590 5890 6180 6460

6720 6980 7460 7910 8340

8780 9180 9560 9920 10300

4400

4520 4790 5050 5640

6180

6440 6680 7140 7570 7980

8410 8790 9160 9510 9840

4440 4860 5250 5610 5950

6270 6580 6870 7150 7420

7680 7940 8180 8420 8650

8870 9410 9920 10400 109ocl

11300 11700 12500 13300 14000

14800 15500 16100 16700

5530

5760 5970 6380 6770 7140

7510 7860 8190 8490 8790

3970 4350 4690 5020 5330

5610 5880 6150 6400 6640

6870 7100 7320 7530 7730

7940 8420 8870 9310 9720

10100 10500 11200 11900 12500

13200 13800 14400 15000

8430 8740 9350 9910 10400

11000 11500 12000 12400 12900

&in. Orifice in %-in. Plate

6270 6870 7420 7940 8430

8870 9310 9720 10100 10500

10900 11200 11600 11900 12200

12500 13300 14100 14700 15400

16000 16700 17200 18800 19900

20900 21900 22800

5810 6360 6870 7350 7880

8210 8610 9000 9360 9720

10100 10400 10700 11000 11300

11600

5430 5950 6430 6870

5120 5610 6060 6480 6870

7240 7600 7940 8260 8570

8870 9170 9450 9720 9980

10200 10900 11500 12000 12500

4630 5070 5480 5860 6220

5320 5750 6150 6520

6870 7210 7530 7830 8130

8420 8700 8960 9220 9470

9720 10300 10900 11400 11900

12400 12900 13700 14600 15400

4670 5040 5390 5720

6030 6320

7680 8060 8420 8760 9090

9410 9720

10000

6550 6870 7180 7470 7750

6870 7130

7380 7630 7860 8090 8310

8530 9040 9530

8030 8290 8550 8790

10600

10900 11500 12100 12700 13300

13900 14400 15400 16300 17200

9030

9270 9830 10400 10900 11300

11800 12300 13100 13900 14700

15500 16200 16800 17500

12300 13000 13600 14200 10400

10900 11300 12100 12800 13500

14200 14900 15500 16100

14800 15400 16400 17400 18400

19400

13100 13600 14500 15400

18100 17100 16200 19000 17900 17000

21100 19700 18600 17700 21900 20500 19300 18300

24500 22700 21200 20000 19000 18100 17300 16600 15500

13-40 PETROLEUMENGINEERINGHANDBOOK

TABLEl3.2a-CAPACITIESOFORlFlCEWELLTESTERS*(continued)

0.60

Specific gravip Y

0.70 0.80 0.90 1.00 1.10 1.20 1.30 1.50 Pressure, in. of water

1 .o 1.2 1.4 1.6 1.8

2.0 2.2 2.4 2.6 2.8

3.0 3.2 3.4 3.6 3.8

4.0 4.5 5.0 5.5 6.0

6.5 7.0 8.0 9.0

10.0

11.0 12.0 13.0 14.0 15.0

1.0 1.2 1.4 1.6 1.8

2.0 2.2 2.4 2.6 2.8

3.0 3.2 3.4 3.6 3.8

4.0 4.5 5.0 5.5 6.0

6.5 7.0 8.0 9.0 10.0

11.0 12.0 13.0 14.0 15.0

9260 %-in. Orifice in %-in. Plate 12300 11600 11000 13400 12700 12000 14500 13700 13000 15500 14600 13900 16500 15500 14700

17300 16400 15500 18200 17100 16300 19000 17900 17000 19800 18600 17700 20500 19300 18400

21200 20000 19000 21900 20700 19600 22600 21300 20200 23300 21900 20800 23900 22500 21400

24500 23100 21900 26000 24500 23300 27400 25900 24500 28800 27100 25700 30000 28300 26900

31300 29500 28000 32500 30600 29000 34700 32700 31000 36800 34700 32900 38800 36600 34700

40900 38500 36500 42700 40300 38200 44500 42000 39800 46200 43500 41300 47800 45100 42800

14200 13100 14400 15500 16600 17600

18500 19400 20300 21100 21900

22700 23500 24200 24900 25800

26200 27800 29300 30700 32100

33400 34700 37100 39300 41500

43700 45700 47600 49300 51100

10500 11500 12400 13200 14000

14800 15500 16200 16800 17500

18100 18700 19300 19800 20400

20900 22200 23400 24500 25600

26700 27700 29600 31400 33100

10000 11000 11800 12700 13400

14200 14900 15500 16100 16800

17300 17900 18500 19000 19500

20000 21200 22400 23500 24500

25500 26500 28300 30000 31700

33400 34900 36400 37700 39100

8960 9810 10600 11300 12000

12700 13300 13900 14400

15500 16800

10500 11400

17900 19000

20000 21000 21900 22900 23800

12200 12900

13600 14300 14900 15500 16100

16700 17200 17770 18300 18800

19200 20400 21500 22600 23600

24500 25500 27200 28900 30400

15500 16000 16500 17000 17500

17900 19000 20000 21000 21900

22800 23700 25300

24500 25300 26100 26900 27600

28300 30100 31600 33200 34700

36100 37400 40000 42500 44800

47200 49300 51400 53300 55200

28300

36400 31200 32500 33700 36200

37500

l-in.01 fifice in YE, .in. Plate

22400 21100 20000 24500 23100 21900 26500 24900 23700 28300 26700 25300 30000 28300 26800

31600 29800 28300 33200 31300 29700 34600 32700 31000 36000 34000 32200 37400 35300 33500

38700 36500 34600 40000 37700 35800 41200 38900 36900 42400 40000 37900 43600 41100 39000

44700 42200 40000 47400 44700 42400 50000 47100 44700 52400 49400 46900 54800 51600 49000

57000 53800 51000 59200 55800 52900 63200 59600 56600 67100 63200 60000 70700 66700 63200

74400 70100 66500 77800 73300 69600 81000 76400 72500 84100 79300 75200 87100 82100 77900

23900 26200 28300 30200 32100

33800 35400 37000 38500 40000

19100 18300 20000 21600 23100 24500

25800 27100 28300 29400 30500

31600 32700

17500 19200 20800 22200 23500

24800 26000 27200 28300 29300

16300 17900 19300 20700 21900

23100 24200 25300 26300 27300

28300 29200 30100 31000 31800

32700

28300 30600 22600

24100 25600

27000 28300 29500 30700 31900

32700 34600

36500 38300 40000 41600 43200

44700 41400 46200 42800 47600 44100 48900 45300

33000 34100

30400 31400

33700 34600

32300 33300 34200

35100 37200 39200 41100 43000

44700 46400 49600 52600 55500

58400 61000

36200 37200

38100 40400 42600 44700 46700

48600

50300

51600 54700 57700 80500 63300

65800 68300 73100 77500 81600

48600

47800 50700 53400 56100 58500

61000 63200 67600 71700 75600

79500 83200

35600

36500 38700 40800 42800 44700

46600 48300 51600 54800 57700

36500 38300 40000

41600

53900 57200 60300

46200 49000 51600

54300 56800 59200 61400 63600

66400 63600 69100 66200 71700 68700 74200 71200

66000 68300


Pressure, in. of water 0 lx -.-_ -

1 .o 43900 1.2 48000 1.4 51900 1.6 55500 1.8 58900

2.0 62000 2.2 65100 2.4 67900 2.6 70700 2.8 73300

3.0 76000 3.2 78500 3.4 80900 3.6 83300 3.8 85500

4.0 87800 4.5 93100 5.0 98100 5.5 103000 6.0 107000

6.5 112000 7.0 116000 8.0 124000 9.0 132000 10.0 138000

11.0 145000 12.0 152000 13.0 158000 14.0 164000 15.0 170000

TABLE 13.2a-CAPACITIES OF ORIFICE WELL TESTERS’ (continued)

0.70 -

Specific gravity

40600 445oa 48000 51400 54500

57400 60200 62900 65500 67900

70300 72700 74900 77000 79200

81200 86100 90800 95200 99500

104000 107000 115000 122000 128000

135000 141000 147000 152000 158000

0.80 0.90 1 .oo 1.10

I %-in. Orifice in h-in. Plate

1.20

38000 41600 44900 48100 51000

53700 56300 58800 61200 63600

65800 68000 70100 72100 74000

76000 80600 84900 89100 93000

96900 101000 107000 114000 120000

126000 132000 137000 142000 147000

32400 35500 38300 41000 43500

45800 48000 50200 52200 54200

58100 58000 59700 61500 63100

64800 68700 72400 76000 79300

82600 85700 91600 92700 1020ocl

35800 34000 39200 37200 42400 40200 45300 43000 48100 45600

53100 55500 57700 59900

62000 64100

67900 69800 71600 76000 80100 84000 87700

91300 98400 101000 107000 113000

119000 124000

50400 52600 54800 56800

58900 60800 62700 64500 66200

68000 72100 76000 79700 83200

86600 89900 96100 102000 107000

113000 118000 123000 127000 132000

112000 117000 121000 126000

31000 34000 36700 39200 41600

43900 46000 48100 50000 51900

53700 55500 57200 58800 60500

62000 65800 69400 72700 76000

79100 82100 87700

98100

103000 108000 112000

1.30

29800 32600 35300 37700 40000

42100 44200 46200 48000 49900

51600 53300 55000 56500 58100

59600 63200 66600 69900 73000

76000 78900 84300 89400 94200

103000 108000

116000 112000 120000 116000

1.50

27700 30400 32800 35100 37200

39200 41100 43000 44700 46400

48100 49600 51200 52600 54100

55500 58800 62000 65100 67900

70700 73400 78500 83200 87700

92000 96200 100000 104000 108000

‘Values are cu 11 m 24 hours ar a base pressure 01 14 65 ps~a. base and flowing temperature. 6O’F. The volumes obtafned from these tables apply only lo Orifice Well Testers manufactured by American Meter Company. For other testers, use the tables recommended by the manufacturer The rates gwen I” the tables for water are based on the following equabon’

where q = rate ,n cub,c fee, I” 24 hours,

C = rate for I I”. given in the tables, and p = pressure I” inches of water.

The rates gwn an the tables for mercury are based on the followlng equation which was determined from a 88118s Of 100 tests covering the limits of the tables.

where q = rate in thousands 01 cubic feet in 24 hours. C = 7 464 for U-!n orifice,

= 13.61 for i-117 orifice, = 23 10 for II/r-in. orlflce.

Pm = pressure in Inches of mercury, 29.32 = assumed atmospherfc pressure I” Inches of mercury

= 14.4 ps,, 0 3 = factor determined by tests, and

7 s = speaffc gravity of gas.

13-42

Pressure, in. of mercury

1 .o 1.1 1.2 1.3 1.4

1.5 1.6 1.7 1.8 1.9

2.0 2.2 2.4 2.6 2.8

3.0 3.2 3.4 3.6 3.8

4.0 4.5 5.0 5.5 6.0

6.5 7.0 8.0 9.0

10.0

11.0 12.0 13.0 14.0 15.0

16.0 17.0 18.0 19.0 20.0

21.0 22.0 23.0 24.0 25.0

26.0 27.0 28.0 29.0 30.0

31.0 32.0 33.0 34.0 35.0

36.0 37.0 38.0 39.0 40.0


TABLE 13.2b-CAPACITIES OF ORIFICE WELL TESTERS’

SDeclfic aravitv

0.60

52 55 58 60 62

64 67 69 71 73

75 78 82 85 89

92 95 98 101 104

107 113 120 126 132

137 143 154 164 173

183 192 200 209 217

225 233 241 249 256

264 271 278 285 292

300 306 313 320 327

334 340 347 353 360

366 373 379 386 392

0.70 0.80 0.90 1.00 1.10 ~ ~ ~ - - U-in. Orifice in %-in. Plate

49 51 53 55 58

60 62 64 65 67

69 72 76 79 82

85 88 91 93 96

99 105 111 116 122

127 132 142 151 160

169 177 185 193 201

208 216 223 230 237

244 251 257 264 271

277 284 290 296 302

309 315 321 327 333

339 345 351 357 363

45 43 48 45 50 47 52 49 54 51

56 53 58 54 59 56 61 58 63 59

65 61 68 64 71 67 74 70 77 72

79 75 82 77 85 80 87 82 so 85

92 87 98 92 103 98 109 103 114 108

119 112 124 117 133 125 142 134 150 142

158 149 166 156 173 164 181 170 188 177

195 184 202 190 209 197 215 203 222 209

228 215 234 221 241 227 247 233 253 239

259 245 265 250 271 256 277 261 283 267

289 272 294 278 300 283 306 289 311 294

317 299 323 304 328 310 334 315 339 320

41 39 37 36 33 43 41 39 37 35 45 43 41 39 36 46 44 42 41 38 48 46 44 42 39

50 48 45 44 41 52 49 47 45 42 53 51 46 47 43 55 52 50 48 45 56 54 51 49 46

58 55 53 51 47 61 58 55 53 49 63 60 58 56 52 66 63 60 58 54 69 65 63 60 56

71 68 65 62 58 73 70 67 64 60 76 72 69 67 62 78 74 71 69 64 80 77 73 70 66

82 79 75 72 67 88 84 80 77 72 93 88 85 81 76 97 93 89 85 79 102 97 93 89 83

106 102 97 93 87 111 106 101 97 so 119 113 108 104 97 127 121 116 111 103 134 128 122 118 109

141 135 129 124 115 148 142 135 130 121 155 148 142 136 127 162 154 148 142 132 168 160 153 148 137

174 166 159 153 142 180 172 165 158 147 187 178 170 164 152 192 184 176 169 157 198 189 181 174 162

204 195 186 179 167 210 200 191 184 171 215 205 197 189 176 221 211 202 194 180 226 216 207 199 185 232 221 212 204 189 237 226 216 208 194 242 231 221 213 198 248 236 226 218 202 253 241 231 222 207

258 246 236 227 211 263 251 240 231 215 268 256 245 236 219 273 261 250 240 223 278 266 254 245 227

283 271 259 249 231 288 275 263 253 236 293 280 268 258 240 298 285 272 262 244 303 289 277 266 248

1.20 1.30 1.50

GASMEASUREMENTAND REGULATION

TABLE 13.2b-CAPACITIES OF ORIFICE WELL TESTERS’ (continued)


1.0 1.1 1.2 1.3 1.4

1.5 1.6 1.7 1.8 1.9

2.0 2.2 2.4 2.6 2.8

3.0 3.2 3.4 3.6 3.8

4.0 4.5 5.0 5.5 6.0

6.5 7.0 8.0 9.0

10.0

11.0 12.0 13.0 14.0 15.0

16.0 17.0 18.0 19.0 200

21.0 22.0 23.0 24.0 25.0

26.0 27.0 26.0 29 0 30.0

31.0 32.0 33.0 34.0 35.0

36.0 37.0 380 39.0 400

Specific gravity

0.60

96 100 105 109 113

117 121 125 129 132

136 143 149 155 161

167 173 179 184 189

194 206 218 229 240

251 261 280 298 316

333 349 365 381 396

411 425 439 453 467

481 494 507 520 533

546 559 571 583 596

608 620 632 644 656

668 679 691 703

0.70 0.80 0.90 1 .oo 1.10 - - ~

97 101 105

109 112 116 119 123

126 132 138 144 150

155 160 165 170 175

180 191 202 212 222

232 241 259 278 293

308 323 338 352 366

380 394 407 420 432

445 457 470 482 494

506 517 529 540 552

563 574 585 596 607

618 629 640 651

714 661

l-in. Orifice in Win. Plate 83 87 91 95 98

102 105 108 112 115

118 124 129 135 140

145 150 155 159 164

168 179 189 199 208

217 226 242 258 274

288 303 316 330 343

356 368 381 393 405

416 428 439 451 462

473 484 495 505 516

527 539 548 558 568

578 589 599 609 619

78 82 86 89 93

96 99 102 105 108

111 117 122 127 132

137 141 146 150 154

159 169 178 187 196

205 213 229 244 256

272 285 298 311 323

335 347 359 370 381

392 403 414 425 435

446 456 466 477 487

496 506 516 526 536

545 555 565 574 583

74 78 81 77 85 81 88 84

91 87 94 90 97 92

100 95 103 98

105 100 111 105 116 110 120 115 125 119

130 124 134 128 138 132 142 136 146 140

150 143 160 153 169 161 178 169 186 177

194 185 202 192 217 207 231 220 245 233

258 246 271 258 283 270 295 281 306 292

318 303 329 314 340 325 351 335 362 345

372 355 383 365 393 375 403 384 413 394

423 403 433 413 442 422 452 431 462 440

471 449 480 458 490 467 499 476 508 485

517 493 526 502 535 511 544 519 553 528

1.20 1.30 1.50

68 65 60 71 68 63 74 71 66 77 74 69 80 77 72

83 80 74 86 83 77 68 85 79 91 88 81 94 so 84

96 92 86 101 97 so 105 101 94 110 106 98 114 110 102

118 114 106 122 117 109 126 121 113 130 125 116 134 128 120

137 132 123 146 140 131 154 148 138 162 156 145 170 163 152

177 170 158 184 177 165 198 190 177 211 203 189 223 215 200

235 226 210 247 237 221 258 248 231 269 259 241 280 269 250

290 279 260 300 289 269 311 299 278 320 308 287 330 317 295

340 327 304 349 336 312 358 345 321 368 354 329 377 362 337

386 371 345 395 380 353 404 388 361 412 397 369 421 405 377

430 413 384 438 421 392 447 430 400 455 438 407 464 446 415

472 454 422 480 462 430 489 470 437 497 478 444 505 485 452

13-43

13-44


Specific gravity

0.60 0.70 0.80 0.90 1.00 1.10 1.20 1.30 1.50

1 .o 162 1.1 170 1.2 178 1.3 185 1.4 192

1.5 199 1.6 206 1.7 212 1.8 219 1.9 225

2.0 231 2.2 242 2.4 253 2.6 264 2.8 274

3.0 284 3.2 293 3.4 303 3.6 312 3.8 321

4.0 330 4.5 350 5.0 370 5.5 389 6.0 407

6.5 425 7.0 442 8.0 475 9.0 506 10.0 536

11.0 565 12.0 593 13.0 620 14.0 646 15.0 672

16.0 697 17.0 721 18.0 746 19.0 769 20.0 793

21.0 816 22.0 838 23.0 861 24.0 883 25.0 905

26.0 927 27.0 948 28.0 969 29.0 990 30.0 1011

31.0 1032 32.0 1052 33.0 1073 34.0 1093 35.0 1113

36.0 1133 37.0 1153 38.0 1173 39.0 1193 40.0 1212

PETROLEUMENGINEERING HANDBOOK

TABLE 13.2b-CAPACITIES OF ORIFICE WELL TESTERS’ (continued)

- - - - - II/bin. Orifice in %-in. Plate

150 158 165 171 178

184 191 197 202 208

214 224 234 244 254

263 272 281 289 297

305 324 343 360 377

394 409 440 469 496

523 549 574 598 622

645 668 690 712 734

755 776 797 818 838

858 878 897 917 936

955 974 993 1012 1031

1049 1068

141 147 154 160 167

173 178 184 189 195

200 210 219 229 237

246 254 262 270 278

285 304 321 337 353

368 383 411 439 464

489 514 537 559 582

604 625 646 666 667

706 726 745 765 784

803 821 839 858 876

894 912 929 947 964

982 999 1016 1033 1050

132 126 139 132 145 138 I51 143 157 149

163 154 168 160 173 164 178 169 184 174

188 179 198 188 206 196 215 204 224 212

232 220 240 227 247 235 255 242 262 249

269 255 286 271 302 287 318 301 333 316

347 329 361 343 388 368 413 392 438 415

461 438 484 459 506 480 527 500 548 520

569 540 589 559 609 578 628 596 647 614

666 632 684 649 703 667 721 684 739 701

757 718 774 734 789 751 809 761 826 783

843 799 859 815 876 831 893 847 909 862

925 878 942 893 958 909 974 924 990 939

120 126 132 137 142 147 152 157 161 166

171 179 187 195 202

210 217 224 230 237

243 259 273 287 301

314 327 351 374 396

417 438 458 477 496

515 533 551 568 586

603 619 636 652 668

684 700 716 732 747

762 777 793 808 822

837 852 867 881 896

115 110 103 120 116 108 126 121 113 131 126 117 136 131 122

141 135 126 146 140 130 150 144 134 155 149 138 159 153 142

163 157 146 171 165 153 179 I72 160 187 179 167 194 186 173

201 193 180 208 199 186 214 206 192 221 212 197 227 218 203

233 224 208 248 238 222 262 251 234 275 264 246 288 277 258

301 289 269 313 300 280 336 323 300 358 344 320 379 364 339

399 384 357 419 403 375 438 421 392 457 439 409 475 456 425

493 473 441 510 490 456 527 507 472 544 523 487 561 539 501

577 554 516 593 570 530 609 585 544 624 600 558 640 615 572

655 629 586 670 644 600 685 658 613 700 673 626 715 687 640

730 701 653 744 715 666 759 729 679 773 743 691 787 756 704

801 770 717 816 783 729 830 797 742 844 810 754 857 824 767

‘Values are thousands 01 c” 11 I” 24 hours at a base pressure ol 14 65 pm; base and flomg temperature. 6O’F.


The orifice used differs from that in the orifice meter. This one is thicker and has a rounded edge. This edge is placed toward the flow, for experience has shown that a sharp-edged orifice does not give reproducible results under critical-flow conditions. In fact, sharp-edged orifices do not conform to existing theories and correlations.

The general equation for a critical-flow prover is

CP qs = (yRT~0,5 , . . . . . . . . . . . . . . .

where p = pressure on prover, psia,

C = orifice coefficient for prover,

9s = rate of flow, thousands of standard cu fit24 hr, measured at 14.4 psia and 60”F,

yK = specific gravity of gas (air= 1 .OO), and

T = absolute temperature, “R.

The critical-flow prover is one of the basic devices used for determining the gas-flow rate in the open-flow testing of gas wells. Values of the coefficient C may be found in Table 13.3.*

Pitot Tube. The pitot tube is another measuring device used extensively for testing flow rates during tests. It works by measuring the difference between the impact pressure at the tip and the static pressure in the flowing stream. This impact pressure results from conversion of the kinetic energy of the flowing gas to pressure. If the conversion efficiency is relatively constant, a conversion between pressure and flow rate is possible. The pitot tube is normally made of L/s-in.-ID pipe and is inserted in the center of a nipple at least eight diameters long.

Table 13.4 gives the flow rate in thousands of cubic feet per day for a pitot tube inserted in different-diameter nipples.

A pitot tube is used largely for temporary flow measurement since it is small and easy to handle. Very few permanent installations are made, for it produces low-pressure differentials, is difficult to calibrate, and often clogs.

A pitot tube measures a point velocity, i.e., the velocity at only one point across the cross section of the pipe. Inasmuch as the velocity varies throughout this cross section, the problem of proper location presents itself. In the absence of unknown disturbing elements such as pipe burrs or undue roughness, the velocity at the center is theoretically 20% greater than the mean velocity. For approximate measurement, the standard tables use this factor to convert readings taken at the center of the pipe to a volume flow rate. For exact work, the mean velocity would be found by taking a series of experimental velocity measurements across the pipe diameter.

A similar method is used for large gas-flow rates and/or where debris produced with the gas makes other methods unfeasible. This consists simply of measuring

‘A set Of constants m Si metric units has not been officially adopted at this tome. Therefore the constants in customary units are used (Table 13.3) and final results are converted to SI metric umts. The readers are advised, however. to refer to the of- facial standards in their area for the appropriate tables and equations presented in this chapter.

13-45

TABLE 13.3-ORIFICE COEFFICIENTS FOR CRITICAL-FLOW PROVERS

Size of orifice (in.)

‘/I 6 3h ‘h %rs %2 ‘/4

=/Is

%

% 6 ‘/2

% 44

‘ii 1

1% IV4

1% 1% 1% 2

2% 3

Orifice Coefficient

2-h. Prover 4-in. Prover

1.524 -

3.355 - 6.301 - 14.47 - 19.97 -

25.86 24.92 39.77 -

56.68 56.01 81.09 100.2 101.8 156.1 154.0 223.7 224.9 304.2 309.3 396.3 406.7 499.2 520.8 616.4 657.5 742.1 807.8 884.3

1,002.o 1,208.O - 1,596.0 - 2,566.0 - 3,904.o -

the static pressure through a side opening four diameters from the end of a nipple at least eight diameters long. Values for this system are also shown in Table 13.4. This is the least desirable of the methods discussed because of the potential error involved.

Further data on the use of these four devices for the measurement of gas flow may be found in U.S. Bureau of Mines Monograph 7.

Other Meters That Use Velocity. Rotameter. This is a variable-orifice meter useful for the indication of flow rates in operations (Fig. 13.5). It is economical in sizes up to 10 cm [4 in.], at moderate pressures. It can transmit electrical or pneumatic signals. Its rangeability is about 1O:l.

Eccentric Orifices. These are used for many two- phase streams including those containing solids (Fig. 13.6). Substantial straight-line pipe sections are required both upstream and downstream.

Segmental Orifices. These possess the same basic applicability as the eccentric orifices with an additional advantage that they do not dam solids on the upstream side of the plate (Fig. 13.7). These are used primarily for large line sizes and low fluid viscosities.

Centrijkgal (Elbow) Meters. This is used for large pipe sizes primarily. It is based on the centrifugal force induced as the fluid changes direction (Fig. 13.8). This centrifugal force is a function of velocity, R and D, as well as fluid properties. A common pipe elbow is generally used. If very accurate results are desired, calibration is necessary. Its rangeability is about 3: 1.

Turbine Meters. The turbine meter has a wide range of applicability except for high-viscosity fluids. Range- ability is about 2O:l. Turbine meters come in various configurations, but they all transmit a signal based on rotation speed, which in turn is a function of flow rate and fluid properties (Fig. 13.9). It is difficult to generalize about this type of meter because some have

13-46

Impact pressure Open flow, McflD

in. of water

in. of mercury psi

0.1 0.2 0.3 0.4 0.5

0.6 0.7 0.8 0.9 1.0

1.25 1.36 1.6 1.8 2.0

2.2 2.4 2.7 3.0 3.5

4.1 4.5 5.0 5.4 6.0

6.8 8.2 9.0 9.5 10.0

10.9 12.0 12.2 13.9 15.0

16.3 17.7 19.0 20.4 21.8

24.5 27.2 29.9 32.6

- - - - - -

-

0.10 0.12 0.13 0.15

0.16 0.18 0.20 0.22 0.26

0.30 0.33 0.37 0.40 0.44

0.50 0.60 0.66 0.70 0.74

0.80 0.88 0.90 1.02 1.1

1.2 1.3 1.4 1.5 1.6

1.8 2.0 2.2 2.4 2.6

2.8 3.0 3.2 3.4 3.6

- -

- - - - - -

- -

-

- -

-

- -

- - -

-

-

0.5

- - - - -

-

1.0 -

-

-

1.5 - -


TABLEl3.4-PITOT-TUBE VALUES

1 in.

10.97 15.52 19.00 21.95 24.53

26.89 29.03 31.02 32.92 34.69

38.78 40.45 43.89 46.56 49.00

51.45 53.74 57.20 60.02 64.91

70.01 73.60 77.57 80.90 84.91

90.48 99.20 104.0 107.0 109.7

114.5 120.1 121.4 129.2 134.2

140.1 145.8 151.4 156.7 161.8

171.7 180.9 189.7 198.0 206.1

214.0 221.6 228.9 235.8

- 242.8

2 m. 3 in. 4 in. 6 In. 10 in. 12 in.

8’: 549 790 1,410 98 613 882 1,570

108 672 968 1,720 116 726 1,050 1,860 124 776 1,120 1,990 132 823 1,180 2,110 139 867 1,250 2,200

155 969 1,400 2,480 162 1,010 1,460 2.590 175 1,100 1,580 2,810 186 1,160 1,680 2.980 196 1,230 1,760 3,140

206 1,290 1,850 3,290 214 1,340 1,930 3,440 228 1,430 2,060 3,660 240 1,500 2,160 3,840 260 1,620 2,340 4,160

280 1.750 2.520 4,480 295 1,840 2,650 4,710 310 1,940 2,790 4,960 324 2,020 2,910 5,180 340 2,120 3,060 5,430

362 2,260 3,260 5,790 9,500 396 2,480 3,570 6,350 10,000 416 2,600 3,750 6,660 10,400 428 2,680 3,850 6,850 10,700 439 2,740 3,950 7,020 11,000

458 2,860 4,120 7,330 11,500 481 3,000 4,300 7,690 12,000 486 3,040 4,370 7,770 12,100 517 3,230 4,650 8,270 12,900 537 3,360 4,830 8,590 13,400

560 3,500 5,040 8.960 14,000 584 3,650 5,250 9,330 14,600 606 3,790 5,450 9,680 15,100 627 3,920 5,640 10,000 15,700 648 4,050 5,820 10,400 16,200

686 4,290 6,180 11,100 17,200 734 4,520 6,510 11,600 18,100 768 4,740 6,830 12,100 19,000 802 4,950 7,130 12,700 19,800 824 5,150 7,420 13,200 20,600

857 1,930 3,420 5,350 7,700 13,700 21,400 887 2,000 3,500 5,540 7,980 14,200 22,600 917 2,060 3,660 5,720 8,240 14,600 22,900 943 2,120 3,770 5,900 8,480 15,100 23,600 971 2,180 3,880 6,070 8,740 15,500 24,300 35,000

99 140 171 198 221

242 261 279 296 312

349 364 395 419 441

463 483 515 540 584

630 662 698 728 764

814 892 936 962 987

1,030 1,080 1,090 1,160 1,210

1,260 1,310 1,360 1,410 1,460

1,550 1,630 1,710 1,780 1,860

176 248 304 351 392

430 464 497 526 555

620 648 702 744 784

823 860 915 961

1,040

1,120 1,180 1,240 1,300 1,360

1,450 1,590 1,670 1,710 1,760

1,830 1,920 1,940 2,070 2,150

2,240 2,330 2,420 2,510 2,590

2.750 2,890 3,040 3,170 3,3ocl

5 in.

274 395 388 475

559 684

8 In.

702 944

1,220

1,100 1,550 1,900 2,200 2,450

2,690 2,900 3,100 3,290 3,470

3,880 4,050 4,390 4,660 4,900

5,150 5,370 5,720 6,000 6,490

7.000 7,360 7,760 8,090 8,490

1,580 2,240 2,740 3,160 3,530

3,870 4,180 4,470 4,740 5,000

5,580 5,820 6,320 6,700 7,060

7,410 7,740 8,230 8,640 9,340

10,100 10,600 11,200 11.700 12.200

13,000 14,300 15,000 15,400 15,800

16,500 17,300 17,500 18.600 19,300

20,200 21,000 21,900 22,600 23.300

24,700 26,000 27.300 28,500 29.700

30,800 31,900 32,000 34,000

GAS MEASUREMENTANDREGULATION

TABLE 13.4-PITOT-TUBE VALUES (continued)

13-47

Impact pressure

in. of mercury osi

3.8 - 4.0 2.0 4.2 -

4.4 -

4.6 -

4.8 -

5.0 2.5 5.2 5.4 -

5.6 -

5.8 6.0 3.0 6.5 -

7.0 3.5 7.5 -

8.0 4.0 8.5 9.0 4.5 9.5 -

10.0 -

10.2 11.2 12.2 13.2 14.3

15.3 16.3 17.3 18.3 19.3

20.4 22.4 24.4 26.5 28.5

5.0 5.5 6.0 6.5 7.0

7.5 8.0 8.5 9.0 10.0

10.0 11.0 12.0 13.0 14.0

15.0 16.0 17.0 18.0 19.0

20.0 21.0 22.0 23.0 24.0

25.0 26.0 27.0 28.0

1 in. 2 in.

249.4 998 255.9 1,020 262.0 1,050 268.4 1,070 274.5 1,100

280.3 1,120 286.1 1,140 291.8 1,170 297.4 1,190 302.7 1,210

308.1 1,230 313.4 1,250 326.0 1,300 336.6 1,350 350.0 1,400

361.5 1,450 373.9 1,500 383.9 1.540 394.2 1,580 404.6 1,620

408.1 1,630 428.0 1,710 447.0 1,790 465.5 1,860 483.0 1,930

500.0 2,000 516.0 2,060 532.1 2,130 548.0 2,190 563.0 2,250

577.6 2,310 605.6 2,420 632.5 2,530 658.0 2,630 683.8 2,740

707 2,830 731 2,930 755 3,020 779 3,120 802 3,210

826 3,310 850 3,400 874 3,500 898 3,590 922 3,690

946 3,780 969 3,880 993 3,970

1,017 4,070 29.0 1,040 4,160

4 in 5 in. 6 in. 10 tn.

2,240 2,300 2,360 2,410 2,470

2,520 2,570 2,630 2,680 2,720

2,770 2,820 2,930 3,050 3,150

3,250 3,370 3.460 3,550 3.640

3,689 3,850 4,030 4,190 4,350

4,500 4,650 4,790 4,930 5,070

5,200 5,450 5,700 5,920 6,150

6,360 6,580 6,800 7,010 7,220

7,440 7,650 7,870 8,080 8,300

8,520 8,720 8,940 9,150

8,980 9,210 9,430 9,650 9,880

10,100 10,300 10.500

3 In. 8 m. 12 In.

3,990 16,000 24,900 35,900 4,090 16,400 25,600 36,800 4,190 16,800 26,200 37,700 4,290 17,200 26,800 38,600 4,390 17,600 27,500 39,500

4,490 18,000 28,000 40,400 4,580 18,300 28,600 41,200 4,670 18,700 29,200 42,000 4,760 10,700 19,000 29,700 42,800 4,840 10,900 19,400 30,300 43,600

4,920 11,100 19,700 30,800 44,300 5,010 11,300 20,000 31,300 45,100 5,220 11,700 20,900 32,600 47,000 5,420 12,200 21,700 33,900 48,800 5,600 12,600 22,400 35,000 50,400

5,780 13,000 23,100 36,200 52,100 5,980 13,400 23,900 37,400 53,700 6,140 13,800 24,600 38,400 55,300 6,310 14,200 25,200 39,400 56,800 6,470 14,600 25,900 40,500 58,200

6,540 14,700 26,100 40,800 58,800 6,850 15,400 27,400 42,800 61,600 7,150 16,100 28,600 44,700 64,400 7,450 16,800 29,800 46,600 67,000 7,730 17,400 30,900 48,600 69,600

8,000 18,000 32,000 50,000 72,000 8,260 18,600 33,000 51,600 74,300 8,520 19,200 34,100 53,200 76,600 8,770 19,700 35,100 54.800 78,900 9,000 20,300 36,000 56,300 81,100

9,240 20,800 37,000 57,800 83,200 9.680 21,800 38.800 60,600 87,200 10,100 15,800 22,800 40,500 63,300 91,200 10,500 16,500 23,700 42,100 65,800 94,800 10,900 17,100 24,600 43,800 68,400 98,600

11,300 17,700 25,500 45,200 70,700 102,000 11,700 18,300 26,300 46,800 73,100 105,000 12,100 18,900 27,200 48,300 75,500 109,000 12,500 19,500 28,000 49,900 77,900 112,000 12,800 20,100 28,900 51,300 80,200 115,000

13,200 20,700 29,700 52,900 82,600 119,000 13,600 21,300 30,600 54,400 85,000 122,000 14,000 21,900 31,500 55,900 87,400 126,000 14,400 22,500 32,300 57,500 89,800 129,000 14,800 23,100 33,200 59,000 92,200 133,000

15,100 23,700 34,100 60,500 94,600 136,000 15,500 24,200 34,900 62,000 96,900 140,000 15,900 24,800 35,700 63,500 98,300 143,000 16,300 25,400 36,600 65,100 102,000 146,000

9,360 16,600 26,200 37,400 66,600 104,000 150,000

6,230 6,400 6,550 6,710 6,860

7,010 7,150 7,300 7,440 7,560

7,700 7,830 8,150 8,460 8,760

9,040 9,340 9,600 9,860 10,100

10,200 10,700 11,200 11,600 12,100

12,500 12,900 13,300 13,700 14,100

14,400 15,100

Open flow, McflD

13-48

Impact pressure

psi

30.0 32.0 34.0 36.0 38.0

40.0 45.0 50.0 55.0 60.0

65.0 70.0 75.0 80.0 90.0

100.0 110.0 120.0 130.0 140.0

150.0 160.0 170.0 180.0 190.0 200.0


TABLE 13.4-PITOT-TUBE VALUES (continued)

Open flow. McflD

1 in. 2 in.

-4,260 1,064 1,112 4,450 1,159 4,640 1,207 4,830 1,255 5,020

1,302 5,210 1,421 5,690 1,540 6,160 1,660 6,640 1,778 7,120

1,898 7,600 2,017 8,060 2,136 8,840 2,252 9,010 2,492 9,980

2,732 10,900 2,970 11,900 3,208 12,800 3,445 13,800 3,681 14,700

3,921 15,700 4,160 16,700 4,399 17,600 4,635 18,600 4,870 19,500 5,108 20,500

3 in. 4 In.

9,580 17,000 10,000 17,800 10,400 18,600 10,900 19,300 11,300 20,100

11,700 20,800 12,800 22,800 13,900 24,700 15,000 26,600 16,000 28,500

17,100 30,400 18,200 32,300 19,200 34,200 20,300 36,000 22,400 39,900

24,600 43,700 26,700 47,500 28,900 51,300 31,000 55,100 33,100 58,900

35,300 62,800 37,500 66,600 39,600 70,400 41,700 74,200 43,900 78,000 48,000 81,800

Fig. 13.5-A typical rotameter.

Fig. 13.6-An eccentric orifice.

Fig.13.7-A segmental orifice

5 in. 6 in. 8 in. 10 in. 12 in.

26,600 38,300 68,100 106,000 153,000 27,800 40,100 71,200 111,000 160,000 29,000 41,700 74,200 116,000 167,000 30,200 43,500 77,300 121,000 174,000 31,400 45,200 80,300 126,000 181,000

32.600 46,900 83,400 130,000 188,000 35,500 51,200 91,000 142,000 205,000 38,500 55,400 98,600 154,000 222,000 41,500 59,800 106,000 166,000 239,000 44,500 64,000 114,000 178,000 256,000

47,500 68,400 122,000 190,000 273,000 50,400 72,600 129,000 202,000 290,000 53,400 76,800 137,000 214,000 308,000 56,400 81,100 144,000 225,000 324,000 62,400 89,800 160,000 249,000 359,000

68,400 98,400 175,000 273,000 394,000 74,200 107,000 190,000 297,000 428,000 80,200 116,000 205,000 321,000 462,000 86,200 124,000 221,000 345.000 496,000 92,000 133,000 236,000 368,000 530,000

98,000 141,000 251,000 392,000 565,000 104,000 150,000 266,000 416,000 599,000 110,000 158,000 282,000 440,000 634,000 116,000 167,000 297,000 464,000 668,000 122,000 175,000 312,000 487,000 702,000 128,000 185,000 327,000 511,000 736,000

proven very good and some very bad in a given application. For this reason, the choice should be based on experience with a given model in a given service.

Bearings can be troublesome. Ball bearings may be susceptible to damage by abrasive solids; sleeve bearings tend to be more trouble-free but rangeability may be reduced.

When properly selected, turbine meters are very acceptable. Their acceptability for custody transfer has been slow in some areas because of lack of proven experience compared with the orifice meter. Liquid turbine meters have been accepted more readily than those for gas service because they could be tested more readily.

Repeatability is often +0.05 % on test and accuracy is within f 1%. Like all meters, these values are a function of meter condition.

The axial flow turbine meter has proved useful for high gas-flow rates. Its rangeability is up to 100: 1 and it has a linear response to flow rate. A summary of performance of these meters is found in Refs. 5 and 6.

Vortex Shedding Meter. This is a newer meter for measuring all fluids that do not have too high a viscosity (Fig. 13.10). It has no moving parts, wide rangeability, and is linear over wide ranges of flow.

Fig. 13.11 shows the principle involved when a fluid flows around a blunt object. The flow is unable to follow the surface of the object and sheds from it at some point to form a continuous series of eddy currents. The frequency of shedding is proportional to flow velocity and inversely proportional to the diameter of the object. This shedding may be sensed by thermistors or by what is called a “shuttle ball” contained in the fixed flow element. Movement of this ball is detected by a magnetic sensor. This ball-type is suitable for low-density, clean liquids.


Sonic Meters. Sonic meters offer many potential advantages if they can be made so that they remain stable electronically and overcome drift tendencies. They possess no moving parts and have linearity.

They contain three components: a transducer assembly, transmitter, and a receiver. The transducer assembly shown in Fig. 13.12 contains two transducers that alternately transmit sonic pulses diagonally across the pipe. The difference in transit times is proportional to flow velocity.

Regulation Principles of Control

The regulation of pressure and temperature flow or level is an integral part of the regulation of gas and liquid streams. A means is necessary to sense the value of the controlled variables and then cause the movement of a valve to maintain the value desired. The communication link between sensing element and valve may use gas (pneumatic), liquid (hydraulic), electricity, and electronics, More than one may be used in a given system by employing a transducer, a device for converting signals.

Historically, most of the systems used pneumatic controls, so they will be used to illustrate the principles involved. However, the use of electrical and electronic controls is increasing rapidly. When a control system is chosen, these also should receive serious consideration.

Many of the routine problems encountered with lease equipment stem from failure or misapplication of controls. This has become even more pronounced in recent years as this equipment has become more complex.

Oilfield instrumentation is less complex than that in plants and refineries but imposes severe service conditions. In such service, the instruments often have only wet supply gas and rare adjustment by expert servicemen and are exposed to the elements, yet must give continuous and trouble-free service. Even with these severe conditions, it must be concluded that many troubles stem from improper application rather than inadequacy of the controls themselves, although many types are not suitable for oil-field use.

Definitions

To choose controls properly, it is necessary to know not only the requirements of the process but also the corresponding characteristics of the controls available. No understanding of controls may be complete without a familiarity with applicable definitions.

Pneumatic controls are those actuated by air or gas (the most common type used in the field).

Diaphragm motor valve is the term applied to a complete valve that uses pressure to open and close it.

Topworks (motor) is that portion of the motor valve that contains the mechanism to open and close the valve.

Inner-valve assembly (trim) includes the stem and seat within the valve body that actually controls the flow of fluid.

Fluid is any liquid or gas being controlled. Proportional control describes the system whereby the

valve opening is proportional to the degree of change in the controlled variable.

Snap action is a mode of control whereby the valve is either wide open or closed (on/off control).

Fig. 13.8-A centrifugal (elbow) meter.

Fig. 13.9-A turbine meter

Fig. 13.10-Karman vortex trail

Fig. 13.11 -Vortex flow pattern.

w,

Fig. 13.12-A sonic flowmeter

Reset is an addition to a proportional-control system to enable the instrument to hold itself at the control point as the process load varies.

Derivative response is a further addition that provides corrective action based on the time rate of change of the deviation from the control point.

Se/j=operated controller is a valve that is actuated directly by the controlled variable.

Pilot is a relay that transforms the controlled variable into an equivalent signal to the control valve, such signal controlling the action of that valve.

Supply gas is that gas necessary in a pneumatic pilot to operate it.

Controlled variable is the pressure, liquid level, temperature, or flow rate being controlled.

Measuring means is the means used to detect any change in the controlled variable.

Smsirivity is the ability to detect small deviations in the controlled variable.

Reproducibility is the ability of an instrument to repeat and measure consistently the values of a static condition

over a period of time. Static error is the difference between the absolute

value of the controlled variable and the measured value. Lag is the period of time by which the measured value

follows the change in the absolute value of the controlled

variable. Sturic condirions occur when all changes in the con-

trolled variable are instantaneous. Dynamic con&ions occur when the controlled

variable is continually changing. Normally closed r’ulve is a valve that is held closed by

a spring or some similar device and is opened by the action of pilot and/or the controlled variable.


Normally open valve is the reverse of the above; one that is closed by the action of the pilot and/or the controlled variable.

Drifr includes reproducibility and the inability to repeat a measurement because of changes in the measured variable.

Process Characteristics

Because the purpose of an automatic controller is to regulate a process, it is fundamental that the properties and characteristics of a process be understood. A process is defined as the collective functions performed in and by equipment in which a variable is controlled. As an example, a field heater that heats well effluent by circulating hot water is a unit of equipment in which the process of heating the wellstream is accomplished. The process consists of a controlled variable (the temperature) and a controlled medium (the wellstream). Other controlled variables could be rate of flow, liquid level, or pressure. A control agent (circulating water) is the medium for effecting the temperature change, and thus regulation of the control agent regulates the controlled variable. The total requirements of the process for the control agent at any one time are defined as the process load. If the rate of flow of the wellstream increases, then more or hotter water is required to maintain the same temperature, and a change in process load has taken place. By the same token, a drop in inlet temperature at the original rate of flow would also constitute a process-load change.

The properties of an entire process include its potential, capacitance, resistance, and dead time. Capaci- tance, not to be confused with capacity, is the change in the quantity of energy or material per unit change in some variable, usually the controlled variable.

Resistance, or opposition to the flow of energy or material, is another process characteristic. The most familiar concept of resistance is in electricity when it is expressed as the ohm.

Potential is most generally recognized in the concept of electricity where it is expressed as a volt. Table 13.5 summarizes the characteristics of a process.

Dead time is the time lag that occurs when energy is being transferred at a constant rate of flow through a given distance. It is equal to the time it takes the energy to move the distance. The dead time is a characteristic of the process, and it is not to be confused with the time lag inherent in the automatic controller itself.

The purpose of the automatic controller is to prevent deviation of a process from a desired standard. This may be accomplished by various modes of control action, each of which incorporates a distinct limit, range, and speed of correction.

TABLE 13.5-PROCESS CHARACTERISTICS

Process Characteristic Thermal Pressure Liquid Level Electrical

Capacity Btu cu ft cu ft Coulomb Potential degree psi ft volt

Capacitance Btuldegree cu ft/psi cu ftlft Farad

Resistance degree psi ft ohm

Btu/sec cu ftlsec cu fVsec


Fig. 13.13-Liquid-level control using displacement-type float.

Proportional Control. The modes used depend on the frequency and magnitude of the process load changes, the degree of control needed, and the dynamic lag inherent in the process and controls used.

Proportional control is the basic action that is employed in all controllers not using snap (on/off) action. Because it is defined as control action in which a continuous linear relation exists between the value of the controlled variable and the position of the valve, the following equation may be written.

%s~, . . . . . . .(lO) dt dt

where dyldt = rate of change of the controlled variable

with time (level, pressure, etc.),

dpldt = rate of change of pilot output pressure or current signal with time, and

S = characteristic constant of the pilot relating

output to input.

This characteristic is represented by the sensitivity usually expressed as a percentage. The proportional band or range, expressed as a percentage, relates the percentage of the full range of the measuring means that the controlled variable has to traverse to stroke the valve fully. This is illustrated in Fig. 13.13.

This figure represents a typical level-control application using a displacement-type controller-i.e., one that does not float on top of the liquid but depends on the varying buoyancy of the liquid as the level changes. Consequently, the control range is represented by the distance between Points A and B, because below A or above B a change in level has no effect on the buoyancy of the float.

Therefore, at 100% proportional control the level would have to move from A to B or vice versa, to stroke the valve fully from full open to closed. At 25 % , a level change of 0.25xdistance AB would fully stroke the valve, etc. The same principles apply with all other modes of control, including gas regulation.

Example Problem 2. A backpressure controller with a range of 0 to 100 psig is set at 50% proportional control,

OPEN 100

CLOSE0 0 0 25 50 75 IO0

PER CENT OF RANGE

Fig. 13.14-Characteristics of proportional control

using a normally open valve. If the valve is open at 100 psig, (a) at what pressure will it be fully closed? (b) what is the opening at 75 psig?

Solution. The control range is 100 psig, which at 50% proportional control means that a 0.50~ 100=50 psig change will fully stroke the valve; therefore it is fully closed at 50 psig. At 75 psig it would be half open.

The amount of fluid that a valve will pass at a given position depends on the size and characteristic of the inner-valve assembly. Ideally, the amount of fluid passed under given conditions, with a given valve, should be directly proportional to the valve opening. Often it is not, however; so that the type of inner valve has a definite effect on the proper proportional setting. Also ideally, the diameter of the inner valve should be such that the range of anticipated process loads may be handled when the valve is between 25 and 75 % open. In the final analysis, the choice of the proportional band to use in a given installation must be governed by experience and adjustment on the job.

This is illustrated in Fig. 13.14. The slope of the lines depends on the proportional band used, each line representing a valve of given size. Under these conditions, each valve may operate only along a single line, in accordance with Eq. 1.

If the instrument was set to control at 60% of the range, Valve A would be 50% open; Valve B, 25% open; and Valve C, 75% open.

Suppose, however, that an instrument set to operate along Line A is in gas-regulation service and the flow rate increases such that a valve opening of 75 % is needed to maintain the pressure. According to Line A, the only way the proportional controller can provide this opening is for the pressure to be at 70% of the range, or 10% above the set point. This 10% is called offset.

Consequently, with proportional control alone a change in process load brings a change in the valve position and some change in the liquid level. The proper proportional setting is therefore one which makes the lines in Fig. 13.14 as perpendicular as possible.

Zero per cent proportional control represents a special case. It is the upper limit of a range where the pressure is controlled by the valve alternately opening and closing. This is normally called snap action or on-off control. At a given snap setting, the valve will stay closed until the liquid level reaches a certain point between A and B and


---I r--* r-’

++$$f;LNT

I L--:

Fig. 13.15-On/off action of a controller.

Figs. 13.15 and 13 16 show on/off action and proportional action, respectively. In the latter, it is seen that the amount of valve movement is proportional to the deviation of the measured variable. By contrast, the amount of correction applied with reset action depends on both the magnitude and duration of the deviation away from the control point. The prime purpose of reset is therefore to prevent “offset” and keep the controlled variable at the control point even as the process load changes.

The differential equation of a controller with proportional plus reset action is

---\ ,-.

VALVE AC&N ‘\ ‘.

$s$+sR(y-r-,)_ . . . . .(11)

Fig. 13.16-Proportional action of a controller.

2suR7

--IT \, $ &$\$

PROPOST!G~&L ‘L VALVE ACTION ’

l\ PROPORTIONAL AND RESET VALVE ACTION

where dpldt, S, and dyldt have been previously defined, R=controller reset constant, and y-y, =duration of the controlled variable from the control point.

In this type of controller, the two corrections occur simultaneously as shown in Fig. 13.17 (this figure also shows whether the actions took place separately).

Fig. 13.18 shows typical curves in a process being controlled by this combination action. When a change in process load takes place the valve returns the measurement to the control point with a minimum of cycling. The original motion of the valve corresponding to the measurement change is caused by proportional action, but the change of the valve to its new position is due entirely to reset. In other words, the valve has moved to a new position to maintain the controlled condition. With only proportional control both the valve and controller would have changed.

Fig. 13.17-Valve action for reset and proportional controllers (at left); valve action for combined propor- tionallreset controller (at right).

PROPORTIONAL AND RESET

Derivative Response. Proportional control plus reset does not provide correction that is rapid enough for certain processes. Derivative response may therefore be added to anticipate a change in process load and transmit a corrective signal to minimize the lag. This action corrects on the basis of the rate of change of the deviation from the desired standard. This term stems from the fact that the first derivative of the change from the desired standard is incorporated into the control mechanism. The equation for an instrument with proportional plus reset plus derivative response is therefore

“1 TIME MINUTES where Fi, is an instrument-adjustment factor. The action of a controller with this response is shown in Fig. 13.19.

Fig. 13.18-Typical control curves when using a propor- tionallreset controller.

The general applications of the various combinations of control action may be summarized as follows:

Proportional. Where process time lag is small in com-

then remain full open until the liquid level drops to the control point. Where rapid process load changes are encountered, such as surging flow through separators, snap action is normally recommended.

parison with the apparatus capacity such as tank heating or large surge vessels, or where “offset” may be tolerated.

Proportional Plus Reset. Where it is necessary to use a narrow band to prevent “hunting” or overcontrol, and as the frequency and magnitude of the process load changes become greater.

Reset. The “offset” obtained with proportional control is sometimes too great to be tolerated and it is necessary to add other modes of control. One such addition is known as reset.

Proportional Plus Derivative Response. In processes involving long time lags and large capacities when small and frequent load changes occur. Does not provide compensation for process load changes.


Proportional Plus Reset Plus Derivative Response. Where long time lags and large capacities are combined with large and sudden load changes.

Because the addition of reset and derivative response increases the investment cost, they should not be used unless their action is needed for proper control.

Liquid-Level Control

The control of liquid level is an integral part of gas processing. It furthermore affords a means of illustrating the general problem of instrumentation and gas regulation. In view of the widespread applications and the variety of conditions encountered, it is not surprising that a number of mechanisms are used. These may be conveniently subdivided as: (1) mechanically operated valve actuated by a float; (2) pilot-operated valve actuated by a float; (3) diaphragm motor valve actuated by (a) displacement- type controller and (b) “floatless” level controller; and (4) external devices, including inverted bucket traps, float traps, etc.

Either snap action or proportional control is normally employed on most lease equipment. Most applications do not justify the cost of reset and/or derivative response since small changes in level with flow rate and time lag are usually not critical problems.

Separator controls, in fact, arc usually set on snap action because this enables the controller to handle a surging condition better. The degree of snap action will depend on the conditions encountered. However, it should be set so that not over 2.5 % of the vessel’s liquid capacity is filled, above the control point, before the valve opens. The inner-valve size should be determined on the basis of 110% of the maximum flow to be encountered. This means that during a surging condition the valve can handle the volume and prevent the separator from filling up.

Some uncertainty always exists, of course, when one tries to estimate future flow rates. On low-pressure separators, sizing is not a serious problem because of the low-pressure drop across the valve. On high-pressure gas-condensate separators, however, where the liquid flow rate is low, too large an inner valve on snap action can blow all the liquid out of the separator and allow gas to enter the stock tank before the valve can close. The author observed one such installation when 1 ,OOO-psig gas hit a lOO-bbl tank, knocked the cover off the vent valve, sprayed oil over the countryside, and lifted the tank off the ground.

The manufacturers of control valves furnish sizing curves, and these should be consulted. In using these, it should be remembered that critical-flow conditions occur when the upstream pressure is approximately twice the valves’ downstream pressure. Consequently, only twice the downstream absolute pressure should be used with these sizing curves if the actual pressure ratio exceeds 2.

When handling very small volumes of liquid, the use of such curves often indicates the use of so-called “metering trim.” Its use is not recommended in field applications, for the small quantities of solid materials often present may clog the valve. For this reason, nothing smaller than %,-in. needle trim is usually recommended in field service.

Float-operated mechanical oil valves are satisfactory at pressures up to 125 psig. Such valves are directly actuated by movement of the float on top of the liquid,

13-53

T I “.? E

----COhlEilNATlON OF THOSE BELOW

-----DERIVATIVE

--- RESET - PROPORTIONAL

Fig. 13.19-Typical control curves for proportional, reset and derivative-action controllers.

through an adjustable linkage. At higher pressure, the leverage supplied by the float is insufficient to provide satisfactory valve action because of the pressure acting on the valve seat. On separators with greater than 125psig working pressure, the use of a pilot-operated valve is advisable. Whenever applicable this type of control is very dependable and simple to adjust and repair.

Float-actuated pilot-operated valves are particularly applicable at pressure to 1,000 psig on vertical vessels. Such control has been “standard” on separators for years because the pilot is rugged and simple and will operate satisfactorily with the “wet” supply gas from the separator overhead.

This overhead gas contains entrained liquid so that a drip pot ahead of the pilot is advisable. With high- pressure separators, the expansion of gas from separator to instrument pressure may cause sufficient temperature drop to condense some water. This in turn presents hydrate or freezing problems, particularly in cold weather. A variety of solutions have been used, including tracing with warm separator gas, use of dehydrator pots, running such gas through a heater or treater if available, and insulation of the lines.

Displacement-type liquid-level controls use the buoyant effect of liquid on the float. The average float movement does not exceed x6 in. As the level varies on the float, the weight changes correspondingly and this change in torque is transmitted to the pilot, which, in turn, controls the valve movement. This type of control is applicable in all pressure ranges but is used primarily at high pressures and/or in horizontal vessels. Because of the small movement and small float diameter, the vessel opening may be decreased. This advantage becomes pronounced at high pressure or on small-diameter horizontal vessels.

The pilots on such controls are more sensitive than those discussed above, which allows one to control the level more closely. Most such pilots also allow the valve action to be changed from snap to varying degrees of proportional control by a simple adjustment.

They are necessarily more complex and expensive. Consequently, their use must be justified by process requirements. The instrument gas requirements are also more critical, because as little as one drop of water may plug the small orifice in some pilots.


Floatless level controls are a relatively new development. With these, the pilot is actuated by the varying liquid head in the vessel. As the liquid rises in the separator, it overcomes the pilot spring and forces the pilot assembly upward, closing the upper seat and opening the lower separator seat, which vents the diaphragm pressure to atmosphere. The separator fluid pressure then opens the valve. When the valve is throttling, the nonbleed three-way valve action of the pilot plug against its seat adjusts the motor-valve diaphragm pressure. This type of control has the obvious advantage of eliminating large vessel openings and offering application on very small vessels.

External devices such as float cages and traps find application particularly on vessels with small and/or infre- quent liquid loads. Some types of float cages offer no particular economic advantage at high pressure. They are most commonly used on low-pressure plant suction and instrument systems. Inverted bucket traps have been successfully used on small glycol absorbers for economic reasons but regular level controls certainly offer advantages.

At high pressures, a choke nipple downstream from the control valve is advisable, for it (1) provides a factor of safety if the valve “cuts out” or fails to close for any reason; (2) reduces the pressure differential across the valve, which improves the valve action and enables it to shut off tighter: and (3) prevents damage to low-pressure equipment downstream if the instrument supply gas fails on a normally open valve.

The control range with a float-operated controller is limited by the tlange diameter and the tloat-arm length. With a displacement element, control is possible only throughout the float length, for a change in level above or below the float does not affect the buoyancy. Therefore, the desired change in level should not exceed the tloat length.

The choice of a liquid-level control is somewhat arbitrary, but It 1s good policy to choose the simplest control that will meet process requirements. Average leasc- operating conditions impose severe service on the control, and the ability to make repairs with field personnel will generally reduce downtime.

Types of Regulators

The regulation of backpressure and pressure reduction in a system may be conveniently divided into three categories when considering the type of system needed. The low-pressure range is usually 0 to 125 psig; the intermediate pressure 125 to 500 psig; and the high pressure greater than 500 psig. The use of these ranges is primarily for convenience since some types of valves can operate satisfactorily in all of them.

All pressure regulators are similar in principle, the specification of type being dependent on the process requirements, pressure drop, variation in flow rate. limitations of the loading device, and the maximum pressure. In either of these services, pressure is regulated by the control of flow rate.

This flow is controlled through movement of the regulator inner valve. which is held either open or closed by some means of preloading. The amount of preloading and the size of the diaphragm used, if any. are such that the inner valve will move to the opposite extreme of

travel shortly after the diaphragm pressure passes the desired working pressure. The control pressure, therefore, is varied by changing the amount of preload, thus upsetting the equilibrium between it and the diaphragm pressure. With most valves, increasing preload increases pressure.

The preloading may be accomplished through the use of spring compression, dead weight, or fluid pressure. Valves a and b in Fig. 13.20 are examples of weight- loaded and spring-loaded valves, respectively. Valve c is a spring-loaded valve incorporating a pilot.

In backpressure service, the preloading normally closes the valve, while in pressure reduction the valve is normally open. With the former, the upstream pressure is introduced under the diaphragm and works against the loading device to position the valve properly. In pressure-reduction service, the downstream pressure is introduced under the diaphragm and tends to close the valve until the inner valve is properly positioned. Many valves may be changed from one service to the other by simply reversing the valve body and the stem, while still others have provision for reversing the diaphragm action.

Some applications require that a certain differential pressure be maintained across the valve regardless of how much the upstream or downstream pressures vary. The chief application of this is with positive- displacement pumps where the maximum pressure ratio is controlled by a valve located on a bypass.

In this service, the valve is normally closed with the downstream (higher) pressure under the diaphragm and the upstream (lower) pressure on top of the diaphragm. The desired differential is then established with the loading device.

Each of the types of preloading has its distinct applications and limitations. The weight-loaded regulator is very sensitive within its pressure range if the tlow rate

does not vary widely. This is true primarily because the inner valve may move through a greater change of position than with a spring-loaded valve. The greater sensitivity is particularly pronounced when the diaphragm is in direct communication with the valve body, which eliminates friction in the stuffing box.

Very few weight-loaded regulators are recommended above 50 psig because at greater pressures heavy weights are necessary. As a result, a small change in controlled pressure is only a small portion of the loading weight. and there is less subsequent movement of the inner valve, which in turn renders control more difficult.

A spring-loaded regulator is generally slightly less sensitive than a comparable weight-loaded regulator under constant-flow conditions. In low-pressure service. (particularly between 0 and 20 psig) it is an ideal type of loading, for the springs used tend to compress in direct proportion to the load, with constant load increase being required for each increment of compression. Consc- quently, spring loading is excellent in low-pressure systems and where the diaphragm is loaded with a pilot. Because each compression increment is a valve-travel increment, the valve travel is always in proportion to the relationship of the controlled pressure to the preloading. At high pressure, this characteristic does not exist because of the heavy springs required. Therefore. non- pilot-operated spring-loaded regulators are not normally


PILOT ASSEMBLY

MOTOR VALVE STEM ASSEMBLY

UPSTREAM PRESSURE

MOTOR VALVE DIAPHRAGM PRESSURE

Fig. 13.20-(a) Weight-loaded backpressure regulator. (b) Pressure-reducing valve. (c) Backpressure valve with self-contained pilot.

recommended above 125 psig because of this loss in sensitivity and their closer approach to the price of pilot- loaded regulators.

With either a direct-acting weight or spring-loaded regulator, a change in the uncontrolled pressure will cause a corresponding change in the controlled pressure if the pressure drop is low. This is true especially if the pressure drop is less than 25% of the inlet pressure. Therefore, when accurate control is necessary under these circumstances, pressure-loaded or pilot-operated valves are recommended.

Both the regulation of flow and tight shut-off should not normally be attempted with the same valve unless absolutely necessary. If this is required by operating conditions, a single-seated valve should be used. Double- ported valves are seldom operated under the actual conditions at which they were manufactured and tested.

Consequently, as temperature changes the distance between the two seats, only one of the two will seal off tight.

Single-ported valves will seal off tight but the unbal- ancing force on the inner valve increases with the square of diameter. Because the diaphragm/spring combination must overcome this force and still control travel, it is logical that the inner-valve size is limited by practical diaphragm size. As a general rule of thumb, direct-acting spring- and weight-loaded regulators up to 2-in. diameter may be used with single-port construction.

Pilot-loaded valves, either pressure-balanced or spring-loaded, offer particular advantage at higher pressure since the pilot presents a means of increasing the control-valve travel with a given change in controlled pressure. This enables one to use a more accurately sized valve. The resultant smaller valve gives better shutoff and closer control because the full range of travel may be utilized.

Low-Pressure Service. It is in this pressure range that weight- and spring-loaded regulators find particular application on the lease. One of the most common applica-

tions is in back-pressure service on low-pressure separators operating at less than 40 psig. All the valves shown in Fig. 13.20 are applicable. One modification of a diaphragm-type weight-loaded valve is used where the weight simply acts to counteract the unbalanced force acting under the inner valve. The choice of this valve is predicated primarily from cost considerations since it is the cheapest of the valves. No diaphragm is used.

The simple weight-loaded valve has the advantage of being cheap and simple. Its primary disadvantage stems not from its operation but from the circumstances surrounding its use. The weights become loose and shift, are lost, or are hindered in movement by outside obstructions. It is not too uncommon to see rocks and other miscellaneous objects used as substitutes for, or additions to, the proper weights. Consequently, these regulators serve as a proper but not necessarily completely satisfactory backpressure control.

Not too many spring-loaded regulators are used on separators because of their higher price, although they are generally satisfactory below 40 psig. The best regulator in this range is Valve c, Fig. 13.20. It gives fine control throughout the range with widely fluctuating flow rates and uncontrolled pressures. Where fine control is necessary, particularly above 40 psig, the extra investment is normally justifiable.

Because of the almost infinite number of different field conditions that arise, the above considerations can at best serve only as a general guide. Unfortunately, the actual choice of type must often stem largely from past experience.

High-Pressure Service. At pressures above 125 psig, it is usually difficult to justify anything other than a pilot- operated diaphragm motor valve for pressure control. Non-pilot-loaded valves are sometimes applicable but

usually only in those circumstances where pronounced load changes are not encountered. At these higher pressures, the cost differential also becomes less. which further encourages the use of pilot-operated controls.


VENT NOZZLE 1

Fig. 13.21-Proportional pilot for pneumatic service.

Most pilots now in use in the oil field are pneumatic in nature and use natural gas as the actuating fluid. Any natural gas that is free of fluids and at a pressure greater than 1.5 psig is suitable in this service.

Fig. 13.21 is a schematic view of a pressure pilot that will give both proportional and on/off control. As the controlled pressure varies, the bourdon tube will change shape and. in turn, raise or lower the curved bar.

The supply pressure is held constant between 1.5 and 20 psig by the pressure regulator. The vent nozzle is so sized that when wide open (curved bar away from it) it will pass more gas than the orifice. Consequently, the pressure on the valve diaphragm and the valve position depends on the opening of the vent, which in turn depends on the position of the curved bar, as fixed by the bourdon tube. If the vent is wide open, the pressure on the diaphragm is zero, while if it is fully closed, the diaphragm pressure equals the controlled supply pressure.

It is necessary that both the vent and orifice be very small to minimize the amount of gas vented. If the curved bar had a fixed pivot rather than a bellows, only on/off (snap) action would be possible because of these small openings. From a purely mechanical standpoint, any small movement of the curved bar would in effect make the vent wide open.

The bellows is used to impart a rotating motion to the curved bar around the end of the vent. When the bar begins to rise off the vent, the bellows contracts, which tends to keep part of the vent opening covered. As a result, more vertical movement of the bar is necessary to open (or close) the valve fully. All points in between then represent some degree of proportional control.

WELLHEAD

The vertical distance that the bourdon tube has to lift the bar to stroke the valve completely increases as the lever arm decreases. This then allows for adjustment of the proportional band.

If the vent is close to the left end of the curved bar, the movement of the bellows imparts little rotation to the bar and on-off action results.

Most pressure pilots have the bar marked to show various percentages of proportional control. The percentage shown indicates that the controlled variable must vary through that percentage of the instrument’s range to open or close the valve fully.

Example Problem 3. If a pressure pilot has a bourdon tube with a range of 0 to 200 psig and it is set on 50% proportional control, how much must the pressure vary to make the valve be fully stroked?

Solution. It must vary (0.50)(200-O)=lOO psig to stroke the valve fully.

From the discussion above it therefore follows that the closer the vent is to the bellows, the higher the percentage of proportional control obtained.

Most pressure pilots are slightly more complex than the schematic presentation in Fig. 13.2 1, to improve the mechanical action and the sensitivity. However, the basic mechanism and principle of operation shown is common to all pneumatic pilots.

In most applications, the valve is spring-loaded, but there are some applications where a large quantity of high-pressure gas must be controlled with a low-pressure drop. Such an application is encountered with the switching valves of a dry-desiccant dehydrator. A large inner-valve assembly is required, which increases the unbalanced forces on the diaphragm, which in turn necessitates a heavy spring. In such cases, the use of a pressure-loaded balanced diaphragm valve is indicated. With this type, greater valve travel may be obtained, with less change of controlled pressure than with any other type. With a constant loading pressure, on a larger balanced diaphragm, the available force to travel the valve with incremental change of control pressure is greater than with a spring-loaded valve because of spring characteristics. However, spring loading is simpler wherever applicable.

The greater the pressure drop across a valve, the greater the unbalanced force that exists. Where this pressure drop is greater than several hundred psi, it is good practice to install a choke nipple downstream from the valve. This choke nipple should preferably be about 10 to 25 % larger than the inner valve so that the latter is the controlling element. The nipple serves two purposes:

LOW-TEMPERATURE

Fig. 13.22-Typical pressure control for a high-pressure well.


Fig. 13.23-Small high-pressure spring-loaded pressure regulator.

(1) it reduces the pressure drop across the valve and increases sensitivity; and (2) it acts as a secondary control in the event the valve fails to close or cuts out.

The pilot shown in Fig. 13.21 may be modified to give what is known as high-low shut-off. A pilot so modified will cause the valve to close if the controlled pressure becomes higher (or lower) than the preset condition. This modification has particular application in high- pressure wells to (1) shut in the well in the event a line break is encountered; and (2) prevent overpressuring of a separator or other piece of equipment downstream from a choke or valve.

In some instances, it is desirable to use an automatic rather than a manual adjustable positive choke. A regular diaphragm-motor-valve topworks may then be combined with the choke body to give a control valve and choke combined. Although it costs more, there is some advantage to using the positive choke as a rough control with a separate valve at the wellhead for safety. Fig. 13.22 illustrates such a scheme where several wells are being flowed to a central low-temperature separation unit, the pressure drop being taken at the unit. The valve at the well protects against line break while the back-pressure valve at the separator holds the separation pressure constant regardless of how the sales line pressure varies. The two control valves fix the pressure drop across the choke, with the flow rate being determined by the choke setting.

Where very high formation pressures are encountered, one or more “storm chokes” in the well proper make a desirable installation.

Fig. 13.23 shows a spring-loaded valve for pressure- reduction service. This is a popular valve for pressure regulation in services where the rate of flow is not large. The pivot point on the arm is so located that it ratios the force on the inner valve down to where the spring can control properly.

Fig. 13.24 shows a double-port diaphragm motor valve that would be controlled by a separate pilot, such as that shown in Fig. 13.21. Fig. 13.25 shows several of the different types of inner valves that might be used in the double-port valve shown. The shape used depends on the service and the type of control desired. The same

Fig. 13.24-Typical spring-loaded motor valve

general shapes of inner valves are also available in single-port valves.

Control of Field Compressors

Most of the controls on a field booster compressor are designed to protect the equipment and facilitate operation with minimum attention; it is theoretically possible to operate the system with little more than manual control. The exact scheme used will of course vary with the application, but that shown in Fig. 13.26 is typical of small-lease operations where low-pressure gas is compressed in one stage.

No control and cooling system is shown for the engine since this will depend on the engine manufacturer. On most small portable units, the engine is self-contained

13-58

Fig. 13.25-Several valves.

types of inner valves used in double-oort

with its own radiator and pump. If two-, or more. stage compression is used, the controls shown for the inlet scrubber would be duplicated for each interstage scrubber. A provision would also have to be made for interstage gas cooling.

The backpressure controller shown on the vent is to prevent overpressuring of the separator in case the compressor plant is down. The other backpressure valve on the suction line is to prevent the plant from lowering the separator pressure by drawing off gas faster than it is produced. It should be set several pounds lower than the valve on the vent line.


The meter run is shown at the plant. although it could be conveniently located at the separator. A check valve (Location 12) is included to prevent the backflow that may result when multiple units feed the same compressor.

A butterfly valve (Location 14) is an optional but desirable item to control the compressor suction pressure. All compressors are rated using a given suction and discharge pressure, and this valve simply prevents the former from getting too low and overloading the unit. This valve is particularly desirable if the deliverability of gas to the plant is widely variant.

The inlet scrubber is primarily a protective device to keep liquid out of the cylinders and should not be used as a primary oil and gas separator. This vessel serves simply as a liquid sump and a convenient device on which to hang certain necessary controls. As a convenient rule of thumb, this vessel should have a cross-sectional area of at least 100 sq in./MMcf/D of gas processed. The minimum size is normally 12 in. A float-operated trap (Location 6) is very satisfactory for controlling the liquid level.

Switches (Locations 1, 2, 7, and 8). constitute the basic safety shutdown system. These switches are normally open, but if the condition at any one goes beyond set limits, it closes, grounding out the engine magneto, which in turn stops the engine. All are simple and inexpensive and afford positive protection. A horn and/or remote indicator may be incorporated that is actuated at

any time the compressor stops. Switches at Locations 1 and 8 are high- and low-

pressure shutdown, respectively. They serve to keep these pressures within the limits set up for the compressor. The switch at Location 2 is a high-temperature shutdown that is actuated by the hot water leaving the cylinder jacket. It shuts down the unit when any failure

occurs in the cooling system.

h4ANIF.m. .OLD FOR

HIGH PRESSURE I I

WATFR m j ti&iKE-Ui 9

-p-T--.(.----------h-2-J (q

TO LOW TENSION WATER

- GAS LINE TERMINAL OF --- WATER LINE ENGINE MAGNETO ---- ELECTRIC LINE

Fig. 13.26-Typical control system for field compression of natural gas (1) High- pressure shutdown switch. (2) Water-temperature safety switch. (3) Low- pressure liquid-relief valve. (4) Water-temperature bypass control. (5) Plug valve. (6) Trap-type level control. (7) High-level shutdown. (8) Low-pressure shutdown switch. (9) Globe valve. (10) Low-pressure backpressure controllers. (11) Relief valve. (12) Check valve. (13) Orifice meter. (14) Low- pressure pilot-controlled butterfly valve.


The switch at Location 8 is a high-level shutdown that operates when the liquid level becomes too high in the scrubber because of either failure of the valve at Loca- tion 6 or an excess quantity of liquid entering the plant.

The bypass valve at Location 5 is desirable because it provides a means of taking the load off the compressor, particularly during start-up. During normal start-up. this valve would be open and gradually closed once the compressor and engine reached running speed.

The bypass relief valve (Location I I) is an optional and, to a certain extent, a luxury item because it duplicates the protection offered by the switch at Loca- tion I, It provides relief in the event the outlet or bypass plug valves are inadvertently closed, without shutting down the compressor. If used, it would normally be set to open 5 to 10 psig before the high-pressure switch closes.

The water-temperature bypass control (Location 4) serves to control the jacket water temperature by govem- ing the amount of water to the cooler. In its simplest and cheapest form, it consists of an automobile-radiator-type thermostat in a three-way cast-iron valve body. It is substantially the same device used on most oil-field engines. A conventional temperature control would of course be satisfactory but its cost cannot be justified unless it is a large installation or is used where the complexity of the system does not lend itself to this type of control. In some cases where several services are handled by the same cooler, it is often more satisfactory to control the temperature by varying the speed or pitch of

the fan. Although no scale trap is shown, it is normally a good

policy to have one to prevent accumulation and possible system damage. Some provision for water make-up is necessary; so a small surge tank on top of the cooler should be provided for this purpose. Drains should be provided so that the system may be completely drained.

The pop-off valve (Location 3) is primarily to prevent overpressure of the system, particularly the radiator. A simple brass valve ‘15 in. in size is normally sufficient,

On large systems, a standpipe may be used ahead of the pump suction to handle surges and serve as a water reservoir.

Nomenclature A? = area of orifice plate opening

b = values for Reynolds number factor, Tables

13.lg and 13.lh

C = orifice coefficient for prover C’ = orifice constant

d; = internal diameter of pipe

d,, = orifice-opening diameter

F,, = thermal expansion factor

F,, = basic orifice factor

F,, = ratio of orifice opening diameter to internal pipe diameter

F,? = specific-gravity factor

F,, = instrument-adjustment factor

F,, = gauge location factor

F,,, = manometer factor F /J/l = pressure-base factor

F,,,. = supercompressibility factor

F,. = Reynolds-number factor

F,/> = temperature-base factor Fd. = flowing-temperature factor

is= force-mass conversion factor

h,,. = differential water head across orifice

K,, = efficiency factor, which includes approach

factor

P= Pfm = PM =

4e = Q= R=

s=

pressure flowing mean pressure

pressure in inches of mercury

gas-flow rate

heat gained or lost by system

controller reset constant

characteristic constant of the pilot relating

t=

T= U=

output to input

time

absolute temperature

internal energy, which includes all energy

such as heat, electrical, chemical, and

surface L’ II

v= w= x=

x, ,x2 =

Y= z=

Y,q =

velocity in conduit

volume

work done by system

0. datum defined in Fig. 13.1

potential energy above the datum plane

defined in Fig. 13.1

expansion factor

height above an arbitrary datum plane

specific gravity of gas (air= I .O)

13-59

Subscripts

f = flowing conditions

g = gas

0 = orifice

s,. = base or standard conditions

References I. Campbell. J.M.: “Gas Conditioning and Proce~\ing,” C;unphell

Petroleum Series, Norman. OK (1984).

2. Or@r Cowto~r Ttrhks. American Gab A\sn., Report No. 3,

revered (1969). Also. ANSI/API 2530.

Chapter 14 Lease-Operated Hydrocarbon Recovery Systems Robert N. Maddox, Oklahoma State u.*

Introduction In lease production of natural gas, the marketing specifications as prescribed by the gas sales contract must be considered when selecting the system for processing wellhead gas for liquid recovery. Natural gas at the wellhead can contain liquefied hydrocarbons, free water. water vapor, acid gases, and other undesirable components. To make wellhead gas merchantable, these components must be reduced to a composition that will satisfy the marketing specifications. The first part of this chapter is devoted to the removal of liquefiable hydrocarbons from natural gas. The latter part of the chapter describes techniques for removal of some of the other components.

The removal of liquefiable hydrocarbons, which are called “condensate,” is necessary for efficient transmis- sjon of natural gas in pipelines. If hydrocarbons condense in the pipeline, additional horsepower is required to overcome the increased pressure drop. Where the heating value of the natural gas is specified by the gas sales contract, there must be control of the condensate removal to satisfy this limitation. A final consideration for removal of condensate is that additional revenue is derived over that from sale of natural gas. In many instances this additional revenue will readily pay out the cost of process equipment required to produce gas of merchantable quality.

Low-Temperature Separation (LTS) Systems Theoretical Considerations Before various methods for removing condensate from natural gas are discussed, some of the physical phenomena involved in the formation of condensate are examined.

Retrograde Condensation. Retrograde condensation is a phenomenon that occurs at the high temperatures and

‘The author of the chapter on thls tqxc I” the 1962 edltlon was Edwm C Young

pressures frequently encountered in condensate hydrocarbon reservoirs. In the retrograde condensation region, as shown in Fig. 14.1, condensate forms at constant temperature with a reduction in pressure or at constant pressure with an increase in temperature, both of which are contrary to the normal expectations for condensation. If fluid in the reservoir existed at Point A, condensation would occur as the pressure in the reservoir declined from A to B. Less frequently encountered is the case shown by Line C-D. Here the fluid is in the single-phase region at Point C. As the temperature is increased at constant pressure from C to D, condensatioh occurs.

Retrograde condensation also can occur at lower temperatures and pressures than those encountered in the reservoir. This phenomenon ’ can occur at normal processing conditions for lease-operated equipment. Gas removed from a separator at typical high-pressure separator conditions (100°F and 1,000 psia) will undergo an increase in dewpoint temperature as the pressure is lowered. In many cases this “bulge” in the dewpoint line at lower pressures can amount to as much as 20°F or more, enough to cause troublesome condensation in lines thought to be at constant temperature as relatively small pressure drops occur.

Fig. 14.2 shows the calculated dewpoint curve for the off-gas from a separator operating at 120°F and 1,000 psia. The maximum dewpoint temperature is 136°F and occurs at about 500 psia. Also shown in Fig. 14.2 are constant mol% liquid lines. Over nominal line pressure drops very little liquid will form but that formed will be primarily the heaviest components in the gas stream.

Cooling. A second phenomenon to consider in condensate removal from natural gas is the cooling that can occur when the pressure on the gas is decreased. This temperature decrease can have one of two causes. When natural gas expands from a high pressure to a lower pressure without heat transfer or work being done (a constant-


i *

Smooth Curve

Temoerature

Fig. 14.1-Pressure-temperature diagram for typical natural gas and showing retrograde behavior.

enthalpy expansion), there is an accompanying temperature drop or refrigeration effect normally referred as to the Joule-Thomson effect. If, however, the expansion occurs through a turbine then work is removed from the gas during the expansion and cooling occurs also. Ad- vantage can be taken of the available pressure drop to lower the separation temperature of the hydrocarbon mixture and cause more liquid to form from the natural gas. Cool- ing from turbine expansion must be modeled along the lines of compression calculations and is not easily correlated. Cooling available from constant-enthalpy expansion can be estimated by charts such as Fig. 14.3. One must be cautious in using charts like Fig. 14.3 because they are composition-dependent and cooling depends on gas composition and amount of liquid formed as well as the initial and final pressures.

Hydrate Formation. A third phenomenon that must be considered is the possible formation of hydrates when

Temperature. “F

Fig. 14.2--Retrograde condensation of separator off-gas.

water is present in the natural gas stream. Hydrates are materials that have fixed chemical compositions but exist without chemical bonds and are called “clathrates.” They form a solid similar to snow at temperatures above 32°F (the freezing point of water) when the gas is under pressure. They appear to be hydrates of a mixture of the component gases and not a mixture of the hydrates of the individual gases. The hydrates form at a temperature that is characteristic of a given gas mixture rather than at the hydrate temperature for the individual components in the mixture. The hydrates normally include several water molecules for each hydrocarbon molecule so that the presence of liquid water is generally considered necessary for the formation of hydrates in sufficient quantity to cause plugging of a line, valve, etc. Turbulence accelerates the formation of hydrates and for this reason they frequently occur downstream from valves, regulators, chokes, orifice plates, sharp bends, etc. Fig. 14.4 can be used to estimate hydrate-forming conditions for different natural

180

170 160 I50

y 140

o I30 $120 6 II0

u IO0

5 90

5 80

2 70 % 60

t 50

40 30

20

IO 0 0 500 1000 15002CCO25003COO35004ooO45005CCO55006ooO

INITIAL PRESSURE, psig

Fig. 14.3-Temperature drop associated with a given pressure drop,

LEASE-OPERATED HYDROCARBON RECOVERY SYSTEMS 14-3

gases. Caution also must be used in Fig. 14.4 because, as shown by the different hydrate-forming lines for O.&gravity gases, there can be considerable difference in the hydrate temperature of gases of the same gravity.

If the composition of the gas is known, a composition- dependent calculation of the hydrate temperature, either by hand3 or by computer, I will give a much better estimate of the hydrate temperature than will Fig. 14.4.

A necessary condition for hydrate formation is the presence of liquid water. Prediction of the temperature where free water will occur will help identify the first point at which hydrates might form. The chart shown in Fig. 14.5 gives the water vapor content of sweet [no hydrogen sulfide (HlS) or CO21 natural gas as a function of temperature and pressure. As the temperature decreases at a given pressure the water content required for saturation also decreases. This will result in condensation of liquid water for a saturated gas stream as it is cooled. As an example, suppose a well is flowing 1 MMscf/D of natural gas at 1,000 psia saturated with water vapor but containing no liquid water at 110°F. The gas is cooled to 60°F because of ground and atmospheric cooling. At 1,000 psia and 1 lO”F, the gas contains 80 lbm water vapor/MMscf and at 60°F it contains only 18 IbmiMMscf. One day of gas production will result in the formation of 62 Ibm of free water because of the cooling. Referring to Fig. 14.4, if the gas flowing has a specific gravity greater than 0.6, hydrates are likely to form in the flow line at some point of turbulence.

Constant-Enthalpy Expansion Systems The constant-enthalpy expansion systems use the rcfrigcration effect that results from a pressure drop taken on a high-pressure wellstream. This expansion occurs across a choke and the resulting rcfrigcration effect ih dependent on the tetnperaturc on the upstream side of the choke. the pressure differential across the choke. and the amount of liquid formed. For obtaining the tnax- imum removal of liquefiable hydrocarbons from the gas stream for a given pressure differential and sales-gas pressure. the lowest possible temperature within reasonable limits should be attained in the separator. This in turn means the lowest possible temperature upstream of the choke. Two basic methods commonly used to accomplish condensate removal are low- temperature separation with or without hydrate in- hibiton. Each method is discussed in the following sections.

Low-Temperature Separation Without Hydrate Inhi- bitor. The basic unit for low-temperature separation without hydrate inhibitor includes essentially a choke, separator. and heat-exchange coils. Assuming that the inlet wellstream contains a minimum amount of free water and is of sufficient temperature to prevent formation of hydrates upstream of the choke, the operation is as follows. The wellstream enters the unit shown in Fig. 14.6 through the heat-exchange coil. where it is cooled through heat exchange with the liquid external to the coil. The wellstream then passes through an adjustable choke used to control the flow rate through the system and establish a means for introducing the necessary pressure drop. The turbulence and temperature drop created by the expansion across the choke cause the formation of hydrates and the

6000 1 I 1

4000-

2000-

5 1000-

2 - eoo-

K

i

E

30 40 60 00 70 80

TEMPERATURE, OF

Fig. 14.4-Approximate hydrate-forming conditions for natural gas mixtures.

condensation of the liquefiable hydrocarbons. The hydrates and condensate are separated from the gas by means of centrifugal force, normally generated by locating the choke tangential to the shell of the separator, and by gravity. The hydrates and condensate collect in the bottom of the separator where they absorb heat from the inlet coil, causing the hydrates to be melted. The liquid level is maintained by a level controller such that the coils are always submerged in the liquid.

Two possible operating problems might occur in this simple system. Either the wellstream could be near the hydrate temperature on entering the coil and further cooling would create hydrates upstream of the choke, or there is an insufficient amount of the liquid bottoms causing hydrates to build up inside the separator. In either case the system will malfunction. To use low-temperature separation successfully, the pressure of the gas upstream of the choke must be approximately twice the pressure in the low-temperature separator. Certainly, the higher the pressure upstream of the choke the lower the temperature that can be achieved in the low-temperature separator.

The common solution to this problem is to install an indirect heater upstream of the low-temperature separator. The indirect heater temperature would be maintained at a level to ensure wellstream gas temperatures above the hydrate temperature. Heat transfer is accomplished


4OOOH CORRECTION FGA SALINITY I--

‘>OLlD’Y IN mrdf %

s 200 Warning: Dashed lines are =k#‘: 2

e rncta-stable equilibrium. f E 5

Actual equilibrium is lower

0 water content. Angle is a

Water contents of nstursl gasss with ~orre~,,on~

for ~almoty and relal~v. denwty

’ W i i i i i.. . . . . . . . . . -KO 40 20 0 .‘o 40 W R” 100 I20 14Cl win I00 700 240 2.90

loo00

0ooo

6ooO

6

Temperature, F

Fig. 14.5-Water dewpoint of natural gas.


by flowing the wellstream through the coils in the indirect heater. The gas temperature is controlled by a thermostat located in the outlet end of the coil. A second thermostat can be located in the liquid section of the low-temperature separator to override the heater controls in the event the liquid temperature is too low.

For the simple system outlined here, the prime concern was to keep the temperature high enough upstream of the choke to ensure continuous operation. One of the necessary elements for hydrate formation, as pointed out previously, is the presence of free liquid water. In cases where free water and liquid hydrocarbons are present in the wellstream, and where these liquids are formed because of the chilling effect in the coil, a high-pressure separator can be placed between the coil and the choke. The liquids separated in this vessel are dumped at a relatively warm temperature into the bottom of the low-temperature separator and combined with the liquids in the separator. This reduces the amount of free water and liquid hydrocarbons passing through the choke and allows more net cooling to occur. Since the sensible heat transfer to these liquids can be used for cooling the gas and condensing additional hydrocarbon vapor, it also reduces the amount of hydrates formed by minimizing the amount of free water passing through the choke.

A further refinement to this system is possible for increasing the condensate removed from the wellstream. This includes the addition of a heat exchanger located between the inlet high-pressure liquid separator and the choke. The high-pressure wellstream flows through the tube side of this heat exchanger, and cold sales gas off the top of the low-temperature separator flows through the shell side. The amount of sales gas flowing on the shell side is controlled by a three-way bypass valve actuated by a temperature controller upstream of the choke. In this manner the upstream temperature at the choke can be maintained at a minimum value but still above hydrate

TEMPERATURE CONTROL %LVE +.....

Fig. 14.6-Typical low-temperature separator

temperature. From the analysis given, the lower this inlet temperature, the lower will be the temperature in the separator and the greater will be the amount of condensate removed.

The complete system is shown in Fig. 14.7. The wellstream flows through the coil in the low-temperature separator where it is slightly chilled, then to the inlet high-pressure liquid separator where free liquids are separated from the gas. The gas then flows through the gas- to-gas heat exchanger, through the choke, and into the low-temperature separator. The cold gas flows from the separator, through the gas-to-gas heat exchanger, and into the sales-gas line. The liquids from the bottom of the low- temperature separator are dumped to some form of stabilization before going into storage.

Fig. 14.7-Typical LTS system with inlet separator and intermediate heat exchanger.


GAS TO GAS HEAT EXCHANGER

GAS-GLYCOL CONDENSATE

PUMP TO STORAGE

Fig. 14.8--Schematic flow diagram of a glycol injection LTS system

In summary, the following considerations are necessary for operation of low-temperature separation systems without hydrate inhibitors. The controlled temperature of the high-pressure stream must be kept slightly above the hydrate temperature to prevent freezing upstream of the choke. The hydrate temperature can be determined by reference to the hydrate curve for natural gas (Fig. 14.4). The temperature of the low-temperature separator can be estimated from pressure-temperature drop curves for natural gas in Fig. 14.3. Allowances must be made for the liquid content, which will slightly reduce the temperature drop if the liquid hydrocarbons are not separated in a high-pressure separator upstream of the choke.

On wellstreams with low flowing temperatures a heater may have to be installed upstream of the low- temperature unit. In such cases the temperature is controlled with the heater and the gas-to-gas heat exchanger may not be required. If a well makes little free water and/or no waxy distillates or free condensate, the inlet high-pressure liquid separator can be eliminated.

This type of system is economically feasible from a liquid-recovery standpoint for gas streams with liquid content of a few barrels up to about 100 bbl/MMscf gas. The optimal recovery pressure will vary with different wellstreams but may be as low as 350 psig.

A rough rule of thumb is that 0.05 bbl additional liquid can be recovered per MMscf of gas for each degree lowering of the temperature. Below 20°F the increase in recovery becomes lower, and in some cases only slight additional recovery of net stock-tank liquid can be real- ized by lower temperatures.

Low-Temperature Separation With Hydrate Inhibitor. The formation of hydrates can be prevented by changing the character of the water in such a way that it will not hydrate with the natural gas. This is accomplished through the use of a substance known as a “hydrate inhibitor.” The most commonly used hydrate inhibitors are glycols and alcohols. Either will function satisfactorily. The system with a hydrate inhibitor is similar to the one in Fig. 14.7 except that the inhibitor is injected between the inlet high-pressure separator and the regenerative heat ex-

changer. The inhibitor mixes with free water formed on cooling and prevents hydrate formation. The advantages and disadvantages of alcohols and glycols as inhibitors are discussed by Campbell,’ as is the required amount of either material necessary for preventing hydrate formation. Further discussion here centers on the use of glycol as a hydrate inhibitor. A typical system using glycol is shown in Fig. 14.8.

The presence of the inlet high-pressure liquid separator is very important in this system. The produced water must be removed in this inlet separator to prevent contamination of the glycol and to keep the regeneration equipment from having to handle this extra vaporization load. The contamination occurs as (1) excessive dilution of the glycol, which reduces its inhibiting qualities; (2) as salt water, which is detrimental to the regeneration system; and (3) as solids, which can cause foaming and other problems in the separation and regeneration systems. If the wellstream contains waxy heavy ends these must be removed and kept out of the low-temperature separator where they could build up and cause the system to malfunction.

Glycol, which has been processed through the regeneration system to remove the water picked up in the system, is pumped from a storage tank to an injection point downstream of the inlet separator. The glycol rate is controlled by setting the speed of the pump. The rate is established by calculating the amount of water to be inhibited while going from the saturated condition downstream of the inlet separator to a saturated condition in the low- temperature separator. The glycol concentration, resulting from dilution with this water to be picked up, must be maintained sufficiently above the freezing point of the so; lution to prevent freeze-up in the low-temperature separator or the regenerative heat exchanger. There are published curves3 that can be used to calculate the necessary quantity of glycol.

The inhibited wellstream passes through the regenerative heat exchanger, to the choke, and into the low- temperature separator. The cold sales gas off the separator passes through the shell side of the heat exchanger. A three-way valve is used to control the amount of heat


Fig. 14.9-Schematic flow diagram of a glycol injection LTS system with low-temperature stabilization.

transferred from the wellstream to the sales gas to maintain a specified temperature in the low-temperature separator. This valve is actuated by a temperature controller located in the low-temperature separator. Since the inhibited wellstream can be cooled below hydrate temperature ahead of the choke, lower temperatures are possible in this system than with the uninhibited system. This allows more liquefiable hydrocarbons to be condensed out

of the wellstream and separated along with the rich glycol. The liquids from the low-temperature separator con-

tain condensate and rich glycol. The step of separating these liquids is difficult because of the low temperature and high viscosity. One method of handling this separation is to pass the liquids through a heater, as shown in Fig. 14.8, to bring the temperature to a level where separation can be achieved. However, this decreases the potential liquids recovery because of the warming up of the liquid hydrocarbons. Since condensates from glycol injection systems are commonly stabilized to give better stock tank recovery, in some cases the glycol hydrocarbon mixture passes through the stabilizer before separation. A flowsheet for this process is shown in Fig. 14.9.

If the gas being processed contains aromatic hydrocarbons, these will dissolve preferentially in the glycol. They may then cause degradation and decomposition problems in the glycol regeneration system.

The glycol from the injection system is rich in water and must be reconcentrated before it can be recirculated. This is accomplished in the glycol reboiler as shown in Figs. 14.8 and 14.9. The rich glycol is fed into a still column where it contacts the steam rising from the reboiler. This serves to preheat the glycol, strip out any dissolved gases, and condense any glycol that may be entrained in the steam vapors. With diethylene glycol the reboiler is operated at around 240 to 250°F to provide outlet glycol concentrations of about 80 to 8.5 wt% The glycol-condensate separator is usually designed to act as a surge to take up the slack in the event that glycol might be temporarily accumulated elsewhere in the system.

Glycols are expensive. so the system must be designed and operated to keep losses at a minimum. Losses occur because of the slight solubility of glycols in condensate

and the vaporization of glycols in the regeneration system. Proper equipment design and sizing for separation in the low-temperature separator and the glycol- condensate separator can keep mechanical losses to a minimum. In considering the type of glycol best suited for an injection system, both the solubility and the vaporization characteristics must be considered. Ethylene glycol has the least solubility in condensates but the highest vaporization losses. The vaporization can be reduced through a more elaborate regeneration system but at a higher initial equipment cost. Triethylene glycol has the least vaporization loss but is the most soluble. Diethylene glycol is usually considered a good compromise. In a typical condensate, 70 to 80 wt% diethylene glycol solution is about 0.4 wt% soluble. More concentrated solutions are more soluble but these are seldom encountered in glycol injection systems. Aromatic hydrocarbons tend to increase the solubility of the glycol in the condensate and higher losses can be expected when they are present. In installations where extreme care is exercised to eliminate glycol losses of a mechanical nature, such as leakage past the pump packing and unnecessary spillage, glycol losses have been found to be approximately 0.2 gal/MMscf of gas processed.

Hydrocarbons separated in the inlet high-pressure liquid separator can be recombined in the low-temperature separator provided they are not of a waxy base. The solution gas flashed in discharging from the high-pressure separator can be recombined with the sales gas and the liquids recombined with the condensate prior to stabilization. Waxy-based hydrocarbon liquids are best discharged directly to the stabilizer, where the temperatures are higher and they are more easily handled.

The low-temperature separation system with hydrate inhibitor eliminates the formation of hydrates and allows the gas to be cooled below the hydrate temperature before expansion. This results in an increase in the amount of condensate removed from the wellstream. The operating costs are higher than for the system using expansion without inhibitor, but the increased recovery will normally more than offset this. Glycol injection can be used more effectively on wellstreams where the pressure drop is low-

14-8

Fig. 14.10-Process flow for expansion process.

er than can the straight expansion system. Approximate- ly 1,000 psi of pressure drop should be available for operation of an LTS system while a glycol injection system can function well with pressure drops as low as 500 psi.

Turbine Expansion Systems The turbine expansion low-temperature liquid recovery system differs from the choke or valve expansion in that the turbine turns a shaft from which work is extracted. A typical turbo-expander process is shown in Fig. 14.10. The gas enters through an inlet separator with any liquid separated at this point being introduced to a low point in the stabilizer tower. The gas then goes through heat exchange with the cold gas leaving the stabilizer. Another separator is installed if sufficient liquid is formed in the gas-to-gas exchanger with the liquid being introduced at an intermediate point in the stabilizer. The cold gas then flows to the expander where the pressure is reduced and low temperature achieved. The gas and liquid mixture


leaves the expander and flows to the separator that normally is on top of the stabilizer column. Sales gas flows back through the exchanger and may be compressed in the direct-connected centrifugal compressor before being put into the sales gas line. Since extremely low temperatures are reached in a typical turbo-expander plant, dehydration normally is a first step though some plants do use alcohol injection. The gas frequently is expanded below sales gas pressure and then recompressed to make use of the work that must be extracted from the shaft of the turbine. The stabilizer is either a demethanizer or de- ethanizer with the mixed hydrocarbon product being sold.

A fairly recent development in gas processing, the turbo-expander process is one of great simplicity and ease

of operation. The favorable operating characteristics allow the plant to run unattended through long periods and its simplicity and relatively low investment cost make it an attractive option.

Mechanical Refrigeration Systems Because mechanical refrigeration systems frequently chill to as low as O”F, they involve essentially the same problems encountered in the low-temperature systems where the temperature is obtained by pressure expansion. The systems really are quite similar. The only essential difference is that the choke in the low-temperature separator systems is replaced by a chiller in the mechanical refrigeration systems.

Although glycol-injection systems are used extensively, there are many installations where inlet dehydration is used to lower the water dewpoint of the gas below the operating temperature in the chiller. There is merit to this system since the glycol does not come in contact with the condensate and glycol losses are, therefore, much smaller. Some of the other operating problems such as separation of glycol and condensate also are eliminated. The dehydrator, if it exists in the flowsheet, would ordinarily be placed between the inlet liquid separator and the chiller. The dehydrator could either be a liquid or granular desiccant

We.

GAS TO GAS COLD LIQUID TO EXCHANGER1 rGAS EXCHANGER

HYDRATE INHIBITOR TO REGENERATION

ACCUMULATOR

W-LEVEL CONTROLLER R HYDROCARBONS

TO STORAGE PC-PRESSURE CONTROLLER TC-TEMPERATURE CONTROLLER

Fig. 14.1 l-Hydrocarbon-liquid recovery system with mechan. lcal refrigeration and stabilization.


TABLE 14.1-COMPARISON OF COMMON REFRIGERANTS’

Evaporator Temperature (“F)

- 80 -70 - 60 -50 -40 - 30 -20 -10 0 10 20 30 40 50 60 Evaporator Pressure ~ ~ Ammonia 5 55 7 67 10 4 139 183 23.7 30.4 38.5 48.0 59 7 73 0 89 2 1075 Propylene 7 20 9.50 12 50 16 1 20 7 26 0 32 1 39.0 48.0 58.0 70.0 82.5 96.0 113.0 131 0 Propane 5 55 7 45 978 126 16 2 20 5 25 5 31 3 38 1 46.4 56.0 67 3 80 0 94 1 110.0

Freon 12 2.80 3.97 5 36 7.12 93 120 153 19.2 23.8 29.3 35.7 43.1 51 7 61.4 72.4

Condensed ltquld temperature 95OF Condenser pressure In psla ammonba 197; propylene 212; propane 177; Freon 12 123.

Pounds of Refrigerant per Minute per Ton of Refrlgeratlon

Ammonia 0 454 0 450 0 446 0.442 0 436 0.435 0.432 0.429 0.426 0 424 0 422 0.420 0.418 Propylene 2.07 2.01 1.96 1.92 1 87 1.83 1 79 1.76 1.72 1.69 1.66 1 63 1.60 1.57 1.54 Propane 2.18 2.11 2 04 1.98 1 93 1.88 1 83 1.78 1.74 1.71 1.67 1 63 1 59 1.56 1 53 Freon 12 5.18 5.03 4 89 4.77 4 65 4.53 4 42 4.32 4.22 4.13 4.05 3 95 3 a8 3.81 3 74

CFM of Refrigerant per Minute per Ton of Refrlgeratlon

Ammoma 20.4 14.9 11.1 8.40 6 45 5.00 3.96 3.13 2.52 2.04 1 69 1.38 1.14 Propylene 27.1 20.2 15 7 12.0 9 18 7.30 5 85 4.74 3.64 3.11 2 53 2 12 1 80 1.51 1.28 Propane 37.4 27.4 20.0 15.5 12.0 9.37 7 29 5.79 4 77 3.87 3 12 2.59 2 13 1.77 1.50 Freon 12 59.9 43.2 31.7 23.7 18.0 13.9 108 8.52 6.79 5.47 4.44 3.63 3.00 2.50 2.09

Brake Horsepower per Ton of Refrigeration

Ammonia 4 31 3.74 3.23 2.80 2 41 2.08 1.78 1.50 1.26 1.03 0.835 0.648 0 483 Propylene 5.00 4.47 3.96 3.51 3.10 2.69 2 35 2.06 1.74 1.46 1.20 1.00 0.830 0 647 0.485 Propane 4.98 4.39 3.87 3.43 3.03 2 67 2 32 2 03 1.75 1.49 1.24 1.01 0 800 0.622 0.458 Freon 12 5.70 4.98 4.33 3.79 3.31 2 86 2 47 2.14 1.83 1.55 1.30 1.06 0 848 0.668 0.490

Condensed liquid temperature 125OF Condenser pressure In psla ammoma 303: propylene 314: propane 260. Freon 12 184

Ammonia 0.492 Propylene 2.67 2 58 2.50 Propane 2.86 2.74 2.63 Freon 12 6 42 6.19 5 98

Ammonia 22.0 Propylene 35.2 26.0 20.0 Propane 50.0 35.6 25.0 Freon 12 74.0 53.0 30.8

Pounds of Refrlgeranl per Minute per Ton of Refrigeration

0487 0.483 0 478 0474 0 469 0.466 0.463 0.460 2 42 2.35 2 28 2 22 2 16 2 11 2 06 2.01 2.53 2.44 2 36 2.29 2.22 2.16 2 10 2.04 5.80 5.61 5 45 5 28 5.14 5.00 4.87 4.75

CFM of Refrigerant per Minute per Ton of Refrlgeratlon

16.1 12.0 9.09 6.97 5.40 4.26 3.38 2.72 15.4 11.5 9.40 7.32 5.85 4.72 3.81 3.08 19.7 15.4 11 a 9.16 7 20 5 94 4.79 3.79 28.8 21 7 166 12.9 10 1 8 05 6 45 5 21

0.457 0.454 0.452 0.450 1.97 1.93 1.89 1.86 1.99 1 94 1.89 1.84 4.64 4.53 4.43 4.33

2.19 1.82 1.49 1.23 2.60 2.18 1.84 1.56 3.14 2.63 2.19 1.80 4.25 3.50 2.90 2.42

Brake Horsepower per Ton of Refngeration

Ammonia 5.68 4.96 4.38 3.81 3.33 2 92 2 54 2 19 1.90 1.63 1 38 1.16 0.952 Propylene 7.49 6.72 5.96 5.32 4.71 4.14 3.66 3.23 2 79 2.41 2.03 1.78 1.55 1.31 1.10 Propane 7.47 6 60 5.85 5.18 4.60 4 06 3.59 3.18 2.81 2 43 2.07 1.77 1 50 1.25 1.03 Freon 12 8.09 7.11 6.25 5.46 4 78 4 18 3.67 3 20 2 78 2 41 2 07 1.77 1 49 1 24 1.02

I Gases consadered satwaled at compressor ~niet Heat of hqutd assumed condenser pressure and mnpera,ure shown 2 Horsepowers are average “BI”BS based on ce”lrlf”gal COmpreSSOr effiCle”CleS wllho”l economlrlng 3 Properltes 01 ammoma tram US Bureau 01 Standards propylene and propane kom Ell~ol Co Bull 11961) P 1 I and Freon 12 from “Thermodynamtc Properws 01 Freon 12

D”PO”l co (1955) 4 All cases shw”” can be handled by one compressor body except ammonia al O’F and colder ,95’F condenscng) and 2O’F and colder (125°F condensing, 5 One ton Of refrlgeralloo = (2 Cm au-tlr

In a refrigeration process, low temperature is achieved by cooling the gas using a refrigerant at low pressure. The heat in the incoming stream vaporizes the refrigerant at low pressure. The refrigerant must then have its pressure increased so that it can be liquefied. The pressure may be raised in two ways-compression or by absorbing the refrigerant in a liquid, which is then pumped to high pressure with the refrigerant being stripped and subsequently condensed. Compression refrigeration systems are more common, but either should work satisfactorily .

Compression Refrigeration Systems A typical compression refrigeration system is shown in

Fig. 14.11. The inlet gas stream flowing through the chiller causes the refrigerant to boil. The cold refrigerant vapors flow out of the chiller and to the compressor. Only one chiller is shown in Fig. 14.11. However, frequently in larger systems two or even three chillers may be used with each operating at a different pressure. The chiller pressures are fixed at interstage compression pressures for the compressor.

After compression, the refrigerant vapors are liquefied by cooling with either water or air. The liquid refrigerant is stored in the receiver until required in the chiller. A number of different materials may be used as refrigerants. Tables 14.1 and 14.23compare properties of a number of commonly used refrigerants.


Refrigerant Number (AR Des~onat~onl

Molecular weight Gas constant, R (ft-lbfllbm-“R) Bohng pomt at I atm.OF

Freezmg punt a1 1 aIrx°F CrItIcal femperature.°F CrltwX pressure. psja

Specific heal of Ilquld. 86OF Speclflc heat of vapor. C,,

6OoF at 1 atm

13738 11 25 74 7

-168 368 0 635 0

0 220

18739 170 93 8 25 9 04 1176 38 4

-31 137 417 4 294 3 495 0 474 0

0 218 0 238

0 156

Spectfrc heat of vapor, C,. 60°F al 1 atm 0 145

Ratlo of C,/C, =K (86OF at 1 atm) 111

12093 86 48 12 70 17 87

-21 62 -41 4

252 - 256 233 6 204 8 597 0 7160

0 235 0 335

0 146 0 149

0 130 0 127

1 14 1 18 1 12 1 09

0 171

0 151

113

Ratlo of speclfrc heats. llquld (105”F)lvapor C, (40°F). saturation pressure

2 04 1 55 2 14 I 47 1 59 1 77

Liqud head (1 psi at 105°F). 11 1 61 1 64 2 04 1 51 1 65 2 10

Saturation pressure at. PSI -50°F

O°F 40°F

105OF

0 52 7 12 11 74 1 35 2 55 23 85 36 79 0 64 5 96 7 03 51 67 83 72 2 66 15 22 25 7 141 25 227 65 1158 50 29

Net refrigeration effect (40 lo 105OF: no subcoolmg). Btu-lbm 67 56 49 13 66 44 54 54 43 46 59 62

Cycle etflcmcy (40 to 105OF) % Carnot cycle 90 5 83 2

Solubihty of water m refrigerant negllglble neglrgrble Miscibihty wth 011 miscible miscible TOXIC concentration, ~01% above 10% above 20%

TABLE 14.2-PROPERTIES OF SIX REFRIGERANTS

lr~chioromono DlchlorOdi Monocholorodl fluoromethane fluoromethane fluoramethane

1, 12 22

CCI ,F CCI F CHCIF..

81 8

negllglble hmlted

The refrigeration load on the chiller for a given gas flow rate is a function of the wellstream analysis, pressure, and the inlet temperature to the chiller. The two factors involved in this cooling load are the sensible heat required to reduce the wellstream from flowing temperature to chiller temperature and the latent heat required to condense the liquefiable hydrocarbons. Obviously, the richer the wellstream the greater the refrigeration load required. The chiller and refrigeration system must be designed to meet the requirements of each individual installation. These requirements are based on the wellstream analysis, pressure, and temperature.

For economical operation the system must be designed to use as much of the cooling as possible. This is done by taking advantage of heat exchange between cold and hot streams wherever possible.

Selective Adsorption Systems On lean wellstreams where economic justification and payout of the equipment necessary for hydrocarbon removal by refrigeration are not possible, a selective adsorption system should be considered. This type of system frequently will be competitive in hydrocarbon recovery with mechanical refrigeration for lean gas streams.

Txhiorotrl- Dichloroletram iluoroelhane lluoroethane

113 114

Ccl, F-CCIF C.,CI .F;

a7 5

negllgtble miscible

84 9

mlsclble above 20%

Areotrope of Drchlorodl-

fluoromethane and

Dlfluoroethane 500

73 800 CCI,F, 26 2%

CH,CHF,

99 29 1557 - 28 0

254 221 1 631 0

27 96 60 94 167 85

82 0

negllglble mlscrble

above 20%

The selective adsorption system consists basically of two or more adsorption towers filled with a solid material capable of adsorbing hydrocarbons. The towers are controlled on a time cycle to be adsorbing, regenerating, and/or cooling. Alternatively, the bed is used to adsorb the desired quantity of hydrocarbon from the gas stream. When the bed becomes saturated with hydrocarbons, the gas stream is switched to another bed. The saturated bed then is regenerated by passing hot gas through the bed. The hot gas vaporizes and drives off the adsorbed hydrocarbons. During the regeneration cycle, the bed and some of the parts of the adsorber become heated and must

be cooled before the bed can again be used on an adsorbing cycle.

The approach used in hydrocarbon adsorption is the same as is widely used in dry-desiccant dehydrators. The basic difference is in the time allowed for the adsorption cycle. Dehydrators normally operate on a 6-hour or longer cycle. Hydrocarbon recovery units operate on a much shorter time cycle and for this reason are frequently called “short-” or “quick-cycle” units.

When a multicomponent mixture like natural gas is adsorbed, the lighter components saturate the bed first. They then are displaced gradually by the heavier hydrocarbons and, eventually, the hydrocarbons would be displaced by


TABLE 14.2-PROPERTIES OF SIX REFRIGERANTS (continued)

Refrrgerant Number (AR1 Desrgnatron)

Warning Properhes Explosrve range, vol% Safety group. u L Safety Group. ASA 89 1 Toxic decomposrtron products

vlscoslty. cp saturated lrqurd 95OF

105OF vapor at 1 atm 30°F

40°F 50°F

Thermal conductwrty. k saturated ltquld 95°F

105-F vapor at 1 atm 30°F

40°F 50°F

Lrqurd circulated. Ibmlm~nlton (40 to 105°F)

Theoretrcal drsplacement cu ttlmmlton (40 lo 105°F)

Theoretrcal hpiton (40 to 105°F)

Performance coetfrcrent 4 ‘Il!(hp/ton) (40 to 105OF)

Cost comoared wth R 11

Trrchloromono-

fluoromethane 11

CCI ,F

Ethereal odorless when mrxed wrth air

none none

Yes

0 3893 0 2463 0 2253 0 5845 0 3420 0 3723 0 2395 0 2207 0 5472 0 3272 0 0101 00118 0 012G 0 0097 0 0108 0 0103 00119 0 0122 0 0098 0 0109 0 0105 00121 0 0124 0 0100 00111

0 0596

0 0581 0 0045 0 0046 0 0046

2 96

16 1

0 676 0 736

6 95

1 00

Dichlorodr- Monocholorodr- Trrchlorotri- fluoromethane fluoromethane fluoroethane

12 22 113

CCI;,F CHCIF, CCl,,FmCCIF,

Areofrope 01 Dlchlorodr-

fluoromethane Drchlorotelra- and fluoroelhane Dilluoroelhane

114 500

73800 CC! F C-Cl F, 26 2Jf

Ct. ,CHF

same as R 11 same as R 11 same as R 11 same as R 11 same as R 11

none none none none none none none none

6 5A 4 to 5 6 1 1 1 1

yes Yes Yes yes

0 0481 0 0573 0 0512 0 0435 0 0469 0 0553 0 0500 0 0421 0 0047 0 0060 0 0037 0 0056 0 0049 0 0061 0 0039 0 0057 0 0051 0 0063 0 0040 0 0059

4 07

3 14

3 02

1 98

0 75

6 29

2 77

3 66 4 62

39 5

0 70

6 74

2 15

9 16

0 722

6 39

1 57

6 52

2 97

water present in the gas. The secret to operating a hydrocarbon recovery unit is determining the point where the largest percentage of desired liquefiable hydrocarbons has been adsorbed. The cycle time normally has to be a compromise since it is done on a time basis.

Fig. 14.12 shows a typical installation for a selective adsorption system. The wellstream flows into an inlet- liquid separator to remove any free liquids and thus cut down on possible adsorbent contamination. The gas then flows through a pressure-reducing valve, used to control the flow of gas through the regeneration and cooling cycles, and into a manifold for switching to the tower on stream by the time-cycle controller. As the gas passes through the bed, the heavier hydrocarbons and water vapor are adsorbed. Stripped dry gas leaves the tower and flows through the gas-to-gas heat exchanger into the sales- gas line.

The tower being regenerated has been saturated through previous contact with the main gas stream. Gas for regeneration is taken off the main gas line upstream of the pressure-reducing valve, and the flow rate through the regeneration system is controlled by the pressure drop across this valve. The regeneration gas flows through the heater where it is heated and then to the tower being

none none

5A 1

yes

0 2150 0 2100

3 35

2 69

0 747

6 31

2 00

regenerated. The heavier hydrocarbons and water vapor that were adsorbed are driven off by the hot gases. The gas and vaporized fluids pass through an air-cooled heat exchanger where the temperature is lowered, condensing some of the liquid. Additional cooling is obtained in the main gas-to-gas heat exchanger with further condensation. The regeneration gas and condensates are passed through the regeneration-system separator where the condensates are removed from the gas. The gas off the separator flows back into the main gas stream downstream of the pressure-reducing valve. The condensates are separated, the water is dumped, and the hydrocarbon liquid typically flows to a stabilizer.

The system just described is commonly called an “open regeneration system.” This means that the regeneration gas flows continuously out and into the main gas stream on either side of the pressure-reducing valve. This requires that a pressure drop be taken in the main gas stream. In applications where presure drops cannot be tolerated, a closed regeneration system can be used. Such a system can be seen in Fig. 14.13.

A blower is used to recirculate the regeneration gas through the regeneration heating and cooling cycles. The blower picks up gas downstream from the regeneration-


INLET

GAS

I

GAS-GAS HEAT EXCHANGER SALES

GAS OUT _

-MAIN GAS FLOW LINES -REACTIVATIMU GAS FLOW LINES

6% OPEN VALVE 84 CLOSE0 VALVE

Fig. 14.12-Typical two-tower open-cycle selective hydrocarbon- adsorption system.

liquid separator and boosts the pressure sufficiently to produce the required circulation rate of the regeneration gas. Although the system shown in Fig. 14.13 is for a three-tower system, the blower will work equally well on the two-tower system shown in Fig. 14.12. The blower discharge would connect into the meter run going to the regeneration gas heater. The pressure-reducing valve and extra piping shown in Fig. 14.13 are included to ensure continuous operation while the blower is shut in for maintenance. Under these circumstances the unit operates as an open-cycle system. To compensate for varying pressures in the reactivation-gas system caused by temperature changes, equalizer lines are connected into the system with check valves. In the event that gas is required in the system, dry gas is taken from the sales-gas end of the system. If the reverse is the case, then gas within the regeneration system is discharged ahead of the adsorber so that

wet gas can be processed before entering the sales-gas line.

One advantage of the closed system is that the regeneration gas never becomes a factor in the drying cycle. Es- sentially the same gas is used over and over again. This results in slightly higher recoveries with an additional operating expense that requires an economic study for justification. The inclusion of the blower in the system adds considerably to the maintenance expense and operating problems encountered over the open system. The main advantage of the closed system is that the pressure drop on the main gas stream can be held to a minimum.

The size of the adsorption beds is a function of the amount of liquid to be removed. This quantity of liquid may be obtained from a small volume of rich gas or a large volume of lean gas. The diameter of the tower is established from the allowable velocity of the gas through

-MAIN GAS FLOW LINES

$2

- ~E$;IVATION GAS FLOW

OPEN VALVE fz CLOSED VALVE

Fig. 14.13-Typical three-tower closed-cycle selective hydrocarbon-adsorption system.

LEASE-OPERATED HYDROCARBON RECOVERY SYSTEMS

the bed. At low pressures this diameter can become quite large if a given capacity must be satisfied. The depth of the bed is determined by the amount of adsorbent required, the allowable pressure drop, or the minimum bed depth. The length of the time cycle affects the size of the bed for a given amount of liquid removed. The amount of adsorbent decreases with decreasing time, but the heating and cooling requirements increase as the cycle is short- ened. The heat required is a function only of the pounds of liquids to be removed and the materials to be heated such as the adsorbent and portions of the towers. The heating and cooling rates are thus inversely proportional to the time of the cycle.

For a given wellstream, the design of the equipment and the operation and control are most important in effecting liquid removal. The time cycle is important and is good only for a range of variation in the wellstream analysis. If the wellstream becomes richer, the cycle may be too long and all liquids are not removed. If the wellstream becomes leaner, the cycle is too short and the adsorbent is not used to the maximum capacity.

The quality of process cooling can also affect the process efficiency. If sufficient cooling is not accomplished, condensation of the liquefiable hydrocarbons will not be complete. In the open cycle this means that the regeneration gas will carry these vapors back into the bed to be adsorbed again. This could result in a considerable amount

of recycling of hydrocarbon vapors. The nonhydrocarbons present in the natural gas will

have some effect on the process. This is especially true of the water vapors that eventually replace the hydrocarbons adsorbed in the bed and allow them to be expelled. The short-cycle unit is an excellent dehydrator as well as a hydrocarbon-removal unit. Presence of Hz S has little effect on the process unless oxygen is present. In this event, the oxygen causes free sulfur to form which will tend to foul the desiccant. CO;!, nitrogen, and other gases occasionally found in natural gas do not affect the process significantly.

When selecting a short-cycle unit, the following can be used as a guide. When the butanes and heavier hydrocarbons to be adsorbed amount to 10 bbl/MMscf of gas or less, then the application of this type of system is normally feasible. Between 10 and 1.5 bbl/MMscf the choice must be made between the quick-cycle and the refrigeration system. Only through an economic analysis of both these systems can the proper choice be made. On wellstreams in excess of 15 bbl/MMscf the refrigeration system is more feasible. The normal application for the quick-cycle unit is downstream of low-temperature separation systems in which the reservoir pressure has decreased to the extent that the refrigeration from expansion cannot effectively remove the liquefiable hydrocarbons.

Hydrocarbon Stabilization The previous discussion covers the processes by which a maximum removal of liquefiable hydrocarbons from the gaseous phase of the well stream is accomplished to satisfy gas pipeline specifications and derive additional revenue from the liquid hydrocarbons. Unless the liquid hydrocarbons are handled properly after separation from the main gas stream, the maximum revenue will not be derived. The maximum is obtained by retaining the maximum volume of separated hydrocarbon liquids in atmospheric

14-13

0 20 80 100 120 140 I60 iso 200 TEt&%AT”k?6E” DEGREES FAHRENHEIT

Fig. 14.14-Vapor pressure-temperature curves for motor and natural gasolines.

storage and salvaging or recovering the maximum amount of light vapors that flash from the separated liquids.

If a liquid is to be stored at atmospheric pressure without vaporization losses, it must have a vapor pressure no greater than the existing atmospheric pressure at the maximum temperature it will reach in the storage tank. The vapor pressure exerted by the liquid is called its “true vapor pressure.” Measuring the true vapor pressure of a liquid is difficult. For this reason a standard vapor pressure, called the “Reid vapor pressure” (RVP) is much more frequently determined for the liquid. The RVP is determined by using a standard ASTM technique in a 100°F controlled-temperature bath. By sampling the liquid hydrocarbon product, the actual RVP of the liquid being produced can be determined.

Fig. 14.14 shows a correlation for approximating the true vapor pressure of a liquid for which the RVP has been determined. Because vapor pressure is composition- dependent, Fig. 14.14 should be considered an approximation only. If the maximum storage temperature were expected to be 6O”F, Fig. 14.14 shows that approximately a 29-psi RVP product could be stored. If the maximum storage temperature were lOOoF, the maximum RVP for the product would be approximately 14 psi.

When the hydrocarbon liquids are dumped from the high-pressure separator, the liquids are at their boiling point for the pressures involved. Each reduction in pressure causes some vapors to be boiled off. If the liquids were dumped directly to a storage tank, violent boiling would take place, resulting in a loss of not only the lighter vapors but also some of the heavier ends. By taking


REFLUX CONOENSER\ t

rl IGC

I II STABILIZER

’ ii 1 i 1 CCiTROLLER r-1 I I

JJ Ll6l.k f--1 SEFAF;(ATOR I I

i/,r--T -I II I SALT’EATH! !

I

4 HEATER_$_$

PRODUCT TQ

STOCK TANK

STANDARD LOW- TEMPERATURE

i STABILIZATION UNIT WITH STANDARD LOW-TEMPERATURE

SEPARATION UNIT , SEPARATION

Fig. 14.15-Stabilization unit with standard LTS system.

the pressure reduction in stages these losses are reduced. Increasing the number of stages of separation results in greater stock-tank recovery. Regardless of the number of stages used, some recoverable hydrocarbons that are normally liquid are lost with the solution vapors from each stage of separation. Considerable volumes of low-pressure gas are released from each stage and are costly to salvage. Recovery of a maximum volume of hydrocarbon liquid stable under stock-tank conditions with a minimum volume of solution vapors removed at relatively high pressure can be accomplished by fractionation of the first-stage separator liquid. This process is commonly called “stabilization process, ” and various systems are outlined.

Fig. 14.15 shows a flow diagram of a stabilizer used with a heated-liquid-type low-temperature separation sys- tern. In this system the condensate feed stream to the stabilizer is preheated. Condensate comes from the low- temperature separator at about 80 to 90°F and is preheated in a heat exchanger with the bottom product to approximately the same temperature as the stabilizer at the point of entry. A temperature controller regulates a bypass valve in the feed line so that a constant temperature can be maintained in the feed to the stabilizer. The hydrocarbon mixture is fractionated and the lighter components pass overhead in the vapor state. The stable liquid is withdrawn from the bottom of the stabilizer. A portion of the overhead vapors is condensed either by cooling water or by the cold gas stream from the low-temperature separator in a heat exchanger. The liquid is separated and returned to the stabilizer as reflux.

Gas from the reflux accumulator is vented for fuel and instrument gas. If the excess is great enough, it may be recompressed and added to the sales gas, but it may have to be dehydrated after compression.

Heat is supplied to the stabilizer by circulating the liquid that has passed through the fractionating section through an indirectly fired salt-bath heater. This provides stripping vapors for the liquid as it flows down the tower. The remaining liquid is automatically discharged by a liquid-

level controller in the bottom of the stabilizer. This liquid is partially cooled in the feed preheater and further cooled with air coolers. Final cooling to atmospheric temperature is done with the cold gas from the low-temperature separator. After cooling, the bottom product is stored in a stock tank.

A number of variations of this system are possible, depending upon the utilities available at the particular location. Where these units are used in conjunction with normal recovery methods, cooling water or an air cooler may be used in place of the cold gas.

Fig. 14.16 shows a flow diagram of a stabilizer used with a glycol-injection cold-liquid-type low-temperature separation system. The system uses cold feed from a glycol-injection system for the feed stream to the stabilizer. Condensate at 0 to 25°F is fed into the stabilizer at the top. This cold feed requires no other equipment for refluxing the stabilizer

Condensate is fractionated by the same process used in the system described by Fig. 14.15. The light components pass overhead as vapor, and the stable product is withdrawn from the bottom of the tower. The bottom product is cooled in a heat exchanger either by the cold gas from the low-temperature separator or by cooling water. Overhead gas may be used for fuel and supply gas. If the volume warrants, it may be recompressed for sales gas. Heat is supplied with an indirectly fired salt-bath heater.

To determine the pressure and temperature at which the maximum recovery will be achieved, Fig. 14.17 is included. The curves are drawn for 14.5.psia stock-tank pressure. The example shown is for a low-pressure stabilizer using a steam generator for a bottoms heater. For temperatures and pressures above this range, the high- pressure stabilizer and salt-bath heater combination are used.

Where the feed to the stabilizer is at low temperature, as in glycol-injection systems, good fractionation can be obtained by feeding the cold liquid on the top tray, causing


LOW-TEMPERATURE SEPARATOR PRESSURE

I I I STABILIZER

~Gd~~:z7L Li”“lV LLlLL

CONTROLLEF; ___ __--_----- -----

pmm ?

REBOILFR I ’ I

GLYCOL INJECTION LOW-

STABILIZATION UNIT WITH GLYCOL INJECTION AND LOW-TEMPERATURE SEPARATION SYSTEMS

Fig. 14.16-Stabilization unit with glycol-injection LTS system

the column to reflux itself. This eliminates the relatively expensive reflux system required on stabilizers where the feed is at higher temperatures. The elimination of reflux condenser, reflux control, and reflux pump greatly simplifies the operation of a stabilizer on an oilfield lease where operational supervision is limited. In most cases sufficient refrigeration is available in the system to chill the liquid from the heated-liquid-type low-temperature separation system. The simplified stabilizer can then be used with the heated-liquid system as well as with the glycol-injection low-temperature system.

By this method, the separator liquid is separated into its heavier components, which will remain liquid at atmospheric pressure and temperatures. and its lighter components, which will become a gas at atmospheric conditions. A minimum of liquid components is lost with the lighter components in this method of separation.

For selective adsorption units the liquefied hydrocarbons are recovered on a batch basis. This means that the liquids build up fast over a short period of time and no more are produced until the cycle switches. This means that an accumulator must be used between the separator and the stabilizer so that a more continuous flow of liquids can be provided to effect eftictent stabilizer operation.

In stage separation all the increased volume of normally liquid hydrocarbons is not reflected in the stock tanks because a larger amount of it is carried out as a vapor during the separation of the lighter components. By the use

of stabilization, almost all the increased volume of normally liquid components in the low-temperature separator is retained in the stock tanks under stable conditions. In addition to producing a larger stock-tank volume of liquid from a given volume of low-temperature separator liquid, the stabilizer produces all the lighter normally gaseous components overhead at one point in the system at relatively high pressure. Gas at this pressure can be compressed back to gas-transmission line pressure at a much reduced compression cost as compared with the low-

pressure vapors from the same system using stage separation. The vapors from the stabilizer are also lean in heavy hydrocarbon components and therefore more desirable in a gas-transmission line than the hydrocarbon- liquid-laden vapors from stage separation.

When the overhead vapors from a stabilizer are compressed into the sales-gas line and the stabilized liquid is stored in a vapor-tight tank, every pound of the produced well fluid, with the exception of that used for fuel gas in lease heaters (if any) and the stabilizer reboiler, can be marketed.

600

STABILIZER PRESSURE,PSIG

Fig. 14.17-Stabilizer operating conditions. Initial settings for determining the point of maximum recovery.


fraction of that component in the vapor, 4’i, to the mole fraction in the liquid, Xi, that would be in equilibrium with that vapor at the temperature and pressure as given by

Vi/V &L-

Xi l,lL . . . . . . . . . . . . . . . . (4)

The value of Ki for a given component in a stream is a function of the temperature, pressure, and composition of the stream. In light hydrocarbon systems containing essentially only the paraffin hydrocarbons, the composition dependency of the VLE constant is slight. For this reason many calculations only approximate the composition dependency of Ki.

By definition the summation of the mole fractions in

, any stream is equal to unity and shown by CX, = 1, Eyi=I, and CZi=l.

I, The operating equations used for equilibrium flash cal- -IT I

culations are derived by assuming that the separator oper-

l’; ates with the vapor and liquid leaving the separator in equilibrium as shown in Fig. 14.18.

Combining Eqs. 3 and 4 and eliminating v, leads to Fig. 14.18-Equilibwm flash separation.

To evaluate the requirements of the lease from an equipment standpoint, the operator must understand the principles of material balance and equilibrium as they apply to multicomponent light hydrocarbon mixtures. Be- fore undertaking such an evaluation, the complete properties of the wellstream must be known. This includes a component-by-component analysis, the temperature, pressure, and the quantities of gas and liquid that can be marketed. These quantities are essential if a payout for the equipment is to be estimated. If sufficient information is available, liquid recoveries can be calculated by using the procedure outlined next.

In making hydrocarbon liquid recovery calculations, compositions and flow rate normally must be expressed in terms of the number of molecular weight equivalent of the material present (moles). The feed to the recovery unit, F, is the sum of the liquid, L, and the vapor or gas, V, as given by

F=.L+v. . . . (I)

Individual component compositions are expressed in mole fractions. The mole fraction of any component in a stream is the moles of that component divided by the total number of moles for all components in the stream. The mole fraction of any component in a liquid stream is denoted x, ; in the vapor phase yi; and in the feed or wellstream ii. A material balance for any component on the same basis as Eq. 1 can be written

z;F=xJ,+y; v . . (2)

or

f;=l,fV,. _._........................... (3)

By definition the vapor/liquid equilibrium (VLE) constant, K;, for a given component is the ratio of the mole

1j= fi . . . . . . . . . . . . . . . . . Ki”

‘+L If li had been eliminated we would get

Kifi v;- K,+L,“’ . . . . . . . .

In either case L=Cli, P’/=CVi. E.qs. 5 and 6 apply at any point in the process step where

the liquid and vapor are in equilibrium. In normal lease processing, there will be many instances where the vapor and liquid are separated. Eqs. 5 and 6 can be used to predict the amount and composition of vapor and liquid present. Solution of either Eq. 5 or 6 is by trial and error because both equations involve V and L, the total vapor and liquid rates leaving the separator. To use the equations, the recommended procedure is to solve for the stream that is expected to be present in smallest amount. A value for that stream is assumed and the appropriate equation used to solve for the moles of the individual components in that stream. These are summed to give the tota- stream flow rate. If the value calculated differs from the value assumed, the next assumption should be even further in the direction of the difference.

Keep in mind that the terms involved in Eqs. 5 and 6 are molar flows and not the normal units used to describe liquid and vapor flow rates.

If a digital computer program is available it can readily be used for solution of Eqs. 5 and/or 6. In most instances the computer program will incorporate an appropriate equation of state (EOS) to evaluate equilibrium constants. The user of such a program should be aware that the results will differ for different EOS’s. Also, two programs written by different people using the same EOS may give significantly different quantities of liquid and vapor for the same separator at the same temperature and pressure


conditions. In addition, the manner in which the unde- characteristics of CO> in the presence of water. Since it fined components (C,, or C7 + fractions) are character- is not combustible, it lowers the heat content of the natural ized also will have an impact on the calculated liquid gas, and this becomes a factor especially where large present. ’ volumes of CO* are present. The same systems that are

To determine the volume of gas at any point in the proc- used to remove the H;?S also can remove the CO*, The ess, the number of moles of gas present may be multi- presence of both these acid gases in the same wellstream plied by 379.5 to determine the number of standard cubic will require larger recirculation and regeneration equip- feet. If the actual volume of the gas under the tempera- ment than if only one were present. ture and pressure are required the procedure followed in Other impurities can be found in natural gas but their determining the gas volume in Chap. 20 must be used. occurrence is not so frequent. Mercaptans. compounds To estimate the liquid production, the number of gallons similar to alcohols but with sulfur replacing the oxygen per mole for each component should be determined from (RSH instead of ROH), are found occasionally. Because a standard reference such as the GPSA’s Engineering Dutu mercaptans contain sulfur, they will form sulfur dioxide Br,~k.~ These densities in gallons per mole should be on combustion and this could present a problem. In addi- multiplied by the number of moles of each component and the results summed to give the total volume of liquid formed.

For the case of hydrocarbon recoveries from a stabilizer or from short-cycle adsorption units, no attempt is made to present the methods of calculations here. They are beyond the scope of this treatment but should be checked carefully for reasonableness and apparent accuracy.

Gas-Treating Systems for Removal of Water Vapor, Cot, and H2S Natural gas can contain any number of nonhydrocarbon impurities in the formation or at wellhead conditions. Some of these are detrimental to efficient pipeline operation, whereas others have no effect on operation but do affect the heat content or Btu rating of the natural gas.

In almost every case natural gas contains water vapor to some extent. The characteristics of the rock in the formation will determine largely the extent to which water occurs. In some cases the gas is supersaturated, which means that free water will be present. In other cases the gas is saturated at reservoir conditions, which would mean that at producing conditions it will be supersaturated. Fi- nally, the water content can be much lower than saturation but higher than specifications to be satisfied for pipeline acceptance. The formation of free water with pressure and/or temperature reduction can result in the formation of hydrates if the temperature falls below the hydrate-forming temperature. This phenomenon is discussed more fully in the preceding section. In addition to the problems of hydrates, the formation of free water because of condensation can add to the horsepower requirements for pipelines because of increased pressure drops caused when water collects in low spots in the line and reduces the flow area of the gas. This condition is also conducive to corrosion in the pipe. Water vapors must be removed from the gas, and various methods used for this removal are discussed in the following.

Sour gas is the name commonly given to natural gas containing HIS. This HIS is found in concentrations varying from a trace on up to 30 mol% . The presence of H 1 S causes severe corrosion to occur when free water is present in natural-gas pipelines. When burned it forms sulfur dioxide, which is very toxic and can be a serious problem on the marketing end of the pipeline. Various methods for removal of H2S are discussed in the following.

tion, mercaptans may present a problem because they are foul smelling. Interestingly, even if mercaptans have to be removed from the gas to make it salable, a small amount of mercaptan will normally be added to the gas intentionally before it is sold. This is done so that natural gas leaks can be detected by smell. Nitrogen is frequently found in natural gas. It has no detrimental effects other than lowering the heat content of the gas. Oxygen is sometimes encountered but the quantities are usually so low as to be negligible. Another impurity that is only rarely encountered is helium. Because manufacture of helium is all but impossible, and because of its importance in many industrial applications, the U.S. government has established helium-recovery facilities. Production of helium is a specialized low-temperature process that will not be discussed further.

Removal of Water Vapor In the preceding sections the removal of liquefiable hydrocarbons is discussed. The basic processes used for this removal of hydrocarbons invariably result in the removal of water vapors. Experience has shown that the water-vapor dewpoint of the gas leaving the low- temperature separator in these processes is from 10 to 12°F below LTS temperature for systems not using hydrate inhibitors and from 20 to 40°F below LTS temperature for systems using hydrate inhibitors. Dewpoint depressions for selective adsorption systems will be discussed further, and the use of liquid desiccants will be covered for the field of gas dehydration in this section.

To understand water-vapor dewpoint and dewpoint depression better, refer to Fig. 14.5. As was explained, this curve shows the saturation conditions for water vapor in natural gas. The water dewpoint temperature is the same as the saturation temperature for a given pressure. As- sume that gas is flowing into a dehydration unit at 1,000 psia and 100°F and that it must be dehydrated to 7 lbm water/MMscf gas to meet contract specifications. At 1,000 psia and 100°F the gas contains 62 Ibm of water. The inlet water-vapor dewpoint temperature is 100°F. From the chart for 1,000 psia and 7 lbm water/MMscf the outlet water-vapor dewpoint temperature is 33°F. The dehydrator must be capable of producing a 67°F dewpoint depression and removing 55 lbm water/MMscf gas.

Dehydration With Organic Liquid Desiccants. Since 1949 the removal of water from natural gas with organic liquid desiccants has become one of the most widely accepted methods of dehydration. Though a large number of organic materials can be used as liquid desiccants, by

A companion to H 1 S is CO2 It occurs quite frequently in natural gas but is not nearly so serious a detriment as H;?S. The main difficulty encountered is the corrosive


FUEL GAS I

GLYCOL

Fig. 14.19-Glycol-absorption gas dehydrator.

far the most generally used are the ethylene glycols. Ethy- lene glycol (EG), diethylene glycol (DEG), triethylene glycol (TEG), and tetraethylene glycol (TRG) all can be satisfactorily used for dehydration of natural gas. How- ever, by far the most widely used and generally accepted is TEG. Figure 14.19 shows a schematic flowsheet for absorption dehydration of natural gas using an ethylene glycol. The flowsheet will vary little from one glycol to the next. Wet gas enters the absorber and flows up through a series of plates coming into contact with the concentrated glycol solution. The glycol removes from 75 to 95 % of the incoming water vapor in the gas, depending on the efficiency of the overall system. This lowers the dewpoint of the gas and the dehydrated gas leaves the absorber top and flows to the sales gas line.

The dilute glycol solution leaves the bottom of the absorber and flows through a heat exchanger and is heated prior to discharging into the still column. The heated glycol flashes off solution gases and enters the still column and comes into countercurrent contact with vapor rising from the reboiler. These vapors are essentially steam but do contain some hydrocarbon vapor and a small amount of glycol. The cooler glycol feed tends to condense the glycol vapors but does not affect appreciably the flow rate of the steam or hydrocarbon vapors which are discharged from the top of the still column. The dilute glycol then dumps into the reboiler where it is heated to the temperature required for producing glycol of high enough concentration to allow the outlet gas to be sufficiently dehydrated. The regenerated glycol overflows the weir in the reboiler and flows into the storage tank. Glycol from the storage tank is pumped to the top tray of the absorber where it comes in contact again with the gas being dehydrated. The glycol flows through the absorber by gravity, removing water vapors from the gas. Since the gas contains more water with each subsequent lower stage, the water picked up on the lower trays is greater than that on the upper trays. The dilute glycol is dumped from the absorber by a liquid level controller (LLC). Some glycol dehydrators use spent glycol and some high pressure gas as the driving force for the diaphragm pump for circulating lean glycol. If this is done, the contactor will not have an LLC.

Fig. 14.20-Equilibrium (minimum) water dewpoint obtainable for a given lean TEG concentration and effective contactor temperature.

Glycol dehydration units have been designed for capacities ranging from a few Mcf/D to those designed for mil- lions of scf/D. Where the dehydration required is within the range of the TEG unit, it generally will be the most economical method of natural gas dehydration.

TEG has gained its position of prominence in gas dehydration because of several advantages. The first is that TEG will give a greater dewpoint depression with less loss of glycol than with any of the others. For equal concentrations, DEG can theoretically produce greater dewpoint depressions than is possible with TEG. However, the temperature to which the glycoliwater solution can be heated without decomposing the glycol is only about 330°F for DEG, whereas TEG begins to decompose at temperatures above about 400°F. This allows for greater concentrations of TEG in the glycol solution with consequently higher dewpoint depressions.

The concentrations of the glycol solution are normally given in weight percent, This is somewhat misleading as the absorption of water takes place on a mole-composition rather than a weight-composition basis; 98 wt % TEG contains almost 15 mol% water. Thus the weight concentrations can be somewhat misleading because of the great difference in molecular weight for TEG and water.

Fig. 14.20 shows the equilibrium dewpoint for natural gas streams in equilibrium with TEG solutions of a given concentration. You will note that the higher the contactor temperature, the higher the equilibrium dewpoint. The effective contactor temperature for a glycol absorber will


Wt% in TEG in dry qlycol outlet

14-19

- _ 98.0 98.5 49.0 99.5

II I II II 111 I I I!I i Ir,,,.,,,,,,,,tiiiiiiii 96.5 97.0 97.5 98.0 98.5 99.0

Wt% TEG in rich glycol feed

Fig. 14.21-Effect of stripping gas and vacuum on TEG regeneration.

be at the temperature of the wet gas coming to the absorber. Some heat is released when the water is absorbed but the masses of water and glycol involved are so small in comparison with the mass of the gas flowing that, in general, the contactor will operate at the gas temperature.

The dashed line on Fig. 14.20 shows the approximate concentration of TEG solution that can be obtained by a reboiler operating at 400°F and atmospheric pressure. This concentration is in the range of 98.5 to 98.7 wt% glycol. If a higher concentration of glycol is required to obtain the desired outlet dewpoint, some additional means of regeneration must be employed. This can either be stripping gas, whose effect on glycol concentration can be estimated by the use of Fig. 14.2 1, or there are a number of patented closed regeneration recycle processes that also give higher TEG concentrations. In addition, a vacuum pump can be used to lower the pressure in the reboiler and still column as indicated in Fig. 14.21.

One important variable in selecting the glycol to be used for dehydration is the loss of solution that can be expected. These losses result from vaporization. mechanical entrainment, and vaporization with the still overhead. TEG is more favorable from a loss or replacement standpoint because the consumption resulting from vapor losses is much less than for DEG or EC for the same operating conditions. The vaporization losses are directly proportional to the vapor pressure of the liquid and the vapor pressure of TEG is considerably less than for the other glycols. Glycol consumption can be kept to losses in the range of 0.1 gal/MMscf gas under normal operating conditions with contactor temperatures of 100°F or lower and

pressures above about 200 psi. Glycol contactors should always contain a well-designed mist pad to remove entrained giycol from the dehydrated gas.

There are practical limitations to using TEG for dehydration from both a dehydration-efficiency and an operating-cost standpoint. The lower the temperature in the contactor, the greater the viscosity of the glycol solution. As the viscosity of the solution increases, the efficiency of the contactor decreases, which results in less water-vapor removal per stage or plate. At temperatures below 50”F, a pronounced reduction in efficiency has been observed. At temperatures approaching the freezing point of water, the TEG solution becomes so viscous that the liquid is all but impossible to move through the system. As the contact temperature increases, the vapor pressure of the solution increases, which results in higher vaporization losses of glycol and higher outlet water content of the gas. If the contactor temperature is increased to 120°F the vapor pressure of a 99-wt% glycol TEG solution will be almost twice what it is at 100°F with consequently larger vapor losses of the circulating solution.

There is no reasonable pressure limitation on the use of TEG for dehydration. However, the maximum dewpoint depression appears to increase with pressure up to about 500 psi and then remains essentially constant for higher pressures. Dehydration units have been designed and operated satisfactorily at pressures as high as 2,000 psi.

The higher the ratio of glycol circulated to the water vapor removed from the gas, the less will be the dilution on the individual trays and the dewpoint depression will,

14-20

Fig. 14.22-Typical flow sheet for a dry-desiccant plant

theoretically, be increased. However, since the dilution on the top tray is relatively small the overall effect of increasing the circulation rate is not appreciable. The overall effect could be obtained by increasing the number of plates or trays in the contactor. From an economic standpoint a circulation rate of 3 to 5 galilbm water absorbed appears best for most applications.

The number of trays in the contactor varies with the dewpoint depression desired. For normal pipeline-quality gas on a typical wellhead installation, four actual plates should be sufficient in the contactor. When higher dewpoint depressions are required, higher TEG concentrations are used and more trays are in the contactor. For high dewpoint depression operations as many as 10 to 12 or more plates may be installed in the contactor.

Various contaminants that enter the absorber under operating conditions can prove troublesome. A small amount of liquid water entering the absorber will have no serious effects on the system. If the water contains salt, however, the salt will be deposited in the reboiler and potentially can cause reduced efficiency of heat transfer and even the formation of hotspots and tube failures. Large quantities of water will overload the regeneration equipment causing inefficient regeneration of the glycol and perhaps other operational problems.

Liquid hydrocarbons are potentially a source of trouble for the system. An inlet separator capable of separating all liquid water and hydrocarbons before introducing the gas to the absorber is a must. Presence of liquid hydrocarbons will tend to clog the filter and flood the still column and reboiler. This may result in a serious fire hazard if the overhead from the still column is not piped away from the unit. Liquid hydrocarbons also can cause glycol foaming in the absorber and seriously reduce the capacity of the unit and increase glycol losses. Heavier hydrocarbon liquids will accumulate in the glycol and not be separated. Hydrocarbons may also decompose and deposit out on the firetube along with the decomposed glycol. This will cause the glycol to become discolored and also to lose its effectiveness for dehydration.

Ordinarily, there should not be serious corrosion in a TEG dehydrator unless CO2 and/or HzS is present. These acid gases do cause serious corrosion and other


operating problems in the dehydrator. Oxygen also causes glycol degradation, For this reason, storage tank and surge drums should be gas blanketed.

Dehydration by Adsorption. Water vapor also can be removed from natural gas by use of a solid desiccant as an adsorption medium. The adsorption process is a very complex phenomenon involving transfer of the component adsorbed from the gas phase to contact with the solid. For this reason the most effective adsorbents have extremely large surface areas per unit of mass. This means that their surface is honeycombed with capillaries that serve to provide the necessary surface area.

A flowsheet of a typical dry desiccant dehydration unit is shown in Fig. 14.22. There must be at least two desiccant beds for continuous operation because adsorption is a batch process. The main gas stream flows through an inlet separator where all free liquids are removed. Any liquids can be harmful to the dehydration process. Free water will reduce the capacity of the unit if not removed. Hydrocarbons can poison the bed if not properly regenerated. The main gas stream then flows through a pressure-reducing valve that controls the flow of the regeneration gas by inducing a pressure drop in the main gas stream. The gas then flows through one of the two adsorption towers where it contacts the desiccant and the water vapors are removed. The main gas stream flows to the sales gas line through a gas-to-gas heat exchanger where heat is removed from the regeneration gas.

The second adsorption tower is regenerated while the other tower adsorbs water vapor from the main gas stream. Regeneration gas (approximately 10 to 15% of the total gas flow) is taken from the main gas stream upstream of the pressure reducing valve and passed through a heater where the temperature is raised to approximately 450°F. The hot gas then passes through the desiccant bed. The water adsorbed through the bed is vaporized and swept out of the bed by the regeneration gas. The hot, wet regeneration gas passes through the gas-to-gas exchanger where it is cooled by the main gas stream. The water condensed from the gas is separated in the regeneration gas scrubber. The regeneration gas flows into the main gas stream downstream of the pressure-reducing valve. Because of the regeneration gas flow, the pressure drop through a solid desiccant dehydration unit will be higher than for a TEG unit and will be approximately 25 psi for the system.

An adsorption dehydration unit is controlled automatically on a time-cycle basis. The most frequently used cycle is 8 hours for the adsorption and regeneration. This requires that the towers be sized to handle 8 hours of flow from a water-vapor-capacity standpoint and that the heating and cooling requirements must be satisfied on the same basis. The time-cycle controller switches the three-way valves to place one tower on stream and the other on regeneration. The three-way valve on the heater system is controlled so that, when the heating cycle is completed, the valve switches to allow cool gas to flow through the system to cool the desiccant and tower prior to placing it on adsorption.

Some of the significant factors that must be considered in the design of these units are gas pressure and temperature, gas velocity, design outlet water content, adsorption capacity of the desiccant, and free liquids to be removed from the main gas stream. For good operation,


low spots in the flowlines and equipment where water might collect during the regeneration cycle must be avoided. Proper sequencing of the valve switching must be accomplished so that no unusual pressure or velocity surges will occur and no wet gas will be allowed to pass from the unit to the sales-gas line. Precautions should be taken to ensure that no high temperatures occur in the main gas stream.

There are a number of materials that can be used satisfactorily for desiccants. Activated aluminas, silica gel beads, and molecular sieves have all been used satisfactorily. Several grades or types of each desiccant are available and the final choice of the one to be used depends on the outlet water dewpoint required and an economic balance.

The capacities of the desiccants to adsorb water vary. However, in all cases, the initial desiccant capacity will indicate that the desiccant is capable of picking up far more water than can be designed for on a long-term basis. The capacities tend to drop off fairly rapidly initially and then gradually taper off until the desiccant becomes ineffec- tive and must be replaced. The capacity and life of the desiccant are strongly dependent on the nature of the application. Under ideal operating conditions, a life of several years can be expected. Under severe fouling conditions the life of the desiccant may be reduced to one year or less. The design capacity to be used for the individual desiccant must be determined through an economic balance between the first cost of the unit and the additional operating cost for occasional desiccant replacement.

The heating required in the regeneration cycle is the sum of the heat of vaporization of the water adsorbed: the sensible heat to raise the desiccant, water, and water vapor to the regeneration temperature; and the heat required to heat the piping, vessels, etc. to regeneration temperature. Fuel requirements are directly proportional to the heat requirement. Cooling requirements are also directly proportional to the heat requirements. Proper design of the unit to minimize heat requirements can have a substantial effect on the size of heating and cooling equipment required.

The desiccant beds must be protected from slugs of liquid water and liquid hydrocarbons. Slugs of water will serve at least to decrease the adsorptive capacity of the unit and at worst can result in loss of the desiccant. Slugs of liquid hydrocarbons can cause fouling and reduce capacity for adsorbing water. The higher-molecular-weight hydrocarbons can plug the pores in the desiccant pellets and seriously affect the adsorptive capacity for water.

Protection of the beds is sufficiently important that an inlet separator must be included as an integral part of the unit. In addition, a guard or protective bed of desiccant placed ahead of the main adsorption vessel or as a top layer of desiccant in the adsorber can serve to protect the main desiccant bed from such items as compressor oil. The heavy oil when adsorbed on the desiccant can seriously affect the adsorptive capacity.

Corrosion usually is not considered a serious problem in dry desiccant dehydration. However, where there are large quantities of CO1 and/or HIS, corrosion may occur in the regeneration gas heat exchanger. At this point free water is condensing from the system and the water in combination with CO> and/or H 2 S can create serious corrosion problems.

14-21

Fig. 14.23-Amine gas desulfurizer.

Removal of Acid Gases The presence of acid gases, H2 S and/or C02, in natural gas is undesirable from many standpoints. Perhaps the principal objection is the corrosion that results when free water is present. For this reason the H 2s and CO1 normally are removed at the wellhead or relatively close to it. There are a number of systems that can be used for removal of acid gases.

Sweetening by Ethanolamines. Perhaps the most widely used type of acid-gas-removal system involves the use of an ethanolamine. A simplified flow diagram of a typical ethanolamine-type desulfurization unit is shown in Fig. 14.23.

In this process a solution of water and ethanolamine that may vary from about 15 to 60 wt% ethanolamine is used for removing HzS and CO? from the incoming gas stream. The process is based on the principle that the acid gases, HzS and CO2, will react with the ethanolamine at ordinary temperatures. The reaction can be reversed by reducing the pressure and heating the solution. The sour gas passes up through the contactor and the lean ethanolamine solution passes downward. The foul solution is discharged from the bottom of the contactor and flows through a heat exchanger before it discharges into the top of the still or regenerator column. The ethanolamine solution is boiled by application of heat in the reboiler. This boiling action supplies vapors, primarily steam, that pass up through the still column sweeping the H2S and CO;? from the ethanolamine solution.

The regenerated ethanolamine leaves the reboiler and passes through the amine-to-amine heat exchanger into a storage tank from which it is recirculated to the contactor with the amine pump. The H2 S and CO? leaving the top of the still column have a large volume of steam with them. To keep down the quantity of makeup water required and to minimize ethanolamine losses the overhead product usually is cooled. The water condensed in this cooling is returned to the regenerator as reflux.

A number of different types of ethanolamine can be used in the process. Monoethanolamine (MEA), diethanola- mine (DEA), diglycolamine (DGA), and methyldieth- anolamine (MDEA) are among those that are the most popular. There are a number of things that can affect the choice of ethanolamine to be used in a given system. MEA is a stronger base and has a lower molecular weight than


the others. This means that lower concentrations should be possible and that the removal of H 2 S and CO1 should be greater. DGA is also a primary amine with good removal properties but has the same molecular weight as DEA. DEA is a secondary ethanolamine and slightly less basic and therefore will not make specification sweet gas at as a low a pressure as MEA and/or DGA. MDEA is said to offer selectivity when only H2S removal is desired. Other sulfur compounds such as carbonyl sulfide (COS) also can affect the choice. COS reacts irrever- sibly with MEA and this requires the installation of a reclaimer to control MEA losses.

In many ethanolamine sweetening units, corrosion is the greatest operating problem. Corrosion can occur in the reboiler, storage tank, still column, heat exchangers, and the contactor. Most of the corrosion problems can be minimized by correct design. Limiting the heat flux in the reboiler and condenser can serve to minimize corrosion there. Maintaining pressure on the sour amine solution until it is passed through the heat exchanger and to the point of introduction to the still column can help minimize corrosion in the heat exchanger and piping. Limiting the amount of acid gas pickup per unit of circulating solution will also help. Limiting the temperature on the stream introduced to the top of the still column will help. The lowest possible pressure should be maintained on the still column because this lowers the boiling temperature of the ethanolamine solution.

Another severe problem that can occur in ethanolamine- type desulfurizers is foaming. This foaming most frequently occurs in the contactor and can result in excessive amine losses. Foaming has many causes but the presence of liquid hydrocarbons in the contactor is a frequent one. In addition, particulate matter can stabilize foam. A good filter should be used to remove any iron sulfide or other solid materials from the circulating solution. If oxygen enters the system it will cause problems. It can cause the

formation of thiosulfates and also cause direct degradation of the amine solution.

Iron Sponge Sweetening. Hydrated iron oxide can also be used for removing Hz S from natural gas. This process is “selective” and removes only the HzS from the gas. It is suitable for removing small quantities (a few grains per 100 scf) of H 2s from natural gas streams. The flow sheet is similar for that of a solid desiccant dehydration unit except for the fact that there is no regeneration gas stream. The iron oxide or sponge is generally suspended on wood chips to disperse it and limit the heat release caused by the reaction of H2.S with the iron oxide. The iron oxide must be kept in a basic environment (pH > 8) so that soda ash or caustic soda solution is normally injected into the bed with the natural gas. The gas leaving the bed has essentially all the H 2 S removed.

Since the iron sponge is consumed in the process and must be replaced frequently, the vessels must be constructed in such a way that the bed can be replaced easily. Iron sulfide will self-ignite when exposed to air, so extreme caution must be used when replacing the iron sponge bed. In addition, disposal of the spent sponge can present a problem because, when it burns, sulfur dioxide is formed.

In all desulfurization units, disposal of the HzS gas presents a problem. Increasingly, government agencies forbid exhausting the H2S to the atmosphere either as H2S or, after incineration or flaring, as SO;?. For this reason disposal of the removed HzS must be an integral part of the planning for any desulfurization unit.

References I. Maddox, R.N. and Erbar. J.H.. “Advanced Techniques and Ap-

plicatmns,“ Gas Conditioning und Procrs.h~. Campbell Petrole- um Series. Norman, OK (1981) 3.

2. Campbell, I.M.: Gus Conditioning and Prows.sing. Campbell Petroleum Series. Norman, OK (1982) 2.

3. GPSA D~gineen’ng Data Book, ninth edition, fifth rewsion. Gas Prw- essor~ Suppliers Assoc., Tulsa (1981).

Chapter 15

Surface Facilities for Waterflooding and Saltwater Disposal K.E. Arnold, Paragon Engineering Services*

Introduction In producing operations it is often necessary to handle brine that is produced with the crude oil. This brine must be separated from the crude oil and disposed of in a manner that does not violate environmental criteria. In offshore areas the governing regulatory body specifies a maximum hydrocarbon content in water that it will allow to be discharged overboard. Currently this ranges from 7 to 72 mg/L depending on the specific location. In most onshore locations the water cannot be disposed of on the surface because of possible salt contamination and it must be injected into an acceptable disposal formation or disposed of by evaporation. On the other hand, it is often desimble to inject water into the producing formation to maintain reservoir pressure or increase recovery through watetllooding. Produced water that is properly treated to remove hydrocarbons and solids can be used for this purpose. In addition, supplemental sources of water from other formations or from surface sources could be used for watefflooding

The purpose of this chapter is to discuss the equipment and design criteria that are employed in common systems for either waterflooding or for saltwater disposal. In both cases the design engineer may be concerned with designing piping systems, selecting pumps, separating solids from water, treating hydrocarbons from water, removing dissolved gases and solids from water, treating hydrocarbons from solids, and overall project management.

Piping System Design In any waterflood or disposal system, it is necessary to gather the water from one or more sources for treatment and then to distribute it to one or more points for injection or disposal. This section discusses criteria for selecting pipe diameter, pipe materials, and wall thickness, as well as general design considerations for cross-country piping systems.

‘Authors of theorigmal chapter on thus topic in the 1962 edition were W F. Ellison and R.H. Lasater.

Pipe Diameter

The choice of pipe diameter depends on the pressure drop available, or on a range of acceptable velocities for fluid flow in the pipe.

Pressure at a Point. The pressure at any point in a system can be determined from Bernoulli’s theorem, if the pressure at any other point is known. This theorem, which is derived from conservation of energy, is given

by

z, +P1+(“1)2=22+)2+L!e+zfl, . ..(I)

PI 2g P2 2g

where Z = elevation above a datum, p = pressure, p = density, v = velocity, g = gravitational constant, and

Zj = head loss due to friction between Points 1 and 2.

Darcy demonstrated that head loss was given by

flV2 zfl=- 2gdi ’

. . . . . . . . . . . . . . . . . . . . . . . .

where f = friction factor, L = length, and

di = pipe ID.

The friction factor is, in turn, a function of the nondimensional Reynold’s number, given by

h&=@ . . . . . . . . . . .(3) FL


where NRe is the Reynold’s number and p is the viscosity. The relationship between Reynold’s number and the friction factor is given in the classical Moody diagram (Fig. 15.1).

Pressure Drop in Liquid Lines. The pressure drop for liquid lines can be derived from Eq. 1 as

4J= 0.0000115flq~)2y~L

(di)5 , . . . . . . (4)

where Ap = pressure drop, psi, qL = liquid flow rate, B/D, ye = specific gravity of liquid relative to water,

L = length of line, ft, and d; = pipe ID, in.

This relationship is shown graphically in Fig. 15.2. For liquid flow in pipelines, a friction factor of 0.02 is

sometimes used for preliminary calculations. In determining the actual friction factor from Fig. 15.1, it is sometimes convenient to use either of the following equations.

NRe =7,734 TLdiv

. . . . . . (5)

or CL

NRe=92.1 YL4L -----$-, . . . . . . . . . . . . . . . . (6)

where v is velocity, ftis, and p is viscosity, cp. The roughness, 6, to use in determining which

relative-roughness, t/d, curve governs in Fig. 15.1 depends on the age of the pipe and the material that lines its inside surface. Cast-iron pipe could be expected to be rougher than bare-steel pipe and bare-steel pipe rougher than plastic-lined steel pipe. Roughness factors for new pipe are given in Table 15.1. These should be increased by a factor of two to four to account for corrosion or incrustation effects that could occur with age.

In the past, the empirical Hazen-Williams’ equation has been used by some engineers for flow of water through pipelines. With the advent of computers and programmable calculators, these empirical equations are no longer recommended. However, for completeness, the Hazen-Williams equation is given as

zj =0.015 (4L) ‘.*5L

(di)4.87(CHW)‘.85 ’ ’ ’ ’ ’ ’ ’ ‘. ‘.

where ZJ = friction head loss, ft of liquid, qL = liquid flow rate, B/D,

L = length of line, ft, di = pipe ID, in., and

CHW = constant with a value of 80 to 140, depending on the inside pipe material and its age.

(7)

In determining the length to be used in either Eq. 4 or 7, it is necessary to include an allowance for valves, ells, tees, reducers, and entrance and exit losses from vessels. The most common way of accounting for these pressure losses is to include a certain additional length of pipe to the actual length of pipe in the value used for L. Table 15.2 shows the length of pipe that should be added for various valves and fittings.

Velocity in Liquid Lines. Although it is necessary that a selected pipe diameter ensures that the pressure drop is not excessive, in many cases the velocity in the line, and not the pressure drop, will determine the pipe diameter. In most of the short liquid lines within the plant, there will be more than sufficient pressure available to transport the liquid from one piece of equipment to another. However, if the entire pressure drop were taken in the piping, and only a marginal pressure drop were taken across a liquid-control valve, the velocities in the pipe would be high enough to cause noise, erosion of products of corrosion, or water-hammer problems. For this reason a maximum liquid flow velocity of 15 ftis usually is recommended.

Consideration should also be given to a minimum velocity necessary to prevent solids buildup in the bottom of the pipe. Experiments have shown that when the liquid velocity falls below a certain value, any solids present will settle in a horizontal bed until an equilibrium velocity is reached over the bed. At this velocity, erosion of the solid particles on the surface of the bed is exactly balanced by the deposition of additional particles. It can be shown that, for situations likely to be encountered in oilfield pipelines, a velocity of between 2 and 4 ftis is required to keep from building up such a bed. For this reason a minimum velocity of 3 ft/s is usually preferred for any liquid piping likely to contain solids.

The following equation has proved useful in calculating velocities.

v=o.o124L (di)2 . . . . . . . . . . . . . (8)

This equation is shown graphically in Fig. 15.3.

Choosing Pipe Diameters in Liquid Lines. The choice of a pipe diameter for a liquid line thus becomes one of choosing a diameter large enough for the pressure available while attempting to keep the velocity between 3 and 15 ft/s. On short lines within a plant, it is usually quicker to choose a diameter based on velocity considerations and then check for pressure drop. On longer lines, or on lines within a plant that flow between atmospheric tanks (low available pressure), it is usually desirable to choose a diameter based on pressure-drop considerations first and then to check velocity.

On lines that experience large variations in elevation, it is desirable to employ Bernoulli’s theorum (Eq. 1) at all high points to ensure that there is sufficient positive pressure so that a vacuum is not created. Although it is possible to operate a line with a high-point vacuum, relying on a syphon effect may make it difficult to restart a line if the syphon ever loses its liquid seal. In addition, at any point where a vacuum exists, there is a very real feasibility of drawing oxygen into the system with resultant corrosion and bacteria problems.

SURFACE FACILITIES FOR WATERFLOODING & SALTWATER DISPOSAL 15-3

.

:

,

6 tb+.*.t*. t

Fig. l&l-Friction factor chart.

Fig. 15.2-Pressure drop in liquid lines.


TABLE 15.1-ABSOLUTE PIPE ROUGHNESS (IN.), NEW PIPE

Unlined concrete Cast iron Galvanized iron Carbon steel Fiberglas epoxy Drawn tubing

0.0 1 to 0.1 0.01

0.006 0.0018 0.0003 0.0001

TABLE 15.2-EQUIVALENT LENGTH OF 100% OPENING VALVES AND FITTINGS (FT)

Weld Thread

Nominal Short Long Pipe Size Globe Valve or 450 Radius Radius Hard Soft

(in.) Ball Check Valve Angle Valve Swing Check Valve Plug Cock Gate or Ball Valve Eli Eli Ell T T -- 1% 2 2% 3 4 6 8

IO 12 14 16

:: 22 24

fi 42 48 54 60

55 26 70 33 80 40 100 50 130 65 200 100 260 125 330 160 400 190 450 210 500 240 550 280 650 300 666 335 750 370 - -

- -

13 7 17 14 20 11 25 17 32 30 48 70 64 120 80 170 95 170 105 80 120 145 140 160 155 210 170 225 185 254 - 312 - -

- -

4 6 7 9 10 11 12 14 15 16 21 25 30 35 40 45

1-2 3-5 2-3 8-9 2-3 2-3 4-5 3-4 10-11 3-4 2 5 3 12 3 2 6 4 14 4 3 7 5 19 5 4 11 8 28 8 6 15 9 37 9 7 18 12 47 12 9 22 14 55 14 10 26 16 62 16 11 29 18 72 18 12 33 20 82 20 14 36 23 90 23 15 40 25 100 25 16 44 27 110 27 21 55 40 140 40 25 66 47 170 47 30 77 55 200 55 35 88 65 220 65 40 99 70 250 70 45 110 80 260 80

900 Miter Bends

Enlargement Contraction

Standard Sudden Reducer Sudden

Equivalent L in Terms of Small d”

Standard Reducer

Two-miter Three-miter Four-miter d/D=% d/D= I/s d/D=3h d/D=% d/D=j/a d/D=,/4 &-,=,,z d/0=3/4 d,D=,,2 d/D=Q

- - - 5 - - - 7 - - - 8 - - - 10 - - - 12 - - - 18 - - - 25 - - - 31 28 21 20 37 32 24 22 42 38 27 24 47 42 30 28 53 46 33 32 60 52 36 34 65 56 39 36 70 70 51 44 84 60 52 - 98 69 64 - 112 81 72 - 126 90 80 - 190 99 92 -

‘d is ID of smaller outlet and D IS ID of larger outlet.

3 4 5 6 8

12 16 20 24 26 30 35 38 42 46

- - - - -

1 1 2 2 3 4 5 7 8 9

10 11 13 14 15

- - - - -

4 1 5 1 6 2 8 2 10 3 14 4 19 5 24 7 28 8 - - - - - - - -

-

- - - - -

-

-

- - -

3 3 4 5

i 12 15 18 20 24 26 30 32 35

-

- - -

2 1 3 1 3 2 4 2 5 3 7 4 9 5 12 6 14 7 16 8 18 9 20 10 23 11 25 12 27 13

- - - - -

-

- - -

1 1 2 2 3 4 5 6 7

- - - -

-

-

- - -

- - - - -

1 2 2 2

- - - - - -

-

- - -


LIQUID FLOW RATE, BARRELS FLUID/DAY

Fig. 15.3-Velocity in liquid lines.

Sometimes, it is not feasible to satisfy minimum velocity criteria during the early stages of a project, where flow velocities are low, without violating pressure drop or maximum velocity criteria at peak flow rates. In such cases, engineering judgment is needed to choose between alternatives such as (1) installing a smaller line initially and either looping the line or installing more pumps at a later date, (2) allowing an equilibrium-solids bed to be deposited initially and relying on it being eroded as flow velocities increase, or (3) allowing a velocity greater than 15 ft/s at peak flow rates.

Pressure Drop in Gas Lines. Although this chapter deals primarily with liquid flow, it may be necessary to size gas lines as part of the project. Source water may come from gas-lifted wells, which would require a gas- lift gas-distribution system, produced water may have flash gas associated with its separation and treating equipment, flotation units and gas strippers require gas lines to operate, and fuel, instrument and utility gas un- doubtedly will be required.

Flow in gas lines is considered isothermal. That is, there is sufficient heat transfer to and from the surrounding air, water, or soil to keep the temperature of the gas in the line from changing as the pressure changes because of friction losses. If we assume steady-state gas flow, an ideal gas (Z = 1 .O), and a constant friction factor over the length of the line, the following equation can be derived.

(~q)‘-(p~)~=2d~ ;&);T6, . . . . . . . . . 1

where p1 = pressure at pipe inlet, psia, p2 = pressure at pipe outlet, psia, qg = flow rate of gas at standard conditions,

MMscf/D,

y&T = specific gravity of the gas at standard conditions relative to air, and

T = temperature, “R.

When the Reynold’s number is calculated to determine the friction factor from Fig. 15.1, it is often convenient to use either

qgyg NRe=20,102- .,......., .__...._._.t (10) diwgf

or

NRe = 335 VgfPldiYg

. . . . . . . . . . . . . . . . t . . (11) TpKf

where vgf is the velocity of gas at specific flow conditions, ft/s, and pd is the viscosity of gas at specific flow conditions, cp.



The viscosity of the gas at flow conditions can be derived from Fig. 15.4. Where

PI -P2 p<O.l,

PI

Eq. 9 reduces to

12)

It is recommended that either Eq. 9 or 12 be used. However, in the past, two empirical equations were developed that have been used extensively. The Weymouth equation2 is used for lines with high Reynold’s numbers. It assumes that the friction factor is merely a function of pipe diameter. That is, flow is occurring in the flat part of the relative roughness curves on the Moody diagram. The Weymouth equation can be written as

(PI)2 -(p2)2 =O.Sl ‘;;;;,;3z. . . . . . . . . 1

(13)

This equation would normally apply to short lines within the plant where gas velocities, and thus Reynold’s numbers, would probably be high. Figs. 15.5 and 15.6 can be used to solve this equation.

For long gas lines, where velocities are likely to be less and the friction factor will depend on both the line size and the flow rate, the Panhandle equation2 has been developed:

0.96 1.96~

(p,)2-(p2)2=0.2 yg g;,,, ) . . . . . . I

(14)

where E is flow efficiency (1 .CKJ for new pipe, 0.92 for average conditions, and 0.85 for unfavorable conditions).

Velocity in Gas Lines. As in liquid lines, there is a velocity consideration in picking a pipe diameter for a gas line. At high velocities, there could be problems with both noise and erosion of the layer of corrosion products on the inside of the pipe. The greater the rate of erosion of these products, the greater the rate of corrosion the line would experience.

From a noise consideration, the velocity in the pipe should be limited to 60 to 80 ft/s at actual flow condition of pressure and temperature Experiments have shown that there is a correlation between velocity and erosion of the products of corrosion, which is given by

vRf=- 2, . . . . . . . . . . . . . . . .

where pg is the density of the gas at actual conditions, lbm/cu ft, and CE is a constant for erosional flow. Eq. 15 can be rewritten as

. . . . . . . . ..I.........

API Recommended Practice 14E” for offshore piping systems proposes using a value of 125 for intermittent service and 100 for continuous service for the constant CE. Recent experimental data indicate a CE as high as 300 may be appropriate for an allowable corrosion rate of 10 mil/yr.

The choice of a value to use between 100 and 300 depends on the judgment of the design engineers as to the corrosivity of the gas and the cost of being overly conservative. Where pressures arc below 1,000 to 1,500 psi, the noise criteria will govern and the erosional criteria can be neglected.

If the gas is saturated to the extent that liquids are likely to condense from the vapor phase because of ambient cooling, it is recommended that a minimum velocity of 10 ft/s be maintained. This will sweep the liquid out of the line. At lower velocities, liquid may accumulate at low spots, accelerating corrosion and potentially leading to liquid slugging in the line.

Choosing a pipe Diameter in Gas Lines. As in liquid lines, the choice of pipe diameter will depend on satisfying both the pressure-drop and velocity criteria. For almost all lines within the plant, the velocity criteria will govern, and the pressure drop will probably not even have to be checked. For long gathering or distribution lines the pressure drop available may govern, or a study of pipe diameter and cost vs. compressor horsepower and cost may be necessary.

Materials

Selection of materials for pipe, valves, and fittings for any piping system must take into account the pressure rating of the application, the corrosivity of the fluid, and the location of the line. There is some economic cost over life for each selection and this must be taken into account in determining the types of material to use for a given application.

Asbestos-Cement pipe. Because of its resistance to corrosion and low cost, asbestos-cement pipe is recommended for use in large-diameter lines on gravity or low- pressure water systems (200-psi maximum working pressure). The joint connection consists of asbestos- cement couplings with rubber rings. The pipe itself is not flexible, but a deflection of 6” can be obtained at the coupling. This proves advantageous in eliminating abrupt bends when it is necessary to lay pipe on horizontal or vertical curves. Care must be exercised in the actual laying of the pipe by providing a conditioned ditch for the pipe. Installation guides have been published by the manufacturers of asbestos-cement pipe and are available to the design engineer on request.

Asbestos-cement pipe can be coated internally with plastic or fiberglass to lower the friction factor and to protect against seepage of any crude oil that may be in the water stream. Asbestos-cement has the disadvantage of being more brittle than steel pipe, which could make it susceptible to damage from external loads. In addition, proper bedding and backfill compaction is required to prevent future pipe movement resulting from external loads, which could cause joint leakage.


m 0 r- X

Fig. 15.5-Gas flow based on Weymouth formula.


20,000

l0,000 9,000 8,000 7mo b.000

5.000

two

3.ooa

1.wo 900 Boo 700

600

500

400

300

200


Plastic Pipe. In recent years, there has been a continuing increase in the use of the various types of plastic pipe especially in moderately low-pressure (300 psi), small- diameter (6 in. and lower) water service. Plastic pipe is not susceptible to either internal or external corrosion, it has a low friction factor, and its light weight makes it easy to install. This is a rapidly developing field and there are numerous proprietary brands. However, in general, plastic pipe can be purchased in accordance with the following API specifications: Spec. 5LE for polyethylene line pipe (PE),4 Spec. 5LP for thermoplastic line pipe (PVC and CPVC),’ and Spec. 5LR for reinforced thermosetting resin line pipe (RTRP). 6

The latter category, which includes fiber-reinforced plastic (FRP) pipe, is the strongest, with PVC next, and PE last. All plastic pipe is sensitive to temperature and must be derated as temperature increases. PE pipe is limited to lOO”F, PVC pipe to 140”F, and RTRP pipe is limited to 150°F unless specific tests are run by the manufacturer.

The pressure rating of plastic pipe also depends on the fluid being handled. PE pipe handling crude oil has 50% of the pressure rating of the same pipe handling water. PVC pipe must be derated to 40 % of its water strength if handling crude oils because of the long-term effect of hydrocarbons on the material. The de&ion of RTRP pipe is specified by the manufacturer.

Because of the superior strength and greater resistance to heat and hydrocarbons, the use of FRP pipe is increasing. Pipe is available in diameters to 12 in. and pressure ratings that exceed those in 5LR by a factor of two. However, plastic pipe has the disadvantage of being extremely brittle. This can lead to damage in installation. More important, if the ditch is not prepared correctly in rocky soil, over time as the line settles, rocks can come in contact with the underside of the line and cause a stress concentration and eventual failure. It also has the disadvantage of becoming brittle with time when exposed to direct sunlight.

Cast-Iron Pipe. Satisfactory installations of cast-iron pipe have been made on systems where pressures are more than 200 psi but less than 250 psi. Cast-iron pipe has a high corrosion-resistant quality. This pipe has a higher initial cost than either asbestos-cement or plastic pipe. While it is less susceptible to impact and temperature effects than plastic pipe, it is more brittle than steel pipe. For this reason, it must have flat-faced flanges and care must be exercised to ensure that it is not connected to a valve fitting with a more standard, raised- face flange.

If cast-iron pipe is subjected to a fire and then hit by a stream of cold water, it is susceptible to cracking. For this reason, it is not generally acceptable for services containing hydrocarbons.

Carbon Steel Pipe. The most commonly used material for piping systems is Grade B line pipe (35,000-psi yield) manufactured in accordance with API Spec. 5L for line pipe. ’ Because of its strength, this material can withstand high pressures and is available in diameters to 64 in.

Steel pipe has excellent impact resistance and flexural strength. Unfortunately, it is much more susceptible to both internal and external corrosion, and its installation is more expensive than the lighter plastic pipes.

Where oxygen is excluded from the system, internal corrosion may not be a problem. External corrosion can be reduced by burying and protecting the pipe with a cathodic-protection system. The system could consist of sacrificial zinc or aluminum anodes, which are attached electrically to the pipe at specified intervals. For long pipelines, an impressed-current system may be considered, which would enable the use of fewer anodes.

External corrosion could also be combatted with a coating system. For pipe that is exposed to salt air, a three-coat epoxy-paint system is often specified. Less- elaborate systems are employed in less-severe at- mospheres. Underground or underwater pipe may be protected by a thin-film-epoxy, coal-tar-epoxy, or extruded-plastic system. Thin-film-epoxy systems seem to be more popular lately, because of their greater toughness to potential handling and installation damage.

Most long pipelines are protected both with a coating system and cathodically. The coating system decreases the current demand, while the cathodic-protection system provides protection for any breaks, gaps, or scratches that develop in the coating.

Where internal corrosion is anticipated, carbon steel can be protected in four ways.

1. Cement lining. Individual joints have a s-in. cement-mortar mix applied by a centrifugal spinning procedure. The field joints are protected by an asbestos welding gasket compressed between the joints. Cement- lined pipe is available in sizes from 3 to 24 in. API RP lOE* specifies cement thickness vs. pipe size. The accepted tolerance is f1/j2 in.

2. Co&-kr-epoxy lining. Typically this is a 3/,,-in. lining that is centrifugally spun into the pipe section after a field joint is made. This technique is limited to large- diameter lines.

3. Plastic lining. Various types of epoxies have been installed in steel lines. These can be installed in a small section at a time, or by use of cylindrical devices (pigs), which are run into the line in sections up to 1 mile long. Some success has been reported with plastic lining used to protect pipe against further deterioration, but the necessity of cleaning the pipe wall before applying the plastic coating makes this very difficult. At the present time, manufacturers claim the smallest pipe that can be internally plastic coated is 1 I/ in. nominal.

4. Use of liners. There have been several successful installations of FRP and other liners in steel pipe. The steel provides the strength to resist high pressures and the liner provides corrosion resistance. Other than the high cost of installation, this system has the further drawback in that because of long-term permeability through the liner, the pressure in the annulus between the liner and steel can reach an equilibrium with the line pressure. When the line is depressured for maintenance, this may cause collapse of the liner if the system is not carefully designed. Liners have been installed in pipe from 2- to 30-in. diameter and for distances as long as 0.5 mile in a single pull.

15-l 1 SURFACE FACILITIES FOR WATERFLOODING & SALTWATER DISPOSAL

TABLE 15.3-DESIGN PROPERTIES AND ALLOWABLE WORKING PRESSURES FOR PIPING’

Namlnal Pipe Welght of Wall ID FIOW Allowable Working Pressures for Temperature~(~F)Not To Exceed

Size Schedule Plfx (OR,

Thtckness d, Area -2OlO (in.] Number (Ibm/ff) (in.) (in.) d: P4 4 100 200 300 400 500 600 700

--ET 'h -- 0651 0.640 -~

0 109 0.622 0.0931 0.00211 2,256 2,256 2,258 2,258- 2,134 -- 1,953 1,863

I'/2

2

3

4

6

8

10

12

14

16

,6

20

24

540

X80 s40 X80 160 XX s40 X80 160 xx 540 X80 160

22 X60 160 xx s40 X80 160 xx s40 xao 160

s"4", X60 xx 160 s40 X60 160 S X 160 10 530 X 10 530 540 IO S X 10 520 x30 10 se0 X

1 131 1474 1.679 2172 2.644 3659 2.718 3832 4.666 6.409 3.653 5022 7.445 9030 7.58 10.25 14.33 18.58 10.79 14.99 22.51 27.54 16.98 28.58 4530 5317 2656 434 724 747 405 54 7 1157 49 6 65 4 ,603 36 7 546 72 1 421 626 828 474 70.6 93 5 52.7 78.6 104.1 63.4 94.6

1255

1.050

1.315

1.900

2.375

3.500

4500

6.625

6825

10750

12750

14.000

16.000

16000

20000

24000

0113 0.824 0.3799

0 154 0742 0 2249

0133 1.049 1.2700 0179 0.957 0.8027 0250 0.815 0.3596 0358 0.599 0.0771 0145 1.810 10.620 0200 1500 7 594 0261 1336 4286 0400 1100 1611 0154 2.067 37.72 0218 1.939 27 41 0343 1.887 13.74 0436 1.503 767 0216 3068 271 BO 0 300 2900 20510 0436 2.624 12440 0600 2300 64 36 0237 4.026 1.0560 0337 3.826 619 a 0531 3438 4603 0674 3.152 311 1 0.260 6 065 8,206 0432 5 761 6,346 0.718 5187 3,762 0864 4897 2,816 0.322 7981 32,360 0 500 7625 25,775 0.675 6875 15,360 0 906 6613 14,679 0.365 10020 101,000 0500 9750 68,110 1 125 8.500 44,371 0375 12.000 248.800 0500 11.750 223,970 t 312 10.126 106.461 0250 13.500 446,400 0375 13.250 406,394 0500 13000 371,290 0250 15 500 894.660 0375 15 250 824.601 0 500 15 000 759,375 0250 17500 1.641.309 0375 17250 1.527.400 0 500 17000 1,419,900 0.250 19 500 2,619,500 0375 19250 2.643352 0 500 19000 2,476,099 0750 23 500 7.167.030 0375 23250 6.793.832 0 500 23000 6.436.300

0.00371 0.00300

0.00600 0.00499 0.00362 0.00196 0.01414 0.01225 0 00976 000660 0.02330 OD2050 0.01656 0.01232 0 05130 004587 0.03755 002685 0.08840 0 07986 006447 0 05419 0 2006 0 1810 0 1469 0 1308 0.3474 0 3171 0.2578 0 2532 0.5475 0.5185 0.3941 07854 07528 0 5592 0 9940 0 9575 09211 1 310 1 268 1 227 1 670 1 622 1575 2074 2 021 1 969 3012 2 948 2 063

1.933 1.933 1.933 1.933 1,827 1,672 1.595 3.451 3.451 3,451 3,461 3,261 2,965 2,647 2,103 2.103 2,103 2,103 1.986 I.819 1.735 3,466 3,466 3,468 3,468 3,277 3,000 2,861 5,720 5,720 5,720 5,720 5.405 4.948 4,719 9,534 9,534 9,534 9,534 9,010 0,247 7,866 1,672 1,672 1,672 1,672 1.580 1.446 1.379 2,777 2,777 2,777 2,777 2,624 2,402 2.291 4,494 4,494 4,494 4,494 4,247 3.867 3,707 7,226 7,226 7,228 7.228 6,631 6,253 5,963 1,469 1,469 1,469 1,469 1,388 1.270 1.212 2,486 2,468 2,486 2.486 2,351 2,152 2,053 4,617 4,617 4,617 4.617 4,363 3,994 3.809 6,284 6,284 6,284 6,284 5.939 5.436 5.185 1,640 1.640 1,640 1,640 I.550 1,419 1,353 2,552 2,552 2,552 2,552 2,412 2,207 2.105 4,122 4,122 4.122 4,122 3,695 3,566 3,401 6,069 6,069 6,089 6,089 5,754 5,267 5,024 1,438 1,439 1,439 1,439 1,359 1,244 1,187 2,275 2,275 2,275 2,275 2,150 1,968 1,877 3.978 3.978 3.978 3,978 3,760 3.441 3,282 5,307 5,307 5,307 5,307 5,015 4,590 4,376 1,206 1.205 1,205 1,205 1,139 1.042 994 2,062 2,062 2,062 2,062 1,946 1.783 1.701 3,759 3,759 3,759 3,759 3,552 3,251 3,101 4,659 4,659 4,659 4,659 4,403 4,030 3,844 1,098 1,098 1.098 1.098 1.037 950 906 1.864 1.664 1.664 1,664 1.761 1,612 1.537 3,554 3,554 3,554 3,554 3,359 3,074 2,932 3,699 3,699 3,699 3,699 3,496 3,200 3.052 1,022 1,022 1,022 1,022 966 664 643 1,484 1,484 1,484 1,464 1.403 1.264 1,224 3,736 3,736 3,736 3,736 3,531 3,232 3,082 976 976 976 976 922 644 805

1,245 1,245 1,245 1,245 1,177 1,077 1,027 3,113 3.113 3,113 3,113 2,942 2,693 2,566 466 466 466 486 460 421 401 607 a07 807 607 763 698 666

1.132 1,132 1,132 1,132 1,069 979 934 425 425 425 425 402 368 351 705 705 705 705 666 609 581 987 967 967 987 933 654 a15 377 377 377 377 357 326 311 625 625 625 625 591 541 516 876 876 876 676 a23 757 722 339 338 339 339 321 293 280 562 562 562 562 531 466 464 767 767 767 767 743 680 649 282 262 262 282 267 244 233 466 467 467 467 442 404 366 660 654 654 654 618 565 539

'ASTM Al06, grade 6 seamless pope-petroleum refinery plplnq code for pressure p!p~ng ANSI 631.3.1976-corrosnn allowance=0 05

Exotic Metals. In some highly corrosive environments, especially those associated with CO;! floods, stainless- steel pipe has been employed. This is extremely expensive. However, it may be the only acceptable long-term solution.

Pressure Ratings for Steel Pipe

Pipe-Wall Thickness. Water-injection lines may have to withstand extremely high pressures. In addition, some of the piping in the plant may include high-pressure gas, oil, or water piping. Thus, steel pipe-wall thicknesses other than “standard” weight may be needed.

The thickness for any pipe depends on the pressure rating, temperature, diameter, and piping code applicable to that pipe. Piping codes generally used are: (1) ANSI B31. l-power piping,’ (2) ANSI B31.3-chemical plant and petroleum refinery piping, lo (3) ANSI B31.4- liquid petroleum transportation piping systems, ’ ’ and (4) ANSI B3 1. S-gas transmission and distribution piping systems. I2

Generally, B31.1 and B31.3 have the same equation for determining pipe thickness. This tends to be more conservative than that contained in the other two codes. Wall thicknesses determined from this equation are shown in Table 15.3. These are primarily used for steam piping at all locations, for all piping on offshore platforms, and in large onshore plants that represent a large capital investment.

B3 1.4 and B3 1.8 have the same equation for pipe-wall thickness. The allowable pressure in a given pipe is dependent on the “construction factor.” which is a measure of the potential cost of failure at a given location. The greater the hazard to the public because of failure, the smaller the construction factor. Allowable pressures for various grades of steel and construction factors are given in Table 15.4 and 15.5.

Generally, oil or water lines can be specified with a construction factor of 0.72. The factor for lines containing gas depends on the location of the line. Table 15.5 indicates, in general, the factors to be used. However, care should be taken, since there are specific definitions


TABLE 15.4-GAS TRANSMISSION AND DISTRIBUTING PIPING SPECIFICATIONS FOR CARBON STEEL AND HIGH YIELD STRENGTH PIPE’

Allowable wchnq PreSS”leS UP to 25o~F

and APISLX pope having the same wafted minimum yleld strength Itcable 10 “Liquid Petroleum Transportation Plpmg Code,” ANSI 831


TABLE 15.5-CONSTRUCTION TYPE DESIGN FACTOR, F

Construction General

Type F A 0.72 B 0.6

C 0.5

D 0.4

Description

oil field and sparsely populated area semideveloped areas and lease

facilities commercial and residential subdivided

areas and compressor stations heavily congested areas with multistory

buildings

in B3 1.8 and in the Dept. of Transportation regulations governing gas transmission lines and gathering lines within subdivisions for what factor to use. These detail definitions should be consulted for lines approaching or crossing roads and railroads, and for lines in built-up areas.

Pressure Rating Classes. Flanges and fittings commonly used in the oil field are purchased in accordance with one of the two following specifications: (1) ANSI B16.5-pipe fittings and flanged fittings I3 or (2) API Spec. 6A-wellhead equipment. l4 These specifications establish dimensional standards, allowable pressure ratings, method of production, material properties, and

inspection and test procedures for specific piping classes. ANSI has seven classes of piping, each one of which has a table similar to Table 15.6 that specifies the pressure rating for fittings of dimensions applicable to that class. The pressure rating is a function of material and temperature. Most oilfield piping falls in Material Group 1.1. The pressure ratings for this group are

presented in Table 15.7. API has seven piping classes, which are listed in Table

15.8. API classes are rated at 100°F and are reduced 1.8% per 50°F increase in temperature to a maximum allowable temperature of 450°F. The API metallurgy and testing requirements are more strict than those for ANSI. As a result, although API 2,000 to 5,000 classes have the same dimensions as some of the ANSI classes, the API classes are rated for higher pressures.

Determination of Pressure Breaks. When piping rated for a certain pressure meets piping rated for a lower pressure, a “piping break” occurs. On one side of this break, a higher pressure is possible, while on the other side a relief valve protects the pressure from exceeding a lower pressure regardless of the manipulation of in-

dividual valves. In determining the location of a pressure break in a piping system, the following assumptions are made.

1. Check valves. will leak, allowing communication back to the upstream side of a high downstream pressure resulting from an upset condition.

2. Control valves will leak, allowing pressure to be equal on both sides if flow is stopped.

3. Shut-in sensors cannot be relied on to keep pressure from building up, unless they are installed as two completely independent systems with two completely nzdun- dant shut-in valves.

4. All block valves must be assumed to be either opened or closed in such a manner as to exert maximum pressure on the pipe.

In determining pressure breaks for a complex facility, it is necessary to have a complete mechanical flow sheet that shows all piping and valves. Potential problems can develop when the wrong valve or valve combination is turned at the wrong time. The only way to protect against this contingency is to be able to see the potential paths of communication from one system to another on a mechanical flow sheet.

General Piping Design Considerations

Design Flow Rates and Pressures. The design flow rates and pressure given by the reservoir studies must be used with some judgment by the facilities design engineer. Many unknowns enter the simulations and assumptions used in arriving at these numbers.

It is the designers’ responsibility to balance the higher cost of providing additional capability to handle greater- than-expected flow rates and pressures with the risk- discounted cost of having to make these adjustments at some future date. This development of a “comfort level” for design is extremely complex and must take in- to account the availability of funds, the length of the project, the cost associated with having a surface facility limitation on potential flow rates, etc.

Care should be taken to include flow surges in the piping design. This is particularly true for gas-lift source wells where instantaneous surges of as much as 50% can occur. However, it is also true for all other piping components where instantaneous rates may exceed average daily rates because of changes in operating conditions, action of pressure control and level control valves, pigging operations, etc.

TABLE 15.6-CLASS 150 PRESSURE (psig)-TEMPERATURE (OF) RATINGS FOR VARIOUS MATERIAL GROUPS


TABLE 15.7-ANSI PRESSURE RATINGS FOR MATERIAL GROUP

1.1. TABLE 15.8-API PRESSURE RATINGS

Allowable Pressure (psig)

Class - 20 lo 1 OO°F 101 to ZOOoF Class

150 285 260 2,000 300 740 675 3,000 400 990 900 5,000 600 1,480 1,350 10,000 900 2,220 2,025 15,000

1,500 3,705 3,375 20,000 2,500 6,170 5,625 30,000

Allowable Pressure ANSI Dimensional Equivalent

at lOOoF Rating at lOOoF (Psi) Class (Psi)

2,000 600 1.480 3,000 900 2;220 5,000 1,500 3,705

10,000 - - 15,000 - - 20,000 - - 30,000 - -

Looped Networks. In large gathering and distribution systems, the potential savings associated with looping the system (i.e., installing pipelines parallel to existing lines to increase the system’s capacity) or installing pumps at different locations must be investigated. Although simple loops may be calculated by hand, the ready availability of many fine looped-network computer models make this a fairly easy choice. In most instances, some simple calculations and assumptions can be made that narrow the choices to a few practical alternatives. An experienced engineer can use these techniques to greatly reduce computer time and costs.

Gravity Systems. In gravity-systems design, careful consideration must be given to pressure drops in the pipe. Valves and fittings must be considered in these calculations since there is normally little room for error. It is absolutely necessary that accurate elevations of all tanks and equipment and an accurate profile be determined along the line of the pipe. The hydraulic gradient must be plotted along this profile for worst-case conditions of working levels in the tanks or operating pressure in vessels. The hydraulic gradient should be higher than the pipe at each point to ensure that a syphon is not developed. Fig. 15.7 shows an example profile.

High-point vents should be installed in the line to keep gas from accumulating and potentially blocking the flow. Gas eliminators such as those used in lease automatic custody transfer (LACT) units can be installed for this purpose.

Pigging. On long lines where paraffin, scale, or solids may be deposited, periodic pigging of the line may be required. A pig is a sphere or cylinder, often containing scrapers, which is injected into the line at the beginning in a “pig launcher” and collected in a “pig trap” at the end.

Launchers and traps must be installed wherever the pipeline changes size and at all junction points. Where pigging is expected, care must be exercised in selecting valves and radius of curvature of the pipe to allow the pig to move through the line. Fig. 15.8 shows a typical pig trap which can be used to remove a pig from a line without having to shut in flow.

Selecting Pumps and Drivers

space, efficiency, flexibility to varying throughput and pressure conditions, and type of prime mover. It is beyond the scope of this chapter to discuss any of these in detail. Instead, a broad description of commonly used pump types and their characteristics, types of drivers and their characteristics, and some comments as to pump piping and installation are presented.

Pump Types

There are many types of pumps in use. However, most of the common ones can be classified as either positive displacement or centrifugal by the action they employ to move the liquid to a higher pressure level.

Positive-Displacement Pumps. Positive-displacement pumps employ a moving piston, plunger, diaphragm, or rotor to move a fixed volume of liquid per revolution of the pump. The amount of liquid pumped per revolution is independent of the speed of the pump or the discharge pressure.

Reciprocating pumps are positive-displacement pumps that operate as a result of the movement of a piston or plunger inside a cylinder. Piston pumps can be double-acting in that the fluid could be forced out of the cylinder into the discharge piping ahead of the piston and liquid drawn into the cylinder behind the piston regardless of the direction of the piston travel. If liquid is pumped during a piston movement in one direction only, the pump is classified as single-acting. Pumps with two cylinders are called “duplex,” three cylinders “triplex,” etc.

Advantages of reciprocating pumps are (1) for a given speed, the rate of discharge is practically constant, regardless of head, and the pump is limited only by the power of the prime mover and the strength of the pump parts; (2) efficiency is high regardlesss of the head and speed; (3) owing to low operating speed and the low velocities of fluids, they are well adapted to handling viscous fluids; (4) they are usually self-priming.

Disadvantages of reciprocating pumps include (1) heavy weight and large physical size; (2) valve trouble can occur, especially when pumping liquid containing solids; (3) pulsating flow in both suction and discharge lines; (4) high net positive suction head requirements; (5) not generally suitable for handling liquids containing solids, abrasives, or dirt.

Pumps are used in all water gathering, treating, and Rotary pumps are positive-displacement pumps that disposal systems. Indeed, the most costly single piece of operate by having a rotating member turn inside a hous- equipment in a water-injection system is often the injec- ing in such a way that it creates one or more cavities that tion pump and driver, The factors to consider in the move from suction to discharge forcing the trapped liq- selection of pumps include capacity, head, suction lift, uid through the pump.

SURFACE FACILITIES FOR WATERFLOODING & SALTWATER DISPOSAL

PROFlL E- MAIN LINE

Fig. 15.7-Profile of the main line in a gravity system.

Advantages of rotary pumps are (1) in general, these are the same as for reciprocating pumps; (2) rotary pumps are relatively inexpensive and require small space; (3) they will operate over wide ranges of capacity, head, and viscosity; (4) rotaries are self-priming and are good vapor-handlers; (5) they deliver relatively pulsation-free flow.

Disadvantages of rotary pumps are (1) close clearances and/or rubbing contacts restrict the choice of materials for construction; (2) close clearances require that liquids to be pumped have lubricating value and be noncorrosive-therefore, they are suitable for oil but not suited for water; (3) rotaries have low volumetric efficiency at low speeds because the slip approaches the displacement. This effect increases directly with the pressure/viscosity ratio.

The diaphragm pump is a type of reciprocating positive-displacement pump that operates by the action of a diaphragm moving back and forth within a fixed chamber. Raising the diaphragm creates a vacuum, drawing liquid (or air) into the pump through the suction check valve. Lowering the diaphragm forces the liquid (or air) out through the discharge check valve. This type of pump will handle clear water or water containing large quantities of mud, sand, sludge, and trash. Its popularity for low-volume applications stems from its ability to operate where the quantity of water varies considerably so that much of the time air is being pumped. The suction effect of the diaphragm motion makes the pump self-priming. For high discharge pressure requirements, diaphragm pumps are limited to very low fluid rates. Although they tend to be easy to repair in the field, the frequency of maintenance required is higher than with other pump types.

Centrifugal Pumps. A centrifugal pump contains a central rotating wheel, called an impeller, which imparts high velocity to the liquid by centrifugal force and then

converts most of this velocity to pressure. The liquid flows from the pump even against considerable discharge pipe pressure. By its very nature, the centrifugal pump operates at relatively high rotative speeds. It is the most common type of pump used today.

Centrifugal pumps can be of radial-flow construction, axial-flow construction, or some combination of the two. In radial-flow pumps, flow enters the center of the wheel and is propelled radial to the outside. Radial construction provides maximum head per stage.

Axial flow pumps develop their head by the propelling or lifting force developed in the fluid by the impeller vanes, which, in cross section, are shaped like airfoils. The flow is parallel to the pump shaft axis. The diameter of the impeller is the same at the suction and discharge sides. Velocity energy is converted to pressure by stationary diffuser vanes.

Advantages of centrifugal pumps are (1) simple construction, quiet operation; (2) inexpensive; (3) small space requirement in relation to capacities; (4) no close clearances, therefore, it can handle liquids containing dirt, abrasives, large solids, etc.: (5) low maintenance, dependable; (6) low net positive suction head requirements; (7) capacity adjusts automatically to changes in head. Thus, capacity may be controlled over a wide range at constant speed.

Disadvantages of centrifugal pumps are (1) cannot achieve high pressures like reciprocating pumps; (2) viscosity effects on head, capacity, and efficiency are appreciable at 200 Saybolt Universal Seconds (S.S.U.) and serious at 500 S.S.U.; (3) low efficiencies when compared to reciprocating pumps; (4) efficiency is a function of flow rate. At throughput rates and pressures less than design, considerable additional horsepower may be required.

Pump Drivers

Depending on the location, type of pump, availability, and cost of natural gas for fuel, pump drivers will be gas


Fig. 15.8—Pig trap.

engines, gas turbines, or electric motors. The amount ofhorsepower required for a given installation can becalculated from

qLApP&‘-z 9LZYL

1714E,3960E, , . . . . . . . . . . . . . . . (17)

wherePbh = brake horsepower,qL = flow of liquid, gal/mitt,AP = differential pressure, psi,Ep = pump efficiency at flow conditions, %

Z = head of liquid, ft, and *ye = specific gravity of liquid relative to water.

Bump efficiencies of between 80 and 90% forreciprocating pumps and 55 and 65% for centrifugalpumps are common.

Natural Gas Engines. The reciprocating internal-combustion natural-gas engine is the leading primemover in the oil field because of its high efficiency,availability, and ease of maintenance. Both two- andfour-cycle engines are in use.

In general, for a given engine, the horsepower outputdepends on engine speed in rev/min and whether anexhaust-gas turbocharger is installed to increase the flow

of air to the power cylinders. At higher operating speed,the engine is capable of producing greater horsepower. Aturbocharged engine will be able to develop morehorsepower than a naturally aspirated engine. Unfor-tunately, maintenance costs and downtime increase as anengine is accelerated above a certain limit and when tur-bochargers are added. A turbocharged engine has theadded disadvantage of being able to load itself tomechanical destruction.

Most engine drivers are built to operate in the 900- to1,400-rev/min range. For sustained operations, mostmechanics do not like to operate their engines above1,000 to 1,200 rpm. Turbocharged engines can be ex-pected to use 7,000 to 8,000 Btu/bhp-hr of fuel whilenaturally aspirated units will use 8,000 to 10,000Btu/bhp-hr. Fuel efficiency is fairly constant over largeranges in bhp.

Care must be exercised with nitric and nitrous oxide(NO,) emissions on large installations (over 2,000 hp)or in nonattainment areas. Catalytic converters areavailable for most engines if this becomes a problem.

Gas Turbines. Gas turbine engines have three basic sec-tions-an air-generation section, a combustion section,and a power-turbine section. In the compressor, or air-generator section, ambient air is drawn into the turbine


and compressed with a combination of radial and axial flow elements for delivery to the combustion section. Fuel is mixed with the air in the combustion section and the combustion products am mixed with additional air to provide a specific temperature at the power turbine inlet. The hot gas expands across the power turbine providing power to drive the air generator and the load.

Advantages of turbines over gas engines include the following.

1. Gas turbines can be made very light and compact in relation to horsepower (‘/4 to ‘/2 Ibm/hp) in jet aircraft. The weight/horsepower ratio varies from 1 to 12 lbm/hp for industrial turbines. High-speed turbocharged gas engines weigh from 15 to 40 lbm/hp.

2. Turbine maintenance cost is normally lower and its availability higher. This is true, however, only for turbines in continuous service. Starting and stopping units has a severe effect on maintenance costs.

3. Turbines reject large quantities of high-temperature heat in their exhaust, which can be used to provide process heat.

4. Turbines are available in larger sizes (100,000 hp and higher).

5. Turbines are clean burning from an air pollution standpoint and do not require catalytic converters.

Disadvantages of turbines are (1) high fuel consumption if waste heat is not needed (9,000 to 11,000 Btu/hp- hr at peak efficiency) and (2) large falloff in fuel efficiency when operating at less than peak load.

Electric Motors. In areas where electricity is available from either commercial sources or onsite generators, electric-motor drives are the least expensive in initial capital and maintenance costs. Their use is recommended where the additional cost of purchasing or generating electricity is not too great.

Pump Piping and Installation Details

Suction Piping. It is essential that a flooded suction be furnished for reciprocating pumps. A pump should never be allowed to run “dry” or “starved.” The net positive suction head (NPSH) recommended by the manufacturer is shown on the performance curves for the pump and must be provided. To furnish this head and ensure a flooded suction at all times, it is necessary that (1) the storage tank or basin supplying the pump be set at a sufficient elevation above the fluid end of the pump and (2) that the suction piping be of sufficient size to minimize friction losses in the pipe between the storage tank or basin and the pump. In cases where it is possible to secure sufficient elevation head between the storage tank and pump, a centrifugal pump, commonly called a “booster pump,” is employed. Normally the booster pump is tied into the storage tank and delivers the water in sufficient quantity to the suction header of the reciprocating pumps to furnish the flooded suction and provide the required NPSH.

The recommended NPSH curve, supplied by manufacturers of reciprocating and centrifugal pumps, is the NPSH that results in a 3% drop in capacity. Cavitation usually starts at a higher NPSH. In cases where no cavitation damage can be tolerated, the NPSH required for no loss in capacity should be used for designing suction and charge systems.

The following features relative to suction-piping installations should be provided.

1. Suction pipe should be as large as or, preferably, larger than the pump suction-inlet size. Table 15.9 indicates acceptable flow velocities in suction lines.

2. Long-radius elbows are recommended to eliminate sharp turns.

3. Suction lines should be laid to a constant grade from the storage tank to pump to eliminate high points where vapor may accumulate.

4. For reciprocating pumps, a pulsation dampener should be installed. These can be elastomer diaphragm or acoustic in design. One common alternative is to install an air- or gas-volume chamber. The chamber allows the pump to fill properly by relieving excessive accelera- tion and deceleration of the fluid with each stroke of the pump. The required size of an air-volume chamber and its air space depends on the type of pump, displacement per revolution, and speed of the pump. The air-volume chamber may vary from two to eight times the piston displacement of a single stroke. A sight glass is also installed for gauging the liquid level in the chamber so that gas or air can be added periodically to replace what is absorbed by the liquid being pumped.

5. The suction piping should be flushed out and cleaned prior to starting the pump to remove slag, mill scale, rust, welding splatter, etc.

Discharge Piping. As in the suction piping, the discharge piping should be well planned with a minimum of turns, fittings, restrictions, etc. The discharge piping should be of sufficient size to minimize friction losses in the pipe to furnish the required pressure of the pump discharge. Other factors to be considered in discharge piping include the following.

1. A pressure-relief valve must be employed on positive-displacement pumps in the discharge line ahead of any other valve or restriction.

2. A pulsation damper, or desurger, should be installed in the discharge line near a reciprocating pump to relieve shock or vibratory forces. These forces are the result of pressure variations, or surges, prevalent in positive-displacement pump operation. They also result from water hammer because of valve closures and restrictions in the line.

The design and theory of pulsation dampers, or desurgers, are based on the concept that a constant pressure can be maintained if the liquid can be accumulated as the pressure increases and discharged as the pressure decreases. Several types of dampers, or

TABLE 15.9-TYPICAL FLOW VELOCITIES (ftlsec)

Suction Discharge velocity velocity

Reciprocating pumps Speeds up to 250

revtmin 2 6 Speeds 251 to 330

revlmin I’/2 4% Speeds above 330

revlmin 1 3 Centrifugal pumps 2 to 3 6 to 9


desurgers, are available. Operational features should be investigated and a damper, or desurger, selected to tit each condition.

3. On installations where the fluid is pumped to a level higher than the pump outlet, a gate valve or check valve should be placed downstream of the pressure-relief valve near the pump in the discharge line. This permits shutoff of fluid backing through the pump while pump valves or plungers are being serviced.

The general practice is to connect pumps in parallel through a common suction and discharge header on systems where two or more pumps are installed. A better but somewhat more costly practice is to use separate suction lines for each pump. Headers should be sized to ensure sufficient capacity for satisfactory operation. Each reciprocating pump should have its own separate pressure-relief valve to ensure protection regardless of which pump may be shut down or out of service.

Foundations. For adequate support to maintain alignment and to reduce vibration, a foundation of sufficient size and strength should be provided. Steel-reinforced- concrete foundations are ordinarily used for rotating equipment and are recommended for onshore locations. The pump and prime mover must be set level to ensure good operation. To accomplish this, the pump and mover are mounted on a skid base; the skid base is set on the concrete foundation, leveled, and approximately 1 in. of grout is used to set the equipment.

The required dimensions of the foundation will vary relative to the size and shape of the equipment. The depth to which the foundation must be carried depends on the soil conditions where the foundation is to be set.

The foundation should have sufficient bearing area, which is calculated by using the allowable soil-bearing capacity. Where unbalanced forces are unknown, a rule of thumb that the weight of the concrete-mass foundation should be from 1.5 to 2 times the weight of the reciprocating equipment is sometimes used. Under certain combinations of poor soil and large unbalanced forces, the foundation weight may have to be as much as four times the weight of the equipment.

The size and shape of the skid base will determine the plan dimensions of the foundation and, with acceptable soil conditions, the depth of the foundation is carried to a point that will furnish the required mass. Temperature stresses in normal concrete foundations for pumps and engines govern the amount of steel reinforcing used. The amount of such steel can be estimated by: area of steel (sq in/sq ft of slab area) =0.002 Xcross-sectional area (sq in. of concrete slab). Maximum spacing between bars should not exceed 18 in. Where unbalanced forces and couples are known and for large offshore installations where weight is important, a more detailed dynamic calculation should be made.

Gravity Settling

Solid particles, because of their heavier density than water and net negative buoyant force, will settle to the bottom with a terminal velocity that can be derived fmm Stoke’s law as

A&)’ “E- 18~L, , . . . . . . . . . . . . . . . . . . . . . . . (18)

where Ap = difference in density of the particle and the

liquid, dp = solid particle diameter, and pL = viscosity of the liquid.

This equation can also be expressed as

v= l.78x10-6Ay(d,)2

, . . . . . . . . . . . . . . . . . . (19) Pw

where dp = particle diameter, microns, Ay = difference in specific gravity relative to

water,

Pw = viscosity of water, cp, and v = velocity, ft/sec.

This equation can be used to size a tank, vertical pressure vessel, horizontal pressure vessel, rectangular sedimentation chamber, or a device of any other configuration to allow a particle of a certain diameter and specific gravity to settle to the bottom.

Legend L, = Inlet dlstrlbubon section L- = Effective settlina action L, = Outlel gathering section

““, = Water velocity

“3 = Settling velocity 1, = Time water in effectwe sectnon of flume 1, = Time pan!cle is falling while 1” effective sectmn of flume

Zf 1s =-

“S

Le fw=-

“W

Separating Suspended Solids From Heater Le”S Yw=-

The water that is being treated may have suspended solids such as produced sand, rust, and scales. These can be separated from the water stream by gravity settling, cyclone desanders, centrifuges, filters, or flotation.

Z f

Fig. 15.9-Model for calculation of sedimentation flume capacity.


where V, is the velocity of water, ftisec, and hf is the height of flume, ft. For practical considerations, b should be between 6 and 20 ft and ratio of hf to b should be between 0.3 and 0.5.

The flume can be concrete lined or constructed as a soil pit. Solids that settle in the bottom of the flume can be cleaned out with a bucket. A mechanical sludge scraper run on a chain could be installed to concentrate the solids in one location for easy removal.

Most sedimentation basins are rectangular flumes with length-to-width ratios of 4: 1 or greater to limit crossflow. The width of the flow channel can be determined by setting the time required for a particle to settle from the top of the flume to the bottom equal to the time required for the water to traverse from the inlet of the flume to the outlet as shown in Fig. 15.9. This can be expressed as

b= 369,,+4w

*y(d,)2L,, . . . . . . . . . . . . . . . . . . . . . . . . (20)

where b = width (breadth) of flow channel, ft,

4w = water flow rate, B/D, and L, = effective length, ft.

Note that the width and length of the settling chamber are independent of its de posal of Refinery Wastes IP

th. The API Manual on Dis- recommends a turbulence and

short-circuiting factor of between 1.3 and 1.8, depending on the ratio of water velocity to solids settling velocity. Using a factor of 1.8. Eq. 20 can be rewritten as

b= 6-%wpL,

ay(d,)‘~,’ . . . . . . . . . . . . . . . . . . ..I.. .(21)

API also recommends that water velocity be limited to 15 times the settling velocity, or 3 ft/min, whichever is less. The settling velocity can be calculated from Eq. 19 and the water velocity can be calculated from

v,=6.5x10-5~, _. . . . . . . . . . . .(22) f

Cyclone Desanders

Cyclone desanders are conical-shaped devices that make use of centrifugal force to separate the heavier particles from the liquid. Fig. 15.10 shows the basic operation of a cyclone desander. Pressurized fluid enters a common inlet manifold, which distributes the stream to individual cyclones. Flow proceeds through a tangential feed inlet, which directs the fluid against the wall of a cylindrical section above a truncated cone. The fluid and solid particles move downward in a spiral pattern forcing the heavy particles to move toward the outer perimeter of the cone. Gravity forces these particles to slide downward and to be rejected at the cone apex and carried away in the “underflow” slurry. The water moves toward the vacuum created at the center of the cone, and is drawn off as the “overflow. ”

The size particle that is separated depends on the pressure drop through the cone. The pressure drop, in turn, depends on the flow rate. Thus, there is a minimum flow and a pressure drop that must be provided for each cone to settle a certain size particle. With pressure drops in the range of 25 to 50 psi, cyclones can be used to remove 99% of the 30- to lOO-micron particles.

overflow p,pe

vortex finder k?- 0

-fee0 front view

I conic01 section

i/l apex slda view

i underflow

Fig. 15.10-Hydrocyclone operation.


Fig. 15.11 -Cartridge filter

Centrifuges

Centrifuges are used on drilling rigs to separate low- gravity drill solids and reclaim high percentages of the heavy solids. They have not found wide use in producing operations because of the high maintenance associated with their use. Normally, if it is desirable to separate particles of diameter less than that resulting from sedimentation or cyclones, filters are used.

TABLE lC.lO-CARTRIDGE DESIGN CONDITIONS FOR BRINE

Particle Size

Flow Rate for 3-h. OD x 3&n.

Element Cartridge Type

Pleated wire screen

(microns)

80

(gallmin)

12

Pleated cellulose paper 5to10 4

Rolled laminated cotton and acrylic excelsior filler 5to10 2to 3

Molded fiberglass 2 6

Filtration To avoid plugging the injection formation it may be necessary to separate small-diameter suspended particles by filtration. Filters cannot handle the volume of solids that can be handled by sedimentation and cyclones. By proper choice of filter element, filters can remove fine solids in the 0.5- to 50-micron range and are used as a form of secondary treatment.

Cartridge Filters. Cartridge filters are the simplest to install, require no backwash, and are capable of removing solid particles of %-micron or larger diameter. Their drawback is that they can take only very low solid loadings. The filter vessel must be taken out of service, depressurcd, and the cartridges replaced whenever the volume of solids trapped causes the differential pressure to exceed a predetermined maximum (usually 25 psi). Some modem cartridge filters can be backwashed.

Fig. 15.11 shows a typical cartridge filter. The cylindrical filters arc encased in a pressure vessel. Flow enters the vessel and flows from the outside of the cartridges to the center, where it enters a perforated pipe that is open on the bottom. A bypass mechanism is included that will automatically allow flow to pass from the inlet to the outlet chambers if the differential pressure exceeds the capacity of the cartridges.

Table 15.10 indicates the particle size that can be separated and the recommended flow rate through various standard-size cartridges. The molded fiberglass has the least solid storage area and the pleated wire screen and pleated paper the most.

Sand Filters. Sand filters have beds of graded sand, gravel, anthracite, or graphite. The beds may be of a single medium or may be graded from coarse to tine media to allow for greater solids loading.

The media are arranged in a pressure vessel for either downflow filtration and upflow backwash as shown in Fig. 15.12 or for upflow filtration and upflow backwash. Conventional downflow filters arc limited to flow rates of 2 to 5 gal/(min-sq ft) and total solids loads (before backwashing) of % to 1% lbmlsq ft. With appropriately designed distribution systems, high-rate filters can be operated at 7 to 15 gal/(min-sq ft). This higher loading forces the solids farther into the bed allowing for solid loadings of between 1 and 4 lbmisq ft.

Upflow filters have a greater capacity for solids loading. Flow tends to loosen the bed allowing for greater penetration of the solids. This allows up to 6 lbm/sq ft of solid loading. However, because of the danger of losing the bed, upflow filters are limited to flow rates of 6 to 8 gal/(min-sq ft). They also require longer backwashing time and more backwash fluid.

Walnut hulls are used as filter media in some new filters. There is a system that removes, cleans, and returns this medium to the filter automatically.

Sand filters are good for separating 25-micron particles. Some manufacturers claim their filters are good for 5- to lo-micron separation.

Diatomaceous Earth (DE) Filters. For filtration of O.S- to l.O-micron particles, DE filters arc used. Typically, these have low solids-loading capabilities [‘/z to 1 gal/(min-sq ft)]. A typical DE filter is shown in Fig. 15.13.


The individual leaves, which are spaced at approximately 3-in. intervals, are made up of a wire screen of corrosion-resistant materials such as stainless steel, Monel@, or Inconel@. A precoat slurry of DE is mixed and flowed through the filter to provide an approximately x6-in.-thick coating to the leaves. Flow then is turned into the filter with a “body feed” of “filter aid” (DE and cellulose fiber or perlite) equal on a weight basis to the amount of solids to be filtered. The body feed helps to build an even, permeable filter cake on the leaves.

When a pressure differential of 25 to 35 psi is reached, the unit must be taken out of service and the filter cake removed. This can be done by backwashing alone, backwashing and vibrating, or backwashing and water sluicing the cake. Water with good filtering characteristics may build up a large permeable filter cake that requires backwashing before a large pressure differential is developed.

Flotation

It is possible to remove small particles by use of dispersed or dissolved-gas flotation devices. These units are used primarily for treating hydrocarbons from water and are discussed in that section. Normally gas is dispersed into the water or released from solution in the water and forms bubbles approximately 30 to 120 microns in diameter. The bubbles form on the surfaces of the suspended particles and create a particle whose average density is less than that of water. These rise to the surface and are mechanically skimmed. Because of the difficulty of predicting particle removal efficiencies with the method, it is normally not used in oil field land operations. However, it is being used increasingly in offshore operations and in some underground disposal systems.

Treating Hydrocarbons From Water Produced water, which may have to be treated, will enter the water treating plant from a three-phase separator, free water knockout, gun barrel, heater treater, or other lease equipment, which arc discussed in previous chapters. This water will contain small amounts (100 to 2,000 mg/L) of suspended hydrocarbons in oil droplets. Since the water flows out of these pieces of equipment through dump valves or pumps, the oil particle diameters will be very small.

Theory

Treating equipment to handle this problem relies on one or more of the following principles: gravity separation of the lighter oil droplets from the water, coalescence of the smaller oil droplets, or gas flotation of the oil droplets. In applying these concepts, one must keep in mind the dispersion of large oil droplets to smaller droplets, which will take place if energy is added to the system.

Gravity Separation. As in the case of settling of sand from water, Stoke’s Law, Eqs. 18 and 19, holds true for the buoyant rise velocity of an oil droplet in a water- continuous phase. Several immediate conclusions can be drawn from this equation.

1. The larger the size of an oil droplet, the greater its vertical velocity. That is, the bigger the droplet size, the

BACKWASH OUTLET

CLEAN WATER OUTLET I

RAW WATER INLET

#

BACK’4 INLET

‘ASH

Fig. 15.12-Sand filter

less time it takes for the droplet to rise to a collection surface and thus the easier it is to treat the water.

2. The greater the difference in density between the oil droplet and the water phase, the greater the vertical velocity. That is, the lighter the crude, the easier it is to treat the water.

3. The higher the temperature, the lower the viscosity of the water, and thus, the greater the vertical velocity. That is, it is easier to treat the water at high temperatures than at low temperatures.

The last conclusion requires some further elaboration. Heat is the primary mechanism used in treating small water droplets from oil in oil-treating equipment because of the effect heat has on the oil viscosity, which prompts more rapid settling, and because of the effect heat has on the emulsifier stabilizing the water-in-oil emulsion. Heat is not commonly used in water treating because (1) the percentage change in viscosity per degree of temperature change is much greater in oil than in water, (2) water-inoil emulsions tend to have a higher percent of the dispersed phase than oil-in-water emulsions, and the dispersed phase tends to have larger-diameter droplets stabilized by heat-sensitive emulsifiers, and (3) it takes twice as much heat input to raise a barrel of water as it takes to raise a barrel of oil to the same temperature.


OS TYPE SEAL ARRANGEMENT

Fig. 15.13-Typical diatomaceous earth filter.

TYPE ff LEAVES SEAL TYPE 1 !SHELL MA r I MUMEGAI- IMDTCftHPl sLuoGEo.lFt I lwrf wEwrl FILTER ARCAl 1 ND LEAVES 1 TECHNICAL DATA

Dispersion. The small oil droplets contained in the water-continuous phase are subject to continuous dispersion and coalescence. An oscillating droplet of oil will become unstable when the kinetic energy is sufficient to make up for the difference in the surface energy between the single droplet and the two smaller droplets formed from it. At the same time that this process is occurring, the motion of the smaller oil particles is causing coalescence to take place. Therefore, it should be possible to define statistically a maximum droplet size for a given energy input per unit mass and time at which the rate of coalescence equals the rate of dispersion.

The size of the oil droplets that will exist is a function of one over the amount of work done on the liquid per unit mass per unit time. This can be shown to be a function of the pressure drop as

&i)max = -k,, . . . . . . . . . . . . . . . . . . . . . (23) &

where

(hi) mm = maximum droplet diameter, t = time,

A, = pressure drop experienced by the liquid in time t and

Cd = dispersion constant.

The greater the pressure drop, and thus the shear forces that the fluid experiences in a given time period, the smaller the oil droplets will be. Large pressure drops.

which occur in small distances through chokes, control valves, cyclone desanders, etc., result in small oil droplets and water that is harder to treat. A pressure dmp of 50 to 75 psi will result in a maximum particle size of 10 to 50 microns.

The dispersion process is theoretically not instantaneous. However, it appears from field experience to take place very rapidly. For design purposes, it could be assumed that whenever large pressure drops occur, all droplets will disperse instantaneously. This is, of course, a conservative approximation.

Coalescence. Within water-treating equipment, where the energy input to the fluid is very small, the process of coalescence takes place. That is, small oil droplets collide and form bigger droplets. Because of the low energy input these are not, in turn, dispersed.

Coalescence can also occur in pipe downstream of pumps and control valves. However, in such instances, the process of dispersion will govern the maximum size of stable oil droplet that can exist. For normal pipe diameters and flow velocities, particles of 500 to 5,000 microns are possible.

The process of coalescence in water-treating systems appears to be more time dependent than the process of dispersion. When two oil dmplets collide, contact can be broken before coalescence is completed because of turbulent pressure fluctations and the kinetic energy of the oscillating droplets.

SURFACE FACILITIES FOR WATERFLOODING 8 SALTWATER DISPOSAL 15-23

The time it takes to “grow” a large droplet from a relatively small droplet, in a “quiet” gravity settler, is approximated by

(4b4 r=- .,,...........,...............

2f vK, ’ where

dd = droplet diameter, fv = volume fraction of the dispersed phase, and K, = empirical settling constant.

While it is impossible to determine K, for an actual installation, the following qualitative conclusions can be

drawn. 1. A doubling of residence time will cause an increase

in droplet size of only 19 % 2. The more dilute the dispersed phase, the greater the

residence time needed to grow a given particle size. That is, coalescence occurs more rapidly in concentrated dispersions.

Gravity Separation Devices

Most water-treating equipment makes use of gravity separation of the oil droplets. Included in this categoty are skim tanks, API separators, plate coalescers, and skim piles. Unfortunately, it is necessary to know both the oil concentration in the effluent water and the particle size distribution to design a gravity separator to meet a certain effluent quality.

This information can be determined from experience- derived relationships such as those shown in Fig. 15.14. Further work is needed to define these relationships. For the present, the design engineer must rely on a judgment factor or on laboratory tests for the particular crude oil and water.

Skim Tanks and Vessels. The simplest form of treatment equipment is a skim tank or pressure vessel. These are normally designed to provide large residence times during which coalescence and gravity separation can occur. They can be either pressure vessels or atmospheric tanks.

Skim tanks can be either vertical or horizontal in configuration. They may be set up for vertical downward flow of the water with or without inlet spreaders or outlet collectors. They may also be designed as horizontal vessels where the water enters on one side and flows over a weir on the far end.

In vertical vessels the oil droplets must flow upward against the downward velocity of the water. For this reason, horizontal vessels are more efficient in gravity separation of the two liquid phases. In spite of this, vertical vessels and tanks are sometimes used for two reasons.

1. Sand and other solid particles can be more easily handled in vertical vessels with either the water outlet or a sand drain off the bottom. Experience with elaborately designed sand drains in the large horizontal vessels has not been very satisfactory.

2. Vertical vessels are less susceptible to high-level shut-downs because of liquid surges. Internal waves resulting from surging in horizontal vessels can trigger a level float even though the volume of liquid between the

DROPLET SIZE

Fig. 15.14-Form of empirical curves for oil concentration and droplet size distribution.

normal operating level and the high-level shut-down is equal to or larger than that in a vertical vessel.

Tracer studies have shown that large skim tanks, even those with carefully designed spreaders and baffles, exhibit poor flow behavior and short circuiting. This is probably because of density and temperature differences, deposition of solids, corrosion of spreaders, etc. In one case, a tank with a design mean residence time of 33 hours had a breakthrough of the tracer with a peak within minutes of tracer injection.


As seen previously, the provision of residence time to allow for coalescence does not appear to be cost efficient. However, a minimum residence time of 10 minutes to one hour should be provided to ensure that surges do not upset the system and to provide for some coalescence.

Horizontal Pressure Vessel Si&zg. The required diameter and length of a horizontal cylinder operating one-half full of water can be determined by use of a model similar to that used for settling solids. The following equation can be derived from the model.

diL,= l~q,cL, *y,,(dd)2 . . . . . . . . . . . (25)

where di = vessel ID, in.,

qw = water flow rate, BWPD,

CL, = water viscosity, cp, dd = oil droplet diameter, micron, L, = effective length in which separation occurs,

ft (for design use 75% of the seam-to- seam length), and

AYOW = difference in specific gravity between oil and water.

While Eq. 25 will govern the design, it is also necessary to check for adequate retention time.

t,=0.7(di)2Le . . . . . . . . . . . . . . . . . . . . . (26)

where t, is the retention time in minutes. Vertical Cylindrical Vessel. The required diameter of

a vertical cylindrical pressure vessel or tank can be determined from

(d;)~=7,oooF qwpw AYow(dd)2 , . . . . . . . ..(27)

where F is a factor to account for turbulence and short circuiting. For small-diameter vessels (48 in. or less), F= 1 .O. For larger diameters, F depends on the type of inlet and outlet spreaders, collectors, and baffles that are provided. Large tanks (10 ft or more in diameter) should be considered to have an F> 2.0, depending on the inlet and outlet conditions. Tanks greater than 10 ft in diameter should be discouraged because of short circuiting.

The height of the water column can be determined from retention time requirements:

rrq, z,=o.7- (di)2 , . . . . . . . (28)

where Z, is the height of the water column in feet.

API Separators. An API separator is the name given to a horizontal, rectangular cross-section, atmospheric oil skimmer that follows the sizing equations and guidelines included in the API Manual on Disposal of Re@ery Wuszes. l5 Fig. 15.15 shows a typical API separator. The equations for sizing and their derivations have been discussed previously in the solids settling section.

Plate Coalescers. The use of flow-through parallel plates to help the gravity separation in skim tanks was pioneered in 1959 l6 as a method of converting existing API separators to treatment of droplets less than 150 microns in diameter.

Various configurations of plate coalescers have been devised. These are commonly called parallel plate interceptors (PPI), corrugated plate interceptors (CPU, or crossflow separators. All of these depend on gravity separation to allow the oil droplets to rise to a plate surface where coalescence and capture occur. Flow is split between a number of parallel plates spaced a short distance apart. To facilitate capture of the oil particles, the plates are inclined to the horizontal.

As shown in Fig. 15.16, an oil droplet entering the space between the plates will rise in accordance with Eq. 19. At the same time, it will have a forward velocity equal to the bulk water velocity. By solving for the vertical velocity needed by a particle entering at the base of the flow to reach the coalescing plate at the top of the flow, the resulting diameter can be determined. A restriction is placed on the Reynold’s number for the water to ensure that turbulence in this flow does not affect the oil sheet on the coalescing plate.

General Sizing Equation. For a plate coalescer with flow either parallel to or perpendicular to the direction of flow the general sizing equation for the droplet size removal is:

@A2 = 4.%wL,Pw

c*s 8 zpb,LAy,, ) . .

where dd = design oil droplet diameter, micron,

4w = bulk water flow rate, BWPD, L, = perpendicular distance between plates, in.

P !a = viscosity of the water, cp, 0 = angle of the plate with the horizontal,

Zp = height of the plate section perpendicular to the axis of water flow, ft,

b, = width of the plate section perpendicular to the axis of water flow, ft.

L = length of plate section parallel to the axis of water flow, ft, and

AY DW = difference in specific gravity between oil and water.

Experiments have indicated that Reynold’s number for the flow regime cannot exceed 400, on the basis of the hydraulic radius as the characteristic dimension. Thus the maximum flow rate is given by

(qw)max= 1562Z,b,

. . . . . . . . I.. . . I.. L P

(30)


J ‘I ii

/ ,L”DU SECTION A -A r(OPPL RS

Fig. 15.15-API oil-water separator.

Parallel Plate Interceptor (PPZ). The first form of plate coalescer was the PPI. This involved installing a series of plates parallel to the longitudinal axis of an API separator (a horizontal, rectangular cross-sectioned skimmer). The plates form a “V” when viewed perpendicular to the axis of flow so that the oil sheet migrates up the underside of the coalescing plate and to the sides. Sediments migrate toward the middle and down to the bottom of the separator, where they are removed.

Corrugated Plate Interceptor (CPZ). The most common form of parallel plate interceptor used offshore is the CPI. This is a refinement of the PPI in that it takes up less platform space (length) for the same particle size removal, and has the added benefit of making sediment handling easier. Fig. 15.17 is a typical design using a corrugated plate.

In CPI’s, the parallel plates are corrugated (like roof- ing material) with the axis of the corrugations inclined to an angle of 45”. The bulk water flow is forced downward. The oil sheet rises upward counter to the water flow and is concentrated in the top of each cor- rugation. When the oil reaches the end of the plate pack it is collected in a vertical channel and brought to the oil/water interface. CPI’s require frequent cleaning of the plate packs where large amounts of sediments are handled.

Crossjlow Devices. Recently manufacturers have

modified the CPI configuration for horizontal water flow perpendicular to the axis of the corrugations in the plates. This allows the plates to be put on a steeper angle to facilitate sediment removal and to enable the plate pack to be more conveniently packaged in a pressure vessel. The latter benefit may be required if gas blowby through an upstream dump valve could cause relief problems with an atmospheric tank.

LARGE DROPLETS RISE i? TO COLLECTION SURFACE 0 /

----OIL SHE

iilL DROPLET

Fig. 15.16-Plate coalescers.

FLO’.+

ET VELOSITY

15-26

Fig. l&17-CPI separator flow pattern.

-QUIESCENT ZONE

- FLOWING ZONE

-OIL RISERS

Fig, 15.18-Skim pile flow pattern.


Crossflow devices can be constructed in either horizontal or vertical pressure vessels. The horizontal vessels require less internal baffling as the ends of nearly every plate conduct the oil directly to the oil/water interface and the sediments to the sediment area below the water flow area. The vertical units, although requiring collection channels on one end to enable the oil to rise to the oil/water interface and on the other end to allow the sand to settle to the bottom, can be designed for more efficient sand removal. Crossflow separators are used where sand is considered a problem and it is not removed in the process upstream of the CPI.

Practical Limitations. Stoke’s law theory should apply to oil droplets as small in diameter as 1 to 10 microns. However, field experience indicates that 30 microns sets a reasonable lower limit on the droplet sizes that can be removed. Below this size small pressure fluctuations, platform vibration, etc., tend to impede the rise of the droplets to the coalescing surface. Thus, the practical limit for sizing plate coalescers is 30-micron removal.

Skim Pile. Skim piles are gravity water-treating devices that are used offshore. As shown in Fig. 15.18, flow through the multiple series of baffle plates creates quiescent zones that reduce the distance a given oil droplet must rise to be separated from the main flow.

Once in the quiescent zone, there is plenty of time for coalescence and gravity separation. The larger droplets then migrate up the underside of the baffle to an oil collection system. Skim piles are used extensively to treat deck drainage of washdown or rainwater that has been contaminated with oil. They have the added benefit of providing for some degree of sand cleaning. Sand tra- versing the length of a skim pile will abrade on the baffles and be water washed. This removes the free oil, which is then captured in a quiescent zone.

Skim Pile Sizing-Deck Drainage. Field experience has indicated that acceptable effluent is obtained with 20 minutes of retention time in the baffled section of the pile. Using this,

(&)*,&=14.3 (q,+O.356 &q,), . . . . . . . . .(31)

where Lbs = length of baffle section, ft, Ad = area of deck, sq ft, and qr = rainfall rate, in./hr.

Intermittent Flow. During periods of no flow, oil droplets rise to the area of the quiescent zone and become trapped and protected from being swept back in- to the flow stream when flow is resumed. The net effect of the baffles is to reduce this rise distance. Each time flow is stopped as the water traverses the baffled section more oil particles are trapped in the quiescent zone.

This phenomenon can be used when it is desired to use a skim pile downstream of a skim tank or CPI for further treating. By use of a snap acting water dump on the influent, intermittent flow is established in the pile.

If t is the time in seconds for the dump cycle,

NC= 41.7 (di)*Lbs

. . . . . . . . . . . . . . . . . . . . . . .(32) qwt


(b)

Fig. 15.19-(a) air flotation process; (b) circular flotation chamber details.

where N, is the number of cycles of no flow a particle sees as it traverses the baffle section.

If t, is the time the valve is closed, the removal efficiency on any cycle of a particular drop size is

E =4.3x10-5(A~,,)(d,)2~, rc . . . . . . . . . . . .

cl&i (33)

The overall removal efficiency of that particle size can then be determined by

E,=l-[l-E,IN,. . . . . . . . . . . . . . . . . . . . (34)

Gas Flotation Units

Flotation units are the only commonly used water- treatment equipment that does not rely on gravity separation of the oil droplets. In fact, the action of these units is independent of the oil droplet size. In gas flotation units, large quantities of small-diameter gas bubbles are injected into the water stream. The bubbles attach to the oil droplets suspended in the stream and cause them to rise to the water surface as a froth. Experimental results have shown that very small-diameter oil droplets in dilute suspension can be removed easily by flotation. High percentages of oil removal are achieved.

Two distinct types of flotation units have been used, which are distinguished by the method employed in producing the small gas bubbles needed to contact the water. These are dissolved-gas units and dispersed-gas units.

Dissolved-Gas Units. Dissolved-gas designs take a portion of the treated water effluent and saturate the water with natural gas or air in a contactor. The higher the pressure, the more gas can be dissolved in the water.

However, most units are designed for a contact pressure of 20 to 40 psig. Normally, 20 to 50% of the treated water is recirculated for contact with the gas.

The gas-saturated water is then injected into the flotation chamber as shown in Fig. 15.19. The dissolved gas breaks out of solution in small-diameter bubbles when the flow enters the chamber, which is operated at near- atmospheric pressure.

Design parameters are recommended by the individual manufacturers but normally range from 0.2 to 0.5 scfibbl of water to be treated and flow rates of treated plus recycled water of between 2 and 4 gal/(min-sq ft). Retention times of 10 to 40 minutes and depths of between 6 and 9 ft are specified.

Dissolved-gas units have been used successfully in refinery operations where air can be used as the gas and where large areas are available. In treating produced water for injection, it is desirable to use natural gas to exclude oxygen. This requires the venting of the gas or installation of a vapor recovery unit. Field experience with dissolved-natural-gas units have not been as successful as experience with dispersed-gas units.

Dispersed-Gas Units. In dispersed-gas units, gas bubbles are dispersed in the total stream either by use of an inductor device or by a vortex set up by mechanical rotors. Fig. 15.20 shows a schematic cross section of such a unit.

Most dispersed-gas units contain three or four cells. Bulk water flow moves in series from one cell to the other by underflow baffles. Field tests have indicated that the high intensity of mixing in each cell creates the effect of plug flow of the bulk water from one cell to the next. That is, there is virtually no short circuiting or breakthrough of a part of the inlet flow to the outlet weir box.


ZONE DESCRIPTION

VAPOR SPACE

AIR OR GAS INDUCTION

FLOTATION

PwOucEo ic:;a INLET

Fig. lS.PO-Dispersed-gas unit with inductor device.

Field tests and theory both indicate that these units operate on a constant percent removal basis. Within normal ranges, their oil removal efficiency is independent of inlet concentration or oil droplet diameter. The design of the induction nozzle or rotor and internal baffles is critical for determining overall efficiency.

Field experiments indicate that most designs can be expected to have efficiencies of about 90%. Because the gas is recycled by the unit, a natural gas blanket can easily be maintained with little or no venting. The low required retention times makes this an ideal choice for offshore facilities where space and weight are at a premium.

Dissolved Gas Removal

Often, produced or surface waters will contain dissolved gases, which must be removed by the water-treating plant. Oxygen in concentrations of 0.05 ppm in hydrogen-sulfide-free water and 0.01 ppm in water containing hydrogen sulfide is generally considered to be sufficient to cause corrosion problems in the facilities and bacteria plugging problems in an injection reservoir. For this reason, attempts are made to exclude oxygen from produced-water systems by maintaining gas blankets on all tanks. However, sometimes these


systems must be designed to handle rainwater, which may introduce dissolved oxygen in sufficient quantities to require removal. The location of surface water intakes can be arranged to minimize the oxygen content in the surface water but, in almost all cases, oxygen may have to be removed.

Some waters may contain ammonia, HZS, or COz, which must be removed. Dissolved gases are commonly removed by either chemical scavenging, gas stripping, or liquid extraction. It is beyond the scope of this book to deal with the design of the complex processes and equipment that can be employed in removing all dissolved gases. Since oxygen is the most common contaminant, we will briefly describe the treatment processes commonly in use to remove dissolved oxygen.

Oxygen Scavengers

Chemical scavengers are used quite often to remove dissolved oxygen from water streams of less than 10,000 B/D. Sulfite is the most common scavenging agent for water treating. The most common forms are sodium sulfite, sodium bisulfite, ammonium bisulfite, and sulfur dioxide. To speed the reaction rate, a catalyst such as cobalt is required.

Gas Stripping

The basic principle employed in gas stripping is that the quantity of oxygen dissolved in the water is directly proportional to the partial pressure of the gas that is in contact with the water (Henry’s law). Since partial pressure of the gas is a function of the mole fraction of that gas, the addition of other gases to the solution will decrease the partial pressure of oxygen and thus the concentration of oxygen in the water.

In a typical gas stripping column, natural gas or steam, if available, is introduced in the base of a packed or trayed column (similar to the familiar glycol contactor used in gas dehydration) and flows upward countercurrent to the water, which is introduced in the top of the column and flows downward.

If natural gas is used, the oxygen-contaminated gas from the top of the tower can be used for fuel, compressed for inclusion in the sales gas stream, or vented, depending on the process design and gas sales contract. Stripping gas usage of between 2 and 5 scf/bbl is common.

It is also feasible to strip oxygen from water by use of a concurrent flow. This is common in cases where lift gas is used as the artificial-lift mechanism for obtaining the water from a reservoir or subsea source. Sometimes, the gas is injected into the water with a static mixer in concurrent flow in a pipe. While this may require more stripping gas, it may be more economical from the standpoint of equipment cost, space, and weight when the value of the stripping gas is low. Stripping gas usage in concurrent flow can be in excess of 10 scf/bbl.

Vacuum Deaeration

Since the partial pressure of oxygen in the gas is a function of the total pressure of the system, the partial pressure of oxygen can also be reduced by applying a vacuum to the water-gas system. Vacuum deaemtors can be combined with either counterconcurrent or concurrent

stripping gas to provide very low oxygen concentrations in the water. Stripping gas usages of a fraction of a cubic foot per barrel are common. Vacuum stripping towers are used where no stripping gas is available, the available stripping gas contains contaminants such as CO;! and H2S, and stripping gas has a high value.

Dissolved Solids Removal

The removal of dissolved solids is of major importance if the water is to be used in steam generation or for some EOR projects. In particular, magnesium and calcium ions in the water may cause boiler scale, react adversely with an EOR chemical, or precipitate in the reservoir as a plugging solid.

Various processes have been used to create chemical reactions to cause the dissolved solids to form precipitates, which can then be settled or filtered out of the water. Aeration has been used to oxidize soluble fer- rous compounds to insoluble ferric compounds, and soluble bicarbonates to insoluble carbonates. However, this process introduces dissolved oxygen, which must then be stripped out of the solution.

The addition of chemicals, such as lime and soda ash, under correct conditions of temperatures and pH can form insoluble carbonates. Alum or other coagulants are then added to help in the settling or filtering of the precipitate. The equilibrium constants for these reactions are usually such that the low total dissolved solids (TDS) required cannot be easily met. For this reason, ion exchange has become the most common process for control of dissolved solids.

Ion exchange can be defined as a reversible exchange of ions between a solid and a liquid in which there is no significant change in the structure of the solid. Ion- exchange solids of various types can be used. The usual ion-exchange material takes the form of granules, or beads, ranging in size from approximately 0.3 to 1 .O mm in diameter.

As a very broad generalization, the synthetic ion- exchange resins can be categorized into four groups.

1. Strong acid resins. These are polystyrene resins that are strongly acid, have a high exchange capacity, and are not damaged by strongly alkaline hot water.

2. Strong base resins. Typical resins in this category incorporate a quatemary ammonium type of structure. The hydrocarbon groups may include methyl groups, polymeric benzyl groups, ethanol groups, and the like.

3. Weak acid resins. These resins usually contain car- boxylic groups as the active, or functional, ion sites. They have a limited use in water treatment, but can be used to remove basic materials from solution.

4. Weak base resins. Typical resins in this category are of the polyamine type. They usually contain a mixture of primary, secondary, and tertiary amine groups and their chemical properties are analogous to amine or ammonium hydroxide solutions. They can be used to remove free acids from solution.

Table 15.11 presents general guidelines for the use of different combinations of resins.

In ion-exchange units the influent water flows through the bed of ion-exchange material. Ions from the bed are exchanged for the undesirable ionic species in the water. When a bed is close to breakthrough, it is regenerated


TABLE 15.11-GUIDELINES FOR CATALYST SELECTION

Feed

TDS Hardness Resin Process

< 2,000 < 700 strong acid single bed 700 to 5,000 < 2,000 strong acid series bed

5,000 to 10,000 < 500 weak acid single bed 5,000 to 10,000 500 to

2,000 strong to weak

10,000 to 25,000 <2,000 > 25,000 < 500

with a strong solution of the ionic species originally contained in the ion-exchange material. The exit solution from the regeneration stage is thus a concentrated solution of the contaminants removed from the water.

Removing Hydrocarbons From Solids Fortunately, most solids that must be handled in a water treating plant are water-wet. Reservoir sand in its in-situ condition is almost always water-wet. As it flows up the tubing and through the process system, it may become coated with an oil layer but, since this layer is on top of a water-wet solid, it is easily removed. The two most common solids cleaning processes arc abrasion and water washing.

In separating solids with a hydrocyclone, the solids rub against the inside wall of the cone and against each other. The high centrifugal velocities involved abrade the oil layer, cleaning the solid. In most applications, one pass through a hydrocyclone is sufficient to clean the solids sufficiently for disposal. Some installations include two or three cyclones, in series, to ensure adequate cleaning. In some installations, an air flotation step is used between cyclones to separate any oil that may exit through the cyclone underflush.

Another cleaning method involves the routing of the undefflush to a batch cleaning vessel. Water is induced in the vessel to agitate and wash the solids. The solid bed is then allowed to settle while the oil is skimmed from the top. The procedure is repeated several times until the solids are clean enough to be educted from the bottom of the vessel and disposed of as a slurry.

On offshore platforms, skim piles have been used effectively to water wash and abrade solids. The solids move down the pile and are abraded as they bounce along the baffle surfaces. These are water-washed in the mixture around the end of each baffle. Oil removal occurs in the normal manner within each baffle section.

Process Selection and Project Management The process selection for a specific project must provide an overall cost-effective system using the individual techniques described previously. In some cases, the process selection and design of the system may be obvious and easily handled by hooking up some standard pieces of equipment in the field with no further engineering design. However, in most cases, several alternative schemes may have to be analyzed, and the process will be complex enough to require the system to be engineered to work as designed. Cost may be significant

acid series operation weak acid series bed chelating single bed

enough to necessitate competitive bids for equipment and installation.

Every project, no matter how small, must proceed through the following steps. On small projects, the engineer alone may handle some of these steps in a matter of minutes. On larger projects, these may take months of analysis and work by teams of individuals.

Conceptual Studies

The first step of any project, the conceptual study, in- vestigates one or more means of accomplishing the objective. An economic and technical assessment and comparison of the various methods or schemes is made. Block diagrams may be used to develop a selected process or alternative into specific descriptions and recommendations for equipment. Equipment type and arrangement are studied and a design philosophy established. An analysis of cost and economic benefit of each alternative is performed and a recommendation made.

Project Definition

The next phase is to define the project for the scheme selected in the conceptual study. Tools used are process flow sheets, layout drawings, preliminary cost estimate, and project execution plan.

The block diagram is converted into a process flow sheet to better define the project. The flow sheet shows all major equipment, main piping, and operating pressures and temperatures. Instrumentation that controls the main process flow is shown and every major line is assigned a stream number. A table is included listing pertinent operating data for these streams.

Layout drawings locate the equipment on the process flow sheets. A well-planned layout is the key to good operation, economical construction, and efficient maintenance. The layout drawing must be integrated with the development of the process flow sheet and must be settled before detailed piping, structural, and electrical design can begin.

A plan of execution for a project begins when the first information is received. This plan must consider the alternatives for breaking the job down into individual work items for bid solicitation. It must balance time and ease of management against cost for such decisions as (1) the scope of work to be included in individual work, (2) degree of engineering to perform prior to bid, (3) potential suppliers’ work load, capability, and competitive situations, and (4) operator sole source preferences.


The preliminary cost estimate is an important tool in the generation of the initial authority for expenditure (AFE). For effective cost control, the preliminary estimate must be made accurately and upgraded when information is received that affects it. Revisions may be necessary because of a change in scope or a realization that the amount of work was over- or underestimated.

Design Engineering

Once the need and extent of engineering assistance is determined, design engineering must begin to translate the process flow sheets into specific objectives and to determine activities required to attain these objectives. The basic item on which all other activities depend and which must be completed early in an engineering design project is the mechanical flow sheet.

Mechanical flow sheets are established from the process flow sheets and show every piece of equipment for the entire facility, including process, utilities, fire-water system, safety systems, spare equipment, etc. Every instrument, valve, and specialty item is shown schematically. Piping should be shown with flow arrows and line numbers indicating size, pressure rating, heat tracing, and insulation.

Vessel and equipment specifications are established for long-delivery items to expedite both the design and purchasing effort. Every facility is designed for a specific function, and thus the criteria by which the equipment is selected are applicable to that facility only. Therefore, this equipment is not normally produced off- the-shelf items. However, an experienced engineer will maximize the use of standard items to minimize cost and delivery time.

Detail Engineering

Once design engineering is completed and major equipment items have been specified and sent out for bid, the next step is to perform the detail engineering. This consists of piping drawings, structural drawings, electrical one-line drawings, instrument data sheets, and control schematics.

Piping drawings translate to the fabrication contractor the piping arrangement, as defined in the mechanical flow sheets. In many cases, a good set of piping drawings is the key to any easy to build and operate facility. In all cases, a good set of drawings is required to speed installation and keep the cost for extras to a minimum.

Structural drawings for an onshore facility detail the foundation site development and road work required, as well as detailing any pipe supports or skids for the production equipment. For an offshore facility, these drawings can include platform drawings as well as those for production skids themselves. The skids could be installed on wooden or concrete piles, steel or concrete barges, or steel jackets with steel decks.

Procurement

A large part of the engineen’ng effort is involved in bidding, evaluating, expediting, and coordinating vendors and vendor information. This is true whether the work is performed by the operator, the engineering consultant, or a turnkey contractor.

Special, long-delivery equipment, vessels, and instrumentation usually are ordered as soon as the

specification can be written, checked, bids obtained and evaluated, and the purchase order issued.

All bids should be opened at one time to minimize any possibility of a bidder receiving an unfair advantage. While it is possible to play one bidder against another by “bid shopping,” any experienced project engineer knows that in the long run this will be counterproductive. If vendors expect the work to be given to the low bidder, they will put their best effort into the price. If an auction is expected, the bid price will reflect this.

Inspection and Expediting

An important phase of the project is inspection. It is the inspector’s responsibility to ensure that the finished equipment or material is of acceptable quality and complies with all requirements of the purchase order.

The inspector witnesses tests on mechanical equipment such as pumps and compressors, observes and ap- proves fabrication methods of vessels, pipe, and structural steel, and generally ensures that the best workman- ship is being performed on the purchased equipment.

The purchaser’s expediter can do much to ensure that the estimated delivery dates will be met by working with both the manufacturer and his own organization. The expediter should seek and study all information that might affect delivery, anticipate delays or bottlenecks, and resolves these with the vendor. He should assist the vendor in obtaining priorities and solving procurement problems by communicating with subsuppliers. If delivery schedules must change, he should advise his employer as early as possible. On small projects, these functions are performed by the inspectors.

Startup

The conscientious preparation of a startup and operations procedure is the best possible check of the practicability, operability, and safety of a system. The procedure may contain only a few pages or may take the form of a book. In any case, it describes (1) overall purpose and design of the installation, (2) operation of the process, (3) details and operating descriptions of all systems, including overall instrumentation (pneumatic and/or electronic) electrical, data transmission, utility and fire- water, etc., (4) instructions for equipment installation, (5) purge and preparation for operation, and (6) procedure for starting.

Project Execution Format

Every project goes through the steps of project management described above. This is true no matter whether the tasks are performed by the operator’s staff, the engineering consultant, or contractors. The project must progress from one step to the next; engineering, procurement, and inspection must be accomplished; and the cost of performing these functions must be borne by the project.

There arc several different project execution formats that must be considered. The choice of a specific format will determine which of these functions will be performed by a particular organization.

Although there are almost as many formats as there are projects, most can be separated into the following four basic types which we shall call turnkey, negotiated tum- key, modified turnkey, and cost plus.


Turnkey. The turnkey format is used when the work is not completely designed. The conceptual study and project definition are normally complete. In this case, the scope of the contractor’s work would include the design engineering, detail engineering, procurement, inspection and expediting, and possibly, startup.

Five advantages are claimed by proponents of this format: (1) the project cost is established before work starts, (2) a single point of responsibility is established, (3) the contractor can design to take advantage of constmction efficiencies, (4) the contractor can speed up equipment delivery by performing engineering in such a manner as to get long lead time equipment on order sooner, and (5) the contractor assumes risk.

Several problems have been experienced with this format.

1. Owner loses design control or can exercise it only at high cost in extra work.

2. Competition is limited to select firms with total design and construction capability.

3. Most large firms have different engineering and construction profit centers. Many large contractors bid the work to outside engineering firms. In either case, the contractor’s engineers may be no more or less aware of construction efficiencies than a third party.

4. Because of the design risks assumed by the contractor, a contingency factor will be added to the price.

5. By necessity, there will be a large number of subcontractors furnishing individual items of equipment to the turnkey contractor. Other than initial approval, the owner has no control over subcontractors.

6. The contractor’s and owner’s interests are not identical. The contractor has an incentive to provide the least costly quality for the fixed price and does not profit or lose nearly as much as the owner with timely delivery. Therefore, the owner must provide inspection and expediting. It is very difficult to do this for subcontractors where no direct commercial relationship exists.

Negotiated Turnkey. The negotiated turnkey format recognizes that, before the detail engineering is complete, it is difficult for a contractor to provide a fixed price for the work while maintaining adequate owner control. In this format, the design and detail engineering are performed (normally for a fixed fee), so that long- delivery items can be placed on order prior to completion of detail engineering. Once the scope of work is defined, a turnkey price is negotiated.

The advantages claimed for this format are the same as those for the turnkey format with the added advantages of maintaining owner control and reducing contractor’s design risk. The disadvantages are identical with the added disadvantage of eliminating much of the owner’s leverage when it comes to negotiating the final contract.

Modified Turnkey. In the modified turnkey format, each work item is separated and bid turnkey as the scope of that work item is defined. In the previous two formats, the prime contractor does this in bidding out items of equipment to subcontractors. The difference in this format is that the owner, or engineering consultant, do the bidding, awarding, and expediting. In addition, those items which are “sole sourced” to the contractors construction arm in the two previous formats must be bid

and evaluated. The main advantages claimed for this format are (1) control of the project is maintained, (2) competition is maximized as individual work items can be bid to firms specializing in such work, (3) owner’s inspectors and expediters have a direct commercial relationship to all suppliers, and (4) contractor’s risk and contingency is controlled to the extent desired by owner. For example, scope of work contingency can be eliminated while weather contingencies are included.

The disadvantages with this format are (1) increased coordination of the contracts is required by either the owner or the engineering consultant, (2) the owner, or consultant, must develop and monitor the plan of execution rather than this being a function of the turnkey contractor, and (3) engineering and project management costs are explicitly determined and not hidden in contract cost.

Cost Plus. The cost-plus format requires the contractor to be reimbursed for all direct costs plus a percentage of costs for overhead and profit. Typically, this format is used where risk is high, or when there is insufficient time to solicit firm bids. Such a case would occur if construction were required within an operating plant, it were necessary to repair storm damage, or a simple field routing job were envisioned. The major disadvantage of this format is that the owner bears the risk of inefficient labor and job organization.

Comparison of Formats. The type of project format to employ depends on the nature of the project, the type of contractors available and their competitive position, and the priorities of the owner. There is no one answer as each project is different and competition and priorities change from time to time.

The author had the opportunity to be involved in two similar projects for the same owner, which occurred almost simultaneously. One was set up as a turnkey format and the other in a modified turnkey format to test the validity of the claimed advantages and disadvantages. The cost of the modified turnkey approach was 15% less, and it took 10% less time to complete because of greater owner control of the schedule. On another recent project, the design was done by the owner and each work item bid out separately. In addition, two turnkey bids were solicited from companies who expressed a desire to construct the complete installation. The additional cost for awarding the low turnkey bid would have been 40% more than the cost of awarding 10 individual bids.

Project Control No matter what the project execution format, it is necessary to implement procedures for controlling both project cost and timing. The most important part of project control begins at the outset of the project with controlling the engineering effort. Priorities set and decisions made at this point will affect project timing and cost throughout the job.

An activities schedule will help in project scheduling. This is a detailed plan of execution for the project’s engineering phase and is similar to, but more detailed than, the overall project plan of execution described previously. When preparing the activities list, an effort should be made to separate activities into logical cost categories.

SURFACE FACILITIES FOR WATERFLOODING 8s SALTWATER DISPOSAL

The main tool for project cost control is the preliminary cost estimate developed during project definition, which lists budget costs for each item in the plan of execution. Budgets should be established for scheduled work packages. It is important for cost control purposes that activities have definite starting and ending points.

Costs are periodically updated as the project becomes better defined. Differences can then be explained and total project cost revised-even before the first item is purchased. As bids are awarded and commitments are made, the total project amount can be adjusted accordingly.

Accounting for costs can be routine, but controlling costs is another matter. Once a project is committed, a large part of final cost is beyond control of the project engineer. It is not unusual for bid items to vary significantly from time to time because of contractor work load, weather conditions, and availability of equipment. Also, difference in price between low bidder and next low bidder can vary 10 to 20%. If the low bidder had not been included in the bid list, item cost would be that much higher. On a recent project, the sum of second low bidders’ prices was 5% higher than that of low bidders.

Early in the project, timing is controlled through the engineering effort to ensure that bids are sent out, evaluated, and awarded in accordance with the project plan of execution. Once bids are awarded, the inspector and/or expediter are responsible for determining if work is progressing on schedule. A schedule thus must be worked out with the successful bidder, preferably before bid award. When delays are spotted, meetings with the contractor are required to return the job to schedule. On occasion, it may be necessary to appeal to higher levels of management in both companies. The sooner a pmb- lem is spotted, the better the chance that corrective action can be agreed on. On large complex projects. the use of computer-based network analysis of activities is sometimes beneficial.

Nomenclature

A, = area of deck, sq ft b = width (breadth), fi

b, = width (breadth) of the plate section perpendicular to the axis of water flow, ft

Cd = dispersion constant CE = constant for erosional flow

C’eW = constant with a value of 80 to 140, depending on the inside pipe material and its age

dd = oil droplet diameter, micrometers

(dd)max = maximum oil droplet diameter, micrometers

di = pipe inside diameter, in. dp = particle diameter, microns

E = flow efficiency, fraction E, = pump efficiency at flow conditions E,, = removal efficiency on any cycle of a

particular drop size E,, = overall particle removal efficiency

f= fv =

F=

friction factor volume fraction of the dispersed phase factor to account for turbulence and short

circuiting hf = height of flume, ft

K,s = empirical settling constant L= length of line or length of plate section

L, = L2 =

Lb.\ = L, = L, = N,. =

parallel to the axis of water flow, ft inlet distribution section outlet gathering section length of baffle section, ft effective settling section perpendicular distance between plates, in. number of cycles of no flow a particle

NR, =

Phh =

PI =

P? = 48 =

sees as it traverses the baffle section Reynolds number brake horsepower pressure at pipe inlet, psia pressure at pipe outlet, psia flow rate of gas at standard conditions,

MMscfiD

qL =

qr =

qH. = t=

t, =

t, =

t.7 =

t M’ =

VRf =

liquid flow rate, B/D rainfall rate, in./hr bulk water flow rate. BWPD time for the dump cycle, seconds time valve is closed, seconds retention time, minutes time particle is falling while in effective

section of flume time water is in effective section of

flume velocity of gas at specific flow

conditions, ft/sec

15-33

v s = settling velocity, ft/sec Zf/ = friction head loss, ft of liquid Z, = height of the coalescer plate section

perpendicular to the axis of water flow, ft

Z, = height of water column, ft

-fR = specific gravity of the gas at standard conditions relative to air

ye = specific gravity of liquid relative to water Ay = difference in specific gravity relative to

water

AY ow = difference in specific gravity between oil and water

8 = angle of the plate with the horizontal pLnf = viscosity of gas at specific flow

conditions, cp

References 1. Daughelty, R.L. and Ingersoll, A.C.: Fluid Mechanics with

Engineering Applications, McGraw-Hill Book Co. Inc., New York City (I 954) 205.

2. “Flow of Fluids Through Valves, Fittings and Pipe,” Crane Co., Houston (1969) Technical Paper 410, l-8.

3. Design and Insrallation of QJshore Production Platform Piping Systems, third edition, API RPl4E, API, Dallas (1981) 22.

4. API Specification for Polyethylene Line Pipe (PE), third edition, API Spec. 5LE, API, Dallas (Nov. 1981).

5. API SpeciJcation for Thermoplastic Line Pipe (PVC and CPVC), fifth edition, API Spec. 5LP, API, Dallas (Nov. 1981).


6. API Specification for Reinforced Thermoserring Resin Line Pipe (RTRP), fourth edition, API Spec. 5LR, API, Dallas (March 1976).

I. API Specification for Line Pipe, 33rd edition, API Spec. 5L. API, Dallas (March 1983).

8. Recommended Practice for Application of Cement Lining to Steel Tubular Goods, Handling Installation and Joining (Tentative), first edition, API RP IOE, API, Dallas (March 19%).

9. American National Code for Pressure Piping, Power Piping, ANSI B31.1, American Sk. of MechanicalEn&eers, New’York City (1977).

10. ASME Code for Pressure Piping, B31, Chemical Plant and Petroleum ReJinety Piping, ANSIIASME 831.3, American Sot. of Mechanical Engineers, New York City (1980).

1 I. American National Standard Code for Pressure Piping, Liquid Petroleum Transportation Piping System, ANSIIASME B3 1.4,

American Sot. of Mechanical Engineers, New York City (1979). 12. American National Standard Code for Pressure Piping, Gas

Transmission and Distribution Piping Systems, ANSI B3 1.8, American Sot. of Mechanical Engineen, New York City (1975).

13. American National Standard, Pipe Flanges and Flanged Fittings I ANSI B16.5, American Sot. of Mechanical Engineers, New York City (1981).

14. API Specifications for Wellhead Equipment, 14th edition, API Spec. 6A, API, Dallas (March 1983).

15. “Oil-Water Separator Process Design,” Manual on Disposal of Refinery Wastes, Volume on Liquid Wastes. API, Dallas (1975) Chapter 5.

16. Brunsmann, J.J., Comelissen, J., and Eilen, H.: “Improved Oil Separation in Gravity Separators,” paper presented to the 1959 API Committee on Disposal of Refinery Wastes, Denver, Oct. 14-16.

Chapter 16

Automation of Lease Equipment G.R. Burrell, Exxon Co. U.S.A.*

Introduction Automation in the oil and gas producing industry covers a broad spectrum of supported functions. In a simple application, automation may be defined as linking together instruments and controls to perform lease-operating procedures automatically in a predetermined manner. Automation in a more complex environment will have digital computers in some form of a supervisory control and data acquisition (SCADA) system. Automation in the industry has tended to evolve as new tools become available and ate accepted by industry and regulatory agencies. Generally, automation has advanced by a more complex linkage of instruments and control devices. Some of the more important tools and/or techniques that have enhanced lease automation are (1) solid-state electronics, (2) lease automatic custody transfer (LACT), (3) tank battery consolidation, and (4) SCADA.

Solid-State Electronics

The development of solid-state electronics first as discrete components and then as integrated circuits has been a key factor in advancing lease automation. Elec- tronics have provided the base for improvements in instrumentation, control elements, communication, and digital computers that form the primary components of enhanced automation facilities. Pneumatic and electromagnetic (relay) logic have been, and will continue to be, used in various forms of automation, but the extent of logic implementation is limited substantially compared with that available for electronics. Pneumatic and/or electromagnetic functions are effective as complementary features to electronic forms of automation and as stand-alone automation for less complex applications.

Microprocessors and their extension to microcomputers are having an impact on automation that may well

‘Authorof theoriginal chapler on this topic in the 1962 edition was Don R. Patterson

exceed that of integrated circuits in the initial form of “hard-wired” logic. Microcomputers combine the advantages of electronic components and program instructions into a flexible, capable, and reliable form that has substantial advantages for automation. These functional advantages have been complemented by a reduction in cost compared to implementation with integrated circuits. Microcomputers are being used in almost every component of automation related equipment from individual instruments through digital computers.

Lease Automatic Custody Transfer (LACT)

LACT is the process of transferring (running) lease- produced oil into a connected pipeline on an unattended basis. LACT includes the capability to determine automatically the quantity and quality of oil being transferred and the control functions to prevent transfer of unacceptable quality and/or volumes. Before LACT, lease oil was produced into a tank, quantity and quality (opening gauge and thieving, etc.) were determined, and a valve was opened to the pipeline to initiate transfer. When the transfer was complete, the pipeline valve was closed and a final (closing) gauge was made as basis for determining net volume transferred. All these steps were manual activities with some related duplication of effort between the lease operator and the pipeline gauger. In addition to being labor intensive, the process was inherently inefficient in use of related treating and storage facilities. LACT is an important tool in the evolution of lease automation. LACT is a significant automation element and has been widely accepted and implemented by industry. In addition, it has become an important building block for other forms of automation in lease operations.

Tank Battery Consolidation

Many oil and gas fields have multiple operators or working interest owners. In addition, most fields consist of a number of separate leases (common royalty ownership,

16-Z PETROLEUM ENGINEERING HANDBOOK

etc.) that require individual oil and gas processing (separation, treating, storage and transfer, etc.) facilities to account for production to each owner.

LACT initially was applied to these separate lease operations. The automatic transfer of produced oil and recycling of unacceptable quality oil [high basic sediment and water (BS&W) content] to treating facilities increased the effectiveness of treating and storage equipment. In addition, tire incremental cost of larger meters, pumps, and related equipment of LACT’s was low compared to the increased oil volume transfer capability. Thus, a technical basis was available to process and transfer much higher produced-oil volumes than present on the average lease.

Historically, individual lease oil-production volumes had been treated to pipeline quality (2 % BS&W or less) before custody transfer from the lease. In the 1960’s, regulatory agencies began to approve operator requests to commingle wet-oil (not pipeline quality) production from multiple leases into a common or central oil processing and custody transfer facility. Oil production from each lease was determined by measuring the wet-oil volume (separator positive-volume or positive- displacement meters, etc.) and correcting for water content with automatic samplers and later with capacitance probes and related net-oil computers. Final sales volume to each separate lease was determined by allocating the custody transferred (sales) volume back to each lease on the basis of its proportion of total wet-oil lease measurements. In other cases, tank battery consolidation was implemented when a field was unitized under a single operator for initiation of secondary recovery activities.

Tank battery consolidation eliminated oil treating and storage facilities on the individual lease. In some fields,

the pipeline trunk lines used to gather individual lease oil volumes were converted to wet-oil gathering lines for the consolidated tank battery operation. Tank battery con-

solidation converts a field’s operation from a number of stand-alone lease functions to a central process with multiple inputs. Oil treating and storage, water treating and disposal, vapor recovery facility, etc., became more efficient and controllable in the consolidated environment.

Supervisory Control and Data Acquisition (SCADA)

SCADA is a common name applied to computer-driven automation systems used in oil and gas production operations. Basic functions generally include status/alarm reporting, production volume accumulation/reporting, well testing, and control. These systems vary from small units that are applied to only a few leases in a single field to large units that serve multiple fields containing several thousand total wells.

SCADA systems are tied directly to the instrumentation and control devices on the process equipment used in oil and gas production. This provides timely and continuous access to the operational information being sensed by the instrumentation. Some SCADA systems emphasize data gathering and reporting to operating personnel for “open-loop” control while others use program logic to analyze input information and initiate control actions directly (“close-loop” control). SCADA

systems may be oriented primarily to operating personnel needs, or they may be multipurpose by providing functions for operating, accounting, engineering, and management groups.

SCADA is a logical extension of automation in the sequence from manual lease operation, to use of LACT, and then to centralized treating, storage, and automatic custody transfer (ACT) with tank battery consolidation. This sequence moved from essentially independent individual lease operations to a single overall process that has a number of closely related functions. SCADA provides the ability to obtain timely operational information for optimization of the interrelated process functions. For example, if a compressor outage reduces the gas processing capacity below a field’s gas-producing rate, well-test information from SCADA allows shutting-in of wells with high GOR’s and thus minimizes reduction of the related oil producing rate. In general, timely and accurate operational information can be used to obtain maximum utility of existing process equipment and minimize need for stand-by capacity.

Some form of automation is used on every lease that produces oil and gas. The extent of practical automation depends primarily on economics. Some of the benefits of automation that may be used in the economic justitica- tion are as follows:

1. Capital investment in lease production equipment is reduced.

2. Operating expenses are reduced through savings in labor costs, maintenance expenses, travel expenses, and power and fuel costs.

3. Ability to initiate and document actions required for regulatory compliance is improved.

4. Surveillance capability of management and support- ing staff groups is improved.

5. The quantity and quality of operational information available for making business decisions is increased, which results in revenue increases and operating cost decreases.

Automation in oil and gas production activities will continue to evolve as additional tools ate developed and applied. Most advances (including improved communications and “smart’ ’ end devices) will take the form of (1) improvements in existing devices, (2) new devices with improved capability/reliability to replace older equipment, and, less frequently, (3) new devices with new features/functions that are applicable to the production process. Much of the basic instrumentation and control equipment (primarily pneumatic-based) that has been used for years with oil and gas processing facilities will continue to be applicable. Implementation of some form of enhanced automation (as described earlier) will be a primary force to increase use of electronic and electronic/pneumatic equipment. On this basis, it appears reasonable to elucidate the topic of automation first by discussing some of the commonly used equipment and then by depicting equipment application in automation systems.

Automatic Production-Control Equipment Automatically Controlled Valves and Accessories

Automatic control valves can be classified in a number of ways, but classification by the energy medium that ac- tuates the valve operator is most pertinent to automation.

AUTOMATION OF LEASE EQUIPMENT 16-3

By using this method of classification, automatic control valves can be grouped into three major categories: fluid- controlled valves, electrically controlled valves, and fluid-electric-controlled valves. In the latter category a fluid energy generally is used to operate the valve and electric energy is used to control the fluid-energy source.

Fhdd-Controlled Valves. The most common types of automatic fluid-controlled valve operators arc diaphragm operators and fluid cylinders. Both of these valve operators can be used on any style of valve body whose inner valve can be positioned by longitudinal displacement of the valve stem. The fluid cylinder operator normally is used with valve-body styles requiring 90” rotation for operation. Diaphragm operators most commonly are applied to valves that have globe, angle, butterfly, and Saunders-type valve bodies. Fluid-cylinder operators are more commonly used with plug valves.

In oilfield applications, the most common fluid used to actuate both valve operators is natural gas, generally taken directly off a separator or heater-treater on the lease. If natural gas is not available, or if for some reason the available natural gas supply is not suitable, a bottled gas (nitrogen, etc.), compressed air, or hydraulic fluid could be used. Diaphragm operators normally require only 15- to 30-psi fluid pressure to actuate the valve. Pressures up to 100 psi and over often are desired for the fluid cylinders because, the higher the fluid pressure available to operate the cylinder, the smaller size the cylinder may be, and consequently the lower the cost. Valves using these types of operators, as a class, are frequently called “pneumatic control valves” even though the control fluid may be something other than air.

Some fluid-controlled valves can be controlled with the fluid flowing in the line in which the valve is located. These types of valves generally use the differential pressure principle for control purposes. A reference control pressure is established by spring-loading or prcssure- loading the valve operator. The control valve is actuated when the line pressure upstream and/or downstream sensed by the valve operator algebraically exceeds the reference control pressure-i.e., the valve can be actuated by pressures either excessively high or excessively low, or both. This feature makes this type of valve particularly suited for use as safety shut-in valves on wellheads.

Electrically Controlled Valves. Automatic electrically controlled valve operators arc of two general types, electric-solenoid (or magnetic) and electric motor. Magnetic operators are used for valves requiring longitudinal motion to position the inner valve. The use of magnetic operators generally is limited to valves 2 in. and smaller in size and of relatively low working pressures. Electric-motor operators can be used with any type of valve but in all cases must include accessories that provide a torque-limiting means and a limit switch to prevent damaging the motor when either extreme valve position is reached. On valves requiring longitudinal motion to seat the inner valve electric-motor, operators also must include a gear rack and pinion to convert the motor’s rotary motion to longitudinal displacement. Because of the relative expense of electric-motor

operators, they normally are used only on large-sized valves and/or valves having high working pressures.

Fluid-Electric Controlled Valves. Self-contained valve operators in the third category are generally hydroelectric-type operators. Operators of this type essentially consist of a self-contained reservoir of hydraulic fluid, a small electric motor, a pump, and a fluid-cylinder device-all within a single housing. The fluid-cylinder principle limits this type of operator to valves requiring longitudinal motion or 90” rotation to seat the inner valve. Valve operators of this type are available for a wide variety of valve sizes and working- pressure ranges.

In addition to the hydroelectric-type operators, any of the fluid-controlled operators mentioned can be made combination-type operators. By the addition of electric- solenoid valves in the fluid-control lines, an electric signal can be used to control the release of fluid energy to the valve operator. Combination-type operators of this kind are commonly called “electropneumatic operators. ”

Valve Switches. It is frequently essential that an automatic control system be able to sense the position of certain valves whether automatically or manually operated. This is accomplished by means of a “valve switch” coupled directly to the stem of the valve in ques- tion. In electrical control systems, the valve switch may be a mercury switch, a microswitch, or a position- sensing switch. In pneumatic control systems, the “valve switch” is a three-way pilot valve. The switches may be adjusted to open or close a circuit as the valve opens and closes. One of the most common applications of valve switches is on tank-run valves, in which case they are generally called “pipeline valve switches.” Another common use of valve switches is to indicate remotely the operational position (open, closed, etc.) of automatic control valves on wellheads, well manifolds, metering-tank inlets and outlets. In these general applications, valve switches normally will be referred to as “limit switches.” When interlocked with an automatic control system, valve switches perform the very important function of preventing subsequent steps in an automatic operation from proceeding unless certain valves are in the proper position. Other type devices can be used to sense “control” valve positions at intermediate points (or continuously) between the open and closed positions.

Automatic Production Programmers

Time-Cycle Controller. Automatic production programmers are scheduling devices that control the particular times and lengths of time that operating functions are performed. The simplest form of automatic production programmer is a time-cycle controller. A time-cycle controller basically consists of a clock with a timing wheel or wheels, containing a number of programming points at regular intervals around its circumference. The clock may be electrically driven, gas driven, or mechanically driven by a spring. It may have a l-, 2-, 4-, 6-, 8-, 12-, or 24-hour rotation period, the rotation period being the time required for the timing wheel(s) to


make one complete revolution. Programming is ac- equipment nearly always is made to be “fail safe”-i.e., complished by positioning the contacts on the timing upon loss of power from the controlling energy medium, wheel(s) such that the rotation of the wheel(s) generates the controls return to the position that will result in the the proper control signal to open or close valves controlled by the time-cycle controller at the proper times.

As commonly applied, a time-cycle controller in conjunction with a diaphragm control valve compose a “stopcock controller” and/or an “intermitter controller.” The primary difference in a stopcock controller and an intermitter controller is in the application. A stopcock controller generally is installed in a well’s flowline at the Christmas tree. It controls the times that the well is opened for production, normally for short intervals several times a day. An ‘ ‘intermitter controller” is installed in high-pressure-gas supply line at the wellhead of a well being gas lifted intermittently. It controls the times that gas is admitted to the well to actuate the gas- lift valves and lift the fluid to the surface. A time-cycle controller in conjunction with any type of automatic control valve may be used to produce a naturally flowing well where it is desired to produce the well less than 24 hours per day and/or 7 days a week.

A time-cycle or percentage-time controller plus a motor starter basically compose an automatic production programmer for an electrically driven rod-pumping unit. The rod-pumping unit controls also generally contain several safety devices: undervoltage relay, which drops out on power failure, overload relays to prevent burning up the motor, lightning arrester, circuit-breaking devices, etc.

The electronic (solid state) timer frequently is used in new installations that require timer functions. Mechanical and electro-mechanical timers will continue to be used in many existing installations.

Any time that these automatic production- programming devices are actuated electrically, the control point for individual wells may be centralized at the well manifolds, a central point on the lease, or even a point remote from the field. The time-cycle controllers, automatic control valves, motor starters, etc. still may remain located at the wellheads while control is exercised remotely. This is not true for similar devices that are actuated pneumatically unless the pilot gas for these devices is controlled electrically. Too much dampening and distortion occurs in pneumatic control signals for effective control when transmitted distances of more than about 150 ft.

Other Programmers

The programmers discussed previously have been used in oil and gas production facilities for many years. Cur- rently, the more complex sequential control will be electronic- and combined electronic/pneumatic-based in many applications. This trend is expected to increase as general-purpose programmable controllers, developed initially in the plant applications, find more use in oil and gas production. The functions of the programmable controller and the remote terminal unit (RTU) may well be combined into a more capable unit for oil and gas production monitoring and control applications.

Production Safety Controls In some respects, virtually all automatic control equipment is also safety-control equipment: automatic control

safest condition. Some automation controls, however, primarily perform safety functions rather than normal operational functions. These include high- /low-pressure safety shut-in valves, excess-flow valves, pressure and temperature switches, and pump-off controls.

Safety Shut-In Valves

High- /low-pressure safety shut-in valves and excess- flow valves are both fluid-controlled valves of the type that is actuated by line fluid. This valve type was discussed briefly in the section on automatic control valves. The use of an excess-flow valve or low-pressure control with a safety shut-in valve primarily safeguards against a flowline break and the resultant loss of oil and surface property damage. High-pressure control with a safety shut-in valve guards against pressures in excess of the allowable limit building up in the flowline. Either of these two kinds of valves normally would be installed at the wellhead.

Pressure Switches

Another means of protecting against excessive flowline pressures and/or flowline breaks is the use of a pressure switch and an automatic control valve. Pressure switches are available that produce either an electric or pneumatic signal, as required, to actuate the automatic control valve. On rod-pumped wells, the control signal from the pressure switch also must shut down the pumping unit-i.e., turn off the switch on an electric motor or ground the magneto on an internal-combustion engine. On rod-pumped wells that have no tendency to “head” or flow when the pumping unit is shut down, the automatic control valve may be omitted.

A “pressure switch” in the sense it is used here consists primarily of a pressure-sensing element, limit pressure contact(s), and an electrical, mechanical, or pneumatic means to transmit the control signal to the object(s) controlled by the pressure switch. The pressure- sensing element is commonly a bourdon tube, though some requirements could necessitate the use of a helical-, spiral-, diaphragm-, piston-, or bellows-type pressure-sensing element. In electrical control systems, the displacement of the pressure-sensing element is made to “make” or “break” an electrical switch, normally a mercury or microswitch, when the line pressure reaches the preset pressure limit(s). In pneumatic control systems, reaching the preset pressure limit(s) may actuate a pneumatic transmitter, relay, or slide valve.

Liquid-Level Controls

Another automatic safety-control device, which also frequently performs an operational-control function, is a liquid-level controller. These devices commonly are used to control liquid levels in separators, heater- treaters, storage tanks, surge tanks, accumulator vessels, metering vessels, etc. They may be used (1) to control high liquid levels to prevent running over a vessel, (2) to control low liquid levels to maintain pump submergence, (3) to control intermediate operational levels to open and close dump valves, to start and stop pumps, etc., or (4) to maintain the interface of two liquids at a given level.


There arc many types of liquid level controls. Some of the more common types of level control devices used in production equipment are float operated, pressure operated, ground-level tank gauges, electric and/or electronic, sonic, and vibration. A float-operated level switch generally consists of a spherical or cylindrical float attached to one end of a mechanical lever with either an electrical switch or a pneumatic relay on the other end. The switch or relay is located in a separate portion of the device housing and is isolated from the float area with a pressure seal. The float is displaced by the rise and/or fall of the liquid level being controlled and the motion is transmitted through the pressure seal to activate the switch/relay. A pressure level control switch may control liquid level on the basis of either differential or static pressure. The differential type devices commonly are used as a form of “pilot operated” dump valves on pressure vessels. The static pressure devices frequently are used for well shut-down service and level control in tankS.

Ground level tank gauges consist of tape, tape drum, and a tank gauge float that are linked to cause the tape to be spooled on and off the tape drum as the liquid level rises and falls within the tank. By extending the tape drum shaft and using appropriate cams, gears, etc., electrical or pneumatic control systems may be activated to control liquid levels in the tank. One type of sonic level control consists of a sound transmitter and receiver that are suitably arranged for separation by the liquid being controlled. The transmitter and receiver are driven with an electronic circuit that can measure the intensity of the sound reaching the receiver. The change in the received sound intensity between airigas and the liquid as the separation material can be used to sense and control a liquid level. Ultrasonic level devices sense the reflection of a sound wave from the gas/liquid interface and use the delay time between transmitting and receiving to determine distance from sensor to liquid level.

Some liquid level devices induce vibration into a detecting element. The degree of vibration dampening caused by the medium surrounding the element can be sensed to differentiate between gas and liquid materials. Electric and/or electronic level controls depend on the different electrical properties (capacitance, conductance, etc.) of the liquid to be controlled and that of the related medium (air, gas, and/or other liquid). An electronic circuit is used to sense the change in electrical property as the liquid level changes and thus controls the level as required.

Other types of level controls are available for specific installation needs. Level controllers should be selected to provide the required function in the least complex and most reliable form. It is advantageous to select equipment that has demonstrated satisfactory application experience within the operating environment that is to be controlled.

Automatic Quantitative Measurement Gross volumes frequently must be adjusted by quantitative measurements such as water content, temperature, pressure, and density before net volumes at standard conditions can be determined. Several of these quantitative pammeters are necessary for automation of lease operations.

Liquid Measurement

There are three types of quantitative (volume) devices commonly used for automatic liquid measurement on the lease: positive-volume meters, positive-displacement meters, and inferential meters. Positive-volume meters are essentially extensions of tank measurement with automatic “filling” and “running” functions. Positive- displacement meters “trap” a fixed volume of liquid within moving elements of the meter. Inferential meters measure liquid by detecting some property of the movin stream that is a basis for determining volume indirectly. Q

Positive-Volume Meters. Positive-volume meters operate on a “fill” and “dump” cycle rather than being a continuous operation. This type meter is essentially the automatic gauging of a tank by using level controls to move a fixed volume through the tank on each complete cycle. The volume that is “dumped” or measured is related to the displacement volume in the meter between the high “fill” point and the low “dump” point in the meter. Various types of level controls are used to control the fluid levels in the meter. Each complete “fill” and “dump” cycle is registered on a counter as a basis for total volume determination. Since the positive-volume meter is cyclic with a separate “fill” and “dump” period, allowance for handling produced volumes while the meter is in the “dump” cycle must be made. Con- tinuous operation is possible by having a pair of meters that are sequenced to have alternating cycles. This essentially requires duplication of facilities and some increased complexity in the controls. A more common arrangement is to provide surge volume capacity upstream of a single meter.

Volume-type dump tanks or meters have been built in a variety of shapes and volumes. Capacities per dump have ranged from less than V bbl to several hundred barrels. The metering chamber may be in a stand-alone vessel or it may be an integral portion of a vessel such as a separator or heater treater. The positive-volume meter is not as compatible with qualitative measurement as the positive-displacement and inferential meters since the volume measurement cannot be separated easily into smaller increments to drive sampling and other qualitative measurement devices.

Positive-Displacement Meters. A positive- displacement meter, regardless of specific type, consists of two primary elements: a stationary case and a mobile element, which acts to isolate within the case fixed volume of fluid each cycle of operation. The mobile element may be a rotor with sliding vanes, rotatable vanes, or rotatable buckets. It may be two rotors that mesh somewhat similarly to two helical or cycloidal gears as they rotate. The mobile element may be a disk that nutates about a camlike follower in three-dimensional motion or a cylinder that oscillates about a cam follower in two-dimensional motion. Or, finally, the mobile element could be a conventional piston such as that found in a power pump. Most positive-displacement meters are, in fact, closely akin to positive-displacement pumps.

Positive-displacement meters rapidly became the standard for ACT use. The positive-displacement meter provided a less costly and less complex facility than the


positive-volume meter. In addition, the positive- displacement meter provides a means to drive samplers and/or net oil computers with signals on a small increment of volume that is more compatible with automatic qualitative measurement requirements.

Care must be exercised in the installation design for a positive-displacement meter. All free gas must be

removed upstream to avoid spinning the meter, which would cause erroneous readings and, possibly, damage to the meter. For greatest accuracy, a constant flow rate should be maintained through the meter and at a rate at least 15% or greater of the rated capacity of the meter. Standards for calibration frequency, methods, etc., are set forth in API Std. 1101. ’

Inferential Meters. The turbine meter and the orifice meter are commonly used inferential meters for liquid measurement. These meters indirectly determine volume by sensing some property of the moving stream that can be related to volume. For example, the rotation of the turbine blades in the turbine meter and the differential pressure developed across the orifice plate in an orifice meter can be used as basis of volume determination.

Turbine meters became important in volume measurement when electronics were accepted as an element of a measuring device. The rotation of the turbine blades can be sensed electronically without need for any mechanical connection to the turbine rotor. This provides a simple arrangement that is inherently reliable and particularly suitable for high-pressure service. Thus, turbine meters initially were applied to measure injection water volumes. However, high-viscosity fluids drastically reduce the range of turbine meters.

Turbine meters are being used for well testing and wet- oil lease production measurements when combined with net-oil computers. These meters tend to be more tolerant of short over-range periods and sandy fluid than are positive volume meters. Turbine meters also arc being used for ACT, particularly for high-volume and/or high- pressure service.

Orifice meters are used more commonly for gas measurement but they have some applications in liquid measurement. Compressible liquids that require pressure correction for volume determination frequently are measured with orifice meters.

Gas Measurement The primary device for lease gas volume measurement has been and continues to be the orifice meter, which initially measured gas volume by using a mercury manometer before development of the bellows-type chart recorder. Orifice meters have these advantages: (1) no moving parts in the gas stream, (2) the ability to handle wide range of flow rates (long term) by means of plate size changes, (3) reliable and nonexternal powered recorder, and (4) a reliable sensor (bellows). The chart recorder is not compatible with automatic data acquisition; other types of gas measurement devices are used with SCADA. These gas measurement devices include positive-displacement meters, gas-flow computers, turbine meters, and vortex meters. ’

Positive-Displacement Meters

The installation of SCADA systems with automatic well

testing generated a need for gas measurement over a wide flow range with direct readout capability. The “rotary” positive-displacement gas meter is similar to the liquid “lobed-impeller” or gear-type meter. The rotary gas meter has the capability to measure gas accurately over a range of about 15 to 1 compared with about 4 to 1 for an orifice meter with a fixed orifice plate size. The rotary meter can be equipped with mechanical compensation on indicated volume for static pressure and flowing temperature corrections. The rotary meter needs to be protected from over-range and liquid accumulation within the measuring elements. These meters are usually applied to low-pressure gas measurement service.

Gas-Flow Computers

Gas-flow computers were developed to use existing orifice-meter runs and to provide a direct readout of gas volume that was compatible with SCADA. These devices use static and differential pressure electrical transducers on a standard orifice meter as a basis for gas measurement. The computer integrates the signals from the transducers and combines with fixed data on meter run size, plate size, etc. to develop a gas volume. The volume readout will be in the form of a switch closure that will register on an internal counter and provide input to electronic counter in an RTU.

Some gas-flow computers can accept a temperature transducer input to measure temperature of flowing gas stream for improved accuracy of volume measurement. Computers also may have capability to use two differential pressure transducers (e.g., 0 to 20 and 0 to 200 in.) and to select input from the unit providing most accurate instantaneous reading for integration. Early gas-flow computers were analog devices. Many current designs are digital units based on a microcomputer. The flow computer integration function also is being done by the microcomputer-based RTU. All designs have integration accuracy compatible with the basic measurement capability of the orifice meter,

Gas Turbine Meters

Gas turbine meters also are used to obtain direct readout on gas volume measurement that is compatible with SCADA. Turbine meters can measure gas volumes accurately over a range of about 20 to 1 at medium pressures. Rangeability tends to increase with increasing static pressure. These meters are usually somewhat less subject to over-range damage than the positive- displacement meter if over-range period is of short duration. Meter proving and checking may require installation of prover loop for a master meter. Gas turbine meters can be equipped with temperature and static pressure volume compensation capability. Many gas (or liquid) turbine meters are destroyed when they are over- ranged while pressuring up the system.

Vortex Meter

Vortex meters have a “bluff body” that spans the flow area through the meter and causes vortices to form in the flowing medium. There vortices are shed off the bluff body at a frequency that is proportional to the volumetric flow rate through the meter. The vortices can be


“counted” with suitable pressure or other flow pattern sensors, which are connected to an electronic component for flow accumulation. Vortex meters have rangeability characteristics similar to positive displacement and turbine meters without the moving parts of these devices.

Temperature Measurement Types of temperature-sensing devices commonly used in oil and gas production include filled-thermal, resistance thermal detection, thermocouple, thermister, and solid- state. The filled-thermal device operates on the basis of the principle that a fluid expands or contracts with changes in temperature. The device consists of a temperature-sensitive bulb connected by capillary tubing to an expansible element that is sensitive to pressure change. The bulb may be filled with a liquid, a liquid and its vapors, or a gas. The expansible element may be a diaphragm, a bellows, or a bourdon tube.

The$lled-thermal device has sufficient output force to be used directly for temperature compensation on positive-displacement meters used for LACT. The bellows assembly is connected to an infinitely variable transmission, which corrects the meter’s volume output to a base temperature of 60°F.

A resistance thermal detector (RTD) works on the principle that a change in resistance of a wine is related directly to a change in temperature of the wire. The device consists of a resistance element (sensing element) and a related electrical circuit, which uses the changing resistance of the element to control an output signal. The output signal can drive recorders and controllers.

A thermocouple works on the principle that heat applied to one end of two strips of metal of different composition, which are bonded at either end, develops an electromotive force (EMF) that is proportional to the temperature. Thermocouples may be of the wire type, in which both elements are in wire form, or of the Pyod type, in which one element is a closed tube and the other a wire welded to the inside bottom of the tube. The thermocouple is connected to an electrical circuit, which senses the generated EMF and develops an output signal that can drive recorders and controllers. Thermocouple devices are used in applications where temperatures may exceed 1,OflO”F.

nermister and solid-state devices exhibit a resistance change with temperature change. Both these devices respond rapidly to a temperature change because of the small mass of the sensing element. Electrical circuits are required to convert the “sensed” resistance change into an output signal that is proportional to temperature.

Automatic Sampler An automatic sampler is a device that removes a representative volume of fluid from a moving stream and retains it in a container for later processing and analysis. Factors that improve probability of obtaining a representative sample include the following: (1) sampling probe should be located in a vertical downrun of pipe at the center of the pipe and with probe opening facing upstream; (2) the total flowstream should be in turbulent flow; (3) sample size and sampling interval should be such that the sample is proportional to the total stream flow; (4) sample metering chamber should be closely coupled to sampler probe and located below the center

line of sample probe; and (5) samples should be col lected and stored at pressures exceeding the vapor pressure of the sample liquid to prevent evaporation and deterioration during storage.

Samplers also are used with positive-volume and positive-displacement meters in well testing and wet-oil (not pipeline quality) lease volume measurements. Representative sampling becomes more difficult with increasing water content in the oil stream. For improved accuracy, fluid mixtures that may have both free water and oil emulsion (oil-external phase) components should be processed through a three-phase separator before the remaining oil emulsion stream is sampled.

The capacitance probe and net-oil computer have not replaced the automatic sampler on most LACT installations because (1) crude oil value frequently is based on gravity (determined on sample volume), (2) the gross-oil volume available directly from LACT counters is satisfactory for daily operating needs, and (3) the potential operating cost reduction by eliminating sampler use is not significant. Most major purchasers will have a recommended (or required) design for automatic sampler installation.

BS&W Monitor Pipelines specify the maximum BS&W content that a crude oil may contain to be acceptable for transfer. The development of LACT equipment required a means to monitor the quality of crude oil as it was being measured and transferred automatically to a pipeline. The commonly accepted device for this function is a capacitance probe BS&W monitor. The dielectric constants of crude oil and water are about 2 and 80, respectively. The BS&W monitor uses a concentric probe, which senses the apparent dielectric constant of a fluid stream by measuring its capacitance between the probe electrodes.

The capacitance probe generally is installed in a vertical riser on the premise that a more uniform mixing of stream components will result and thus, that a more accurate sensing of water content is ensured. Temperature compensation is required since dielectric constants of oil and water vary with temperature. The BS&W monitors used with LACT generally have a 0 to 3 % BS&W range. The monitors have a variable set-point that is used to divert the oil stream back through the treating facilities when the selected BS&W content is reached and/or exceeded. In this application, a BS&W monitor controls the crude oil stream being transferred to the pipeline, but it does not determine the BS&W content as basis for gross volume adjustment.

Net-Oil Computer The application of SCADA systems to production operations increased the need to read directly the net-oil content of an emulsion stream for well testing and lease oil production measurements. The basic principle of the capacitance probe used in BS&W monitors was extended to oil emulsions of higher water content by probe redesign.

The capacitance probe, with modification to give a linear output, provides a technique to obtain instantaneous value of water content in an oil emulsion stream. By combining the probe output with volume output from a positive-displacement or turbine meter, the net-oil and


water volumes in an emulsion stream can be determined. The device that combines the capacitance probe and meter volume information to obtain net-oil and water volumes is designated as a net-oil computer.

The net-oil computer can determine the oil/water content of an emulsion stream with reasonable accuracy (I to 2% in oil measurement to about 35% water). The capacitance probe will continue to indicate water content above this value (if in an emulsion form), but the decreasing oil percentage of the total stream causes increased error in the measured oil fraction. A limitation of the capacitance probe is that any “free” water moving across a probe with oil emulsion will distort the indicated water content and cause substantial errors. Applications with water cuts above 35% can be measured by using three-phase separators (for well testing and/or lease oil production). Treating chemicals can be used with three- phase separators, if necessary, to keep water content in the emulsion to 35% or less. With these procedures, well testing and lease oil production can be processed with total water production of more than 99 %

Supervisory Control and Data Acquisition (SCADA) Systems SCADA systems applied to oil and gas producing operations vary in function and overall capability. Some systems am oriented primarily to local operating personnel needs and may only monitor/control a few wells in a single field. Other systems are applied to multiple fields that have several thousand total wells. Even with the wide variations, SCADA systems consist of the following basic elements: (1) supervisory control/data acquisition equipment, (2) field instrumentation and cabling systems, (3) communication facilities, and (4) digital computer systems.

SCADA Equipment

This equipment functions to interconnect digital computer systems and instrumentation and control devices that are related to the oil and gas producing process. The equipment consists of a communication adapter and RTU’s. A communication adapter is attached directly to the digital computer by a high-speed data link and attached indirectly to RTU’s by communication circuits. The communication adapter serves as a data concentrator for the digital computer, which can drive several communication circuits simultaneously. A number of RTU’s generally share a common communication circuit.

The communication adapter and the RTU’s are designed to use a specific message protocol or structure for the transmission of information. This message structure generally will contain the RTU address, function being requested, function modifiers, and supplemental information used for checking validity of message transfer. A primary feature shared by the communication adapter and the RTU’s is the checking for valid message transmission between the devices. Manufacturers use different techniques to ensure receipt of valid messages and most of these have a high probability of detecting invalid transmissions. Earlier communication adapters were designed with “hard-wire” logic and were separate devices. Many of the current designs are microcomputer- based units that may be an integral part of the digital computer rather than a separate unit. These units fre-

quently can handle multimessage protocols to allow mote flexibility in attached devices.

RTU’s are electronic devices that connect the SCADA system directly to the oil and gas production facilities that are being automated. An RTU has the capability to store information from several input points and to transmit this information in a serial mode over a single communication circuit to a digital computer on demand. The RTU also may receive control information from the computer that it routes to a selected control point. The RTU generally is located within a few thousand feet of its connected instrumentation and control equipment but may be up to several hundred miles from the computer location.

RTU’s commonly sense input information related to status/alarm (on-off, etc.), volume accumulation (oil and gas meter counts), and instantaneous analog values (temperature, pressure, flow rate, etc.). The RTU can provide control output in the form of relay activate (on- off, start-stop) and a set-point value. The set-point is commonly a 4 to 20 mA or 10 to 50 mA signal that is compatible with control devices. Although these basic RTU capabilities may appear limited, most data input and control output functions related to oil and gas production can be implemented directly. Some specific functions may require supplemental logic in local control panels for implementation.

The basic RTU functions described previously are common to units designed with “hard-wired” logic. Many of the current and most future RTU designs will be microcomputer-based. The microcomputer allows substantial increases in RTU functions with relatively small incremental hardware expansion and cost. For example, RTU’s are being used to replace stand-alone gas- flow computers simply by adding to the microcomputer program logic that integrates static pressure, differential pressure, and temperature transducer data from an orifice meter. The microcomputer in the RTU can handle all basic RTU functions and process gas-volume accumulation for 30 to 50 m runs without timing constraints. With continuing developments in electronic technology and improving program support, the outlook is for significant expansion in RTU Capability.

Initial RTU designs that incorporated microcomputers tended simply to implement “hard-wired” logic in the microcomputer. In addition, microcomputers used with RTU’s generally did not have any internal error checking such as parity on memory, data bus, and address bus transfers. These conditions resulted in microcomputer- based RTU’s that were difficult to trouble-shoot when malfunctions were experienced. Self-diagnostic capabilities should be the first functional extension of microcomputer-based RTU’s. Good diagnostic features usually will require a combination of extended hardware and software. Diagnostic checks should be run at startup and at frequent intervals during the regular operation of the RTU. RTU’s should have a “watch-dog” timer that will attempt to restart the microcomputer if it stalls as a result of hardware and/or software malfunction. The restart feature also should provide a means to document its use.

The RTU as an electronic device has two primary environmental factors that adversely affect operational reliability. These factors are heat and electrical tran-


sients. All solid-state electronic components suffer decreased life with increasing operating temperatures. Many users of RTU’s routinely have placed these units in air-conditioned buildings to decrease the mean time between failures. More recent introduction of low-power electronic components such as complementary metal- oxide silicon (CMOS) may minimize the need to provide cooling below normal ambient temperatures.

Solid-state electronic components are subject to damage by extremely short-duration (microseconds) higher voltage electrical transients (e.g., noise, spikes). These transients may enter an RTU by means of primary power or by the many signal loops connecting the RTU to the instruments and control devices associated with the production process. RTU’s frequently are provided battery backup power to allow for continued operation with primary power outages. Common procedure is to operate the RTU continuously on battery power with primary power driving a battery charger. This arrangement provides relatively good isolation for the RTU from power line transients.

All RTU input/output connections to the field cable system also must be protected from voltage transients. Status/alarm and accumulator input points frequently use optical isolation between field cables and internal circuitry. Each field signal loop also should use a gas tube (or similar device) to route induced voltage transients to earth ground. Lightning can induce sufficient energy to literally evaporate protective components in extreme cases but many of the otherwise damaging transients can be suppressed with proper protection equipment. Elec- trical noise also may be caused by inductive devices such as relay and solenoid valve coils. Inductive components should have suppression diodes placed across the input terminals to minimize electrical transient generation.

Field Instrumentation and Cabling Systems

Field instrumentation and control devices are selected primarily to meet needs of the oil and gas producing facilities. If these devices are also to interface with the RTU of a SCADA system, some additional features may be required. RTU status/alarm and accumulator inputs require an electrical switch closure to convey information to the RTU. For example, a separator high-level float control may have a pneumatic switch for control valve activation. An electrical pressure switch may be added to the pneumatic control line to allow the RTU to sense the position of the high-level control. In general, reliability of interface increases when the primary instrumentation has a direct electrical connection compared to the example that required a pneumatic to electrical conversion.

Meters for liquid and gas measurement also need to have an electrical switch activation to indicate some increment of volume accumulation. Liquid meters, for example, frequently will have an electrical switch closure at nominal 1-bbl volume increments. The switch closure must be maintained for some minimum time increment (about 50 milliseconds) to ensure that electrical tmn- sients will not cause invalid volume counts. The RTU will have a separate signal loop (wire pair) and internal electronic counter associated with each meter being monitored.

Pressure, temperature, flow rate, position, and similar

operational parameters are sensed by the RTU as an analog input value from electrical transducers. Nominal electrical transducer output ranges of 1 to 5,4 to 20, and 10 to 50 mA usually can be processed by the RTU. A current output transducer is preferred since the signal is less susceptible to electrical transient distortion than a voltage output.

Multiconductor cables are used to connect the RTU to the instrumentation and control devices associated with the production process. The signal cables consist of individually insulated copper wires genemlly of 19 or 22 gauge that commonly are used for commercial telephone service. The cables usually are buried to minimize probability of mechanical damage and electrical noise intm- sion. Cables can be damaged when buried and repairs should be made immediately to prevent further deterioration. Cable lengths are controlled by cost relationships between installing more RTU’s at separate locations and the amount of cable required for interconnection. Most cable systems will be limited by economics to a few thousand feet.

Radio communication links between RTU’s and a central location within a field can reduce overall cabling costs substantially. This technique can use a master radio station in a “polling” mode to communicate with the individual RTU’s. Separate transmit and receive frequen- cies enhance the radio communication procedure. In this arrangement, the master radio transmits an outbound message to all RTU’s and the RTU’s will decode the address portion of the message. The addressed RTU then will decode the function portion of the message and reply to the master radio by bringing up its transmit radio link for the period required by the return message. The master radio is linked to a regular four-wire communication circuit for communication outside the field area.

The low-energy signals used in the cable system require careful connection of wiring to instrumentation. Any damage to wiring insulation or collection of moisture at connection points may result in sufficient signal leakage to cause invalid sensing of operational information. Some installations are made with the signal loop power supply “floating” or not connected to earth ground. Maintenance checks then can be made to determine current leakage to the positive and negative terminals of the power supply from earth ground as a measure of signal leakage. With selective isolation of the signal loops, leakage problems usually can be identified. Common practice is to design alarm-signal loops to be in normal condition when the sensing device has a closed electrical switch. This feature makes the alarm signal more fail-safe in that any failure of the signal path will cause indication of an alarm condition also. Less critical “status” loops frequently are open in the normal state to minimize total power needs.

Communication Facilities

SCADA systems require capable and reliable communication facilities to connect the communication adapters on the digital computer system with the RTU’s that are located in fields being automated. Most SCADA systems use dedicated or nonswitched communication circuits that have a four-wire configuration. The four- wire designation provides two independent communication paths that will support simultaneous data transfer in

16-10

CHRISTMAS TREE

-I I- a EXCESS FLOW

SHUT-IN CONTROL VALVE

c-- I

-1 -c%+

I

b---f

b HI-LOW PRESSURE SHUT-IN

CONTROL VALVE

-,z,E

c AUTOMATIC CONTROL VALVE WITH

PRESSURE SWITCH

Fig. 16.1-Automatic wellhead safety controls.

two directions (full duplex). Data transmission in SCADA generally is operated in one direction at a time (half duplex) with transmit over one path and receive over the other path but at different time periods. Data are transmitted over the communication circuits with the aid of modems. Modems are electronic devices that convert voltage or current level from the communication adapter into analog signals that can be transmitted over the communication circuit. A similar modem at the RTU receives the analog signals and converts them into voltage or current levels that are compatible with the electronic circuitry in the RTU. A data transmission from an RTU to the communication adapter operates in the reverse cycle.

SCADA can use commercially available voice-grade (telephone) communication circuits. Communication speeds of 1,200 bits/see can be used with long term reliability over these circuits by using conventional modem equipment. Higher communication rates can be obtained by using more complex modems and higher quality communication circuits if the application being monitored requires more critical timing responses. In- creased communication rates tend to require more maintenance time or tine tuning of communication equipment to achieve a constant reliability of performance.

SCADA communication circuits may require multiple ownership and maintenance responsibility links to reach particular locations. Circuits with divided maintenance responsibility tend to have more reliability problems. In these cases, adequate test facilities at the computer site can aid in defining the particular link causing problems. The four-wire circuit is compatible with testing since it can be turned around to receive a transmitted test signal from the computer site. If remotely operated turn-around


devices are located at each link interconnection, the circuit can be tested and problems isolated to the responsible maintenance group.

Digital Computer Systems

SCADA became possible with the development of process-control-type computer systems in the late 1950’s. The process control computer was a special hardware implementation that provided for direct connection to plant instrumentation and control equipment. This same hardware also provided a means to interconnect with the communication adapter of a SCADA system. The SCADA equipment then allowed the process control computer to be connected indirectly to oil and gas facility instrumentation at remote locations.

Process control computers had software operating systems with program execution control that was compatible with SCADA needs. The operating systems were designed to recover automatically from minor program malfunctions to minimize computer operator needs in the continuously operating SCADA applications. Early process control computers had limited memory size that tended to require assembler language routines for some functions. Frequently, bulk storage (disk or drum) devices also were limited in capacity. Most earlier SCADA systems were based on a central computer system concept.

Currently, SCADA systems have extremely capable computer systems available for system implementation. Process-control-type computers are available with few practical limitations on memory size, speed of execution, and random access storage capacity. In addition, many of the general-purpose-type computers (business and scientific applications) now have extensive communication capabilities and operating system features that are compatible with most SCADA requirements. Distributed computing, which uses multiple computers at remote sites, became practical in the mid- to late 1970’s as manufacturers developed capable network software. The network software provided capability to develop application programs at central locations and download through communication links into computer systems at field locations. Increased use of distributed computers in SCADA is anticipated.

Basic application software is generally available from manufacturers of the SCADA systems. The rapidly changing technology, however, has made it difficult for suppliers to develop general purpose application software that is transportable to new computer systems. Most larger SCADA systems have tended to be specialized software applications that have continuing program development to meet the changing operational needs. Application software for both small and large SCADA systems should have flexibility to allow changes/additions to the implemented functions.

Typical Automatic-Control Installations

Automatic Well Control

Automatic Wellhead Controls. Wells may be controlled at the wellhead or at the well manifold. Frequent- ly, it is necessary or desirable to control them at both places. Fig. 16.1 depicts three different types of


i

TUBING PRESSURE GAUGE TIME CYCLE CASING PRESSURE GAUGE CONTROLLER

t-GAS METER 7

Fig. 16.2-Automatic controls for rod-pumped wells. Fig. 16.3-Typical automatic control of gas-lift well.

places. Fig. 16.1 depicts three different types of automatic control valves that could be installed at the wellhead immediately adjacent to the tubing wing valve. Although Fig. 16.1 pictures a naturally flowing well, the same types of automatic controls would be applicable for artificially lifted wells of all types, if required. The excess-flow valve shown in Fig. 16. la generally is used only to protect against flowline breaks when the wells ate choked and controlled at the well manifold. The high/low pressure shut-in valve in Fig. 16.lb may be used whether the well is choked at the wellhead or at the well manifold. It is protection against both flowline breaks and chokes cutting out or plugging. The control valve and separate pressure-sensing element shown in Fig. 16.1~ perform exactly the same functions as the high/low pressure shut-in valve.

Rod-Pumped-Well Control. The typical automatic- control system for a rod-pumped well is shown in Fig. 16.2. The high-low pressure safety shut-in valve is necessary only when the well has a tendency to flow natumlly when the pumping unit is shut down. The excess-flow valve, again, is protection against flow-line breaks. Some operators use them; others do not. They am not always effective unless line pressures are high enough, and the size of the break large enough, to create a substantial pressure drop. The pressure switch is the most common automatic safety control used with rod- pumped wells, particularly where the wells are remotely controlled. Regardless of which of these three types of controls are used, when the control pressure is reached, that automatic control must furnish a signal to shut down the pumping unit. This is accomplished by grounding the magneto of a gas engine or shutting off an electric motor. A pump-off control, if used, would be installed in the flow line immediately adjacent to the pumping tee.

Gas-Lift-Well Control. Fig. 16.3 shows a typical arrangement of controls on the gas-supply line to a well that is produced by an intermitter-type gas-lift installation. The “time-cycle controller” shown on the right is an automatic production programmer. It automatically opens and closes the diaphragm control valve, to which it is connected by instrument lines, according to the schedule manually created in the programmer. The “flow-control valve” is a manually actuated valve used to control the mte at which gas is admitted to a well. Automatic-control valves on the flow line, if required, would be one of the types shown in Fig. 16.1.

Hydraulic-Pumped-Well Control. Fig. 16.4 depicts the typical automatic controls for a hydraulic pumping

POWER OIL TANK m

CONTROL PANEL

OW PRESSURE T-DOWN SWITCH

MANIFOLD TO SUMP

Fig. 16.4-Typical hydraulic-pumping-system control.

system. A high/low pressure switch protects the triplex pump and its prime mover against overloading from ab- normally low suction pressures and/or high discharge pressures. In either case, the pressure switch would shut down the prime mover. Automatic Control Valve V-l in the manifold bypass is generally a diaphragm-type regulator valve which is normally closed. It would open at a pressure slightly under the setting of the high/low pressure shutdown switch and divert sufficient power oil back to the power-oil tank to maintain system pressure at a safe level. Automatic Control Valves V-2 and V-3 are in the power-oil lines to individual wells. They would be closed automatically in the event the pressure switch shut down the triplex and prime mover. With an appropriate programmer, they could also be used to produce individual wells selectively on an intermittent schedule.

Automatic Well Manifolds. In Figs. 16.5 and 16.6 are shown two typical automatic well manifolds designed for controlling the wells at the well manifold rather than at the wellhead. The single-wing well manifold shown in Fig. 16.5 has a maximum of flexibility to meet any system of stage separation and/or treating that might arise. It has another advantage in that it could be installed initially on new leases as they are developed without the automatic-control valve and still add the valve later with a minimum of expense. The dual-wing well manifold shown in Fig. 16.6 is limited to situations where all production in a well manifold is processed through a single vessel or a common sequence of vessels. Both these manifold designs would require flow lines capable of withstanding full wellhead pressure and


PRESSURETE (fti~~~,

F;;;a TESd d i PRODUCTION’ I’

ONE REQ’D FOR EACH SEPARATION AND/OR

TREATING VESSEL

Fig. 16.5-Typical single-wing automatic well manifold.

PRESSURE GAUGE 3-WAY 3-POSITION CONTROL VALVE

PRODUCTION

Fig. 16.6-Typical dual-wing automatic well manifold.

would choke and control the wells at the manifold. All manifold designs should provide for monitoring valve leakage.

Some operators prefer to use three-way, two-position control valves in the well manifold and to control the wells at the wellhead. Automatic-control valves of the types shown in Fig. 16.1 then would be installed at the wellhead in addition to the control valves in the well manifold. Other operators would prefer to use a two-way control valve in each riser to each pressure vessel served by the well manifold.

Automatic Well Testing A typical automatic well-testing system is illustrated in Fig. 16.7. The sequence control logic for conducting the test and for calculating test results may be self-contained in the control panel or it may originate in a SCADA system that is remote fmm the site. In either case, a three-way control valve in the test/production manifold is activated to divert the selected well to the test vessel when the related signals are received from the control panel. The test vessel may be either a separator or a heater treater. The liquid-metering elements commonly will be either positive-displacement or turbine meters. The oil meter will be combined with a capacitance probe and net-oil computer to provide measurement of net-oil and emulsion-water volumes. The free water volume will be measured by a separate water meter. Well-test control logic will combine the emulsion-water and free water measurements to obtain total test water volume.

Fig. 16.7-Typical automatic well-testing system

The test-gas volume can be determined with a positive-displacement meter, turbine meter, or gas-flow computer. The three-way control valves should be equipped with position switches that can be used to confirm that only the selected well is in the test vessel. The test vessel normally will have a high-level float switch that will automatically divert the on-test well back to production status when a high liquid level is detected. The same switch can be used to notify the SCADA system of the test vessel malfunction.

Some operators have extended the use of measurement equipment, as described for well testing, to total lease production. For example, with more logic in the control panel and manifolding of dump lines from test and production vessels, the oil-measurement equipment can be time-shared to measure both test and production oil volumes. The control panel determines (one at a time) which vessel dumps through the oil-measurement equipment and mutes the volume counts to the related counter. Separate gas-measurement facilities are required for test and production volume determination.

The time-sharing concept has been extended further in cettain instances. For example, a number of separate leases, each having only a few wells, were arranged to have all production separators located at a common site. The oil production from each separate lease production separator was determined by time-sharing a single oil- measurement facility. In addition, a single test separator was used to test all wells at the site. The site control panel muted test volume counts to the test counters and to the appropriate lease counters related to the well on test.

LACT

The first efforts to design an acceptable LACT installation tried to incorporate the existing equipment on the lease and familiar operating principles insofar as possible. Thus, understandably, the first officially accepted LACT system was a weir-tank system installed by Gulf Oil Corp. on its Ames Lease in the Bloomer field, KS, in 1955. Shell Oil Co. also pioneered in the development of the weir-tank-type LACT system on its leases in the Antelope and Wasson fields in Texas, beginning unof- ficial experiments as early as 1948. The next significant development in LACT system design was the meter- tank-type system.


Fig. l&8-Typical weir-tank-type automatic-custody-transfer system.

The meter-tank-type system is closely akin t6 the weir- tank system in many respects, but perhaps it deserves consideration as a separate category because these vessels were designed solely as measuring vessels. The great similarity between measurement with the larger- sized positive-volume dump meters and conventional lease tanks assured their early acceptance. Officially accepted LACT installations of this type of the Phillips Petroleum Co. and Amoco in Oklahoma and Texas followed closely on the heels of the Gulf weir-tank-type system in Kansas. The third type of LACT design which has gained wide acceptance to date is the positive- displacement-meter-type system. Exxon Co. U.S.A. was perhaps the strongest early proponent of this type of system, and they performed much development work on their leases in south Texas.

The positive-displacement-meter-type equipment rapidly became the LACT standard primarily because of substantial cost and operating reliability advantages over the other two units. The positive-displacement-meter LACT assembly was skid-mounted for relatively simple installation within existing production facilities. Figs. 16.8 and 16.9 are included for reference to the historical development of LACT as an important element of lease automation.

EMERGENCY HIGH LEVEL FLOAT SW ITCH

t TRANSFER PUMP&- L-^. r-

GAS ELIMINAT”R”’

RETURN LINE DIVERTING VALVE’

Fig. 16.9-Typical automatic-custody-transfer system using metering dump tanks.

Positive-Displacement-Meter LACT System. Fig. 16.10 shows a positive-displacement-meter LACT installation. Option B indicates a two-meter arrangement that was recommended strongly in earlier installations for improved measurement reliability. Operational experience found the single meter arrangement (Option A) to have satisfactory reliability and most LACT’s were installed with the single meter. Some LACT’s have two meters but most of these operate the second meter in a stand-by mode rather than simultaneous measurement through both meters. Most major purchasers have recommended (or required) certain designs for LACT installations.

The routine operation of crude-oil transfer is controlled by the normal operating high-level float switch with an override to shut down on emergency with the low-level float switch. The BS&W monitor will divert fluid stream to the treating facility on detection of high BS&W content. The strainer is used to protect the meter from particles that could damage the meter. The meter is temperature-compensated to indicate oil volume at 60°F standard condition. Daily and monthly volume limit switches prevent overrun of lease on either daily or monthly allowable volumes. A calibration loop is located downstream of the meters to allow convenient

OPTION “A’ SINGLE PD METER SYSTEM

SAMPLER CALIBRATION

,TICKET PRINTER 8 COUNTER

MlLY VOLUME LIMIT SWITCH

MONTHLY VOLUME LIMIT SWITCH

OPTION ‘8”

TWO P 0 METERS IN SERIES

Fig. l&10-Typical automatic-custody-transfer systems using positive-displacement meters.


Fig. 16.11-Automatic backwash of rapid sand nlters in water- treatment plant.

Fig. 16.12-Typical method of controlling injection-pumping rate.

meter-proving or calibration with a master meter. A sampler is used to collect sample volumes for basis of BS&W correction and oil gravity determination. The back-pressure valve is used to minimize gas break-out or “flashing” in the crude stream prior to being metered. A check valve (not shown) should be located downstream from the back-pressure valve (or combined with it). API Standards 1101, 2502,253 1,2542,2544, and 2546, and supplements thereto provide recognized industry codes for liquid petroleum measurement. 2

Automatic Lease Process Control

There are many normal lease-operating processes that may be automated or that already are automated completely and not recognized as such by the average individual. Space limitations will permit only a brief look at a few of these control systems to show what can be and is being done.

Automatic Water-Treating Plant. Fig. 16.11 depicts a rather elaborate automatic backwash system for rapid sand filters in water-treating plants. Normal flow is from the precipitator and oil remover through the sand filters into the accumulator. When the differential pressure across the bank of filters reaches a predetermined value, the backwash cycle is initiated automatically. Valves

Fig. l&13-Automatic control of water-supply wells.

P-3, F-l, F-3, F-4, and A-l are closed; Valve A-2 is opened; and the backwash pump is started. After a slight delay, Valves F-l and B-l are opened and Filter 1 is backwashed into the backwash-settling tank for a predetermined, but adjustable, time. At the end of this time, Valves B-l and A-2 are closed and Valves P-3 and F-5 arc opened. Normal flow is permitted through Filter 1 into the backwash-settling tank for a short interval to settle the filter bed. At the end of this interval, Valves P-3 and F-5 are closed, Valves A-2 and F-4 are opened, and the backwash pump is started. After a slight delay, Valve F-3 opens and the same cycle is completed for Filter 2.

Four float switches are required in the accumulator. Float Switch FS-1 is an emergency low-level float switch, which maintains a flooded pump suction and prevents gas from locking the pump since the accumulator normally will be gas-blanketed. The volume between Float Switches FS-1 and FS-2 is large enough to backwash the filters. Normally the fluid level is not permitted to fall below FS-2. If it should, the injection pump is shut down if it has an electric prime mover, or Valve A-6 is opened and the injection-pump output is bypassed back into the accumulator if the injection pump has a gas-engine prime mover. When the fluid level reaches Float Switch FS-3, the normal injection process is resumed. Float Switch FS-4 is an emergency high- level control that would cause Valve S-l to close to prevent running over the accumulator.

Valve A-3 is shown as a backpressure valve that is intended to keep a constant head on the backwash pump. The backwash pump probably will be a centrifugal pump because of the high rate normally recommended for proper backwashing. If the injection process ceases for the backwash cycle, then gas pressure applied to the top of the accumulator can be used in place of the backwash

pump. After all the filters have been backwashed and settled

into the backwash-settling tank, the solids in the water are permitted to settle out in the bottom of the tank. Then Valve B-2 is opened automatically to permit the solids to be washed out to the pit. Then Valve B-2 closes, Valve B-3 opens, and the transfer pump returns the clear water back through the system. Float switch FS-5 is for the purpose of maintaining a flooded suction on the transfer pump, and Float Switch FS-6 is to prevent running over the vessel.

The system just described represents a composite situation. All automatic water-treating plants would not have to be this elaborate. For example, a backwash- settling tank would be desirable only where it was necessary to keep the water going into the pit at a minimum. On the other hand, by a few small changes,


D/P CELL B - REGENERATION FLOW

Fig. 16.14-Automatic cycling of desiccant beds in dry-desiccant-type gas dehydrators

the systems shown in Fig. 16.11 could also be altered to provide for continuous through-put and injection while one of the filters was being backwashed.

Automatic Control of Injection-Pumping Rate. Fig. 16.12 shows a typical method of controlling the injection-pumping rate of a gas-engine-driven injection pump. A float switch and pilot valve, acting through a snap-active pneumatic relay, controls the position of (1) a low-level bypass regulator, and (2) a bellows-operated pneumatic motor that adjusts the engine throttle linkage relative to the position of the float switch in the clean- water tank. In the event the fluid level drops too low in the clean-water tank, the pilot valve sends a signal to a low-level shutdown switch, which grounds the magneto on the engine and shuts it down.

Automatic Control of Water-Supply Wells. Fig. 16.13 shows a method of using float switches to control the operation of several water-supply wells to maintain an adequate supply of water in the raw-water tank. As always, high- and low-level emergency float switches are provided. Additional float switches are included to program the addition and subtraction of supply wells feeding the system to assure an adequate volume of water is always in the raw-water tank. The well-on float switches would be actuated by a dropping fluid level, and the well-off float switches by a rising fluid level.

At the water-supply wells, there are several ways to control the pumping equipment. If the pump is driven electrically, it might be necessary to start up and shut down the pump motor only as called for. If the water- supply well were artesian or had a tendency to flow naturally, it would be necessary, of course, to furnish a shut-in valve. If the pump is driven by a gas engine, it would be necessary either to provide the engine with an electric ignition system (battery or electric motor) and a startup sequence programmer or to install a diverting valve and leave the engine running while diverting production back into a casing annulus.

Automatic Control of Dry-Desiccant-Type Gas Dehydrators. As a final example, let us consider a lease

process that has been fully automatic since its inception, but one which is rarely thought of in terms of automation: the automatic cycling of desiccant beds in dry- desiccant-type gas dehydrators. Fig. 16.14 is a schematic of a typical dry-desiccant-type dehydrator. The wet gas stream enters the horizontal separator and is divided, with a part of the gas going to the regeneration stream and the remainder continuing in the main gas stream through the dehydrating tower. The proportioning of the flow between the two streams is controlled by the regeneration-rate controller. The rate of flow in the regeneration gas line is measured by a differential- pressure cell and transmitted to the regeneration-rate controller. The regeneration-rate controller, in turn, acts to position Automatic-Control Valve V-10 to maintain the predetermined rate of flow of gas through the regeneration system.

Automatic-Control Valve V-l 1 in the regeneration gas line is controlled by Controller B. Each time the controller rotates until the pin on the next trip clamp unlatches the pilot arm, Valve V-l 1 reverses its position. In the one position, it diverts gas through the heater to provide hot gas for expelling moisture from the desiccant bed of the tower being regenerated. In the other position, it bypasses the heater and provides unheated gas to cool the desiccant bed in the same tower before placing it back in- to service. By the time the controller rotates one more position, the regeneration valves on the towers will have switched so that the hot gas goes to the other tower.

Automatic-Control Valves V-l, V-2, V-3, and V-4 control the flow of regeneration gas to dehydrating towers, and Valves V-5, V-6, V-7, and V-g control the flow of the main gas stream. Valves V-l, V-3, V-6 and V-S are always in the opposite position from Valves V-2, V-4, V-5, and V-7. The main-stream valves and the regeneration-stream valves are manifolded in such a manner that only one tower at a time receives the main- stream gas and that tower is blocked off from the regeneration gas. The other tower receives the regeneration gas and is blocked off fmm the main-stream gas. The position of all these valves is controlled by Con- troller A acting through relay Valve V-9 and the pilot-


gas-control manifolds. Each time the controller rotates until the pin on the next trip clamp unlatches the pilot arm, instrument gas will flow through the bleed orifice instead of flowing to the Relay Valve V-9. Relay Valve V-9 then will reposition itself and allow instrument gas to flow to the other pilot-gas-control manifold and vent the gas from the pilot-gas-control manifold supplied in its original position. This, in turn, will cause each of these control valves to reverse their positions, and thus the flows of main and regeneration gas streams will also reverse. The length of each cycle is controlled by the spacing of the trip clamps on the controller.

References 1. API Manual of Petroleum Measurement Standards, first edition,

API, Dallas (1981) Chaps. 1, 5, and 6. 2. “Specifications for Lease Automotive Custody Transfer (LACT)

Equipment,” second edition, API Spec. 1 lN, API, Dallas (March 1979).

General References Anderson, G.L. and Reed, G.A.: “Automation in the South Swan

Hills Unit,” J. Cdn. Pet. Tech. (Oct.-Dec. 1981) 20, 105.

Atkinson, M.H., and Newberg, A.H.: “Cmde-oil Measurement 1s Going Automatic.” Oil and Gas J. (June 4, 1956) 102.

“Automatic Custody Transfer.” Oil and Gas J. (July 11. 1956) 110.

“Automatic Sale Slated,” Oil and Gas J. (Feb. 13, 1956) 90.

“AutomatIon,” Per. Week (Nov. 16, 1956) 71

Barrett, M.L. Jr.: “Meter Proving,” Oil and Gas J. (Feb. 24, 1958) 153 (March IO, 1958) 201 (March 24, 1958) 213 (April 21, 1958) 179 (May 5, 1958) 133.

Bayless, C.R., and Mlkeska, F.J.: “Automatic Control of Produc- tion,” Oil and Gas J. (June 4, 1956) 78 (June 11, 1956) 129 (June 25, 1956) 110.

Beach, F.W.: “Fail-safe LACT Unit; Here’s How It Works,” World Oi[ (Nov. 1957) 133.

Brainerd, H.A. and Piros, J.J.: “New Controller Recorder Gravitometer,“ Oil and Gus J. (Dec. 2, 1957) 78.

Case, R.C. and Fritsch, D.R.: “Automation for the Ekofisk Offshore Operation,” paper presented at the 1976 Automation in Offshore Oil Field Operations Symposium, Bergen, Nonvay, June 14-17.

DeVerter, P.L. and Scovill, W.E.: “Pan I: Continuous Automatic Sampling,” Oil and Gas J. (April 2, 1956) 125. Warren, F.H.: “Part 11: Continuous Automatic Sampling,” Oil and Gas J. (April 9, 1956) 124. Johnson, R.P.: “Part III: Continuous Automatic Sampling,” Oil and Gas J. (April 23, 1956) 119. Berglund, J.H.: “Part IV: Continuous Automatic Sampling,” Oil and Gas J. (April 30, 1956) 210.

Doble, P.A.C.: “Computer-Assisted Operations in a Northern North Sea Operation,” J. Pet. Tech. (April 1983) 701-08.

Dunham, C.L.: “A Distributed Computer Network for Oilfield Com- puter Production Control, J. Pet. Tech. (Nov. 1977) 1417-26.

EnDean, H.J.: “Oil Field Watchman Checks BS&W Content,” World Oil (Nov. 1957) 151.

“First LACT System for Low Gravity, Viscous Crudes.” Oil and Gas J. (Dec. 2, 1957) 82.

Foster, K.W.: “Centmlia Water Flood: Preplanned Automation Pays Off,” Pet. Engr. (March 1958) B-l 16.

Hebard, G.G.: “Automatic Lease Custody Transfer,” Oil and Gas J. (Nov. 5, 1956) 86.

Hill, R.W.: “Factory-built LACT Unit Is Gas Operated,” Oiland Gas J. (May 6, 1957) 98.

Hubby, L.M.: “Automatic Production Controls,” Paper API 926-l-C presented at the 1956 Southern District Spring Meeting, Division of Production, San Antonio, Mar. 9.

“LACT: A Youngster Now, Soon a Giant,” Oil nnd Gas J. (Sept. 22, 1958) 74.

“LACT Is for Stripper Leases, Too,” Oil and Gas J. (Dec. 15, 1958) 70.

“Lease Automatic Custody Transfer,” Bull. 2509A, API, Dallas (Aug. 1956)

“Lease Is Fully Automatic,” Pet. Week (Feb. 15, 1957) 11

LeVelle, J.A.: “New Production Programming System,” Pet. Engr. (April 1956) B-30.

Matheny, S.L. Jr.:“Computer Production Control Expands,” Oil and Gus J. (March 23, 1981).

McGhee, E.: “Automatic Switching, 9-mile Radio Link,” Oil and Gus J. (Sept. 10, 1956) 114.

McGhee, E.: “How Cities Service 1s Using P.D. Meten for LACT,” Oiland Gas J. (Jan. 13, 1958) 74.

McGhee, E.: “How Shell’s Antelope LACT Works,” Oil and Gas J. (June 3, 1957) 90.

McGhee, E.: “It’s Automatic-Even to Sample Taking,” Oil and Gas J. (Feb. 13, 1956) 104.

McGhee, E.: “LACT-And Why We Like It,” Oil and Gas J. (Jan. 20, 1958) 131.

McGhee, E.: “When a Field Ourgrows Its Facilities,” Oil and Gas J. (Apr. 15, 1957) 108.

McKinley, D.C.: “P.D. Meters Get the Job Done,” Oil and Gas J. (Oct. 1, 1956) 87.

Meyers, D.C.: “How Shell Designs an Automatic Lease,” Oil and Gus J. (Oct. 17, 1955) 111.

Northern, T.P.: “Automatic Lease Operations-Weeks Island Field,” J. Pet. Tech. (Jan. 1954) 21-24.

Packard, H.C., Kelley, H.S., and Newburg, A.H.: “Automatic Custody Transfer of Crude Oil. Part I: General Considerations. Part 11: From the Producer’s Viewpoint. Part III: From the Pipeliner’s Viewpoint,” paper presented at the 1956 API Annual Meeting, Chicago, Nov. 13.

Patterson, D.R.: “Production Automation,” Pet. Engr. (Jan. 1959) B-31.

Pope, S.H. and Stutz, R.M.: “Lease Automatic Custody Transfer Becomes a Reality,” Oil and Gas J. (April 23, 1956) 96.

“Pneumatic LACT System,” Pet. Engr. (Jan. 1957) B-104

“Principles of Lease Automation,” Pet. Equipment (1957) 21

“Production Automation Forges Ahead,” Pet. Wee& (July 26, 1957) 21.

Reese, C.P.: “Automatic Control of the Wattenberg Gas Field-Colorado,” paper SPE 11111 presented at the 1982 SPE An- nual Technical Conference and Exhibition, New Orleans, Sept. 26-29.


Resen, L.: “Humble Tries LACT, Gives A Stamp of Approval.” Oii and Gas J. (March 4. 1957) 94.

Saye, H.A.: “Automatic Well Testing, ” CM and Gus J. (Jan 6. 1958) 102.

Scott. J.O.: “Automation Pays Off in Big Mineral Producing Open- tions,” Oil and Gus J. (Sept. 19, 1955) 114.

Scott, V.B.: “Automatic Lease Operation.” paper presented at West Texas 011 Lifting Short Course, Texas Technological C., Lubbock. April 11-12, 1957.

Shatto, H.L.: “Comments on the Status and Future of ACT from the Production Viewpoint,” paper presented at the 1958 ASME Mechanical Engineering Conference. Denver, Sept. 21-24

Shatto, H.L. and Hall, A.H.: “Greater Rewards from LACT.” Oii and Gas J. (April 7. 1958) 133.

Stormont, D.H.: “Tank Bottoms Are Recycled,” Oil and Gas .I. (Nov. 12, 1956) 173.

Taylor, D-M.: “New Auto-pneumatic Lease Programming System,” Per. Engr. (Dec. 1956) B-28.

Todd, M.: “Automation Applied to Flooding at Naval Reserve Pool,” Oil and Gas J. (March 4, 1957) 84.

Travis, R.H.: “Complete Automation in Water lnjcctwn.” Pvt. Enx’. (Feb. 1957) B-76.

Warren, F.H.: “Automatic Gaging, Sampling. and Testing.” 011 oncl Gas. J. (Nov. S, 1951) 271.

Wasicek, J.J., Kleppinger, K.B., and Grownburg, W W : “An In- tegrated Design of Lease Programming and Custody Transfer Facdities, paper 1125-G presented at the 33rd Annual Fall Meeting of the Society of Petroleum Engineers. AIME. in Houston, Tex., Oct. 3-8. 1958.

Water-flood Project Is Fully Automatic,” Oil and Gas J. (July 7, 1958) 135.

Weaver, E.G. and Hildebrand, S.M.: “Unique Automation System Monitors South Florida Production Operations,” J. Per. Tech. (June 1982) 1307-12.

“Weight MeasuresFlow in New Unit,” Oiland Gas J. (Sept. 8, 1958) 66.

“What’s Ahead for Oil in Automation,” Oil and Gus J. (June 27. 1955) 62.

Wrightsman, L.S.: “Experience with P.D. Meters and Fixed-volume T&measurement Procedures in LACT,” paper presented at the 1958 ASME Mechanical Engineering Conference. Denver, Sept. 21-24.

Chapter 17

Measuring, Sampling, and Testing Crude Oil Don G. Chantey, ARCO oil & C&S CO.*

Introduction In the early days of the petroleum industry, crude oil was sold by the producer in various-size containers called “the barrel.” It was not until 1866 that the Pennsylvania Oil Producers had a meeting and established the size of a standard oil barrel to be 42 U.S. gal at 60°F. As the value of the oil increased and as the size and number of sales grew, it became necessary to establish a set of codes or standards by which the oil could be measured accurately and a net volume determined.

The American Petroleum Inst. (API) first established its Code 25 in April 1929, under the tentative title “API Code on Measuring Field Production and Field Tanks. ’ ’ This code was superseded by API Standard 2500, “Measuring, Sampling, and Testing Crude Oil” in 1955. API now publishes a two-volume, comprehensive Manual of Petroleum Measurement Standards, which represents all branches of the industry and is the recognized standard. It serves as the unified method and

practice in measuring, sampling, and testing crude oil. As the manual is revised and as new chapters are completed, the standard designations will be replaced by chapter designations. The manual is being updated continually and care should be taken that the current standard or chapter is used.

Procedure for Typical Measuring, Sampling, and Testing

The procedure given here is the API method currently in effect for running a field tank of crude oil. Since there are many variables that would alter this method, this procedure is applicable on/~) under specific conditions.

1. The tank is vertical, nonpressurized, fixed roof with side outlets, and it is to be gauged by the innage method.

‘The wginal chapter on the topic m the 1962 edltion was taken from API Standard 25w

2. The oil is less than 100 seconds at 100°F (Saybolt universal viscosity) and is a liquid at atmospheric temperature and pressure.

3. A cup case thermometer is used to read temperature of the oil in the tank.

4. A thief is used to obtain fluid samples from the tank.

5. The API gravity scale hydrometer test method is used to determine the API gravity of the oil; the temperature of the oil has to be near 60°F (* 5°F).

6. The water and sediment in the oil is to be determined by the centrifuge method with a 203-mm [8-in.] cone-shaped tube.

The following outline gives the sequence of steps to be taken and the key points to be noted at each step.

1. Isolate the tank to be checked. 2. Use face mask and fresh air bottles if H2S hazard

exists. 3. Ground your body to stair railing or tank shell

before reaching the top. This prevents static electrical discharge in a hazardous area.

4. Stand to the side of the hatch when opening to permit wind to blow gas away from face.

5. Measure temperature. Suspend thermometer in oil tank. Thermometer should be 12 in. or more from tank shell and must be immersed in oil for 5 minutes.

Use an ASTM-approved, wood-back or corrosion- resistant metal cup case. If atmospheric temperature differs by more than 20°F from that of the liquid in the tank, the cup case should be given at least two preliminary immersions. Empty the cup case after each immersion.

Rapidly withdraw the thermometer and read and record the temperature to the nearest 1°F.

Note: The number of temperature measurements varies with the depth of the liquid. ’


In a tank containing more than 15 ft of liquid, three measurements should be taken: (I) 3 ft below the top surface of the liquid; (2) middle of the liquid; and (3) 3 fi above bottom of the liquid.

In a tank containing 10 to 15 ft of liquid, two measurements should be taken: (1) 3 ft below the top surface of liquid; and (2) 3 ft above the bottom surface of liquid.

In a tank containing less than 10 ft of liquid, one measurement should be taken in the middle of the liquid.

For tanks over 10 ft high with a capacity of less than 5,000 bbl, one measurement in the middle of the liquid should be taken.

6. With a thief, take sample(s) for basic sediment and water (BS&W) centrifuge test.

12. Convert relevant corrcctcd value to standard temperatures. USC Table SA” for crude oils.

13. Take bottom thief sample for BS&W height. Lower the clean, dry thief slowly into the oil to the desired depth, trip thief and raise slowly to avoid agitation. When sample is taken, the top of the thief must be two inches above the bottom of the pipeline connections.

14. Determine and record BS&W height in the tank. Pour remaining thief sample over a test glass until contamination appears. Measure and record (as the BS&W height) the distance from the bottom of the thief to the top of the contamination in the thief. If BS&W height is less than 4 in. from the bottom of the pipeline connection, do not run the tank.

Note: The number of samples to be taken for BS&W determination varies. 2

In tanks larger than l,OOO-bbl capacity that contain more than 15 ft of liquid, equal-volume samples should be taken (I) 6 in. below the top of the liquid, (2) at the middle of the liquid, and (3) at the outlet connection of the merchantable oil, in the order named. This method also may be used in tanks up to and including a capacity of 1,000 bbl.

15. Gauge the tank.4 Do not gauge a boiling or foaming tank. Use steel innage tape with innage plumb bob. Always make contact between gauge line and hatch while running tape into tank.

Gauge the tank only at the reference point on the gauging hatch. On tanks of 1 ,OOO-bbl capacity or less, read to the nearest % in. On tanks of 1,000 bbl or more, read to the nearest ‘/s in.

In a tank larger than l,OOO-bbl capacity that contains more than 10 ft and up to 15 ft of liquid, equal-volume samples should be taken (1) 6 in. below the top surface of the liquid and (2) at the outlet connection of the merchantable oil, in the order named. This method may be used on tanks up to and including a capacity of 1,000 bbl.

Record the reading immediately. Repeat the above until two identical gauges are obtained.

16. Saturate solvent with water. 5 Toluene is approved solvent. It is flammable and toxic. Care should be taken when using toluene.

Fill a 1-qt or 1-L glass bottle with a screw top with 700 to 800 mL toluene. Add 25 mL of either distilled or tap water. Screw cap on. Shake vigorously for 30 seconds. Loosen cap.

In a tank larger than 1 ,OOO-bbl capacity that contains 10 ft or less liquid, one sample may be taken in the middle of the column of liquid.

7. Place BS&W composite sample in sample container. Sample should be a blend of the upper, middle, and lower (if three samples were required) containing equal parts from the samples taken.

8. Seal sample container. In the lower 48 states, with the exception of California, the sample is ready to be tested for BS&W as described below in Step 17. In California, the container should be labeled and delivered to the laboratory for BS&W determination.

9. With a thief, take sample for gravity deterrnina- tion.3 The sample should be taken midway between oil surface and pipeline connection. Hang the thief in the hatch. Remove bubbles. Place hydrometer in oil sample.

10. Determine and record sample temperature to nearest 0.5”F. Hydrometer must float away from wall of cylinder; temperature of surrounding media should not change more than 5°F.

Place bottle in bath for 30 minutes. Maintain bath at a constant temperature of 140°F + 5°F.

Remove, tighten cap and shake vigorously for 30 seconds. Repeat three times.

Allow bottle of water/toluene mixture to sit in bath for 48 hr before using. Be sure no free water is left in bottle.

17. Shake sample container until sample is well mixed.

Fill each of two 203-mm [8-in.] cone-shaped centrifuge tubes with 50 mL of sample.

Use pipette to add 50 mL of toluene. Toluene should be water saturated at 140°F. Read top of meniscus at both the 50 and 100 mL marks.

Add 0.2 mL demulsifier if necessary for clean break in oil/water contact.

Depress hydrometer two scale divisions and release. Read hydrometer to nearest 0.05”API on scale at

which surface of liquid cuts scale. 11. Read and record sample temperature to nearest

0.5”F. Repeat gravity reading if temperature of sample before and after reading of gravity has changed more than 1°F. Apply any relevant correction to observed hydrometer reading (correction scale on bulb) to nearest 0.1 “API. Record the mean temperature reading observed before and after final hydrometer reading to nearest 1 “F.

Note: Hydrometer scale readings at temperatures other than calibration temperatures (60°F) should not be considered more than scale readings since the hydrometer bulb changes with temperature.

Stopper the tube tightly. Invert tube 10 times to ensure that oil and solvent are uniformly mixed.

18. Loosen stopper slightly. Immerse tube to the 100 mL mark in bath for 15 minutes. Bath must maintain 140”Fk 5°F; by contract agreement the bath temperature may be 120”Fk5”F.

Remove tube from bath and tighten stopper. Invert tube 10 times to ensure that oil and solvent are uniformly mixed.

19. Place tubes in trunnion cups on opposite sides of centrifuge. Spin for 10 minutes while maintaining minimum relative centrifuge force of 600.

Following spinning, read and record the combined volume of water and sediment at the bottom of each tube. Read to the nearest 0.05 mL for oil from 0.1 to 1

MEASURING, SAMPLING & TESTING CRUDE OIL 17-3

mL graduation. Read to nearest 0.1 mL above 1 mL graduation. Estimate to nearest 0.025 mL below 0. I-mL graduation.

Return tube to centrifuge without agitation. Spin for 10 minutes at same rate. Repeat this operation until the combined volume of water and sediment remains constant on two consecutive readings.

20. Record final volume of water and sediment in each tube. The sum of the two admissible readings is the percentage by volume of water and sediment in the sample.

After the tank has been run, the following closing data should be obtained.

2 1. Closing oil temperature: no closing temperature is necessary on tanks of 5,000 bbl or less; on tanks 5,000 bbl or more, always read to the nearest 1°F.

22. Obtain closing gauge reading at the same point and in the same manner as the opening gauge reading.

23. Obtain bottom thief. If BS&W level is lower than the opening gauge, report to supervisor.

Abstract of API Manual Publications and Distribution Section, American Petroleum Inst.) 2101 L St. NW, Washington, DC 20037. Since the API Manual of Petroleum Measurement Stan- dards is the industry’s standard on this subject and since the manual is quite long, this chapter is a reference to the API manual.

The following table of contents of the API manual lists the chapter titles, the API stock number, and a short abstract of the content of each chapter. These chapters can be ordered from the API.

API Chap. l-Vocabulary First edition (April 1977). The words and terms used throughout the entire Manual of Petroleum Measurement Standards are defined and described in this vocabulary.

API Chap. a--Tank Calibration Measurement and Calibration of Upright Cylindrical Tanks. API Publication 852-25500, API Standard 2550, jrst edition (Oct. 1965). This standard describes the procedures for calibrating upright cylindrical tanks larger than a barrel or drum. It is presented in two parts. Part I (Sets. 8 through 41) outlines procedures for making necessary measurements to determine total and incremental tank volumes; Part II (Sets. 42 through 58) presents the recommended procedure for computing volumes.

Measurement and Calibration of Horizontal Tanks. API Publication 852-25510, API Standard 2551, first edition (Oct. 1965). This standard describes external measurement procedures for calibrating horizontal aboveground stationary tanks larger than a barrel or drum. It is presented in two parts.

Part I (Sets. 7 through 2 1) includes procedures for the measurement of horizontal, and tilted, cylindrical aboveground tanks with various types of heads; descriptions of tank-measuring equipment and procedures for the calibration of that equipment for which calibration is required; and suggestions for the orderly and complete recording of field measurement data, including tank measurement record forms.

Part II (Sets. 22 through 44) includes procedures for calculating the incremental tank capacities from the field data, suitable for the preparation of incremental capacity gauge tables. Typical examples of calculations are included in Appendix I, as are convenient tables of shape factors for determining contained liquid volume in horizontal cylinders and in formed heads at any liquid depth.

Measurement and Calibration of Spheres and Spheroids. API Publication 852-25520, API Standard 2552, first edition (Oct. 1965). This standard describes the procedures for calibrating spheres and spheroids that are used as liquid containers. It is presented in two parts. Part I (Sets. 2 through 10) outlines the procedures for the measurement and calibration of spherical tanks; Part II (Sets. 11 through 20) outlines the procedures for the measurement and calibration of spheroidal tanks.

Measurement and Calibration of Barges. API Publication 852-25530, API Standard 2553, first edition (Oct. 1966). This standard describes procedures for calibrating barge tanks. It is presented in two parts.

Part I (Sets. 7 through 9) includes procedures for determining the required field measurement data, description of tank measurement equipment, and suggestions for the orderly and complete recording of field data.

Part II (Sets. 10 through 13) includes procedures for calculating the total and incremental tank capacities from the field data, suitable for preparation of the capacity gauge table. Typical examples of calculations are included in Appendix II.

Measurement and Calibration of Tank Cars. API Publication 852-25540, API Standard 2554, jrst edition (Oct. 1965). This standard describes the procedures for calibrating tank cars. It is presented in two parts: Part I (Sets. 2 through 21) outlines procedures for nonpressure-type tank cars; Part II (Sets. 22 through 3 1) outlines procedures for pressure-type tank cars.

Method for Liquid Calibration of Tanks. API Publication 852-25550, API Standard 2555 (Sept. 1966). This standard describes the procedure for calibrating tanks, or portions of tanks, larger than a barrel or drum by introducing or withdrawing measured quantities of liquid.

Correcting Gauge Tables for Incrustation. API Publication 852-25560, API RP 2556,jirst edition (Aug. 1968); supersedes supplement No. I to API Standard 2500. This recommended practice defines incrustation, describes the materials involved and recommends methods to correct the observed volume gauged.

API Chap. 3-Tank Gauging

Method of Gauging Petroleum and Petroleum Prod- ucts. API Publication 852-25450, API Standard 2545 (Oct. 1965). This standard describes the procedures for gauging crude petroleum and its liquid products in various types of tanks, containers, and carriers. Sets. 3 through 58 are applicable for measuring quantities of liquids having Reid vapor pressure (Rvp) under 40 psig;


Sets. 59 through 64 are applicable for measuring liquefied petroleum gases and other products having Rvp of 40 psig or more. The determination of temperature, API gravity, and sediment and water are not within the scope of this standard; however, brief descriptions of portable equipment used for this purpose are included in Sets. 8 and 9.

API Chap. 4-Proving Systems API Publication 852-2315. first edition (May 1978).

This publication serves as a guide for the design, installation, calibration and operation of meter proving systems. All types of proving systems commonly used by the petroleum industry are covered, including field standards, pipe provers, tank provers, master meter provers, meter proving, and analysis of sphere position repeatability.

Proving systems covered in former API Standards 1101, 253 1,2533. and 2534 (Measurement of Petroleum Liquid Hydrocarbons by Positive Displacement Meter; Mechanical Displacement Meter Provers; Metering Viscous Hydrocarbons; and Measurement of Liquid Hydrocarbons by Turbine Meter Systems) are combined in this publication, which supersedes the former standards.

This publication is intended primarily for use in the U.S. and, therefore, is related to the standards, specifications, and procedures of the (U.S.) Natl. Bureau of Standards (NBS). When it is desired to use the publication in other countries, the appropriate national standards organizations and their specifications and procedures apply.

API Chap. %-Metering

Sec. l-Foreword, General Considerations, and Scope of Chap. S-Metering. API Publication 852-30101, first edition (Nov. 1976). This is an overall introduction to Chap. 5, Metering.

Sec. 2-Measurement of Liquid Hydrocarbons by Displacement Meter Systems. API Publication 852-30102, first edition (Jun. 1977). This section specifies the characteristics of displacement meters and gives rules for systematically applying appropriate consideration to the nature of the liquids to be measured, to the installation of a metering system, and to the selection, performance, operation, and maintenance of the same. It does not apply to two-phase fluids. Special precautions should be taken when used in mass measurement systems.

Sec. 3-Turbine Meters. API Publication 852-30103, first edition (July 1976). This section specifies the characteristics of turbine meters and gives rules for applying appropriate considerations to the nature of the liquids to be measured, to the installation of a metering system using a turbine meter. and to the performance, operation. and maintenance of the same in liquid hydrocarbon service.

Sec. 4--Instrumentation or Accessory Equipment for Liquid Hydrocarbon Metering Systems. API Publica- tion 852-30104, first edition (July 1976). This publication serves as a guide for the instrumentation of liquid

hydrocarbon meters to obtain optimal accuracy and service life of metering systems. Selection of any piece of accessory equipment described herein depends on the function, design, purpose, and manner in which a measurement installation is to be used.

The application of this publication is limited to instrumentation by accessory equipment made essentially to enhance the usage of liquid hydrocarbon meters. Thus, all valves, manifolding, vents, etc. are not included. Thermometers, hydrometers, and pressure gauges

are discussed but only so far as certain minimum requirements must be met.

Sec. %-Fidelity and Security of Flow Measurement Pulsed-Data Transmission Systems. API Publication 852-30105, first edition (June 1982). This chapter provides a guide to the selection, operation, and maintenance of pulsed-data, cabled transmission systems for fluid metering systems to provide the desired level of fidelity and security of transmitted data.

API Chap. &Metering Assemblies

Sec. I-LACT Systems. API Publication 852-30121, first edition (Feb. 1981). This chapter serves as a guide for the design, installation, calibration, and operation of lease automatic custody transfer (LACT) systems.

Sec. %-Metering Systems for Loading and Unloading Marine Bulk Carriers. API Publication 852-30125, jirst edition (July 1980). This section deals with the operation and special arrangements of meters, provers, manifolding, instrumentation, and accessory equipment used for measurement in loading and unloading marine bulk carriers.

The information provided in this section is applicable to shore-to-carrier and carrier-to-shore measurement of crude oils and refined products. These procedures are not intended to apply to hydrocarbons and other materials such as liquefied petroleum gas (LPG) or liquefied natural gas (LNG), which require specialized measurement and handling equipment.

Sec. 6-Pipeline Metering Systems. API Publication 852-30126, jrst edition (Aug. 1981). This section provides guidelines for the selection of the type and size of meter(s) to be used to measure pipeline oil movements. Types of accessories and instruments that may be desirable are specified, and the relative advantages and disadvantages of three methods of proving meters (by tank prover, by pipe prover, and by master meter) are discussed. This section also includes discussions on obtaining the best operating results from a pipeline meter station.

Sec. 7-Metering Viscous Hydrocarbons. API Publication 852-30127, first edition (Jan. 1981). This section serves as a guide for the design, installation. operation, and proving of meters and their auxiliary equipment used to meter viscous hydrocarbons.

Measurement of Petroleum Liquid Hydrocarbons by Positive Displacement Meter. API Publication 852-l IOIO, API Standard 1101, jirst edition (Aug. 1960). This section covers the installation of positive

MEASURING, SAMPLING & TESTING CRUDE OIL 17-5

displacement meters, their auxiliary proving equipment, and other accessories. All types of meter installations must meet certain fundamental requirements. These include accurate proving facilities; adequate protective devices, such as strainers, relief valves, and air or vapor eliminators; and dependable pressure and flow controls. A further fundamental installation requirement is that physical conditions during operations and proving should be identical.

API Chap. 7-Temperature Determination

This chapter is in preparation. It will cover the sampling, reading, averaging, and rounding of the temperature of liquid hydrocarbons in both the static and dynamic modes of measurement for volumetric purposes.

The following API standard now covers the subject of temperature determination.

Method of Measuring the Temperature of Petroleum and Petroleum Products. API Publication 852-25430, API Standard 2543 /Oct. 1965). This standard describes the thermometer assemblies and temperature levels used in various tanks and carriers of petroleum.

API Chap. I-Sampling

Sec. I--Manual Sampling of Petroleum and Petroleum Products. A PI Publicrrtim 852-30161 (ASTM D 3057). fir.\t edition (Oct. 1981). This section covers the procedures for obtaining rcpresentativc samples of stocks or shipments of uniform petroleum products. except clcctrical insulating oils and tluid power hydraulic tluids. Sampling crude petroleum and nonunil’orm pctrolcum stocks and shipments also arc covjcrcd. although the representative nature of thcsc sampling methods is in doubt.

API Chap. 9-Density Determination

Sec. l-Hydrometer Test Method for Density, Relative Density (Specific Gravity), or API Gravity of Crude Petroleum and Liquid Petroleum Products. API Ptrbliwtiort 852-30181 (ASTM D 12981, first ed-

rio/r (JI/Hc, /YE/). This section describes the mcthodx and practices relating to the determination of the density. rclativc density. or API gravity of crude petroleum and liquid petroleum products by using the hydromctcr method (laboratory dctcmlination).

Sec. 2-Pressure Hydrometer Test Method for Densi- ty or Relative Density. API Pubkation 852-30182, jrst edition (April 1982). This section provides a guide for determining the density, relative density (specific gravity), or API gravity of light hydrocarbons, including liquefied petroleum products, using a pressure hydrometer.

API Chap. IO-Sediment and Water

Sec. l-Determination of Sediment in Crude Oils and Fuel Oils by the Extraction Method. API Publication 852-30201, j7rst edition (April 19811. This section specifies a method for the determination of sediment in crude oils and fuel oils by extraction with toluene.

Sec. 2-Determination of Water in Crude Oil by the Distillation Method. API Publicutiot~ 852-30202, ,jrst

edition (April 1981). This publication specifies a method for the determination of water in crude oil.

Sec. 3-Determination of Water and Sediment in Crude Oil by the Centrifuge Method (Laboratory Procedure). API Publication 852-30203, first edition (April 1981). This method describes the laboratory determination of water and sediment in crude oil by means of the centrifuge procedure. This centrifuge method is not entirely satisfactory. The amount of water detected is almost always lower than the actual water content. When a highly accurate value is required, the revised procedures for water by distillation (API Chap. 10.2) and sediment by extraction (API Chap. IO. I) must be used.

API Chap. ll-Physical Properties Data

Chap. I1 includes the physical data that have direct application to volumetric measurement of liquid hydrocarbons. It is presented in table form, in equations relating volume to temperature and pressure, computer subroutines, magnetic tape, and microfilm.

Chap. 11.1, “Volume Correction Factors, 1980,” (ASTM D 1250) (IP 200) is available on magnetic tape and includes a paper edition of Vol. X. [APT Pubhcation 852-27150, first edition (Aug. 198011. The tape is nine- track, 1,600 bites/in., EBCDIC, unlabeled, fixed block and requires a 32-bit or higher machine. Chap. I I. I is also available as a FORTRAN card deck (includes paper edition of Vol. X) [API Publication 852-27151,jrst edition (Aug. 1980)] and on microfiche (includes paper edition of Vol. X) [API Publication 8.52-27152, first edition (Aug. 1980)].

The following is a list of tables and computer subroutines to found in Chap. Il. I through I I .3.

Chap. 11.1, Vol. I, 1980. ASTMpublication (ASTM D 1250). (ASTM publications can be ordered from ASTM, I916 Race St., Philadelphia, PA 19103.) Table SA-Generalized Crude Oils. Correction of Observed API Gravity to API Gravity at 60°F.

Table 6A-Generalized Crude Oils. Correction of Volume to 60°F Against API Gravity at 60°F.

Chap. 11.1, Vol. II, 1980. API Publication 852-27015, first edition (Aug. 1980). Table SB-Generalized Prod- ucts, Correction of Observed API Gravity to API Gravi- ty at 60°F.

Table 6B-Generalized Products. Correction of Volume to 60°F Against API Gravity at 60°F.

Chap. 11.1, Vol. III, 1980. ASTM publication (ASTM D 12.50). Table 6C-Volume Correction Factors for In- dividual and Special Applications, Volume Correction to 60°F Against Thermal Expansion Coefficients at 60°F.

Chap. 11.1, Volume IV, 1980. API Publication 852-27045, first edition (Aug. 1980). Table 23A- Generalized Crude Oils, Correction of Observed Relative Density to Relative Density at 60160°F.

Table 24A-Generalized Crude Oils, Correction of Volume to 60°F Against Relative Density 60/60”F.

Chap. 11.1, Vol. V, 1980. API Publication 852-27060, ,first edition (Aug. 1980). Table 23B-Generalized Pro-


ducts, Correction of Observed Relative Density to Relative Density at 60160°F.

Table 24B-Generalized Products, Correction of Volume to 60°F Against Relative Density 60/6O”F.

Chap. 11.1, Vol. VI, 1980. API Publicurion 8.52-27085, first edition (Aug. 1980). Table 24C-Volume Correction Factors for Individual and Special Applications, Volume Correction to 60°F Against Thermal Expansion Coefficients at 60°F.

Chap. 11.1, Vol. VII, 1980. API Publication 852-27100, first edition (Aug. 1980). Table 53A- Generalized Crude Oils, Correction of Observed Density to Density at 15°C.

Table 54A-Generalized Crude Oils, Correction of Volume to 15°C Against Density of 15°C.

Chap. 11.1, Vol. VIII, 1980. API Publication 852-27115, $rst edition (Aug. 1980). Table 53B- Generalized Products, Correction of Observed Density to Density at 15°C.

Table 54B-Generalized Products, Correction of Volume to 15°C Against Density at 15°C.

Chap. 11.1, Vol. IX, 1980. API Publication 852-27130, first edition (Aug. 1980). Table 54C- Volume Correction Factors for Individual and Special Applications, Volume Corrections to 15°C Against Thermal Expansion Coefficients at 15°C.

Chap. 11.1, Vol. X, 1980. API Publication 852-27145, jrst edition (Aug. 1980). This volume gives the background, development, and computer documentation for all tables listed in Chap. 11.1 Vol. 1 through Vol. IX.

Chap. 11.1, Vol. XI, 1982, “Petroleum Measurement Subsidiary.” ASTM Publicution (ASTM D 1250).

Table 1 -Interrelation of Units of Measurcmcnt. Table 2-Temperature Conversions. Table 3-API Gravity at 60°F to Relative Density

60160°F and to Density at 15°C. Table 4-U.S. Gallons at 60°F and Barrels at 60°F to

Liters at 15°C Against API Gravity at 60°F. Table g--Pounds per U.S. Gallon at 60°F and U.S.

Gallons at 60°F per Pound against API Gravity at 60°F. Table 9-Short Tons per 1000 U.S. Gallons at 60°F

and per Barrel at 60°F Against API Gravity at 60°F. Table lo-U.S. Gallons at 60°F and Barrels at 60°F

per Short Ton Against API Gravity at 60°F. Table 1 I-Long Tons per 1000 U.S. Gallons at 60°F

and per Barrel at 60°F Against API Gravity at 60°F. Table 12-U.S. Gallons at 60°F and Barrels at 60°F

per Long Ton Against API Gravity at 60°F. Table 13-Metric Tons (Tonnes) per 1000 U.S.

Gallons at 60°F and per Barrel at 60°F Against API Gravity at 60°F.

Table 14-Cubic Meters at 15°C per Short Ton and per Long Ton Against API Gravity at 60°F.

Chap. 11.1, Vol. XII, 1982, “Petroleum Measure- ment Subsidiary.” ASTM Publication (ASTM D 1250). Table 21--Relative Density 60/6O”F to API Gravity at 60°F and to Density at 15 “C.

Table 22-U.S. Gallons at 60°F to Liters at 15°C and Barrels at 60°F to Cubic Meters at 15°C.

Table 26-Pounds per U.S. Gallon at 60°F and U.S. Gallons at 60°F per Pound Against Relative Density 60160°F.

Table 27-Short Tons per 1000 U.S. Gallons at 60°F and per Barrel at 60°F Against Relative Density 60160°F.

Table 28-U.S. Gallons at 60°F and Barrels at 60°F per Short Ton Against Relative Density 60160°F.

Table 29-Long Tons per 1000 U.S. Gallons at 60°F and per Barrel at 60°F Against Relative Density 60160°F.

Table 30-U.S. Gallons at 60°F and Barrels at 60°F per Long Ton Against Relative Density 60160°F.

Table 31-Cubic Meters at 15°C per Short Ton and per Long Ton Against Relative Density 60160°F.

Table 33-Specific Gravity Reduction to 60°F for Li- quefied Petroleum Gases and Natural Gasoline.

Table 34-Reduction of Volume to 60°F Against Specific Gravity 60/60”F for Liquefied Petroleum Gases.

Table 51-Density at 15°C to Relative Density 60160°F and to API Gravity at 60°F.

Table 52-Barrels at 60°F to Cubic Meters at 15°C and Cubic Meters at 15°C to Barrels at 60°F.

Table 56--Kilograms per Liter at 15°C and Liters at 15°C per Metric Ton Against Density at 15°C.

Table 57-Short Tons and Long Tons per 1000 Liters at 15°C Against Density at 15°C.

Table 58-U.S. Gallons and Barrels per Metric Ton Against Density at 15°C.

Chap. 11.1, Vol. XIII, 1982. API Publication 852-27185, first edition (Jan. 1982). Table 5D- Generalized Lubricating Oils, Correction of Observed API Gravity to API Gravity at 60°F.

Table 6D-Generalized Lubricating Oils, Correction of Volume to 60°F Against API Gravity at 60°F.

Chap. 11.1, Vol. XIV, 1982. API Publication 852-27200, first edition (Jun. 1982). Table 53D- Generalized Lubricating Oils, Correction of Observed Density to Density at 15°C.

Table 54D-Generalized Lubricating Oils, Correction of Volume to 15°C Against Density at 15°C.

Chap. 11.1.77, Extrapolation of Table 6A to -5O”F, Volume Reduction Factors, 1976. API Publication 852-25393. This chapter is a computer printout of Subroutine Table 6 in extrapolated form.

Standard Tables for Positive Displacement Meter Prover Tanks, 1966. API Publication 852-25410, API Standard. These tables provide multipliers for converting to 60°F the volumes of petroleum and petroleum products measured at temperatures between 0 and 125°F in insulated, mild steel prover tanks for positive displacement meters.

Chap. 11.3.2.1, Ethylene Density. API Publication 852-25650, jrst edition (1974). This chapter is a computer subroutine and includes subordinate subroutines “rubin” and “taint” on one FORTRAN IV card deck. It

MEASURING, SAMPLING 8. TESTING CRUDE OIL 17-7

will produce either a density (Ibmicu ft) or a compressibility factor for vapor-phase ethylene over the temperature range of 65 to 167°F and the pressure range of 200 to 2,100 psia.

Chap. 11.3.3.1, Propane Compressibility Table. API Publication 852-25654, first edition (1974). This FOR- TRAN IV subroutine is applicable to liquid-phase propane in the following ranges: relative density, 0.500 to 0.510: temperature, -20 to 120°F; saturation pressure to 1,500 psia. The subroutine computes the following two values: average compressibility per psi {this factor would be applied in the same manner as compressibilities in the current API Standard 1101) and the ratio of volume at flowing temperature and pressure to volume at flowing temperature and saturation pressure.

Chap. 11.3.3.2, Propylene Compressibility Table. API Publication 852-25656, first edition (1974). This FORTRAN IV subroutine is applicable to liquid-phase propylene in the following ranges: temperature, 30 to 165°F: and saturation pressure to 1,600 psia. The subroutine computes the following two values: density in lbm/cu ft at flowing temperature and pressure, and the ratio of density at flowing conditions to density at 60°F and saturation pressure.

API Chap. 12-Calculation of Petroleum Quantities

Sec. 2-Calculation of Liquid Petroleum Quantities Measured by Turbine or Displacement Meters. API Publication 852-30302, first edition (Sept. 1981). This publication defines the various terms (words or symbols) employed in the calculation of metered petroleum quantities. Where two or more terms customarily are employed in the oil industry for the same thing, this publication selects what should become the new standard term-for example, “run tickets,” “receipt and delivery tickets,” and the like are herein simply “measurement tickets.”

The publication also specifies the equations that allow the values of correction factors to be computed. Rules for sequence, rounding, and significant figures to be employed in a calculation are given. In addition, some tables, convenient for manual as well as computer calculations, are provided.

Field Manual, Sec. 2--Instructions for Calculating Liquid Petroleum Quantities Measured by Turbine or Displacement Meters. API Publication 852-30303, first edition (Sept. 1981). This document is the user’s field manual. It is addressed to those who need instructions without explanations as to why a particular course of action is necessary to achieve the same answer from the same data, regardless of who or what does the computing.

The user’s field manual is an instruction document. Those who wish or need to know more about the background to a set of instructions should obtain Chap. 12, Sec. 2, which is an instruction and explanation document.

API Interim Chap. 13-Measurement Control

Charts and Statistical Methods for Petroleum Meter- ing Systems. API Publication 852-25342, API Standard

2534, Appendix B, first edition (March 1970). The more accurate petroleum measurement becomes, the more its practitioners stand in need of statistical methods to express residual uncertainties. This chapter covers the application of statistical methods to petroleum measurement and sampling. Chap. 13 is in preparation.

API Chap. 14-Natural Gas Fluids Measurement

Sec. l-Measuring, Sampling, Testing, and Base Conditions for Natural Gas Fluids. API Publication 852-30341, third edition (March 1975). This chapter presents recommended practices that cover the production, transportation, and custody transfer of natural gas and the products recovered excluding LNG.

Sec. 3-Orifice Metering of Natural Gas. API Publication 852-30343, first edition (Sept. 1981). This standard provides guidance on the measurement of natural gas flow. It provides the standards for construction and installation of orifice plates and associated fittings and instructions for computing the flow of natural gas through orifice meters. Also included are the necessary tables providing the basic factors to apply to adjust for expansion, Reynolds number, temperature, pressure, specific gravity, and supercompressibility.

Sec. %-Calculation of Gross Heating Value, Specific Gravity, and Compressibility of Natural Gas Mix- tures From Compositional Analysis. API Publication 852-30345, jrst edition (Jan. 1981). This publication outlines a procedure for calculating from compositional analysis the following properties of natural gas mixtures: heating value, specific gravity, and compressibility factor.

Sec. 6-Installing and Proving Density Meters Used To Measure Hydrocarbon Liquid with Densities Be- tween 0.3 and 0.7 g/cm3 at 1556°C [60”F] and Satu- ration Vapor Pressure. API Publication 852-30346, first edition (Sept. 1979). This publication provides a method for installation and accurately proving density meters that measure light hydrocarbons used in static or dynamic conditions.

API Chap. 15-Guidelines for the Use of the Intl. System of Units (SI) in the Petroleum and Allied Industries. API Publication 852-25640, second edition (Dec. 1980).

This publication specifies the API preferred units for quantities involved in petroleum industry measurements and indicates factors for conversion of quantities expressed in customary units to the API-preferred metric units. The quantities that comprise the tables are grouped into convenient categories related to their use. They wet-c chosen to meet the needs of the many and varied aspects of the petroleum industry but also should be useful in other, similar process industries. Chap. 58 of this edition of Petroleum Engineering Handbook is the SPE Metric Standard.

API Chap. 16-Measurement of Petroleum by Weight (in preparation)

The purpose of this chapter is to provide references to model regulations promulgated by the National Con-

17-8

ference of Weights and Measures regarding commercial weighing, tolerances, and other technical requirements, and to describe the recognized practices of the petroleum industry when products are handled on a weight basis.

API Chap. 17-Marine Measurement

This chapter provides guidelines that suggest the actions to be taken in measuring and reporting quantities of crude oil or petroleum product marine transfers by shore terminal operators, vessel personnel, and other parties involved in cargo transfer measurement and accounta- bility operations.

Sec. l-Guidelines for Marine Cargo Inspection. API Phlicution 852-30401, jirst cdi~io~r (April 1982). This chapter provides guidelines to encourage uniform marine cargo inspection practices and to simplify the making of contracts that can be interpreted clearly and cxccuted between parties.


References Method of Muusrtrin~ the Tmprrurure of Prtrokm md

Petroleum Products, API Standard 2543. API. Washington, DC

(Oct. 1965).

API Chap. 8, Manual r!f Pm&urn Meusurrmrnt Stundurds,

Sampling; Sec. I-Manual .Sm~p/in~ of Petrolrum und Pm&urn

Products. API, Washington, DC (Oct. 1981)

API Chap. 9, Manual of Perrolrun~ Me~.wrmm~ Sttmdard.~, Do-

~ifi Determination, Ser. I-Hw~rornet~r Test M~~rhodfor Drnviry.

Relative Density (Specific Grrr~‘it~). or API Grrnity (f Crudr

Petroleum andLiquid Petroleum Products. API, Washington. DC

(June 1981).

Method of Gauging Petroleum and Petrokum Product.\. API Stan-

dard 2545. API, Washington, DC (OCI. 1965).

API Chap. IO, Momal of Perroleurn M~crwrem~w Srundurd\.

Sediment and Water; Sec. 3-Determination of Wurrar md Sedim

ment in Crude Oil by the Centrifuge Method (Laborctro~ Pro-

cedure), API, Washington, DC (April 1981) Appendix A

Chapter 18

Offshore Operations William H. Silcox, Chevron Corp.

James A. Bodine, Chevron Corp.

Gerald E. Burns, Chevron Corp.

Carter B. Reeds,* Chevron Corp.

Donald L. Wilson, Chevron Corp.

Edward R. Sauve, Chevron Corp.

Introduction Offshore petroleum operations emerged in the 20th century and brought new dimensions of challenge and ex- citement to oil exploration and production. When a structure taller than a lOO-story building is launched from a barge, or when a small city is built and placed offshore in 2 years, those involved deserve their feelings of pride and accomplishment.

In nearly every corner of the globe, thousands of offshore installations with payloads from 5 to 50,000 tons are producing gas and oil today in water depths from 10 to 1,000 ft. Although subjected to winds and waves up to hurricane intensity, earthquakes, sheet ice, severe tides and currents, or shifting foundations, surprisingly few structures have succumbed to the environment despite the difficulty in predicting environmental forces, equipment failure, or reservoir behavior.

This chapter can only scratch the surface of offshore operations; detailed procedures for design and construction of structures, equipment, and facilities would require volumes. Furthermore, such volumes would be obsolete before they were published. Because there is no concise reference or set of references, this chapter describes the fundamentals of standard practice in several disciplines and offers guidance for the selection of appropriate offshore codes of practice and technical references.

Historical Review In 1859, Col. Edwin Drake drilled and completed the first known oil well near a small town in Pennsylvania. This well, which was drilled with cable tools, started the modern petroleum industry. Drilling methods and tools remained in their infancy for more than 40 years, until hydraulic rotary drilling techniques were first used to drill

*Droeascd

the Spindletop well in 1901. By then, the petroleum industry was already moving offshore.

In 1897, near Summerland, CA, H.L. Williams extended an onshore oil field into the Santa Barbara Channel by drilling a submarine well from a pier. This first offshore well was drilled just 38 years after Col. Drake’s well. Five years later, more than 150 offshore wells were producing oil. Production from the California piers continues even today.

From this start, offshore drilling actually turned inland with activity in the Great Lakes, Caddo Lake in Louisiana, and Lake Maracaibo in Venezuela. Initially, wells were drilled from shore-connected piers and later from wooden single-well platforms. During this period of inland offshore drilling, platform technology remained basic. The one step forward was the change from wooden platforms to concrete structures in Lake Maracaibo.

In the late 1920’s, steel production piers that extended a quarter of a mile into the ocean at Rincon and Elwood, CA, were built and new high-producing wells stimulated exploration activity. In 1932, a small company called In- dian Petroleum Corp. determined that there was a likely prospect about 1/2 mile from shore. Instead of building a monumentally long pier, they decided to build a portion of a pier with steel piles and crossmembers. Adding a deck and barging in a derrick completed the installation. By Sept. 1932, the 60x90-ft “steel island” was completed in 38 ft of water with a 25-ft air gap. This first open-seas offshore platform supported a standard 122-ft steel derrick and associated rotary drilling equipment. Successful drilling with largely unsuccessful results was carried on intermittently on the “steel island” until 1939, when the third well was completed on the pump at 40 B/D.

In Jan. 1940, a Pacific storm destroyed the steel island. During the subsequent cleanup, divers were used for the first time to remove well casing and to set abandonment plugs. ’


Meanwhile, the first offshore field was discovered in the Gulf of Mexico in 1938. A well was drilled to 9,000 ft off the coast of Texas in 194 1. With the start of World War II, however, offshore activities came to a halt. Ac- tivity did not resume until 1945 when the State of Loui- siana held its first offshore lease sale.

At the end of the war, surplus Navy ships and barges became available to the oil industry. At first, Navy landing craft (LST’s) were converted into tenders to support drilling operations on offshore platforms. By installing mud systems and electrical generation equipment, and by storing consumables on the tender, engineers reduced drilling platform payloads by a factor of 10.

The development of tender-supported platform rigs pointed the way toward mobile exploratory rigs that could move on and off location, thereby eliminating the cost of fixed drilling platforms. During the late 1940’s and early 1950’s, a number of mobile rigs were developed in rapid succession.

First was the posted barge, which consisted of a submersible barge with the drilling rig mounted on steel columns. The barge was sunk on location with the drilling rig clear of the water. Next came the submersible, with large vertical columns that provided enough buoyancy to transport the drilling rig while floating. These rigs were sunk on location with the drilling rig and deck remaining above water. Finally came the jackup rig. This rig consisted of a barge hull fitted with vertical legs that could be jacked down until they contacted the ocean floor, thus raising the barge, which supported the drilling rig, clear of the water. While the bottom-supported drilling rigs were being developed for the shallow waters of the Gulf of Mexico, floating drilling vessels and techniques were being developed for offshore California. There, water depths in excess of 500 ft were found inside the 3-mile limit.

Civil and structural engineers were largely responsible for the development of submersible and jackup rigs, but naval architects and marine engineers were called on to convert military ships for the drilling industry. Me- chanical engineers from the oil fields designed the specialized subsea and shipboard drilling equipment.

The first floating drilling vessels were converted mine sweepers with A-frames over the side for handling pipe and jet bits. The pipe was jetted into the ocean floor, and core barrels were dropped through the pipe to get cores from the bottom of the hole. Next, Navy patrol boats were converted into drillships with “over-the-side” masts and rotary tables. The first rotary floating drilling vessel went into service in 1953 and was capable of drilling in 400 ft of water to depths of 3,000 ft.

The adverse motion characteristics of these ship-shaped vessels, combined with the “over the side” rotary table, encouraged offshore drillers and engineers to find ways to reduce vessel motions. In 1955, innovative drilling engineers moved the drilling rig from over the side to the center of the ship to reduce the effects of vessel motion. A center well, or moon pool, was installed vertically through the hull, and the drilling rig was mounted over it. This breakthrough led the way to modern-day drilling vessels. Technological advances in subsea systems, vessel station-keeping systems, moored and dynamic positioning, motion compensators, control systems, and navigation systems have all contributed to the success of

drillships during the past 30 years. They will be discussed in more detail later in this chapter.

While ship-shaped vessels were being developed for California waters, a different approach to improving vessel stability was taken in Gulf of Mexico waters. The semisubmersible, or column-stabilized drilling vessel, was developed by addition of buoyant hulls to a submersible so that it could drill while floating instead of sitting on the seafloor. These rigs exhibited superior motion characteristics and now are used extensively in such rough- water areas as the North Sea and off the east coast of Canada.

While mobile drilling rigs were being developed into today’s sophisticated drilling systems, platform technology was keeping pace. In 1947, the first platform “out of sight of land” was built off the coast of Louisiana in 20 ft of water. From then until the 1970’s, the gulf coast dominated offshore petroleum activity with the installation of more than 5,000 offshore drilling or drilling/producing structures. During the 1970’s, the North Sea captured most of the offshore attention with the advent of huge payload requirements, and concrete gravity structures competed with the steel “template. ” Eighteen concrete structures have been installed in water depths from 240 to 540 ft with payloads up to 40,000 tons.

Meanwhile, steel-structure technology competed successfully for smaller payloads in the North Sea and regained favor as deeper U.S. waters were explored. In 1976, “Hondo,” a pile-supported two-piece jacket, was installed in 850 ft of water off the coast of California. In 1978, “Cognac” was installed in three pieces in 1,025 ft of water in the Gulf of Mexico. Single-piece structures became viable for deeper water as launch barges and transportation technology developed. “Garden Banks” was installed in one piece in 680 ft of water in the Gulf of Mexico in 1976. “Cerveza,” in 935 ft, and “Liguera,” in 915 ft, were installed in the gulf in 1981 and 1982. Designs for steel jackets for up to 1,200 ft of water are in the final design stages for placement in the Santa Barbara Chan- nel and the Gulf of Mexico.

Many other specialty structures have been installed. In 1966, a steel gravity-oil-storage structure was placed in service in the Gulf of Mexico. Three 500,000-bbl steel storage domes that resemble inverted champagne glasses were installed in the Arabian Gulf in 1969, 1971, and 1972. Buoyant articulated columns were installed in the North Sea in the 1970’s to serve as tanker mooring devices for loading out crude oil. Tankers and drilling vessels have been moored by various means to support gas/water/oil

separation facilities and to provide temporary oil storage. Breast mooring and single-point mooring systems have been installed in water depths exceeding 100 ft to accommodate a supertanker’s draft. A steel gravity structure with storage capacity of 1 million bbl of oil and a deck payload of 30,000 tons has been installed in the North Sea as an alternative to the concrete structures. A guyed

tower was installed in 1,000 ft of water in the Gulf of Mexico in 1983. A tension-leg platform, the commonly favored concept for water depths of more than 1,200 ft, was installed in 485 ft of water in the North Sea in 1984. Each of these special-purpose structures represents an ad-

vance in ocean engineering technology and forward- thinking business management to support untried ideas.

OFFSHORE OPERATIONS 18-3

Progress is not always the result of new ideas or con-cepts but often a step-by-step improvement in existingtechnology. For example, the skirt pile that is currentlypart of most steel deepwater structures was first im-plemented in 1955, but the idea had been patented in the19th century. The North Sea gravity structure had a prece-dent in a gravity platform constructed offshore in Califor-nia more than 30 years ago. The guyed tower was patentedbefore the turn of the century. The tension-leg platformwas invented during World War II as a seadrome or float-ing airport. Current improvements in computerized de-sign, transportation, and installation equipment, coupledwith an ever-increasing need for new oil supplies, is thedriving force for technological advance.

During the evolution of offshore platforms, the newocean engineering discipline also evolved. Ocean engi-neers are versed in structural engineering, soil mechan-ics, the hydrodynamic effects of waves and currents,structural dynamics, statistical analysis methods, and relia-bility analysis techniques.

The equipment, methods, and techniques for complet-ing, producing, and maintaining wells on the ocean floorhave also undergone tremendous advancements since thefirst subsea wells were completed in the late 1950’s. Earlysubsea Christmas trees were made up of the same con-ventional valves and flanges as trees for land wells. Theone concession to underwater operations was fail-safehydraulic actuators on remote-control valves. These ear-ly trees were usually diver-installed and connected byBowlines to shore. One company developed a swimminghydraulic wrench that was fitted with television camerasand maneuvering thrusters. This system, integrated intothe wellhead system, was successful to a degree. It wasthe first attempt to eliminate divers from subsea opera-tions. Over the past 25 years, there has been a continu-ous effort to reduce dependency on divers, but divers arestill a very important part of the offshore oil industry.

Complex multiwell systems have been installed on theocean floor. Single-well completions have been made in1,300 ft of water. Control systems that involve hydraul-ic. electronic multiplex, and acoustic signal transmissionsystems are now common. Unmanned, remotely operat-ed vehicles now are being developed that will become anintegral part of the subsea completion system. Much hasbeen accomplished in the past 25 years, but with explora-tory drilling being done in 6,500 ft of water, even moreremains to be done in this area of subsea completions.

The search for offshore oil and gas reserves has directedthe petroleum industry to the ice-covered waters of theArctic. In 1963, the first commercial oil field was discov-ered in the upper Cook Inlet of Alaska. For the first time,ice driven by extreme tidal currents produced loads onthe production facilities far in excess of other environ-mental forces. By the end of 1968, 14 platforms wereproducing oil and gas from the inlet.

The onshore oilfield discoveries of Prudhoe Bay in 1968and Kuparuk in 1969 established the Alaska North Slopeas an oil province. In 1977, construction of the Trans-Alaska Pipeline System was completed, and oil beganflowing directly to the ice-free port of Valdez. This de-velopment has inspired extensive exploration activity inthe Arctic offshore continental shelves of the U.S. andCanada.

Fig. 18.1– Typical floating drilling arrangement.

The industry has constructed 26 sand and gravel islandsfor exploratory drilling in water depths to 100 ft since1972. Several caisson-retaining systems have been im-plemented to speed construction and to reduce the fill re-quirements for the islands. Beyond 100 ft, drillships havebeen used, but they operate only during the ice-free sum-mer season. In 1983, a floating conical drilling unit wasdeployed in the Canadian Beaufort Sea. The unit is capa-ble of resisting early winter ice loads, hence extendingthe drilling season to 6 months a year.

At the current time, at least four major Arctic marineprojects are in the planning phases: the Arctic Pilot Proj-ect in the Canadian Arctic Islands, the Arctic MarineHydrocarbon Production Project in the Canadian BeaufortSea, the Endicott Development nearshore U.S. BeaufortSea, and the Hibernia Development off the east coast ofCanada. Permanent production platforms, subsea pipe-lines, icebreaking tankers, supply vessels, and evacua-tion systems are a few of the facilities being developed.

In summary, though the offshore industry has come along way since the wooden pier days of Summerland, thetechnological requirements have barely been addressed.

Offshore DrillingThe Introduction brought us quickly from the very earlydays of the oil industry to today’s jackup drilling units,semisubmersibles, and drillships. This section will dis-cuss the planning, preparation, and equipment necessaryto conduct a typical floating drilling operation (see Fig.18.1). Focus will be primarily on floating drilling becauseoperations from jackups, submersibles, and platformsgenerally follow land drilling practices. The last portionof the section will be devoted to special considerations,such as deepwater and high-current drilling and consid-erations for cold and hostile environmental conditions. Fora general discussion of the technology of offshore drill-ing, completion, and production, see Ref. 2.

Planning and Preparations

Site Conditions and Considerations. The culminationof the sometimes arduous and complex task of geologicevaluation of a potential offshore play is for the explora-tion geologist to put a finger on the map and say “drill

1 a-4 PETROLEUM ENGINEERING HANDBOOK

here.” This decision sets in motion a series of actions that will eventually lead to the drilling of an offshore well.

The first major step is to select a rig to drill the well. Because all rigs have specific operating criteria and limits, however, certain data must be known about the drillsite and surrounding area. Basic rig selection criteria consist of water depth, expected environmental conditions during the forecasted drilling period (wind, waves, current profile, and climatological conditions), distance from nearest dock facility, and availability of consumable supplies (such as drilling mud, cement, pipe, rental tools, and spare parts).

Water Depth. A rough idea of the water depth is an important criterion for rig selection. If the water depth does not exceed approximately 350 ft, any of the three ma.jor rig types can be considered. Jackups can handle a water depth range from their shallow draft limit of 20 to 30 ft to a maximum depth of 350 ft. The maximum depth limitation is a function of other environmental constraints, such as wind, wave, and current conditions at the site. Severe conditions tend to lower the jackup rig’s maximum water-depth capacity.

Drillship water depths range from approximately 100 to 8,000 ft with today’s technology. The shallow side is limited by clearance between the bottom of the hull and the subsea blowout preventer (BOP) equipment. Maxi- mum water-depth limits occur because of riser-system limitations and other constraints that will be discussed later.

Semisubmersible water depths range from approximately 150 to 8,000 ft. The semisubmersible must stay in slightly deeper water than a ship because of the clearance between the submerged hull (60 to 90 ft below the water surface during normal drilling operations) and the subsea BOP equipment. Until 1978, semisubmersible maximum water depth was limited by the practical depth of conventional mooring systems-approximately 2,200 ft. One dynamically positioned semisubmersible that required no conventional mooring system, thus extending the design working depth to 8,000 ft, was commissioned in 1978. Today, several dynamically positioned semisubmersibles are under construction or in service.

The industry water-depth record currently stands at 6.848 ft for a well drilled off the U.S. east coast during the summer 1983.

Expected Environmental Conditions. Wind, waves, and current are all important site-specific data to help in rig selection and in determination of vessel heading, mooring pattern, mooring line tensions, riser tensions, subsea equipment selection, and equipment operational limits.

Wind, wave, current, and climatological data are generally the responsibility of an oceanographic consulting firm or your own company’s oceanographer. Many sources of environmental data are available-the marine climatic at- las, ship observations, U.S. Navy publications. private- ly funded oceanographic studies, and university-sponsored research. Converting these data into useful site-specific wind, wave, and current information is the scientific specialty of oceanography.

The oceanographer must have specified coordinates of the location and the time of the year (with some cushion on both ends) in which operations are expected. With that, he can develop the expected wind, wave, and current conditions for the location. For an exploratory location, the oceanographer may provide environmental data for oper-

ational weather, seasonal one-year storm, and seasonal IO-year storm. With that information, the drilling engineer and technical support staff can accomplish several tasks necessary in planning the well.

1. A preliminary rig selection can be made based on water depth, wind, wave, and current information.

2. A preliminary estimate of vessel heading can be determined. Before a final heading is specified, however, local knowledge of the area should be considered. Local conditions-such as swell, tide-generated currents, and rapidly changing wind directions-frequently can affect the optimum vessel heading significantly. The primary objective of optimum vessel heading is to minimize vessel motion (primarily pitch, roll, and heave) while keeping the vessel’s mooring line forces within acceptable limits and providing a lee side (calm-water side) for supply and crew boats to tie up.

3. To assist in vessel selection, a vessel motion or downtime analysis can be run. Computer programs that compare a particular vessel’s motion characteristics with the predicted wind and waves are available. The result indicates vessel motion. The resulting motion can be compared with a previously established set of acceptable operating limits (by computer analysis or manually) to determine an approximate downtime to be expected. This analytical tool is most useful in comparing two rigs for a particular location.

4. After the vessel is selected, mooring and riser analyses can be run to determine whether the vessel is adequately equipped for the location. In addition, both mooring and riser operating tensions can be determined. Both are necessary after the rig arrives on location. Typi- cally, the mooring system is analyzed with a one-year seasonal storm to determine what operating tensions should be pulled on the anchor lines. A IO-year storm can be analyzed to determine the level of proof test to pull on each mooring line. With reasonable risk considered, if each line can withstand a IO-year storm proof test, normal operations should be safe without the fear of slip- ping an anchor or breaking a mooring line. Drilling riser top tensions are developed to minimize ball-joint angles and riser sag while keeping riser-pipe stresses within acceptable limits.

For jackup rig evaluation, comparing water depth, current, wind, and tides with the maximum recommended criteria established by the rig designer is extremely important. In water depths nearing the rig’s maximum capability, strong current or other environmental factors may reduce the acceptable water depth.

Soil or foundation competency at the site must be known for jackup operations also. At an exploratory location with unknown soil consistency, soil borings generally will be required before the rig’s arrival on location. They are useful in determining depth of leg penetration and to ensure that the soil can adequately support the rig.

Logistics Considerations. Logistics must also be considered in rig selection. Remote locations require substantially more planning and preparation than do locations adjacent to established bases and supplies. Consideration must be given to (1) frequency of consumables supply; (2) distance from supply base (length of boat run); (3) number of people the rig can accommodate; (4) availability of spare parts: and (5) shipment delays caused by customs regulations.

OFFSHORE OPERATIONS

Floating rigs’ (ships and semisubmersibles) variabledeck-load capacity must be considered and compared withfrequency of consumable supplies required. Ships, as anexample, have much greater variable deck-load capacitythan semisubmersible drilling rigs (15,000 vs. 3,000 tons).If the location is in an extremely rough environment, how-ever, the semisubmersible is more stable in rough seasthan the ship. Trade-offs and compromises are necessaryingredients in rig selection.

Availability of pipe, mud, fuel, water, and other con-sumables must be carefully determined during the plan-ning effort. Helicopters to transport personnel and lightequipment in routine and emergency situations are a nec-essary part of most floating drilling operations. Those lo-cated within a few minutes of the coastline and supportbases are sometimes exceptions.

Climatological conditions have a major effect onhelicopter operations. Fog and impaired visibility condi-tions will ground flight operations and, depending on theirextent, can have a major effect on the resupply of con-sumables, transportation of crews to and from supportbases, and overall rig operations. Floating ice, low tem-perature, and high currents offer special considerationsthat are discussed at the end of the Offshore Drilling sec-tion of this chapter.

Seismic and Other Location Studies. Preparation todrill an exploratory location will include running andevaluating a suite of location surveys. Site surveys gener-ally are run by seismic companies specializing in prespudsite studies. These companies will conduct the surveys,evaluate the data, and prepare formal reports that pre-sent the data that will be useful in selecting the exact lo-cation, in preparing the mooring plan, and in determin-ing how the top hole will be drilled.

For exploratory drilling in federal offshore waters, theU.S. Mineral Management Service issued a set of guide-lines that require certain surveys to be performed and an-alyzed before it will issue a permit to drill. Theseguidelines cover studies on shallow geological hazards,culture and archaeology, and biology.

The operator or lease holder must cover a minimumprescribed grid of traverse lines in carrying out thesestudies. In addition, certain minimum instrumentation isrequired to be run during the surveys. These include spar-ker, uniboom, sub-bottom profiler, side-scan sonar, andfathometer for surveys of shallow geological hazards. Ifthe drilling equipment is to be on board a floating vessel,no bottom sampling is required. If a bottom-setting jack-up barge is to be used, then a bottom sample or core mustbe obtained. Side-scan-sonar, magnetometer, and fathom-eter are required for cultural and archaeological surveys.For biological surveys, box-core samples of hard-bottomareas and ocean-floor photography or TV view of hard-bottom areas are required.

The shallow-hazard surveys are required for all sites.The grid must be at least 8,000 ft on a side, centered onthe proposed location, and surveyed on 1,000-R grid lines.The cultural surveys need to be run only in waters of lessthan 400-ft depth. The biological surveys must be run inareas where endangered species exist or hard-bottom sedi-ments might be disturbed. Navigation and location of thesurvey grid during the water-borne surveys must be ac-curate to within 50± ft.

Fig. 18.2– Jackup rig.

Rig-Selection ConsiderationsRig-selection criteria and rig types were discussed brief-ly earlier. In this section, we will discuss the differencesin four rigs that are used for offshore drilling: jackups,submersibles, semisubmersibles, and ships. We will alsoconsider drilling equipment, mooring systems, and proce-dure manuals.

Rig types. Jackup rigs (see Fig. 18.2) consist of barge-shaped hulls with three or four (sometimes more) struc-tural or tubular legs. Jackups must be towed to locationor loaded on specially built ships for major moves. Shiptransportation of jackups is becoming more frequent asnew special transport vessels become available. Shiptransport is considerably faster for long moves (6 to 8 vs.2 to 3 knots) and much less risky. Loading and offload-ing the jackup requires a calm-water site at both ends ofthe move. Once the jackup is in its “jacked-up” position,drilling proceeds in a way similar to land or platform op-erations. However, several subtle differences should bementioned.

First, water conditions must be relatively calm—generally less than 6- to 7-ft waves-before the rig canjack its hull out of the water. Major concerns are impactand lateral loading on the legs just as it comes in contactwith the ocean floor. If the rig is rolling and pitching be-yond specified limits, the jacking operation must be sus-pended until calmer conditions prevail. The same logicapplies when the rig is jacking down.

Second, once the rig is jacked up to working positionwith a safe air gap between the ocean surface and the un-derside of the hull, primary concerns are lateral loadingon the legs and scouring around the leg mats caused bycurrent. Excessive current can cause troublesome vibra-tion, and scouring can lead to foundation failure. Bothconditions are monitored closely, and corrective actionsare taken when necessary.


Fig. 18.3– Submersible rig

Third, the drilling operation is similar to a land opera-tion after the outer casing is driven or drilled and cementedin place. Surface BOP and conventional drilling equip-ment are used.

Fourth, the casing extending from the ocean floor tothe rig is a structural member and should be analyzed be-fore installation. Wall thickness and strength of the pipeshould be specified (and will vary if a mudline suspen-sion system is used) to ensure that it will withstand thelateral loads of the current and the axial loads of the sur-face BOP and successive casing strings.

Submersible rigs (see Fig. 18.3) are limited to shallow-water drilling. Once the rig is on location and ballastedto sit on the ocean floor, drilling operations proceed ason a land site. Foundation considerations are as impor-tant here as in jackup operations. Logistics and supplyconsiderations are common to all offshore operations, sojackups and submersibles can be just as severely hamperedby fog and bad weather as floating drilling rigs.

Semisubmersible rigs (see Fig. 18.4) evolved from sub-mersibles. Some semisubmersibles can operate when rest-ing on the ocean floor or in their normal semisubmerged

Fig. 18.4– Semisubmersible rig


Fig. 18.5– Drillships. Fig. 18.6– Vessel-motion terminology.

position. The major advantage of a semisubmersible isthat it provides a stable floating drilling platform. Roll,pitch, and heave are minimized because minimum struc-ture is exposed to the water plane. The rig’s main disad-vantage is that variable deck-load capacity is limited byits reserve buoyancy or the amount of watertight volumeabove the water line. A semisubmersible with four 50-ft-diameter columns breaking the water plane displaces about62 tons of seawater for each foot of displacement of thecolumn. An equivalent 400-ft-long by 60-ft-wide ship dis-places 756 tons for each foot of hull displacement. Be-cause semisubmersibles are sensitive to variations in deckload, they are outfitted with extensive ballasting systemsthat are capable of shifting ballast rapidly to maintainproper trim and of deballasting or ballasting as cargo isloaded or offloaded. The semisubmersible is highlyregarded as the year-round drilling vessel for the open-sea environment because it is very stable in pitch, roll,and heave. 3

Drillships (see Fig. 18.5) are noted for their mobilityand high storage capacity. Drillships have a definite ad-vantage over semisubmersibles because of their size andspeed. Most drillships are designed to pass through themajor canals of the world, thereby substantially reduc-ing the distance between oceans. The distance from theGulf of Mexico to the U.S. west coast by the way of thePanama Canal is 4,500 miles. The distance around SouthAmerica to the U.S. west coast (the route a semisubmer-sible must travel because it is too large to pass throughthe Panama Canal) is 15,000 miles. The cost of movingthe ship to the west coast is generally much less than thatof moving a semisubmersible because of time savings (lessday rate) and distance savings (less fuel). Ships general-ly can travel at a higher speed than a semisubmersible(12 to 13 vs. 8 to 9 knots) for even more time savings.As pointed out in the semisubmersible discussion, thedrillship can carry a much larger variable deck load, whichoffers the advantage of less frequent resupply.

The very nature of drillships (long, narrow hulls withlarge water planes), however, dictates their sensitivity tosea conditions in pitch, roll, and heave. Operations canbe carried out with minimum weather downtime, how-ever, by working drillships in protected waters at seasonswhen conditions are best for open-sea drilling. Clearly,the biggest disadvantage of a drillship working in severeenvironments is its motion characteristics, especially inpitch, roll, and heave.3

Motion Characteristics. To compare the advantage ofone drilling vessel over another, their relative motioncharacteristics must be considered carefully. Vessel mo-tions for ships and semisubmersibles can be analyzed bydetermining the rig’s response in the six degrees of free-dom (pitch, roll, heave, yaw, surge, and sway) relativeto the uniform waves (see Fig. 18.6). All vessels shouldhave a set of motion-response curves. The curves gener-ally are obtained for each rig configuration in a modelbasin. Each hull shape has a unique set of curves. Rolland heave generally control the limiting operation. Withcurves like those shown in Figs. 18.7 and 18.8, vesselmotion in roll and heave can be determined for a particu-lar set of wave data representing the drilling period. Oceanwaves represent a spectrum of wave heights and waveperiods. Computer programs are available to calculatevessel motion by entering wave data and the rig’s motioncurves. The result will be a motion history of that partic-ular rig for a specific drilling period.

Performance Evaluation. The next step is to comparethe performance of the two rigs. One performance yard-stick is the weather-related downtime the rigs will sufferunder the same environmental conditions. Downtime anal-ysis can be particularly useful when comparing availabledrilling vessels for a one-well project or a complete drill-ing program. While one vessel may appear to be more


20 I

.

Fig. 18.7-Vessel response-roll.

economical because it has a lower day rate, it may cost more to complete the job because of weather-related downtime.

The key to weather-related downtime is identifying the maximum limit in degrees of roll, feet of heave, degrees of pitch, etc., that can be tolerated during each discrete operation of floating drilling. The maximum level may be based on equipment operating limits, safety considerations, work efficiency, potential for damage, or other factors. Although such a limit is seldom concise, it can be a fair comparison to evaluate relative rig performance. Implementing an operating limit by shutting down an operation on the rig is completely a judgment call with many variables to be considered on the spot. Each rig should have its own set of operating limits established from experience with the rig or from experience of the rig operating personnel. Table 18.1 is an example of limiting vessel motions for most floating drilling operations.

With the appropriate operating limits, the percent of the time each applies, and the rig’s motion history, weather-related downtime can be calculated. A number of papers have been published on downtime analysis. Var- ious techniques (both manual and computer-aided analyses) can be applied to calculate weather-related downtime. 4

TABLE l&l-DRILLING VESSEL OPERATING LIMITS

Heave Limit

Operation ft m -- Anchoring, running riser,

landing BOP 6 1.8 Running casing, coring,

well testing IO 3.0 Drilling, tripping, logging 12 3.6 Circulate and condition mud 20 6.1

Time Criterion

Roll Applied Limit Per Well

(deg.) (04

3 10

3 40 6 30

10 20

Fig. l&8-Vessel response-heave

An additional item normally not included in the motion- related operating limits is wind. High winds frequently result in shutdown because the rig crane cannot safely handle casing or riser. This is a valid input to the rig’s overall performance and should be included in the final downtime comparison. Occurrences other than severe weather also cause operating downtime. Equipment breakdown and repair downtime (sometimes the result of severe weather, but not always) must be determined from experience and operating history with a particular rig or company. This increment of downtime is unpredictable and difficult to estimate.

Mooring Systems (Stationkeeping). Once the engineers are satisfied that a particular rig or group of rigs is capable of handling the environment of a specified offshore location, other equipment systems must be evaluated and compared.

Mooring equipment provided to keep the rig on location is of major significance. Major questions to be answered regarding mooring equipment include the following: (1) is the mooring line (chain, wire, or a combination of chain and wire) strong enough to withstand the loads during the strongest anticipated storm; (2) does the rig have sufficient wire or chain on board or available for the water depth at the specified location; (3) do the anchor handling or supply boats that are being considered have adequate pennant-wire-handling equipment on board (lengths must be greater than the water depth and sufficiently strong to handle the 30- to 40-ton anchors and can approach 2.5 to 3 in. diameter); (4) does the vessel have adequate instrumentation to monitor mooring- line loads; and (5) does the rig have adequate chain-locker capacity to hold the desired amount of chain, or must part or all of the chain be stored on supply boats? (Vessels that don’t carry their own chain have greater in-transit deck-load capability but normally will require longer to moor up because of the additional chain-handling requirements.)


_I VESSEL OFFSET

1, / ,,- ZERO ANGLE

Fig. 18.9-Optimum vessel position.

These questions must be answered to specify an adequate mooring system properly. Mooring analysis, which is necessary to answer several of the questions, will be discussed later in this section.

Adequate stationkeeping (keeping the vessel within acceptable limits on the location) is a result of a properly designed and operated mooring system. Why is stationkeeping important? Ideally, the vessel should be located directly over the well. However, wind and current forces can cause the vessel to take an offset downstream from the wellhead location. Waves cause the vessel to oscil- late around that offset position.

the top of the BOP can cause rapid and excessive wear ‘:&!

” if the angular offset exceeds 1 to 2” for an extended length of time; (3) excessive vessel offset can cause increased*. riser sag, compounding both the ball-joint offset and the wear problems. Proper monitoring of the ball-joint angle and adjustment of the mooring system will result in a vessel offset upstream of the current and wind that will minimize the lower ball-joint angle. Optimum vessel offset would yield a zero ball-joint angle (see Fig. 18.9).

There are many variations in mooring patterns. Differ- ently shaped vessels will require different mooring patterns (see Fig. 18.10).

It is important to keep the vessel reasonably close to One criterion in mooring-system design is that the the wellhead position for several reasons: (1) the subsea restoring forces should be able to withstand nearly the drilling equipment can accommodate angular offsets of same storm conditions from any direction.5 The moor- up to lo”, but beyond that the equipment mechanically ing pattern is designed to fit the vessel and particular en- locks up; (2) drillpipe that is rotating in the ball joint at vironmental conditions anticipated at the site.

A \ 400

A \

>

c 3

d AJ SYMMETRIC NINE LINE

SYMMETRIC TEN LINE

30”-70” EIGHT LINE

SYMMETRIC EIGHT LINE

44

45O-90° EIGHT LINE & 45”-90“ TEN LINE

30°-60” EIGHT LINE

Fig. 18.10-Typical spread mooring patterns, 4.:


r-@---i VESSEL VESSEL

I f -(9f f I

/’ LEGEND WATER LINE

- I --

@ HORZONTAL FORCE AT VESSEL.

0 S TOTAL VESSEL MOVEMENT FROM ZERO

I’

HORlZONTAL LOAD TO SPECIFIED HORZONTAL LOAD.

ANCHOR LINE , ,’

@ TOTALhNE REMAINING ON BOTTOM

@ @

/’ @ ANGLE OF LINE FROM HORIZONTAL

AT ANCHOR. /

/ /

BL /

0 E VERTlCAL FORCE ON ANCHOR.

/ /

/ 0 ANGLE OF ,.,NE FROM HORIZONTAL

D / AT VESSEL

\ NOTE’ D B E AREZERO UNTIL C

ANCHOR BECOMES ZERO

Fig. 18.11-Typical catenary configuration.

The restoring forces are generated by the niboring line. Environmental loads acting on a vessel displace it horizontally until an equal and opposite horizontal force (restoring force) is developed by the anchor and mooring lines. As the vessel is displaced, tension in the anchor line increases because of additional line being lifted off the ocean floor and because the vertical component of a&nor line tension, which increases as line is lified off-bottom, is affected by the angle in the anchopline at the vessel (see Fig. 18.11). .

Vertical or uplifting [orces on the anchor are zero as. long as line-remains on botto proFrly designed and operated mooring system sho ways have line remaining on bottom during maximum storm conditions. If all the line comes off-bottom,+the chances of dislodging an anchor are high.

W&h a spread mooring system, vessel excursion in moderate weathei conditions can be restricted to 2 to 3 7% of water depth by‘pulling initial operating tensions in each line. Fig. 18.12 shows the nonlinear behavior ofborizon- tal force (horizontal component of line tension) and vessel disphcement for a typical spread mooring. If the vessel _ _

9 , i I

1 i’

._-,_--; -

0 10 . 20 30 40 50

DISPLACEMENT, FT.

NOTE: KS1 = PSI X 1000

Fig. 18.12-Single-line catenary horizontal force vs. horizontal displacement.

had two opposing mooring lines and could pull tension on each line initially, vessel displacement could be greatly reduced for the same environmental loads because the line would operate in a much “stiffer” region of its horizontal force vs. displacement curve.

/

Initial operating tension, however, does affect the maximum line tension that will be required in maximum storm weather. The same environmental loads on the vessel are produced during maximum storm weather regardless of the value of initial operating tension. This force must be balanced by one or more mooring lines. This restoring force is in addition to most of the horizontal components of the initial tension in the line. The vessel will probably not be displaced enough to reduce the initial tension in the leeward lines completely. In actual operations, leeward lines can be slacked off during maximum storm weather to reduce maximum line tension and vessel offset. In general, the higher the initial tension, the higher the maximum line tension during maximum storm conditions. Too little initial tension ,pwever, will result in unacceptable vessel offset during operating weather conditions. Table 18.2 identifies desirable stationkeeping criteria.

Dynamic positioning is another method of stationkeeping where no mooring lines are used. These systems require acoustic positioning beacons, multiple thrusters on the vessel’, and an on-board computer system and are primarily for deepwater drilling. Dynamic positioning will be discussed briefly in the last section, Special Consider- ations.

Drilling-Equipment Considerations. Rig-selection considerations should include a review of the vessel’s drilling equipment. Much of the drilling equipment found on board floating drilling vessels is identical or similar to equipment on land drilling rigs. This discussion will be limited to equipment unique to floating drilling.

Fig. 18.13 identifies the major components of the subsea drilling system and related shipboard ,systems. The figtire sho& some of the components of the drilling system’that have been developed to accommodate vessel motion and water ‘depth. The components to be explained


TABLE 18.2-DESIRABLE CRITERIA OF STATIONKEEPING

Operational: Minimal weather Maximum vessel excursion

Nonoperational, But Riser-Connected: Maximum weather condition Maximum line tension Minimum line remaining on bottom Maximum vess%l excursion

Riser-Disconnected: Weather conditions Maximum line tension Minimum line remaining on bottom

are the BOP, the flex joint, riser, riser slip joint, riser and guideline tensioners, drillstring motion compensator, guidelines, and control system.

BOP. The subsea BOP stack is a major change from land or platform drilling operations. Drilling riser, extended kill and choke lines, remote hydraulic and electrohydraulic control systems, and subsea wellhead equipment are all product modifications needed because the BOP was relocated on the ocean floor. The well’s major pressure-containing components were put on the ocean floor because of the need to compensate for vessel motion.

A BOP stack, whether located on the surface or subsea, is considered a last resort for preventing a well kick from becoming a blowout. Several steps are taken to control unusual well conditions before use of the well shut- in device (BOP). If the previous steps have failed and it becomes necessary to shut the well in, the shut-in equipment must be highly reliable. BOP equipment is designed with reliability as its ultimate criterion. Because of its relative inaccessibility, the subsea BOP requires additional redundancy and reliability.

The BOP stack is a combination of individual BOP’s designed to shut in a well under pressure so that formation fluids that have mov&l into the wellbore can be circulated out while continuous control of the well is maintained.

A description of the BOP stack components is included below (see Fig. 18.14).

Rum Preventers. The massive steel rams have rubber seals, and are hydraulically actuated to seal off the wellbore. Pipe rams seal the annulus around the drillpipe and are designed so that an entire string of drillpipe and collars can be suspended from a pipe joint landed on a ram. The ram seals must be the correct size to seal; 3-in. seals cannot be used for 5-in. drillpipe. Conventionally, three pipe rams are used. A fourth ram, a blind-shear ram, is used to seal over the open hole and to shear drillpipe when necessary: Shearing pipe is, of course, one of the last resorts in an emergency situation. 5 Variable-bore rams are an option that is offered$or tapered drillstrings.

Annular Preventers. Annular preventers are comprised 9 of specially designed, reinforced rubber elements that can seal around any tubular or near-tubular objects that &ill go through the BOP’s. They will also seal over the open hole and can pass drillpipe tool joints without severely

Drilling operations can be carried out That which results in ~3~ lower

ball-joint angle, generally 2 to 3% of water depth

Seasonal l-year storm % breaking strength

500 ft That which results in ~5~ lower

ball-joint angle, generally 5 to 6% of water depth

Seasonal 1 O-year storm I/Z breaking strength

100 ft

GUIDE LINE

TENSIONER -

4 EA TYP 4. 6, OR 6 EA TYP

TOP FLEX JOINT

c STORAGE REEL

TO KILL 8 CHOKE

FLEXIBLE HOSE

a CONDUCTOR CASING

SURFACE CASING

l

Fig. 18.13-Floating drilling system.


in addition to the spring or fail-safe close feature. Twovalves in each line should always be used for redundan-cy. They should be located as close to the stack as possi-ble for mechanical protection.5

Unitized BOP Stuck. The unitized BOP stack that con-sists of two hydraulic connectors, three or four rampreventers, one or two annular preventers, four K&Cvalves, one flex joint, and a control system is generallyhandled in one or two pieces on board the rig. The com-plete assembly can weigh from 200,000 to 400,000 lbmand stand 25 to 30 ft high.3

Handling and moving the BOP stack from its storageposition to the moonpool and back presents unique prob-lems. Generally, either special overhead cranes or hydrau-lically actuated carts are used to move the stacks.

BOP maintenance is extremely important. The only timeavailable for routine maintenance is between locations.On short field moves, this can present problems. LandBOP systems are frequently broken down and sent to theshop for maintenance between wells, but that is virtuallyimpossible to do without causing major delays on a float-ing drilling rig. A few rigs are equipped with backup BOPstacks to minimize the chance of major delay.

BOP testing is done in two steps. The stack must becompletely function-tested (each of the 30 to 40 hydraul-ic functions actuated to verify that each works) before run-ning. It must also be completely pressure-tested beforeit leaves the deck. Each pressure-containing component(rams, annulars, and K&C valves) must be tested to apressure specified by the operator. API RP 53 on BOP’s6

identifies testing procedures as a minimum safe guideline.After the BOP has been run and latched on to the subseawellhead, it must again be pressure-tested. Followingprocedures defined by regulatory agencies, periodic func-tion and pressure-testing must be done on the BOP equip-ment during the course of a well. A complete deck andsubsea BOP testing checklist simplifies frequent testingrequirements.

Flex Joints. 3 A flex joint is installed between the lowerend of the riser and the BOP stack. This joint essentiallyacts as a pinned connection to minimize bending stressesin the riser as the drilling vessel is moved by wind, wave,and current action.

The first flex joints were made from bag-type annularBOP’s fitted over a mandrel flanged to the top of the BOPstack. The rubber element in the preventer was inflatedagainst the mandrel to a pressure high enough to keepdrilling fluid in the riser from leaking past it. This typeof flex joint, which was not positively locked to the BOP,worked fine in shallow waters (200 ft or less) where ten-sion was not pulled on the riser.

The next flex joints were the pressure-balanced balljoints. These joints came into existence when operationsmoved into deeper waters and it became necessary to pulltension in the riser through the ball joint into the BOPitself. With this positive pull upward on the ball joint, itwas necessary to provide a pressurized oil pad betweenthe male and female halves of the ball joint to minimizewear. Pressurized oil was provided through a line fromthe surface and was contained between upper and lowerO-ring seals within the ball joint. The balancing pressureon the ball joint was determined by dividing the tensionpulled through the ball joint by the projected horizontalarea between the ball-joint seals.

Fig. 18.14– BOP stack

damaging the sealing element. Annular preventers are ac-tuated by an annular piston that squeezes the seal into thebore. The piston area is large relative to the other func-tions on the stack and, except for initial closure, shouldbe operated at pressures lower than the other stack func-tions. This decreases the possibility of extruding the rub-ber seal out of the preventer. 5 Frequently, two annularpreventers are used. One normally will be located abovethe upper hydraulic connector so that it can be retrievedwith the riser.

Hydraulic Connectors. 3 Hydraulic connectors providethe main pressure seal between the wellhead housing andthe BOP and between the top of the BOP and the lowermarine riser package (LMRP–usually contains the topannular preventer, flex joint, control system, and cross-over to the bottom riser joint). The high-pressure well-head housing is the male portion of the connector. It willbe a mandrel or a hub type. The connector is the femaleportion and consists of a series of hydraulic cylinders thatactuate locking dogs into grooves machined into the well-head housing or collet fingers that clamp over the well-head housing hub. Both types of connectors use metal-ringseals. This provides continuous metal-to-metal sealing upthrough the BOP.

Kill-and-Choke Valves. These valves are the subseashutoff of the high-pressure kill and choke (K&C) linesthat run from the BOP’s to the choke manifold on the rig.K&C valves are hydraulically controlled from the sur-face and are designed to close by spring action when open-ing pressure is released. Some valves close hydraulically

OFFSHORE OPERATIONS

Steel-laminated elastomers now are replacing ball joints as riser flex joints. These joints are longer-lived and require less maintenance than the pressurized ball joints. They also eliminate the need for the pressure source and hydraulic lines.

Some operators also require the installation of a flex joint between the upper end of the riser and the slip joint. Pressurized ball joints and elastomeric joints have been used successfully in this application. Most flex joints are designed for an angular travel of f 10” for a total included angle of 20”.

SZip Joints. 3 All floating drilling vessels, ship-shaped or semisubmersible, heave up and down as swells go by. A slip joint is the link between the riser fastened to the bottom of the ocean and the heaving drilling vessel. The slip joint, similar in action to a trombone, consists of an inner and an outer barrel. The outer barrel is connected to the riser and the inner barrel to the ship. As the ship heaves up and down, the inner barrel strokes in and out of the outer barrel. A pair of inflatable rubber elements mounted on the upper end of the outer barrel serve as the seal between the barrels to prevent loss of drilling fluid. The second seal is for redundancy.

Riser Tensioner. For a drilling riser to survive, two things must happen.3 First, the drilling vessel must be kept within prescribed limits as it moves about in surge and sway. Second, the riser must be tensioned properly so that it will not sag and ultimately be overstressed in bending.

The controlling criterion is not vessel position relative to the well on the ocean floor, but the angle between the axis of the lower end of the riser and the vertical axis of the BOP stack. This angle is called the lower riser angle. During drilling, this angle should be kept at less than 3”. A greater angle will cause the rotating drillpipe to cut into the flex joint and BOP stack. In extreme cases, lost circulation has resulted from a worn-through flex joint. In normal drilling, the riser angle is kept to less than lo. If it exceeds 3”, drilling is stopped until the vessel can be repositioned.

To keep the lower riser angle as near 0” as possible in areas where ocean current is a factor, the drilling vessel may have to be located up-current from the well.

If the drilling vessel is located up-current, as shown in Fig. 18.9, but inadequate tension is pulled on the riser, the riser could sag, as denoted by the dotted line. If the drilling vessel is moving about and there is heavy drilling fluid in the riser, the angle at the flex joint could exceed 10” and put bending stresses in the riser. If this situation is not corrected, the riser ultimately will fail.

Hydropneumatic tensioning units were developed to keep constant tension pulled on the riser. Determination of the tension required is a complex problem in which water depth, riser size, mud weight, ocean current, vessel motion, and sea conditions must be considered. A number of computer programs, both time and frequency domain, have been developed to determine the tension needed. Many oil companies that operate offshore have their own riser programs or have access to them. These programs give the riser tension required and the desired vessel offset.

The tensioner system works on the principle that displacement of a relatively small amount of hydraulic fluid against a large pressurized volume of air results in a very

18-13

Fig. 18.15-Riser tensioner unit.

small change in the hydraulic pressure. Variation in tension on the riser can be kept to less than 5% by proper design.

The tensioning unit (see Fig. 18.15) consists of a series of large air storage tanks that are connected to the air or gas side of an accumulator that serves as the interface between the air and hydraulic systems. The tensioner is a cylinder/piston arrangement that has wire-rope sheaves mounted on the lower end of the cylinder and on the upper end of the piston rod that extends out of the cylinder. A wire rope that is dead-ended on a storage reel is reeved through the sheaves over alignment sheaves and is attached to the outer barrel of the slip joint. As the drilling vessel heaves up, it pulls on the line, which pulls the piston into the cylinder, displacing fluid into the accumulator against the large volume of air. The air is precharged to give the desired tension. Similarly, when the vessel moves down, the gas pressure displaces hydraulic fluid against the piston, extending the piston rod and maintaining a constant pull on the riser.

Guideline tensioning systems, developed to keep constant tension in the guidelines, operate in much the same manner as the riser tensioners. The only difference is that they are smaller because less tension is required on the guidelines.

Drillstring Motion Compensators. Without drillstring motion compensation, 3 the drill bit would be constantly lifting off and banging down into the bottom of the hole as the drilling vessel heaves up and down. Weight control on the bit under these conditions without some type of motion compensation is next to impossible. Bumper subs (trombone-type slip joints) in the drillstring above the drill collars were used initially to provide some relief from vessel motion. However, with bumper subs, once the drillstring was in the hole, the weight on the bit (WOB) (weight of the drill collars) was fixed and could be changed


tor. When the vessel heaves down, the piston is forced up by the pressurized hydraulic fluid from the accumulator. The gas side of the accumulator is connected to large- volume gas bottles. The small volume of fluid displaced by the piston against the large volume of gas gives a low compression ratio. This means that there is very little change in the gas pressure and the hydraulic pressure, resulting in an almost constant WOB.

At the start of drilling, the gas pressure is adjusted so that it will barely support the weight of the drillstring. WOB is increased simply by reducing the gas pressure. This transfers weight from the drillstring motion compensator to the bit. As a hole is made, the blocks are lowered to keep the compensator at midstroke of the piston. To reduce the WOB, the gas pressure is increased. Large air bottles are kept charged with high-pressure air for this purpose.

Re-Entry Systems. Re-entering a 3-ft-diameter hole in the ocean floor in shallow waters without too much current, say less than half a knot, isn’t too difficult. 3 If that same hole is put under half a mile of water in an area with l- to 2-knot currents, the problem obviously is more difficult.

Almost from the beginning of floating drilling, wire- rope guidelines have been used to guide drillstrings,

Fig. 18.18-Drillstring motion compensator

only by pulling the drilling assembly and changing the number of drill collars. Another disadvantage was that the early bumper subs were not hydraulically balanced. Mud pressure in the drillstring that was higher than the external pressure in the drillpipe/hole annulus, as when jet bits are used, caused the bumper subs to “pump” open and to become as stiff as the drillpipe, making them in- effective.

Balanced bumper subs that have the internal pressure routed to both sides of the stroking member were invented to solve this problem. These solved one problem and created another. When working in sandy drilling fluids, the balanced bumper subs’ seals wore out after relatively short runs, making it necessary to come out of the hole with “green” bits to replace the worn-out, leaking subs. Because of their short lives, the worn-out bumper subs were repaired on board, which required an inventory of spare parts and personnel trained in their repair.

These considerations led to development of drillstring motion compensators (see Fig. 18.16). These hydropneumatic units are installed either between the traveling blocks and the hook or in the crown block at the top of the derrick. These units have been successful for both drilling- bit weight control and running and landing heavy subsea equipment (such as 400,000-lbm BOP stacks). They are common on most drilling vessels today.

Drillstring motion compensators are similar to riser tensioners in the way they function-i.e., a small volume of hydraulic fluid is displaced against a large volume of pressurized gas. The weight of the drillstring is supported on a vertical piston inside a cylinder that is connected to the rig blocks. The piston is supported by pressurized hydraulic fluid between the piston and cylinder. As the vessel heaves up, the piston is pulled down into the cylinder, displacing hydraulic fluid into a gas-charged accumula-

casing, BOP stacks, and riser pipe into or onto subsea wells. In most instances, the guidelines are anchored to the ocean floor by the temporary guidebase. In some cases, when the hole for the structural pile is spudded without a temporary guidebase, the mud pumps were run at full capacity as the bit entered the ocean bottom. This washed a large conical hole in the ocean floor that, with luck, could be re-entered without guidelines. However, under these conditions, when the structural casing and the permanent guidebase are run, the guidelines are attached to the permanent guidebase for subsequent re-entry operations.

With the advent of dynamically positioned drillships, guidelineless re-entry systems were developed. These systems still had temporary and permanent guidebases; however, instead of using guidelines and guideposts, they were fitted with guidecones that provided a large target for the tools or casing being run. TV cameras were run through the drillpipe, casing, riser, or BOP stack (depending on what was being run) to provide guidance into the hole or back onto the BOP stack. Combinations of TV and sonar also have been used for re-entry guidance. With the dynamic-positioning system, the driller can take control of the drilling vessel from his station and position it as required for re-entry . Re-entry by means of these systems has been made in waters as deep as 6,800 ft.

Marine Risers. The first floating-drilling systems did not use marine risers for mud returns. 3 Hoses that were connected below a rotating packer mounted on top of the BOP stack brought mud returns back to the drilling vessel. The rotating packers, which sealed around the drillpipe, were very short-lived and allowed drilling fluid to leak into the ocean when they failed. It was the failure of rotating packers that led to development of today’s marine risers.

As may be seen in Fig. 8.14, the marine riser extends from the BOP stack on the ocean floor up to the drilling vessel. The marine riser, in the parlance of land drilling, is just a very long pitcher nipple. In addition to serving


as a pitcher nipple or mud-return conduit, the marine riser serves another useful purpose: it guides the drillstring through the BOP’s and into the hole being drilled in the ocean floor.

The riser joints can be ordered in any length desired, but the length normally is determined by the geometry of the drillship. Normal riser joints are 50 ft long, and at least one riser made of 75-ft joints is in service.

In the beginning, riser couplings were simply threaded collars. Cross threading of couplings being made up on a moving vessel led to the development of clamp-type couplings, piloted union-type couplings, and finally the radi- ally driven dog-and-groove couplings. Riser couplings now being developed for waters in excess of 7,000 ft are of a piloted-bolted type.

As drilling entered deeper waters and drilling vessels ran out of space to install more and more riser tensioners, it became necessary to reduce the weight of the riser by adding buoyancy material. Syntactic foam was used first. Later, air cans were installed around the riser joints to make them air buoyant. Both types of buoyancy are now in everyday use. The cost of the dense syntactic foam that is required for deeper waters is offset by the cost of high-pressure air compressors for air-buoyancy risers. Air-buoyant risers do have one advantage over the foam riser package: the air in the buoyancy cans can be dumped so that the riser will plumb bob vertically below the drilling vessel and not tend to drift off with the current.

For ultradeep waters (deeper than about 10,000 ft), free- standing risers are visualized. Work done in conjunction with the Natl. Science Foundation’s proposed Advanced Ocean Drilling Program indicates that to provide the means for rapid disconnect of the drilling vessel from the well, it will be necessary to establish a disconnect point in the riser at about 1,000 ft below the ocean surface. To support the riser vertically below the disconnect point after a disconnect, IO-ft-diameter buoyancy cans will be fitted to an appropriate number of riser joints. The disconnect point will include shear rams to cut the drillpipe if an emergency disconnect becomes necessary. This intermediate disconnect point is essential because it is estimated that pulling 10,000 ft of riser could take from 7 to 10 days, well outside our weather-forecasting capability.

K&C Systems. On land rigs, the K&C outlets3 on the BOP stack are plugged directly into the K&C manifolds on the rig floor. In floating drilling, where the BOP’s may be from several hundred to several thousand feet below the rig floor, K&C lines must be provided to bridge the water depth.

In the early days in relatively shallow waters, high- pressure hoses were connected to the BOP stack and, as the stack was lowered, were paid off hose reels. When the stack was landed on the wellhead, the hoses were connected to the K&C manifolds. As water depths became greater, the hose reels became too large for convenient use, and another way to bridge the water depth had to be found. This was done first by installing guide funnels at about 15-ft intervals along the length of the riser as it was run. These funnels were lined up with receptacles immediately above the K&C valves on the BOP stack. With the riser in place, screwed-pipe K&C lines were run down through the guide funnels and stabbed into the receptacles on the stack. Their upper ends were connected into the K&C manifolds.

This system, while functionally satisfactory, was time- consuming to run and test, so another method was developed. This method was to make the K&C lines integral with the riser. The tops of the K&C joints were fitted with a female seal pocket filled with chevron packing, and the lower end fitted with a male seal nipple. When the riser was run, the seal nipples dropped vertically into the female seal assembly. No rotation or screwing was required. The joints were held together by the riser couplings.

Some manufacturers of flexible high-pressure pipe now are proposing to provide long K&C lines that would be stored on reels and paid out with the BOP stack when it is run down to the ocean floor. On the larger vessels now in service, there is space for the large hose reels required.

Control Systems. The simplest way to operate an actuator in a hydraulic control system is to connect hydraulic lines from a pressure source through control valves directly to the actuator. 3 Some actuators require two lines to complete the control cycle; others, such as spring-return fail-safe actuators, require only one line.

Subsea BOP’s were controlled this way during the start of floating drilling. An early stack consisting of a hydraulically actuated connector top and bottom, K&C valves, four ram preventers, an annular preventer, and a pressure- balanced ball joint would require as many as 17 control hoses. These hoses, bundled together, were stored on a large hose reel. All hoses first were connected directly to their function on the stack, then pressure- and function- tested before the stack was run. Improperly tagged hoses led to many hours of troubleshooting to get the stack to work properly. This time-consuming job had to be done each time the stack was run.

Eventually, male and female multifunction stab plates were developed that reduced some of the hookup time, but the same problem of larger hose reels in deeper waters resulted. In addition, as BOP’s became more complex, as many as 30 to 40 hoses were included in the hose bundles, doubling and tripling their size. To solve the problem of large hose reels, multihose bundles, and their slower actuator response times in deeper waters, two new types of control systems were developed: the piloted all- hydraulic control system and the direct-wired electrohydraulic control system.

Backup Control Systems. In spite of the best-laid plans and even with two control pods providing 100% redundancy, problems or failures still occur in the most modern control systems. It is desirable to have reliable backup systems if the primary controls fail. This has led to development of two types of backup control systems: the acoustic control system and the last-chance hydraulic stab system. 3

The acoustic backup system uses acoustic signals through the water as the control link between the drilling vessel and the BOP stack on the ocean floor. Energy to power the acoustic signal receiver and to position control valves is provided by dry-cell batteries. Hydraulic energy to power selected functions on the BOP stack comes from accumulators mounted on the BOP stack. These accumulators are kept charged because they are part of the normal control system. Typical functions are to close shear rams, to close pipe rams, and to disconnect the riser at the lower marine-riser package.

An acoustic transmitter located on a surface vessel, drilling vessel, work boat, or other vessel is used to send

18-16

MSSEL CHARACTERISTICS

PRIVARY MSSEL

KIIRING LINE LOCATIIX? CHARACTERISTICS BATtM?3RY

I I HORIZOifIAL FORCE KIRIZONIAL DISPlAC'd%V

rl

WEtiTtER DATA

4 KORING REW:+t3.UATlON:

!V, ,,IU,;W MXl f'D3RING LINE TO DC"3Y ,:i,T,,?i OFEUTING TtUSlON

,PRCUF OR lEJ7 TEXiON

Fig. 18.17-Mooring analysis method.

a signal that is coded for the desired function, down to the receiver on the BOP stack. This signal is interpreted and the proper control valve is actuated, directing hydraulic fluid from the accumulators to the desired function. Acoustic backup systems now are installed on most deepwater drillships. Solid-propellant gas generators also have been tested successfully as backup subsea energy sources.

The last-chance hydraulic stab system provides the means for actuating several selected functions when all else has failed. A hydraulic stab that is ported to accommodate the desired functions is run down to the BOP stack on drillpipe. It may be guided down guidelines or direct- ed by sonar or TV. Hydraulic hoses are connected to the stab and are run in with the drillpipe as the stab is lowered down to the receptacle on the stack. Once the stab is in place, control is accomplished by pressuring up the appropriate hydraulic line. The stab also can retrieve the lower-marine-riser package or the complete BOP stack in the event of a failed riser. The stab receptacle is connected with shear bolts to a mounting plate on the lower- marine-riser package. The receptacle also is attached to the lower-marine-riser package with a heavy wire-rope bridle. The stab contains a connector that, when lowered into the receptacle, latches the stab to the receptacle. To retrieve the lower-marine-riser package, for example, the stab is run in on drillpipe and is stabbed and latched into the receptacle. After the lower-marine-riser-package disconnect is actuated, the drillpipe is picked up, the shear bolts sheared, and the load transferred to the wire-rope bridle. The piece then is recovered by pulling the drillpipe.

Extended- Water-Depth Capability. Occasionally, a drilling vessel is considered that has a maximum-water- depth capability just short of the wellsite water depth (1,300-ft water depth with a 1 ,OOO-ft capacity rig as an example). To ensure that the rig is adequate for the location, consider additional riser availability and storage space; additional riser tension (or added buoyancy); lengthened control hoses and TV cable; additional guideline length; mooring system adequacy (mooring lines and


pennant wire); size of control hose reels (large enough to hold additional hose; ease of installing larger ones); size of guideline winch drums (large enough to handle additional line); and substructure strength (enough to support the added tension requirement).

Generally, the added water depth can be accommodated, but each rig and each site should be considered separately.

Operating Manuals and Emergency Procedures

Rig selection considerations should include a review of each drilling vessel’s operations manual and emergency procedures plan. The operations manual will include drilling operations and equipment-handling procedures. Nor- mal operating limits for discrete drilling operations will be specified. The emergency procedures plan should cover detailed responses and courses of action to be followed during marine emergencies, well emergencies, and bad weather situations. Disconnect and hang-off procedures must be identified, and special equipment should be on board to accomplish the suspension under adverse conditions. An agreement on well-control procedures should be reached between the drilling contractor and the oil company personnel. The drilling contractor personnel will implement the procedure, so if it is different from their previous procedures, additional training should be conducted.

Mooring and Riser Analyses

Mooring Analysis. Mooring systems and the objectives of station-keeping have been discussed briefly. The concept of the catenary and horizontal restoring force were mentioned. Combining these forces with the wellsite water depth, physical description of the rig’s mooring equipment, and environmental data is the task of a mooring analysis. Several commercial computer programs are available to perform mooring analysis. Some companies have developed their own programs. Mooring-analysis methods are documented in numerous articles and papers. Two are referenced at the end of the Floating Drilling section. In addition, API RP 2P discusses mooring analyses. ’

Fig. 18.17 describes the basic procedure followed in mooring analysis. Combining vessel characteristics and mooring-equipment specifics with bathymetry and weather data yields the length of mooring line to deploy, the initial operating tension, and the proof or test tension. The results can be obtained for a number of mooring configurations to determine which is optimum or simply to verify a recommended configuration.

Riser Analysis. Marine or drilling risers were described earlier. Accurate performance of drilling risers can be determined only by analysis. In floating drilling operations, the riser behaves as a string. It gains all of its structural integrity from tension. The single most important parameter in operation of the system, therefore, is riser top tension. Insufficient top tension can result in operational problems associated with large ball-joint angles and,

if low enough, buckling of the riser pipe body. Overten- sioning, however, produces high stresses in the riser that can result in a shortening of its life because of fatigue cracking. For each combination of environmental conditions, mud weight, riser weight, and vessel offset, there

is an optimum range of riser top tension.


Commercial programs are available to do riser analysis. As with mooring analysis, some companies have developed their own programs. API RP 2Q addresses riser design* and API RP 2K discusses riser use and maintenance. 9 Papers written on riser-analysis procedures are referenced at the end of the Floating Drilling section. The following discussion covers riser-analysis criteria and operational considerations, but not details of the complex analysis.

The items considered in riser analysis are riser stress, ball-joint angle, top tension, riser top angle, tensioner line angle, sheave friction, and riser pipe collapse.

Riser Pipe Stress. Static and dynamic stresses in riser pipe are calculated by the riser-analysis program. Static loads are caused by the riser weight, mud weight, current- induced hydrodynamic forces, the applied top tension, and deflection of the top of the riser. Deflection of the top of the riser is caused by vessel offset. Dynamic loads result from wave-induced water-particle motion and vessel surge/sway motion. Wave-induced surge/sway motion produces dynamic riser deflections and hydrodynamic forces because of the relative motion of the riser and the water.

The criteria for acceptable static and dynamic stress levels is shown in Fig. 18.18. For purely static loads (no dynamic load applied), stresses up to 50% of the pipe- material yield strength are allowed for normal operations and stresses up to 67% of yield strength for limited or emergency operations. These allowable stresses have factors of safety of 2.0 and 1.5, respectively. For purely dynamic stresses, the allowables have been reduced to 25 % of the pipe-material yield strength and 25 % of the pipe- material ultimate strength because of fatigue considerations. Combined static and dynamic stress states must fall within the recommended range indicated on the graph. High stresses occur in the pipe-to-connector weld and at the base of the groove in the connector pin. These two areas should be inspected frequently.

Bull Joint Angle. To minimize wear by the drillpipe, the angle of approach of the riser to the BOP stack should be kept as small as possible. Problems are minimized if this angle is maintained to less than lo--a goal readily attainable in a mild environment. With moderate to severe environments, establishing an allowable ball-joint angle of 3” is a compromise between wear problems and the application of criteria too restrictive to permit economical drilling operations.

The lower ball-joint angle is affected by many variables. Of these, rig personnel can readily adjust only riser top tension and vessel position. The rig’s riser-angle indicator should be monitored continuously and the vessel position and/or riser tension adjusted accordingly. Chang- ing the vessel location relative to the wellhead is the best method of minimizing ball-joint angle. The lower ball- joint (flex-joint) angle is the most important operating criterion to maintain.

Top Tension. For long-term operations, it is not desirable to work riser-tensioner systems at more than about 75 % of their rated capacity. To do so will result in premature failure, generally in the tensioner lines. Tension requirements can be reduced by the use of buoyancy.

Sufficient tension/buoyancy should be specified to prevent drastic consequences should one tensioner fail. Af- ter ball-joint angle, this criterion is the most restrictive

Fig. 18.18-Recommended stress ranges.

on tension requirements. When operating at the recommended tension, failure of one tensioner should not cause increases in ball-joint angle past 3’) and stress should remain in the recommended range for normal operations (see Fig. 18.18).

Minimum operating tension should always be sufficient for emergency disconnect. An overpull at the lower- marine-riser package connector of about 50,000 lbf is recommended to ensure that the lower-marine-riser package and riser will retract sufficiently to clear the top of the BOP.

Increasing riser top tension within the specified range can reduce bottom ball-joint angle. Increased tension beyond the maximum recommended, however, will significantly increase pipe stresses and have very little effect on decreasing ball-joint angle. At that point, the vessel must be moved to correct excessive ball-joint angle.

Riser Top Angle. Although the lower ball-joint angle is the most critical, the top angle must also be controlled. Tensions selected for drilling operations include a top angle of less than 4”.

Tensioner Line Angle and Sheave Friction. Variation in tensioner line angle generally has very little effect on riser tension. Sheave friction, however, may be substantial in some systems. If so, its effect should be compensated for in the tensioner control system so that the desired tension is maintained on the top of the riser.

Riser Pipe Collapse. Riser-pipe material strength and wall thickness should be sufficient to prevent collapse owing to seawater hydrostatic pressure when the riser is completely void. Reduction in collapse strength because of axial loading and bending stress should be included. In general, collapse considerations become important in water depths greater than 800 ft.

The objective of riser analysis is to specify recommended top tensions that keep the system within safe working limits under all anticipated conditions, as described in Table 18.3.

Field Operations With the major planning and preparation completed, we will now discuss the sequential steps of drilling a well from a floating vessel. The sequence of events in this description is not necessarily followed for every well drilled from a floating vessel, but it is a method that has worked in the past and will work in the future.

18-18

TABLE 18.3~-RECOMMENDED OPERATING LIMIT

Ball-Joint Angle Range

oto 10

1 to 30

3O and increasing 5O and increasing

Comments

Maintain ball-joint angle within these limits, if at all possible.

Maximum limit for normal operations. Preferably should be in this range only temporarily.

Start drillpipe hang-off procedure. Drillpipe hung off. Start riser disconnect

procedure.

Establishing Location

The location has been plotted on the map, the seismic work reviewed, and the drilling program written. Mate- rials have been delivered (especially subsea wellhead equipment, 30- and 20-in. pipe), and it is time to survey the location. Surveys must be accurate for several reasons. Locations near lease boundaries must be accurately placed from a legal or ownership viewpoint. Well location relative to seismic mapping is critical. If the rig has moved off location, getting back on location requires good survey data.

Today’s techniques can provide accuracy within 10 ft. In well-established areas-such as around the perimeters of the U.S., Canada, and in the North Sea-radio trian- gulation systems are used. In remote areas, satellite navigation (SAT NAV) systems, with receivers located on the floating vessels, are used. SAT NAV systems are accurate to within 3 ft. Depending on the well location relative to available satellites, however, it will take multiple satellite passes and approximately 24 hours to achieve that accuracy. The site may be marked with a buoy or the rig may be surveyed indirectly. Once on location with the mooring system set and tested, drilling is ready to begin. See Fig. 18.19 for the sequence of operations.

Spudding The Well

Step 1 in drilling from a floating vessel is to lower the temporary guidebase to the ocean floor. 3 The temporary guidebase is generally 12 by 12 ft and is outfitted with a bull’s-eye that is observed by TV for levelness determination. This base is run on drillpipe connected to it with a J-tool or hydraulic connector. Four wire-rope guidelines are attached to the subbase before it is lowered. The base may be loaded with weighted rotary mud so that necessary tension can be pulled in the guidelines when drilling equipment is lowered to the ocean floor. With the base on bottom, the drillpipe running string is marked at the kelly bushing, averaging out the vessel heave, so that the water depth may be determined by measuring the pipe when it is pulled. This measurement, corrected for tide, is the water depth from the kelly bushing to the ocean tloor that is used in all subsequent drilling, logging, casing, and testing operations.

Step 2 consists of running the drilling assembly-the bit, drill collars, bumper subs, and drillpipe-down the guidelines, through the temporary guidebase, and into the

ocean floor to drill the hole for the structural casing. This hole must be drilled carefully to ensure that it is kept within lo of vertical because it later will control how vertical the BOP stack will be when it is landed on the wellhead.


Several single shots should be taken while this hole is drilled. When drilling is completed, heavy mud is spotted in the hole to prevent sloughing or caving in. This hole generally has a 36-in. diameter and is 100 to 200 ft deep. The structural casing probably will be 30 in. in diameter with a 3/4- or 1-in.-thick wall. It is called structural casing because it serves as a foundation to provide lateral support for the BOP’s during subsequent drilling operations. In addition to drilling a hole for the structural casing, it may be driven or jetted in. In these instances, some drillers elect not to use the temporary guidebase.

Running 30-in. Casing

In Step 3, the structural casing (with the permanent guidebase attached and the guidelines threaded through the guideposts) is run down the guidelines into the hole. 3 Care must be taken while the 30-in. casing is run to ensure that it is tilled with water. Should this be overlooked, it is possible to collapse the 30-in. casing. It is run on, and cemented through, the drillpipe lowering string. Ce- ment returns are taken on the ocean floor and may be observed on TV. To ensure a good cement job for this critical casing string, the cement may be overdisplaced by as much as 100%.

In Step 4, hole is drilled for the conductor casing. If the soil the structural casing was set in is competent, a riser is run and latched into the permanent guidebase, and drilling returns are taken on the drilling vessel. Riser top tension should be sufficient only to minimize lower ball- joint angle. Too much top tension could result in pulling the 30-in. casing out of the ground. If the soil is not competent, the riser is not run and returns are taken on the ocean floor. If the riser were used with incompetent soil, the hydrostatic head of the drilling fluid could break down the soil at the shoe of structural casing, resulting in lost circulation. The conductor hole may range in depth from 298 to 499 ft. The size of conductor casing commonly used has a 20-in. OD. If the riser has been used, it is pulled before the conductor is run.

Running 20-in. Casing

In Step 5, the 20-in. conductor is run down the guidelines on drillpipe, landed in the housing on the structural casing, and cemented back to the ocean floor. 3 The top of the 20-in. conductor is fitted with a 163/4- or 185/,-in. high-pressure wellhead housing prepared internally to receive subsequent casing strings. The size of the casing head is determined by the size of the BOP’s on the drilling vessel. The external profile on the upper end of the casing head is prepared to match the type of wellhead connector installed on the BOP stack. A metal-on-metal seal ring provides the pressure seal between the connector and wellhead.

Running the BOP

Step 6 consists of function- and pressure-testing the BOP stack on the deck, then running it down the guidelines and latching it to the casing head.3 The BOP’s, which can weigh as much as 400,000 lbm and range in height from 30 to 40 ft, normally are run on the drilling riser. Depending on water depth, running the BOP stack can be a short or long procedure (from a few hours to several days). Each riser joint must be carefully inspected as it


LEVELING

BULLSEYE \

MUDLINE.

f

/

I

iI

I

a-

GUIDE LINES

.;OWERING

STRING

-FLEX JOINT

TEMPORARY

“GUIDE BASE 1

STEP 1 LANDING TEMPORARY GUIDE BASE

STEP 4 DRILLING HOLE FOR CONDUCTOR

I DRILL

STRING

, GUIDE ARMS

I

STEP 2 DRILLING STRUCTURAL CASING HOLE

STEP 5 RUNNING CONDUCTOR WITH CASING HEAD

GUIDE _

POSTS

RUNNING

TOOL

-PERMANENT

GUIDE BASE

-LOWERING

STRING

STEP 3 RUNNING PERMANENT GUIDE BASE AND STRUCTURAL CASING

MARINE

RISER

FLEX JOINT

LOWER MARINE

RiSER PACKAGE

BLOWOU 1

PREVENTER

STACK

STEP 6 BOP’S INSTALLED READY TO DRILL TO T D.

Fig. 18.19-Floating drilling-subsea systems.


Fig. 18.20– Dynamically positioned drillship.

is run. Each riser connection must be checked for cor-rect makeup. The total weight of the BOP is supportedby each connection. Integral K&C lines (and hydraulicsupply line if it is hard-piped) should be pressure-testedevery second or third joint to avoid an unnecessary pull-ing of the BOP for leak repair. In addition to careful riserinspection, riser handling tools and the riser spider shouldbe inspected for cracks or damage. The riser handling toolsupports the total weight of the BOP and riser each timeanother joint is added to the string. The top of the riseris fitted with a slip joint to accommodate vessel heave andoffset. The slip joint lands in a diverter housing immedi-ately under the rotary table. Riser tensioning lines are con-nected to the slip-joint barrel so that tension can be appliedto the riser when the stack is landed on the wellhead. Oncein place, the stack is function- and pressure-tested, andall is ready to begin drilling to total depth (TD). As holeis drilled, additional casing strings may be run throughthe riser and BOP’s. Periodic BOP testing after the firstcasing string is landed in the wellhead housing must bedone carefully. A leaking casing-hanger-seal assemblycould collapse the casing if test pressure exceeds casing-collapse pressure. API RP 536 includes testing guide-lines. If the well is to be put on production, the tubingstring also is run through the BOP’s and hung off in thecasing head.

Drillstem TestingDrillstem testing (DST) requires specific equipment notnormally installed on floating vessels. An inside-the-BOPproduction tree includes redundant master valves and asurface-actuated hydraulic latch for emergency discon-nect, high-pressure piping and valves from the chokemanifold to production-equipment area, test trap withmetering equipment, storage tanks with transfer pumps,and a flare boom. This equipment requires considerablespace. Storage-tank location may require beefing up thelocal deck structure. Flare booms require special foun-

dations. Installation and operation of production well-testequipment requires planning and significant rig-up time.

The BOP production tree space-out is critical. The treelands in the subsea wellhead housing, and BOP pipe ramsseal the casing tubing annulus. The tree height, includ-ing the emergency disconnect mandrel, must not extendabove the bottom of the blind shear ram. In case of emer-gency disconnect, the blind-shear-ram area must be clearfor shut-in.

DST is a critical operation. It must be conducted undercarefully controlled conditions. If H 2 S is possible in theproduction, special precautions are necessary. Local regu-lations and API RP 49 cover H2.S requirements. 10

Plug and AbandonmentIf the decision is made to abandon the well, it must beplugged first. Local regulations dictate plugging details.Abandonment plugging generally consists of laying ce-ment plugs in the wellbore at specified intervals to justbelow the ocean floor. The plugs are pressure tested asthey are installed.

Next, the subsea wellhead and bases must be recovered.This is accomplished by cutting the casing strings approx-imately 15 ft below the mudline. Cutting can be done withmechanical cutters (tools are available that can cut133/8-in., 20-in., and 30-in. strings in one step) or withexplosives. If explosives are used, the rig may have tobe moved several hundred feet away from the wellhead,depending on water depth, so that the explosion shockwave won’t damage the vessel’s hull. Retrieved wellheadequipment and bases can be reconditioned and reused.

Special ConsiderationsDeepwater DrillingMaximum water depths continue to extend. During 1983,a well was drilled in approximately 6,800 ft of water offthe U.S. east coast. Locations in water depths greater than


_,- . . -I R1

SC-in. HIGH STRENGTH CASING ASSEMBLY

JACK-UP AT WELL SITE

Fig. 18.21-Thirty-inch casing with helical strakes.

2,000 ft should be considered deep water. Mooring becomes more difficult, subsea equipment is heavier, collapse under hydrostatic conditions becomes critical, equipment performance under emergency disconnect conditions, and well-control procedures require additional considerations.

The major differences in shallow vs. deepwater floating drilling equipment are station-keeping (moored vs. dynamic positioning), riser design (material strength vs. buoyancy needed), BOP control systems (hydraulic vs. multiplex), and backup systems (diver vs. unmanned).

Deepwater operations require longer calm-weather periods and improved weather forecasting to accomplish specific tasks. Running and retrieving BOP’s can take just a few hours in shallow water. In deep water, 2 or even 3 days may not be unreasonable. Relatively calm sea conditions are required during that time.

Dynamic-positioning (DP) systems have extended station-keeping capabilities to depths of more than 10,000 ft. DP systems consist of acoustic beacons located on the ocean floor, hydrophones mounted on the vessel’s hull, thruster units fore and aft, and an on-board computer system for control. Dynamically positioned drilling vessels are equipped with from 12,000 to 20,000 hp for stationkeeping. Increased fuel consumption while operating in the DP mode is a major cost increment in a rig’s day rate (see Fig. 18.20).

Cold Environment

A few special cold-weather drilling vessels are available today. However, most rigs operating in cold environments today were not designed with low-temperature steel requirements. Highly loaded or highly stressed components -such as the derrick substructure, lifting subs, riser running tools, riser connectors, and elevators-must be fabricated from steels with low-temperature resistance (Charpy impact values comparable to the temperatures encountered) if they are to function safely. Material specifica-

tions of highly loaded components should be checked and verified before an unknown rig is taken into a low- temperature work situation.

Other cold-weather equipment considerations include quarters, insulation and heating, work-area heating, control-fluid freeze protection, water-system freeze protection, water-tank heating, and superstructure de-icing capability.

The American Bureau of Shipping specifies requirements for ice-class rigs. ” Those specifications include hull strength in the ice zone, thruster or propeller protection from ice chunks, and other specific requirements that must be met before a rig (ship or semisubmersible) can be certified to work in ice areas.

High-Current Drilling

The first concern in drilling high-current locations is stationkeeping. Does the mooring system have adequate strength, or the dynamic-positioning system adequate horsepower, to keep the drilling vessel on the location? With the mooring analysis previously discussed or with more sophisticated techniques to evaluate dynamic- positioning stationkeeping, we can determine the adequacy of the proposed rig’s stationkeeping system.

The next concern is possible fatigue failure of 30- and 20-in. casing strings (generally in a connector) owing to vibration. High currents can cause vibrations that induce failure in hours under the right conditions. Any surface current of more than 3 knots should be considered high current. Casing strings of 30 and 20 in. have been fabricated with helical strakes to break up the vortices that cause the severe vibrations (see Fig. 18.21).

Vortex shedding, which causes high-amplitude vibration at 90” to the direction of the current, can create severe problems in drilling risers also. Riser fairings have been developed and used on sever,1 occasions to eliminate the troublesome vortex-shedding vibrations success-

fully (see Fig. 18.22). Specia! equipment, in addition to


NOTE: KT = KNOTS

D = DIAMETER

Fig. 18.22-Riser vortex shedding.

strakes and fairing, can be installed on the drilling vessel to allow successful drilling operations in high-current areas. A floating rotary table and a moonpool riser- centralizing system have been developed to accommodate the high loads and high angles imparted on pipes and risers during high-current periods. If current direction is fairly consistent, installing early-warning current-meter strings 2 to 3 miles upstream from the rig can greatly assist in coping with the oncoming current conditions.

Structures Background and Philosophy

As exploration and production encroaches into deeper water and harsher environments, the challenges of structural design increase. Environmental load predictions, transportation analyses, and installation procedures are as important to understand as the more obvious structural- frame analysis. Seldom is a designer afforded the luxury of optimizing a structure on the basis of in-place stress analysis. More often, the transportation and installation (lasting a few weeks out of perhaps a 20-year structure life) will dictate the major framing patterns. Equally dis- gruntling to the structural designer is that most of his accomplishment is seen by only a handful of people, especially once it is in place. For the structure’s lifetime, it is expected to support drilling and/or production operations, and the operator cares little beyond that.

So the structural engineer’s job can be simply defined as getting it designed, built, and in place as quickly and economically as possible, while ensuring functionality and providing for minimal maintenance.

Structure Classification

As indicated in the Historical Review, many types of offshore structures are in service. Some are better suited to certain environmental and operational criteria; some are limited by availability of construction sites; and some are chosen simply by subjective preference of an owner/operator. Selecting a structure type is the first major structural design task after environmental and operational criteria have been defined and might require preliminary design of several concepts before a choice is made.

Template/Jacket. “Template” was derived from the function of the first offshore structures: to serve as a guide for the piles. The piles, after being driven, are cut off

above the template, and the deck is placed on top of the piles. The template is prevented from settling by being welded to the pile tops with a series of rings and gussets. Hence, the template carries no load from the deck but merely hangs from the top of the piles and provides lateral support to them.

Some companies prefer to place packers in the bottom of each template leg and to grout the annular space between the leg and pile from bottom to top. The structure and piles share the axial load from the deck and the com- pressive and tensile loads from the overturning moment produced by lateral wave loads. The grouted pile also provides additional strength to the tubular joints where horizontal and diagonal bracing are welded to the legs. Drawbacks to this system are the difficulty in ensuring that the grout is adequately placed and of sufficient strength to be counted in the analysis and the additional difficulty in platform removal.

Although both top-hung and grouted structures are loosely called templates, some prefer to call the latter a jacket to distinguish the difference in load path. This path is substantially different for the overturning moment as well as axial loads. The top-hung template requires that moment from lateral wave loads be transmitted up the structure to be resolved into axial pile loads. The grouted jacket has a direct downward load path for shear and moments. The novice designer would do well to learn this distinction early in his career.

When steel structures are designed for deeper water (in excess of 250 ft), pile-leg grouting is prevalent. Deep- water jacket designs are heavily influenced by lateral wave loads that produce high base shear and overturning. Piles placed through the legs of the jacket are not always sufficient to transfer these loads to the soil, so “skirt piles” are added, normally in clusters around the corner legs. This adds a new dimension to the installation procedure. Pile guides are required up to water level, and a removable “follower” must be used during pile-driving operations. Grouting procedures for the skirt sleeve-to-pile must

recognize that grout placement and inspection will be done remotely.

Template/Jacket Construction. A typical construction sequence for a template or jacket calls for yard constructing the unit on a pair of skid rails, skidding the structure onto a barge with matching skid rails, towing the barge to location, and launching. After the structure comes to


rest, usually floating horizontally, it is upended by ballasting members at the lower end. Once upright, it is moved onto the final site and lowered to the seabed by continued ballasting.

To date, the maximum water depth feasible for a single- piece jacket appears to be about 1,000 ft. The constraint is a result of construction equipment and facility limitations, although contractors are quick to point out that capability can be extended quickly if the business climate warrants. Meanwhile, an alternative for deeper water is the multipiece structure. These structures are built in either two or three pieces that are launched from separate barges. The pieces can be mated while floating horizontally, like the Hondo t* platform, or stacked vertically and grouted together with pin piles, like the Cognac ” platform. Mat- ing of large structural sections has proved to be expensive and involves added risk over single-piece structures, so plans are under way to extend single-piece construction to water depths beyond 1,000 ft.

Yard facilities are tailored to build structures in the horizontal position. Lifting equipment is huge in both capacity and reach because a deepwater structure might have a base width of more than 200 ft, which would require lifts to be placed to the height of a 20-story building! A structure for 1,200 to 1,500 ft of water could have a base width of more than 300 ft, which is beyond the reach of today’s equipment.

Since 1975, several large transportation barges have been built with lengths of more than 600 ft and capacities of 40,000 tons. Such a barge was used to place the single- piece Cerveza I4 platform in 935 ft of water. Several installation contractors now have plans for “super barges” with lengths of 900 to 1,000 ft and the capability of transporting jackets more than 1,500 ft long.

The increased pile loads on deepwater structures have necessitated advancements in pile-placing technology. In 1960, driving a 48-in. pile to 300 ft penetration for an axial capacity of 2,500 tons was a major feat. In 1984, a designer can plan for an 84-in. pile driven to about 400 ft to develop an axial capacity of 15,000 tons. Most of the advancement has come in the area of pile hammers. Hammers can be built to energy levels of 1.2 million ft- lbf, nearly 10 times that of the hammers used in 1960. Also, underwater hammers are becoming more dependable, which has the advantage of eliminating pile guides and followers for skirt piles and saving the energy normally lost through the follower.

Concrete Gravity Structures. As the name implies, these structures have large mat foundations instead of piles and are heavy enough to resist overturning and base shear from lateral wave and wind loads. Whereas steel template/jacket technology was largely a product of the U.S., the concrete gravity structures were European designs. Three reasons are offered for the emergence of these designs: (1) the European countries with offshore oilfields demanded a high percentage of national content-i.e., design and construction money had to be spent within their borders; (2) European design expertise, construction facilities, and construction skills leaned heavily toward concrete and away from steel; and (3) template/jacket technology was not prepared for the huge payloads and severe environment of the North Sea. The first major steel structure met with severe design changes and inadequate

fabrication and transportation facilities and was finally placed with high cost overruns. The high cost overruns of the first steel structures opened the door to concrete structures. Although these structures also ran into severe cost overruns, they had the advantage of not being payload-sensitive.

The concrete structures were necessarily massive to resist overturning. Furthermore, the concrete members were often controlled more from pressure than from axial forces of the deck load. Hence, changes in the payload requirement that occurred during construction and even for years after placement could be accommodated with relative ease. This proved valuable as the prolific North Sea fields were developed and facilities were continually added to existing structures.

Most concrete gravity structures were designed to store crude oil until it could be loaded on a transport tanker. This was required in the early stages of development because building an adequate pipeline network for the North Sea took many years. Only a few of the structures still are used for oil storage, and tanker-loading’systems are active only in Norwegian fields where a deep coastal trench has delayed pipelines from coming ashore.

Concrete structures have two basic shapes. The French design generally consists of a large, vertical, cylindrical structure surrounded by a perforated wall. The internal structure is capable of storing oil and supports the center of the deck structure. Typically, the perforated wall has 6.6-ft-diameter holes at 13-ft spacing that allow seawater to flow through as the wave passes, thereby reducing the wave force. The wall ends about 33 ft above the water level, and bracing extends up to support the edges of the deck.

Gravity Platform Construction. Concrete structures are nearly always built vertically. The lower portion of the base is built in a graving dock. The dock is flooded to sea level, thereby floating the base. The gate or dike is removed and the structure is towed to a protected deepwater site in a fjord or firth. The remainder of the structure is built by use of slipform methods and increases in draft as construction continues. If the platform has been designed for 450 ft of water, the draft of the completed substructure might be 300 ft.

Meanwhile, the deck is built, outfitted, and skidded onto one or two barges, depending on the configuration. Sub- structure and barge-mounted deck are towed to a mating site where water depth exceeds the design depth. The substructure is ballasted until only a few feet are above water. The barge-mounted deck is towed carefully over the substructure, and the substructure is deballasted until it picks the deck off the barges.

After the mating is complete, the platform is deballasted to minimum draft for towing to location. Minimum draft is still deep-perhaps 350 ft-and is limited by platform stability; further deballasting would raise the platform until it became top-heavy and toppled over.

The tow to site is nearly as critical as the mating operation. The tow route must be carefully surveyed, chart- ed, and marked. Deviation from this route could result in stranding the structure on a sand bar or hitting a submerged canyon wall. A fleet of the world’s largest tug- boats is required to tow the platform safely. Power, steerage, and standby tugs must all respond to the direction of the tow master.


gle of the tower can be restrained to 2 or 3°. The Lenaplatform in the Gulf of Mexico is the only guyed towerbuilt and in service to date. Twenty-four guylines sup-port the structure; their diameter varies between 5 and

Fig. 18.23– Tension leg platform

Once on site, ballasting will set the structure on bot-tom, and grouting fills any voids under the base. Offshoreconstruction and hook-up time can be held to a minimumbecause onshore deck construction should have includedcommissioning most of the systems.

Guyed TowersThe guyed tower is intended for use in deep water—perhaps 1,000 to 2,000 ft. It applies the principle of “com-pliancy” to wave forces. Wave forces are cyclic in na-ture, pushing a structure first in the wave direction andthen against it. By pinning the structure to the seafloorinstead of making it a fixed cantilever, the guylines al-low the structure to sway back and forth with the waveforce, transmitting only small loads to the foundation. Theguyline system holds the tower upright and resists thesteady forces from wind and current. Depending on thesize and configuration of the guyline system, the tilt an-

53/, in.The vertical support system resembles the template in

that the tower is welded to the top of the piles and hangsin tension. The piles carry the deck load directly and, inaddition, the vertical component of the guyline tension.

An additional, undesirable, axial load is caused by thetilting of the tower. The piles on one side compress asthe tower tilts, and on the opposite side they stretch. Theresulting stress cannot be eliminated, only controlled. Thepiles are clustered near the center of the tower to reducethe stretching and compressing. The system can be de-signed successfully by limiting the tilt angle and provid-ing sufficient pile length to absorb the compression.

As the tower tilts, the piles must bend through the tiltangle in a relatively short distance near the mudline. Thestress can be controlled in this case by locating the guidesproperly and determining the stiffness of the soil. The in-fluence of the tilt angle on the pile stresses must be evalu-ated before the guyline system is sized.

Guyed Tower Construction. The guyed tower is lighterand more slender than a jacket would be for the samedepth. Therefore, yard and transportation equipment canhandle a guyed tower for depths greater than that for atemplate. Installation procedures, however, are morecomplex because of the heavy guyline system and the largepiles.

After the tower is upended and set on the seafloor,buoyancy must hold the structure upright until the guy-lines are installed. A derrick barge is needed for drivingpiles, placing the deck, and setting equipment modules.This barge normally relies on its own mooring system forstation keeping, but in this case, some modification willprobably be required to prevent the tangling of mooringlines with guylines. The entire installation sequence mustbe planned carefully to ensure safety of the structure, guy-line system, and floating equipment.

Tension Leg PlatformThe tension leg platform (TLP) is another structural sys-tem based on the “compliancy” principle. The platformis composed of a deck structure and a buoyant hull thatis made up of a series of vertical cylindrical columns,horizontal submerged pontoons, and tubular member brac-ing. The platform is tied to the seabed by a number oftendons that are kept taut by excess buoyancy in the hull.Fig. 18.23 shows a TLP designed for 1,600 ft of water.Four vertical 52-ft-diameter columns are tied together with28-ft-diameter pontoons, and four tendons at each cor-ner column (16 total) tie the platform to the seafloor foun-dation template.

To date, the Hutton platform in the North Sea is theonly TLP in service. It was placed in only 485 ft of water,although the TLP is considered by many as the mostpromising structure for water depths from 1,000 to4,000 ft.

Because the TLP is lightly restrained horizontally,steady forces from wind and current will cause large ex-cursions of about 5 to 10% of water depth. Second-orderwave forces, insignificant on fixed platforms or guyed


towers, can cause substantial steady offset and “slow-drift oscillation” at the natural surge period of 60 to 120 seconds.

The tendon system and well system on the TLP are the most structurally critical and set the limits on horizontal excursion. Often containing flex joints or tapered cantile- vered pipe sections, these systems require careful analysis both for maximum stress and for fatigue life.

The foundation system is subjected to cyclic loads superimposed on a high tension. This system also requires careful analysis based on the best possible geotechnical data for the site.

TLP Construction. The TLP is structurally similar to the common semisubmersible drilling structure, and most efforts to design it differently or to different standards result in extreme cost and schedule overruns. The size of the TLP is related directly to the payload required and the environment it will withstand. This normally means that the displacement (buoyancy) will be between two and five times that of a semisubmersible. Therefore, many shipyard facilities used for semisubmersible construction will be too small for a TLP.

Two distinctly different construction methods are available. The first is separate construction of the deck and hull, requiring a mating sequence similar to that described for a concrete gravity platform. The second method is single-piece construction where the deck is built onto the hull and equipment placed on the completed platform. The first method offers reduced construction time; the second allows for intermediate structural bracing to support the deck because clearance is not required for a barge. The choice between these two construction methods must be made early in the planning stage to take advantage of the benefits afforded by either concept.

The most difficult and uncertain part of TLP construction is the placement of the seafloor template or templates and the piling. Again, several options are open-a single template for the foundation and well system or separate templates. Tendons may be attached to the template or directly to the piles. Again, each system has advantages and disadvantages, and all aspects of allowable tolerances, seabed levelness, potential for settlement, and preference for separate structural systems should be considered.

Special-Service Structures

Many structures have been built for a specific location or specific purpose and are essentially one of a kind or perhaps one of a select few. Self-floating jackets, steel gravity platforms, tanker-mooring articulated towers, mooring dolphins, and single-well caissons are some special-service structures. Each of these concepts and the innovative effort that went into them deserves a chapter in a book, but unfortunately there is not room here. The author recognizes the effort needed to learn the special hydrodynamics, structural dynamics, stress analysis, and construction methods needed for a unique design. Many unheralded innovative engineers have gone virtually unnoticed because their structure, though successful in the purpose for which it was intended, simply was not needed anywhere else.

Structure Selection

From the preceding discussion, it should be obvious that the chosen structure is not always the one that theoreti-

cally has the least concrete or steel. Construction facilities and skills in the area are important, and often the operator’s preference will prevail. Here are some guidelines, though, on the four concepts discussed.

Template/Jackets. Around the U.S., these structures are the norm for water depths from 10 to 1,000 ft. Most U.S. offshore structures have a payload based on less than 50,000 B/D production. These small payloads are conducive to templates and jackets. Future technology might push these structures to 1,500 ft of water; limitations will be based on the structure’s natural period and cost.

Concrete Gravity Structures. These structures require deepwater construction sites in protected water. Norway and England have such sites; the U.S. does not. These structures can readily carry the equipment for up to 200,000-B/D production. The North Sea will probably need more structures with these production rates; we hope the U.S. will also. Depth limitation is probably around 700 ft, unless new construction methods are developed.

Guyed Towers. These structures should be good for deep water and heavy payloads. Because deckloads are carried directly by the piles and only indirectly affect the tower and guyline system, heavy payloads should not cause a substantial increase in structure costs.

TLP. These structural systems should be relatively economical for deep water and light payloads. Increased payload directly affects buoyancy requirements, which in turn directly increases waveloads and tendon and foundation requirements. Conversely, deeper water requires only longer tendons. With a design based on realistic understanding, this concept could be extended to a 4,000-ft water depth.

It will be interesting to see how the future will bring about changes.

Structural Design Process

As previously indicated, there are many types of offshore structures, but the following example will be confined to a barge-launched jacket. The considerations and procedures, however, have a parallel in the design and construction of any structure.

Methods for determining environmental loads and for analyzing structures are outlined in API RP 2A, “API Recommended Practice for Planning, Designing, and Constructing Fixed Offshore Platforms. ” This document is one of a series of recommended practices that are reviewed frequently and reflect the state of the art.

Fig. 18.24 shows a rather large offshore jacket for about 500 ft of water. Drilling and production facilities are assumed to be in modules for placement on top of the module support frame.

The structure must be designed for a variety of conditions, including fabrication, transportation on a barge, launching, upending, placement, and operation. Each condition controls the size of some members. The members controlled by in-service conditions may require several separate analysis procedures to size them adequately.

Field Development Plan. To put the design process in a time frame, Fig. 18.25 shows that portion of a possible field-development program that relates to the jacket,


Fig. 18.24-Jacket members as sized by design conditions.

recognizing that facilities design and construction are parallel. The economic studies are to determine the via- bility of the program and must include a preliminary design of the jacket to determine a cost estimate. Analysis will probably be limited to a single topside load and the maximum storm wave, using a first guess at water depth, wind speed, wave height, and current velocity.

Environmental Criteria. As the design progresses through two or three of the cycles shown in Fig. 18.25, more exact environmental data will be needed, and the oceanographers probably will be working to firm the data up in parallel with the design. Close coordination between the designer and the oceanographer is critical. The information must be available on time, sufficiently detailed, and in a format compatible with the design procedures.

Storm directionality might determine the orientation of the platform. Because of transportation requirements, a jacket is rarely symmetrical, so it will resist lateral wave loading better along one axis than the other. The pattern of well conductors is often dictated by the reservoir characteristics, however, so aligning the strong axis with the most severe storm direction is not always possible.

In addition to maximum storm criteria, statistical wave data and scatter diagrams are needed for fatigue, transportation, and launch analysis, and for determining weather windows for derrick barges that will be used for pile-driving and lifting modules. This information might point out a substantial seasonal variation in weather statis- tics that will dictate the time of year for installation. The designer should anticipate the required magnitude and format of environmental data and pass this requirement to the oceanographer early in the project.

Fig. 18.25-Jacket design-key environmental data inputs.

In-Situ Analyses. The in-situ static procedure in Fig. 18.26 shows the loading systems applied to the jacket. The stiffness analysis assumes these loads to be static, and the results must be combined with the wave data and the natural periods of the structure to derive a dynamic amplification factor. The “combination processor” must combine static and dynamic loads to determine member and joint forces, moments, and displacements. These are then converted to stresses and compared with those permitted by the design codes. If the design is not acceptable, members and joints must be resized and the process repeated.

The linear stiffness analysis usually is adequate for the steel framework, but must be adjusted for the pile/soil interaction that is commonly nonlinear when extreme- event loads are applied. Various linearization techniques or iteration procedures are used to arrive at the correct pile displacement and rotation. This is probably the most difficult and critical area in the analysis because improper stiffness assumptions, either too high or too low, can result in a nonconservative structure.

Transportation and Launch. The approximate location where the structure is to be fabricated must be known to

estimate the tow route and duration. Also, the season must

be established so that sea-state predictions are appropri-

ate. For Gulf of Mexico structures, tow routes normally

are short, and weather forecasts often can cover the entire duration of tow and launch. For the west coast, fabrication might be in the orient, with a tow duration of 6

to 10 weeks. Thus, early construction planning is required

for adequate design and analysis.


Fig. 18.26-In-situ “static” analysis.

The analysis procedure for both tow and launch is shown in Fig. 18.27. A wave-scatter diagram is used to define significant wave height and period combinations for the various sea states to be encountered. Barge motions can be determined either by analysis or model tests. Member accelerations, dead weight. and tiedown reactions can then be fed into the jacket stiffness analysis to determine stress levels for checking against design-code allowables.

The launch analysis requires a step-by-step simulation of the jacket tipping off the barge on the rocker arms until it is floating freely. Loads and reactions. including hydrodynamic loads on submerged members, must be retained for a stiffness analysis

Fatigue Analysis. There are two distinct methods of fatigue analysis that are used in industry. The first and most easily understood is the deterministic analysis (shown in Fig. 18.28). This procedure will be described in some depth. The second method is spectral analysis, which involves a statistical prediction of stress magnitude and cycle distribution in each sea state and over the life of the structure. This method is gaining favor in industry but is too complicated to treat adequately in the context of this chapter.

The basic principle of fatigue damage is straightforward. although its accuracy is often debated. It states that for a given stress level, the fatigue damage, F,,. is equal to the number of applied cycles, n. divided by the number of cycles required for failure, N, or F,, =tdN. For example, if a steel can withstand 2.5 x 10’ cycles at 10 ksi before failure occurs and 6.1 x lo6 cycles are applied, then the damage is F,, =(6.1 x 106)/(2.5 x 10’)=0.244. This indicates that the member is 24.4% damaged. The common procedure for checking total damage is Miner’s rule, which requires that the summation of all damage be less than 1.0. The recommended fatigue curves and al-

Fig. 18.27-Analysis procedure for tow and launch.

lowable cumulative damage can be found in API RP 2A. The fatigue curves show stress, S, plotted against the number of cycles to failure, N. and are often called S-N curves.

If the natural period of the structure exceeds 3 seconds, API RP 2A recommends that a dynamic amplification factor (DAF) be included in the analysis. The DAF accounts for increased stress because of structural vibration. A complete dynamic analysis may be performed in lieu of using DAF’s, but this can become extremely expensive.

Fig. 18.29 shows a spectral fatigue analysis that can be compared to the deterministic analysis in Fig. 18.28. Ref. 13 outlines this procedure.

The previous discussion covers most of the major analyses required for a platform, but don’t be misled by the brevity. Between 75 and 100 complete stress analyses usually are required to determine stresses for in-place fatigue. Transportation fatigue might require twice that many. If seismic analyses are included, the total bill for computer time might exceed $1 million for a thorough final analysis.

Other structures will have unique design and analysis problems. Generally, the same types of analyses arc required, but they differ in areas of dynamic response or hydrodynamic loading. Ultimately. the extent and choice of the analysis procedure falls back to engineering judgment, and there is absolutely no substitute for experience.

Offshore Production Operations Offshore production installations can be either very similar to or radically different from land installations. The purpose of this section is to provide a general overview that will acquaint inexperienced persons with typical designs and requirements for completing and producing wells offshore and describe alternative installations that are currently available to engineers and operators for meeting production objectives under varying conditions.

PETROLEUM ENGINEERING HANDBOOK 18-28

Fig. 18.28-Deterministic procedure analysis.

Platform Production

Well systems and crude-oil and gas process facilities installed on platforms account for more than 99% of current offshore production capacity. A small number of wells are completed on manmade islands. From a design and operating standpoint, these island wells are handled the same as platform wells. Wells that are completed subsea number fewer than 300 worldwide and will be discussed separately.

Well Completions. Except for a few innovative installa-

tions, wellheads and Christmas trees on platforms are bas-

ically the same as for land wells (Fig. 18.30). Control

valves, safety valves, and piping outlets are configured

the same and use the same or similar components. Some

of the valves probably will have pneumatic or hydraulic

actuators to facilitate remote and rapid closure in an emer-

gency. Also, some Christmas trees may have composite

block valves instead of individual valves flanged together.

The major difference, however, between land and platform well completions is the economic incentive on platforms to reduce equipment weight wherever possible and to minimize space requirements. Simply put, lighter, smaller equipment and more compact installations result in less-expensive platforms. A good example is the use of composite block valves to reduce Christmas-tree size and weight. Another example is the spacing of wellheads as close together as drilling operations will permit, with just enough room for safe and efficient operation of tree valves, control valves, and well-workover equipment. Typically, this means centerline distances of 6 to 10 ft between wellheads.

Where only one drilling rig is on a platform, all the wellheads usually are located in one bay. Larger platforms that are designed to accommodate two drilling rigs may have two well bays (one for each rig) with two or more rows of wells in each bay.

EACH WA STATE EACH mnEcTloN

Fig. 18.29-Spectral fatigue analysis procedure

Process Equipment. The primary function of process equipment, whether on a platform or on land, is to stabilize produced fluids and to prepare them for shipping or disposal. Well production is separated into components of oil, gas, and water (and sometimes condensate). The separated fluids are measured and then either shipped, injected back into the reservoir, or flared.

Differences between the process equipment (oil and gas separators, free-water knockouts, gas scrubbers, pumps, compressors, etc.) installed on a platform and those installed on land are minor (Fig. 18.3 1). Where possible, consideration is given to using vessels and machinery that are compact and lightweight-e.g., electric motors are commonly used instead of gas engines for driving pumps and compressors. Vertical clearance between decks may impose height limitations and dictate, for example, the use of horizontal instead of vertical separators.

There is a major difference, however, in the way equipment is packaged. If it is to be installed offshore after placement of the platform jacket and decks, process equipment usually is built in modules at a land site. The module assemblies then are barged offshore, lifted onto the platform, and hooked up. This significantly reduces expensive offshore installation and hookup time. In any event, the equipment and its piping, wiring, and controls are installed as compactly as possible. The extra engineering and fabrication costs needed to reduce deck area to an absolute minimum are more than offset by savings in platform structure cost.

Well Servicing and Well Workovers. On relatively small platforms with no more than 5 to 10 wells, it is common practice in some areas to drill all the wells before any of them is placed on production. The drilling rig is removed after the last well is drilled, and future well workovers are performed with a portable workover rig or well- pulling unit. Downhole work that does not require pulling tubing (e.g., replacing safety valves, gas lift valves. or standing valves) normally is accomplished with a wireline unit.


Fig. 18.30– Platform well bay

On larger platforms with more wells, drilling and pro-duction operations generally are carried on concurrent-ly. In this case, well workovers are performed by thedrilling rig if it is still on the platform. Depending on theurgency of the workover and on economic considerations,the work typically is scheduled to follow completion ofthe well that is currently being drilled. Wireline repairscan be performed without interfering with drilling oper-ations unless the two wells involved are too close togetherfor safety considerations.

Even on the largest platforms, drilling rigs normallyare removed after all scheduled wells have been drilled.Depending on the number of wells and the amount ofdownhole work anticipated, a special workover rig maybe installed permanently on the platform. An economiccomparison between using a portable rig and using a per-manently installed rig should be the basis for selection.

Crude-Oil Disposal. In the great majority of cases, crude-oil production is “shipped” from platforms by subseapipelines. Because most offshore producing areas involvemultiple platforms and more than one operating compa-ny, the pipelines are generally common carriers.

In the simplest of installations, where a pipeline trans-ports only one type of crude oil from a single platform,an optimum pipeline design and installation can be in-volved and expensive. Numerous factors must be evalu-ated, such as seawater temperature, seafloor profile andgeologic features, water currents between surface and

seafloor, hazards from commercial fishing equipment andboat anchors, the possible need for a weight coating suchas concrete to ensure negative buoyancy, cathodic pro-tection, corrosion-prevention coatings, water depth, thepossible need for burial beneath the seafloor, the type ofbeach crossing that will best protect the pipe and also beenvironmentally acceptable, the best riser to use at theplatform to afford protection from corrosion and physi-cal damage from boats and waves, total length of pipe-line, pumping rates and pressures, the need for periodicpigging and inspection, safety shutdowns to prevent orto minimize pollution in the event of failure or accident,and the crude-oil properties and rheology. When crudeoil is shipped from more than one platform, a more de-tailed study of rates and pressures will be required, andif crudes from different reservoirs are being pumpedthrough the same pipeline, a much more detailed studyof oil properties and rheology will be necessary.

Depending on pipe diameter, length, need for burial,need for coatings and cathodic protection, water depth,and various construction considerations, an offshore pipe-line can be the single most expensive element of an off-shore installation, sometimes far exceeding the cost of oneor more platforms. In the great majority of cases, how-ever, piping is still the safest, most economical way totransport crude-oil production to a land site.

Occasionally, an offshore oil field is too remote, pro-duction rates are too low, or the field is too short-livedto justify a pipeline economically. The alternative is to


Fig. 18.31-Platform deck layout for process facilities.

ship the oil by tankers. This usually requires some type of loading system installed 1 to 2 miles from the platform, such as a moored buoy or articulated loading tower. A seafloor pipeline connects the loading facility to the platform, and a tanker is moored to the loading facility during the transfer of oil.

The two most important drawbacks of a tanker-loading operation are sensitivity to weather and the need for separate oil storage. Tanker loading is best suited to mild- weather areas to minimize downtime from storms. Oil storage requirements will depend on total field producing rates and reservoir characteristics (can the wells be shut in for short periods without loss in productivity) as well as tanker downtime. This has led to the development of permanently moored storage tankers.

Gas Disposal. Disposition of gas from an offshore production site will depend on a combination of reservoir and economic factors. If well production is primarily oil, the gas may be handled as a byproduct and be disposed of in the most economical way. Piping the gas to a land site for sale and use as fuel is generally preferred if it can be done economically. Injection back into the producing formation is a common alternative. This helps to maintain reservoir pressure and conserves the gas for possible fu-

ture sale. In some areas, gas flaring is still acceptable, but many countries now forbid it except for short test periods and for the disposal of small amounts of residual waste gas.

Water Disposal. Produced water is normally cleaned so that it may be either discharged offshore in accordance with governmental regulations or reinjected. In either case, a combination of mechanical and chemical means may be used to condition the produced water before disposal. Tankage and filtration are used to remove oil and other contaminants from the water. Chemical treatment is common to control bacteria and corrosion in injection wells.

Subsea Completions

Seafloor well completions occupy a very small niche in the offshore petroleum industry, but they attract a lot of attention. Their primary use has been as single satellite wells producing to a nearby platform. They are a means of producing field extremities that cannot be reached by directional drilling from an existing platform and where the economics do not justify the installation of one or more additional platforms. Some multiwell templates and piping manifolds have been installed that go beyond the satellite well concept.


The main benefit of subsea completion efforts is that the petroleum industry and political governments now recognize and accept them as a technically viable means for producing offshore oil and gas wells. This, in turn, means that future installations can be evaluated on the basis of actual field experience and realistic economic considerations. Subsea completions probably will increase in popularity in coming years, especially if significant discoveries are made in deep water where conventional platforms are either extremely expensive or impractical.

Wet vs. Dry. Seafloor installations are made either with the equipment protected by a dry, one-atmosphere pressure chamber (Fig. 18.32) or with the equipment exposed to the sea environment (Figs. 18.33 and 18.34). The dry chambers are large enough for workmen to install and to repair valves, flanges, and control systems in a shirt-sleeve environment. Access is by way of a diving bell lowered from a work boat. Successful installations have been made in water depths greater than 500 ft.

Most subsea completions are of the wet type and require varying amounts of diver intervention during installation and removal of the seafloor equipment. Minor repairs and trouble-shooting can sometimes be accomplished in place, but major equipment or control-system repairs are made above water after the faulty equipment is removed. Running tools operated from a floating drilling or workover rig are used for installation and removal of well completion equipment in much the same way that drilling operations are conducted when running and retrieving a BOP stack for a subsea well. Wet well completions have been made in water depths greater than 1,300 ft, and a satellite system has been constructed in 2,500 ft of water. The latter is designed to use tethered, remotely operated vehicles with manipulators instead of divers for trouble-shooting installation problems and for assisting with repairs.

Single Satellite Wells. Functionally, seafloor well completions are no different than wells on platforms or land. Completions in shallow water where divers are used extensively for installing the equipment may even look like land wells.

In deeper water, however, where diving requires expensive saturation systems for the divers, more reliance is placed on equipment that can be installed and removed with special tools that are run and operated from a floating drilling rig. Hydraulically actuated wellhead connectors are used instead of flanged or clamped connectors. Tubing hangers that can be locked in place and tested remotely are used. Precise equipment systems that can be remotely connected and disconnected and that permit personnel on the drilling rig to test and to function all of the wellhead equipment are required. Hydraulic controls are generally used for this work. The result is a Christmas tree that may be 20 to 40 ft tall when combined with its drilling bases and that may cost several times more than a land tree.

One very important requirement for subsea-well- completion equipment is that it be totally compatible with the drilling equipment that is used in drilling the well (Fig. 18.35). This requires extensive pre-engineering and preplanning between persons responsible for the drilling, completion, and producing operations.

Fig . 18.32~Subsea tree-dry type with diving bell in place.


Fig. 18.33– Subsea tree–wet type. Fig. 18.34– Exploded view of diverless subsea tree and runningtools.

Multiwell Templates. A seafloor template with guideposts and wellhead receptacles for more than one well issuitable for drilling a number of satellite wells in closeproximity to one another. Minor cost savings are real-ized during drilling operations because the drilling ves-sel does not have to be moved and reanchored betweenwells. Any slight adjustment in position can be accom-plished by taking in or letting out opposing anchor lines.A major savings in flowlines may be possible by com-mingling well production at the template and transferringit through one large pipeline instead of individualflowlines. Well testing can still be accomplished by in-stalling one separate flowline with a valve manifold forswitching wells. Multiwell templates offer additional op-portunities for reduced costs for control systems, gas-liftpiping, and water-injection piping. Obviously, potentialcost savings will be the greatest with a large number ofwells. Multiwell template designs should consider poten-

tial hazards from accidentally dropped tool joints or otherheavy equipment during drilling operations. This may bea deterrent to producing completed wells until all wellsin the template have been drilled. For a large templatewith many wells, this would have a significant effect oncash flow. A detailed economic analysis of all relevantfactors is essential to determine the optimum size and con-figuration of a multiwell template.

Manifolds. One technique for combining some of the ad-vantages of single satellite wells with the piping savingsfor multiwell templates is to produce moderately spacedsatellite wells to a central, seafloor manifold installation(Fig. 18.36). The manifold would include valves and con-trols to commingle or test each well selectively and wouldreduce overall piping- and control-system costs if plat-form process facilities are located several miles away.Both dry and wet manifolds have been successfully in-stalled and operated.


Fig. 18.35-Sequence drawings for drilling and completing a subsea well and for well re-entry workover.

Flowlines and Control Lines. Subsea satellite wells may be installed with either one or two flowlines, depending on well conditions and operating requirements. The production flowline is usually 2 to 6 in. in diameter. Size is dictated primarily by flow rate, flowline length, and wellhead pressure. A second flowline frequently is installed for communicating with the tubing/casing annulus. This line is effective for monitoring annulus pressure and for circulating kill fluid if needed. It can be hooked up to pump pigs, paraffin scrapers, or through flowline tools. It can also be used as a second production line.

In some cases, a decision to bury flowlines below the

seafloor can be made because of local regulations or obvious hazards. Generally, however, the pros and cons for burial and the overall costs and economics should be evaluated carefully before a decision is made to bury flowlines. If the answer is not clear-cut, leaving the lines unburied is probably best. Unburied lines cost less to install, are less expensive to repair, and are easier to inspect for leaks and damage.

Conventional welded steel pipe is used for most flowlines. It can be protected against external corrosion by either anodes, a corrosion coating, or a combination of both. Cathodic-protection methods should be compat-

Fig. 18.38-Seafloor manifold for satellite subsea wells.


ible with corrosion-prevention designs for equipment at both ends of the flowline (wellhead equipment and usually a platform jacket), as well as other nearby or crossing pipelines.

In recent years, flexible pipe made of laminations of steel wires and other materials has become popular for flowline service. Although the material cost for flexible pipe is usually very high in comparison with conventional pipe, this may be more than offset by savings in the installation cost. Work boats or special-purpose vessels equipped with large-diameter reels can lay long lengths of flexible pipe in short periods of time. Flexible pipe normally does not require a separate coating on the outside. but it may require cathodic protection of the end connections.

All flowlines should be protected from abrasion and physical damage from other crossing pipelines, and expansion loops may be necessary if the installed configuration does not allow for expansion and contraction from temperature changes. Other considerations include possible scour damage or vibration fatigue where bottom currents exist and high stresses where the line bridges low spots on uneven seafloor. Installation methods for flowlines are generally the same as for other subsea pipelines. These are discussed in the Offshore Pipelines section.

Control lines, both hydraulic and electrical, for subsea well completions are discussed in the Electrical, In- strumentation and Control Systems section.

Well Servicing-Wireline vs. Through Flowline. There are two common techniques for performing downhole work on subsea wells when the work tasks do not require removing the Christmas tree and tubing. The most common procedure is to install a workover riser between the surface vessel (drilling or workover rig) and the top of the Christmas tree above the swab valves and to install a wireline lubricator on top of the riser. Conventional wireline tools and equipment can be used to remove and to install safety valves and gas-lift valves, to shift sliding sleeves, or to make temperature and pressure surveys. Vessel heave must be compensated for, but otherwise the procedure is the same as for land and platform wells.

To overcome the delays and high cost associated with wireline work on subsea wells (vessel availability, high daily rates, and weather delays), a procedure for servicing wells remotely from the process platform has been developed. It is called “through flowline” (TFL). It is basically a set of tools that are inserted into the flowline at the platform and hydraulically pumped through the flowline, through 5-ft-radius flowline loops (bends) at the subsea Christmas tree, and down the tubing. A complete hydraulic loop is required between the platform and the well to pump the tools down and back. This means that a second flowline is necessary for work to be performed at or immediately below the wellhead and a second tubing string is required, with controllable communication between the tubing strings downhole to perform work downhole. TFL tools are available to fit common tubing sizes and to perform virtually all the same tasks as wireline tools. Numerous technical papers have been written on TFL tools and techniques.

Gas wells generally are not suitable candidates for TFL servicing because a gas-free hydraulic loop is required

for circulating the tools. A decision favoring either wireline- or TFL-servicing procedures for an oil well should be based on a full evaluation of operating conditions, anticipated downhole service requirements, the availability of trained personnel, and an economic comparison of installed costs and servicing costs.

Well Workovers. Work on subsea wells that requires the tubing to be pulled or is otherwise beyond the scope of wireline or TFL tools is a major undertaking and requires extensive planning and preparation. Unless a special- purpose vessel is available that is suitable for workovers, a semisubmersible or ship-shaped drilling rig must be scheduled and mobilized with a workover riser and running tools specially designed for the Christmas tree and tubing hanger. This equipment is needed in addition to a regular drilling riser and BOP stack, which must have a hydraulic connector that is compatible with the wellhead. Space limitations on some drilling vessels may preclude their use, which further complicates the workover.

Floating Production Facilities

For many years, floating drilling rigs (semisubmersibles, ships, and barges) have conducted drillstem tests and short-term production tests of newly drilled wells. These wells were nearly always drilled for exploratory or deline- ation purposes. Common practice was to abandon the wells temporarily or permanently after testing. Follow- ing the installation of a platform, development wells were then drilled and produced. Because brief production tests frequently provided insufficient reservoir data, and because delaying well production until platforms could be fabricated and installed resulted in poor cash flow, the concept of floating production facilities (FPF’s) was developed. This concept requires some type of floating vessel (ship, barge, or semisubmersible) that is equipped for processing crude-oil production instead of for drilling. The vessel is either moored in place with multiple anchors, or is connected to some type of single-point mooring (SPM) or articulated tower. Crude-oil production from one or more seafloor wells is produced to the FPF either directly through individual pipe risers or through a seafloor manifold center and multiple-bore riser assembly (Figs. 18.37 through 18.39).

Applications. Two applications for FPF’s were mentioned above: long-term production tests and accelerated (early) production. The use for long-term testing nearly always involves only one well and may be for a duration of anywhere from 2 to 3 months to a year or longer. Some reservoirs, particularly limestone as opposed to sandstone, cannot be evaluated from a short-term test, and full field development may be unacceptably risky without extended test data. The simplest FPF installation for production tests is a vessel moored with multiple anchors directly over a seafloor well connected by drillpipe or tubing to a subsea test tree installed in a BOP on the wellhead. A similar installation generally preferred for longer-term testing would use a flexible pipe riser instead of drillpipe or tubing and a conventional subsea Christmas tree instead of a BOP with test tree.

The same installations described above can also be used for early production if only one or two wells are involved. For more wells (up to 5 or lo), a seafloor manifold center


Fig. 18.37-FPF with flexible pipe risers and floating loading hose for two subsea wells.

may be used with a multiple-bore riser and some type of economical. Particularly where reuse of the facility may

SPM or articulated tower. An SPM or tower has the added be a factor, the ease of moving an FPF to a different lo-

advantage of permitting the process vessel to weather-vane cation or field may have a significant impact on overall into the wind and seas and thus to mitigate the effect of economics.

bad weather on process equipment. A third application for FPF’s is permanent process fa-

cilities for a small or short-lived oil field where a conventional platform is either uneconomical or marginally

Semisubmersibles vs. Tankers. Most FPF installations have been converted semisubmersible drilling rigs or con- vetted oil tankers. Semisubmersibles are characteristically

Fig. 18.38-FPF with seafloor manifold, composite riser, and seafloor pipeline to loading buoy and tanker.


Fig. 18.39-Flow diagram for FPF process equipment for one-well, long-term test.

better suited to severe weather areas such as the North Sea, because of their superior stability. Their main drawback from an operating standpoint is the lack of sufficient oil storage capacity to prevent shutdowns when tanker- loading operations are disrupted. Loading disruptions are not uncommon and can result from either equipment failure or bad weather.

In relatively mild weather areas, converted tankers perform satisfactorily as FPF’s, especially when moored to SPM’s or loading towers that permit the tanker to weather- vane. Large oil-storage capacity, inherent to tankers, greatly facilitates the scheduling of shuttle tankers to transport the crude oil to refineries or terminals.

The choice between semisubmersibles and tankers for FPF service is heavily influenced by the availability of surplus vessels, the operating conditions, and the geographic area. A careful investigation of the used-vessel market is essential to an economic decision. At the time of this writing, new special-purpose vessels designed specifically for FPF service are being promoted. These include semisubmersibles with some storage capacity to offset that drawback and ship-shaped vessels with anti- roll devices to improve stability. A wide selection of FPF designs and vessels probably will make their use more economically and operationally attractive in coming years.

Disposal of Oil, Gas, and Water. Technically, there is no reason why oil cannot be shipped from an FPF by pipeline. Most installations to date, however, have transported oil with shuttle tankers. Loading operations can be accomplished by a floating hose directly between the FPF and the shuttle tanker. This inexpensive approach is practical in mild weather areas, especially with low production rates. For higher production and transfer rates and for adverse weather conditions, a seafloor pipeline or hose between the FPF and a dedicated loading buoy is safer and generally will result in less downtime.

Water produced with crude oil can be treated and disposed of the same as on a platform. It can be cleaned and

treated for reinjection into the reservoir, or, where local regulations permit, it can be cleaned and discharged into the ocean. Depending on downstream terminal or refinery conditions, shipment of small amounts of water production with the crude oil may be possible.

Small amounts of gas production have to be flared in a typical FPF installation. If flaring is not permitted, or if economics favors reinjection, compressors can be installed for this purpose. Use of FPF’s in fields producing large amounts of gas may require additional facilities for gas treating, processing, and disposal.

Offshore Pipelines

The design and installation of subsea pipelines bear only slight resemblance to their counterpart activities on land. Preliminary sizing of lines can be based on general- purpose pressure-drop curves as long as the effects of the ocean environment on fluid rheology are understood. Also, preliminary cost estimates can be made on the basis of either estimating manuals prepared for this purpose or historical data for similar installations. Final pipeline designs, detailed plans, and cost estimates for fabrication and installation, however, are best handled by pipeline contractors or consultants who specialize in this activity.

Flowlines. Flowlines for subsea wells range in size from 2 to 6 in. As indicated previously in the Subsea Comple- tions section, conventional steel pipe is used for most installations. It is readily available and does a good job when protected against corrosion and physical hazards. Flexi- ble pipe made of laminations of steel wires and other materials is now available from several sources and has been used in a number of instances. It is made in a range of pressure ratings and a variety of materials that are suitable for most applications.

Welded flowlines sometimes are made up on the beach and then towed to the point of placement. Towing can be at or near the surface of the water with the pipe supported by buoys or other buoyant material, or it can be just off-bottom by a combination of buoys and chains for


SIDE VIEW

FRONT VIEW IWZ-~ODAT~ON DFCK LAWUT

Fig. 18.40-Conventional pipe-lay barge.

buoyancy control. More common methods of installation, however, are either by conventional lay barges (Fig. 18.40) or by reel barges (Fig. 18.41). The former make up straight lengths of pipe on the barge and feed it into the water by way of a curved stinger as the barge is winched along the flowline route. The purpose of the stinger is to control radius of curvature as the pipe is lowered into the water and thus to prevent buckling. Buck- ling and overtensioning of the pipe as it contacts the seafloor are prevented by maintaining a predetermined amount of tension on the pipe as it leaves the barge and by controlling the forward movement of the barge.

Probably the most popular method of installing flowlines, both conventional steel pipe and flexible pipe, is with special-purpose reel ships or reels mounted on large work boats. Depending on pipe diameter, several miles of pipe can be reeled onto one or more large reels’ at a shore-mobilization site and then rapidly reeled off at the placement site. The main advantage of this t&h- nique is the speed of installation. Fast installation reduces not only the number of offshore construction days but also costly interruptions caused by bad weather. A job that might require a week of offshore construction time with a conventional lay barge is much more susce

P tible to

weather downtime than a job that can be camp eted in 1 or 2 days with a reel ship.

Flowline connections at platforms generally are made by pulling the line up through a curved conductor pipe called a J tube and then securing the line at the platform deck opening with a flange or clamp. Several procedures are used for connections to subsea wells, depending on Christmas tree configuration and whether the flowline in-

stallation starts at the tree (a first-end connection) or ends at the tree (a second-end connection). Most tree connections have been a pull-in type, where the flowline is first laid on the seafloor and then pulled into a receptacle on the tree base with a wire rope. One advantage of this is that it can be performed either on a first-end or second- end connection. A pull-in procedure can also be used as the flowline is being lowered to the seafloor, making a connection at the tree base before laying the flowline on

Reel Capacity

Fig. 18.41-Work boat with pipe-laying reels


RUN FLOW LINE POSITIONER 0

-_r

LAND FLOW LINE POSITIONER

REMOVE GUIDE LINES,;;;:1 STINGER. BEGIN LAYING FLOW LINES

RETRIEVE FLOW LINE POSITIONER RUNNING TOOL

ACTUATE FLOW LINE CONNECTOR.

Fig. 18.42-J-lay method for installing flowlines away from subsea well (first-end connection).

the seafloor. A procedure particularly applicable to subsea wells in very deep water is the J-lay first-end connection (Fig. 18.42), where the flowline is run vertically to the tree base from the drilling rig, stabbed into a receptacle, and then laid down into a horizontal position as the drilling rig or pipe-lay vessel moves away from the wellsite toward the platform. A trunnion-type assembly on the end of the flowline permits the line to be laid down without bending. The mating of the flowline to the Christmas tree is made after the line is fully horizontal.

Many different devices and procedures have been developed for making the actual connection between the Christmas tree and the flowline. In shallow water, divers can install a piping spool with flanges or couplings on each end. Diverless connections usually have some type of hydraulically actuated device that is operated remotely from the drilling rig.

Larger Pipelines. Pipeline diameters from 8 to 36 in. or more are used extensively for the transfer of crude oil and gas from offshore fields to land sites. Installations typically have been made with conventional pipe-lay barges as described previously, but reel barges also are used extensively for the smaller sizes. Both bottom tows and surface tows are used in limited applications where logistics favor them. As discussed previously, a pipeline can be the single most expensive element of an offshore installation, sometimes exceeding the cost of one or more platforms. Numerous factors must be considered when

designing a line and planning its construction to minimize installation difficulties and to ensure satisfactory operation throughout the expected life of the line. Many technical papers and magazine articles have been written about this subject and are excellent sources of further information-e.g., proceedings from the annual Offshore Technology Conference.

Arctic Production operations in the offshore Arctic regions are within the reach of existing technology. Procedures used onshore and offshore in less hostile regions, however, must be modified to meet the challenges of the harsh climatic conditions in these remote locations. I5

In the last decade, the major areas of industry interest have been the offshore regions of Alaska I6 and Canada. The environmental conditions vary significantly in each of these regions. Major factors that affect normal offshore operations are extremely cold temperatures, fog, gusty winds, short open-water seasons, permafrost, and the persistence of ice. The specific production system that is selected must be tailored to each unique combination of these factors to ensure safe oilfield development.

Environmental Conditions

Ice Characteristics. Sea ice is the principal environmental factor in all of the offshore Arctic areas. I7 The most abundant type of sea ice that is encountered offshore is less than 1 year old. This first-year ice begins to form


during fall and grows to a thickness of 4 to 8 ft during the winter. Sheets of ice close to shore become landfast and remain locked in place throughout the winter. Beyond the landfast zone, the ice is kept in constant motion by wind, currents, and, in some areas, the influence of the Arctic polar pack. This dynamic movement causes shearing and impacting between ice features that produce ridges of ice several miles in length. Ice ridges formed in this manner are called pressure ridges. Localized ridging around a grounded ice feature, the shoreline, or a structure is considered a rubble pile. In areas of extremely cold winter temperatures, the ice blocks within a ridge or rubble pile begin to refreeze into a contiguous feature. De- pending on the conditions, the refrozen consolidated thickness could become several times larger than the first- year ice thickness.

Sea ice that survives one or more melt seasons is considered multiyear ice. The predominant source of multiyear ice features is the polar pack. The pack consists primarily of floes 2.5 to 50 ft thick with embedded ridges 50 to 100 ft deep. During the summer, northerly winds break off portions of the pack and push them toward shore. These multiyear floes are commonly 1,000 to 2,000 ft in diameter.

The other major type of ice is not formed from seawater but is freshwater ice from the glaciers of Northern Canada. In the Arctic Ocean, the glacial fragments are called ice islands. These tabular-shaped features are several thousand feet in diameter and more than 200 ft deep. Because of the enormous size and slow movement rates of these features, they can be tracked for several months before encroachment upon a given area. Most of the areas of interest for oilfield development in the Arctic Ocean are also in relatively shallow water. This shallow bathymetry causes these freshwater ice features to run aground before they are a threat. In the North Atlantic, similar glacial features are called icebergs. Icebergs that weigh more than 50 million tons have been observed in water depths beyond 1,600 ft. Again, the local bathymetry dictates the maximum size iceberg that could encroach upon a given area.

Ice Loading. Ice exerts the predominant forces on Arc- tic offshore structures. Extensive laboratory and field tests have been conducted on small- and large-scale specimens to determine in-situ strength characteristics for design. From the results of these tests, the mechanical properties of ice are predicted that consider its salinity, temperature, crystallographic structure, and loading rate.

Newly formed sea ice is relatively warm-only a few degrees above seawater temperature-and high in salinity. As the ice sheet grows, the temperature at the surface reduces to the ambient air temperature, while the bottom remains near the seawater temperature. The salt in the crystal structure of the sheet begins to consolidate into brine droplets. These droplets migrate down through the thickness of the sheet, creating drainage channels and lowering the overall salinity. Fresh water, from precipitation or melting snow cover, fills these channels and refreezes, thereby further reducing the salinity. The result is an ice feature with varying strength, strongest at the surface and reducing through the thickness with increasing salinity and warmer temperature. For multiyear ice, this process ap-

proaches equilibrium. The multiyear ice floe is near zero salinity and has a relatively cold temperature through its thickness. This results in an ice strength several times greater than a first-year ice sheet.

Another parameter that influences the strength of sea ice is the loading rate. When ice is loaded at a very slow strain rate, it exhibits a plastic behavior. Loaded rapidly, it behaves as a brittle material. Empirical equations have been developed that relate the ice movement rate and shape of the structure or indenter to the strain in the ice feature.

The shape of the structure is also a primary factor in producing a crushing, buckling, or flexural failure of the ice feature. For narrow structures relative to the ice thickness, crushing is the predominant failure mode. As the width increases, a combination of crushing and buckling of the ice field around the platform results in the development of a rubble pile. This rubble pile will then shield the structure from direct impact of subsequent ice floes and ensure failure of the ice mantle away from the production facility. And finally, sloping-sided structures normally force a flexural ice failure. Because ice flexural strength is 20 to 40% of the crushing strength, an appreciable reduction in ice forces can be achieved when a bending failure is induced.

Waves. The wave conditions in the Arctic are similar to other offshore areas, and the design of structures or islands against wave loading is well established. Nearshore sea states can be defined by determination of the open- water area along the storm route or fetch and the water depth. In the Arctic Ocean, the presence of sea ice and the polar pack limits the open-water fetch for storms to generate and consequently reduces the design wave height. Breaking wave conditions exist in most shallow-water- depth locations and around gravel islands. This diffraction, shoaling, and refraction of the waves produces highly irregular sea states. Because the interaction of the waves and structure is dependent on structural geometry, the forces the design wave exerts on the structure should be determined by model testing or approximated by linear diffraction analysis.

Scour of the foundation around the structure caused by waves must also be considered and appropriate protection methods implemented. The scour protection could consist of rock, sand bags, or precast concrete elements placed around the structure after installation. For artificial islands, this becomes slope protection and can be used to reduce wave and ice run-up by providing an artificially rough surface.

Permafrost. Permafrost is soil at a temperature below 32°F with partially or completely frozen pore water. Drill- ing and producing operations in areas with permafrost have been well defined from the experience of the Prud- hoe Bay field. In most nearshore areas of the Arctic, permafrost has been found at or near the mudline. These soils are very stiff and can make excavation for pipelines or driving of piling nearly impossible. Permafrost normally is soil bonded by ice and is very susceptible to changes in temperature. This can result in significant changes in the soil characteristics and must be considered in the design.


Fig. 18.43A– Protected-slope production island.

Production Structures 16

Artificial Islands. Artificial islands already are used inmany shallow-water areas throughout the world for per-manent drilling and producing facilities. The islands thatare currently being used for drilling in the Arctic consistof either unretained or retained beach slope systems, asshown in Figs. 18.43A and 18.43B. Because of the shortsummer construction season and, in some areas, the lack

of island fill material, the quantity of fill required for theisland should be minimized. The minimum island work-ing surface is determined by the area required for drill-ing and production operations. To reduce the quantity ofisland fill, the steepest side slopes that the mode of con-struction and fill material will allow should be provided.The minimum side slopes of unretained islands dependon whether the island is constructed by summer dredg-

Fig. 18.43B– Caisson-retained production island


+ermafrost cement

Fermafrost cement

-+ 9000’ TV0

i 9700' TVD

] KeStage collars

ydraulic set liner

-Class G cement

--Class G cement

BP Alaska hole-casing program

hangers

hanger

cement

Permafrost cement

Non freezing fluid

Safety valve

Non freezing insulating fluid

Stage collar

Permafrost cement

Class G cement

--- Safety joint

+ Hydraulic set packer

Arco hole-casing program

Fig. 18.44-Hole casing programs, Alaska.

ing or winter transport of onshore borrow material over the ice to the desired location. The side slopes for summer dredging are approximately 1 : 20 (vertical to horizontal), and for winter construction are 1 : 3. On completion, sandbags or concrete mats are placed on the exposed slopes of the island to prevent ice and wave erosion. Sandbags, stiffer soils for embankments, or caisson units are used on retained islands during construction to reduce the required volume of fill. The caisson units typically consist of vertical walled concrete or steel units. The caissons also provide easy access to the island as a dock for resupply and could be used for storage of consumables or oil.

Artificial islands must be designed to withstand the horizontal forces exerted by ice. The potential failure modes of the island consist of slope instability, bearing failure, or horizontal shearing of the island near the waterline. Each of these failure modes can be predicted by classic geotechnical analysis. The only variable in the analysis is the properties of the island fill material. During winter construction, the fill is delivered to the site at the cold ambient temperature and dumped into the sea. Ice forms on the granular material and inhibits consolidations. As the island surface thaws, considerable settlement may take place. To minimize the effects of thaw settlements, thermal analysis of the freezing and thawing interface should be conducted to determine the proper gradations of fill material.

The design of production facilities placed on an island is similar to that on land. Equipment foundations must be designed and insulated to reduce the potential for frost heaving, pile jacking, and thaw settlements from seasonal thawing and freezing of the island surface and, in some

areas, subsea permafrost. To prevent thaw settlement, an artificial refrigeration system for the fill material could be installed. Placement of equipment and accommodation modules should account for predominant wind, ice movement, and wave directions to ensure safe year-round operations.

The well systems should be vertically drilled through the frost-susceptible island surface and permafrost and then directionally drilled to true vertical depth (TVD). Wells should be spaced close together to minimize the overall size of the island surface and to reduce the effects of thaw subsidence. The casing program should be designed to withstand freeze-back loading during periods of well inactivity and to accommodate differential movement in the tubing string owing to the thermal effects of the drilling fluids. Two casing programs in the Prudhoe Bay field are shown in Fig. 18.44, I8 which shows that both systems incorporate the use of a permafrost cement and provide a safety valve in the production tubing string below the permafrost for emergency shut-in.

Gravity Structures. Various types of gravity structures are being proposed for use in the Arctic. Many of the conventional gravity structures that are used in the North Sea are being adapted for the deepwater and moderate-ice- concentration areas. In the more hostile areas of the high Arctic, vertical- and sloping-sided gravity structures are being proposed. These structures provide the large deck load and space requirements, protection of the wells within tower shafts, and storage of oil. Because of the extreme winter ice conditions in many areas, the production facilities will have to operate 9 months without major resupply.


Fig. 18.45-Vertical-sided structure.

The vertical-sided structures (Fig. 18.45) are proposed for the shallow, nearshore areas in the Arctic. These structures typically are rectangular or hexagonal and are capable of being installed directly on the seabed or subsea berm. Production equipment can be placed directly on the working surface of the top slab or integrated into the hull of the structure. Wells are drilled and produced directly from the deck of the structure. Because of the large width of this concept, the structural integrity of the system is

Fig. 18.46-Arctic mobile drilling structure: sloping-sided structure.

not sensitive to local discontinuities in the seabed from ice gouges or settlements in the foundation from local degradation of permafrost.

Conical, sloping-sided structures (Fig. 18.46) are being proposed for the deeper-water, dynamic-ice-movement areas. This geometry induces flexural failure of the ice features and is relatively transparent to pack-ice movements. The deck is fully outfitted with processing equipment before it is mated with the structure. Wells are confined to a central moon-pool area in the cylindrical throat. Consumables and oil can be stored in the base.

Piled Structures. Piled steel structures have been developed primarily for the Bering Sea area of offshore Alas- ka. These structures are similar to conventional template or jacket concepts but must be modified to resist annual sheet-ice loading. A typical geometry is shown in Fig. 18.47. The platform concept consists of four or eight main pile legs with intermediate bracing of the legs omitted in the ice-loading zone near the waterline. Well conductors and oil-transport lines are positioned within the legs of the platform for protection from ice loading. This requires close spacing of the wells and, in some cases, completion of the wells at different levels of the deck. Diver- access tubes may also be located in the legs to facilitate the repair and inspection of subsea components of the platform during complete ice coverage.

In most other Arctic areas, pile structures are not practical. Subsea permafrost makes pile installation nearly impossible. The short construction season also does not accommodate the installation, pile driving, and placement of the topside modules in one season. Also, the hookup and commissioning of the production equipment modules would be very expensive in these remote areas.

Transportation Systems

Pipelines. Offshore pipelining is the predominant mode of crude-oil transport proposed for Arctic regions. The pipelines will interconnect platform facilities, mooring

OFFSHORE OPERATIONS

structures, and land-based facilities in much the same manner as in conventional offshore locations. The principal factors affecting the construction and operations of Arc- tic pipelines are the short open-water seasons, subsea permafrost, scour, and ice gouging of the seafloor.

Pipelines can be placed directly on the seabed in deep water, in trenches in areas susceptible to ice gouging or scour, and on causeways or elevated bridges at shore crossings. In areas of extreme ice gouging, redundant lines may be used to lower the risk of interrupted production.

In deepwater, moderate-ice areas of the Bering Sea, the lines will be laid directly on the seabed by conventional lay-barge methods. The 6- to 9-month open-water season, extreme summer wave conditions, and logistics of operating several hundred miles offshore reduce the efficiency of the construction process. The distance from shore or an offshore loading terminal may also require pump stations along the pipelining route.

Marine pipelines in the Arctic Ocean are considered feasible but will require the greatest challenge to existing technology. The open-water season lasts approximately 1 to 3 months a year. In shallow-water locations, the keels of ice features may gouge the seafloor for several miles in a single ice-movement event. Pipelines must be buried in trenches that are deep enough to ensure no damage. Offshore permafrost may also cause difficulty in obtaining the desired trench depth and may require the use of cutter-suction dredges to remove the ice-rich soil. Once the pipeline is installed, refrigeration of the trench may be required to ensure no subsequent thawing of the permafrost. Nearshore pipelines may also have to be designed for wave-induced erosion or strudel scour. Strudel scour, which is common at the mouth of rivers, is the process in which river-water outwash flows over the nearshore sheet ice and floods down through holes and cracks in the ice. This jet of water could create large, eroded pockets in the seabed and produce long, unsupported spans of pipe.

Marine Terminals and Tankers. Transportation of crude-oil products from many of the remote Arctic oil fields also can be accomplished by offshore terminal and tanker systems. The systems could consist of an interfield pipeline-gathering network with processed crude shipped directly to shore. Once onshore, the crude could be stored until it is loaded onto tankers and shipped to market. Another form of marine terminal system is direct shipment of processed crude from offshore platforms to tankers through single-point mooring systems.

The design, construction, and operation of these systems have been proved in many sub-Arctic areas. How- ever, some of the components must be modified for the cold temperatures and persistence of ice. Loading arms must be designed to ensure that hoses remain elevated above the ice mantle and can accommodate loading of tankers from any location around the terminal. Tanker- mooring systems must be designed for both wave- and ice-loading conditions. Extreme ice areas may require offshore loading directly from fixed platforms or from subsea facilities, making maneuvering and stationkeeping very difficult. Icebreaker assistance vessels may also be required to ensure safe access and departure from the terminal by tankers.

To travel through the offshore Arctic regions, purpose- built icebreaking tankers may be required. The amount

18-43

Fig. 18.47-Piled structure.

of hull stiffening will be dictated by the ice conditions for its area of operation. The modifications could range from simply thickening the hull plate to the requirement of an icebreaking bow and turbine-powered propulsion system. Internal bulkhead arrangements should also be arranged to ensure that oil storage tanks are adequately stiffened and not susceptible to direct ice impact.

Special Considerations

Ice Management. A critical support system, unique to the Arctic, is the ice-management system. The key component of this system is instrumentation of the structure, foundation, and ice field around the structure to monitor both local and global ice loading. Other elements of ice management include icebreaker support vessels for tankers or supply boats, tractor-mounted ditch diggers for slot- ting the ice to reduce loads, and water-pumping units to flood areas around a platform to stabilize ice movement.

Electrical, Instrumentation and Control Systems Offshore production facilities have the same basic requirements for electric power and control systems as onshore facilities. These are a power source with a reliable distribution system, instrumentation to control and to monitor production operations, and a safety shutdown system tailored for the installation.

Although offshore and onshore installations have similar needs in these areas, the two operations are significantly different in other respects. As pointed out earlier, deck size and payload significantly affect offshore structure costs. Consequently, offshore electrical facilities must


be designed for minimum weight and space while still offering a high degree of flexibility, reliability, access, and maintainability. Because of deck space and layout limitations, hazardous area considerations are more complex and often are governed by different regulatory codes depending on the type of structure, its location, and the responsible governmental agency. Additionally, environmental considerations are much more demanding because of the generally salt-laden atmosphere and the possibility of saltwater washdowns and sea spray.

Several wiring methods are approved for offshore installations, but the method used frequently depends on local practices and personal preferences. There is no single correct method for wiring an offshore installation as long as the appropriate codes are satisfied and the craftsman- ship is of average quality or better.

Offshore-production instrumentation and control systems also tend to differ from their onshore counterparts. Offshore oilfield operations are confined to a relatively small spot on the ocean rather than being spread over several acres. Consequently, controls tend to be much more centralized. This trend toward centralization has become more pronounced with increased use of computer- based production-monitoring and control systems.

Safety shutdown systems onshore and offshore are roughly equivalent. However, offshore requirements regarding functionality and reliability are somewhat more stringent because of the greater potential for spills and resultant pollution. Increased concern for personal safety and environmental protection creates a more stringent atmosphere for approval, testing, and inspection by outside agencies, as well as thorough documentation of the entire process and facility by the operator.

Alternative methods of addressing and satisfying these and other considerations peculiar to offshore operations will be discussed. The intent is to lay out the problems, to highlight areas of concern, and to discuss possible solutions, rather than to present specific, detailed engineering methods for design and installation of electrical and/or control systems.

Codes and Regulatory Authorities

Offshore electrical installations are governed by one or more codes and regulations, depending on the type of offshore structure involved and its location. In U.S. waters, electrical designs are governed by local or state electrical codes and in most cases the Natl. Electrical Code. I9 Generally, the installation should be in accord with the more stringent requirements of the applicable codes. The guidelines given in API RP 14F*’ provide direction for accepted good practice in accordance with most applicable codes. If the facility falls under U.S. Coast Guard jurisdiction (e.g., TLP’s, mobile offshore drilling units and FPF’s), all or portions of the electrical system also must be designed, installed, and operated in accordance with the applicable portions of U.S. Coast Guard Regu- lations 46 CFR Chapter I, Subchapter IA, Mobile Off- shore Drilling Units, Part 108 and Subchapter J-Electrical Engineering, Parts 110-l 13. *t Overseas installations frequently fall within the jurisdiction of local/national codes or regulatory agencies. This situation occurs in the North Sea where Lloyds of London or Det norske Veritas frequently are named as certifying authority, and their requirements must be met. If there is no lo-

cal certifying authority, it generally is good practice to design all systems to meet normal U.S. requirements for that type of facility.

Platform Loads

With the exception of the hotel loads that serve accom- modations and personnel needs, normal platform loads are not greatly different from the normal assortment of onshore loads. Processes are likely to be concentrated in a single location offshore, and the loads are appropriately higher. Loads usually consist of a mix of transfer pumps, compressors, fans, and heaters as well as utilities (air compressors, sump pumps, fire water pumps, water makers, sewage facilities, etc.). Other typical loads include pumps and gas compressors associated with shipping, artificial lift, secondary recovery, and pressure-maintenance operations. Depending on platform type and production facilities and rates, the loads on any given platform in a field can range from less than 25 kVA to more than 40 MVA.

Layout of Facilities

Considerations that govern layout of offshore electrical facilities are similar to those for land installations, but the options are more restricted because of the space limitations. The primary requisite is to separate to the maximum extent possible the sources of ignition from the process facilities. Electrical equipment should be kept out of hazardous areas when economically feasible. Primary electrical switch gear is frequently grouped in a pressurized, central electrical-equipment room. Remote motor- control centers (MCC’s) sometimes are used to reduce the amount of platform wiring. Installing MCC’s near load concentrations and supplying the MCC’s with high- voltage feeders shortens branch circuits that feed individual loads. Remote MCC’s frequently must be purged or pressurized because of their location. The design approach appropriate in each individual case must be based on prac- ticality and economics.

Detailed load analyses must be made early in the design stage. Each analysis must consider both initial and anticipated future load. Results of the analysis should control design of primary power facilities, layouts of conduit or cable ways, switchgear space, and distribution- equipment layouts. Designs must allow for expansion in each of these areas. The designer must carefully consider possible future system expansion and pay particular attention to possible future artificial-lift and water-injection requirements. Hydraulic pumping and electric submersibles can add significantly to the ultimate electrical load. It is important to consider the potential for future changes in facilities as reservoir conditions change and possible increased demand on the electric power system.

Primary Electric Power

Typically, offshore facilities either generate power locally or they are fed by submarine cables. Generated power ranges from 480 to 4,160 V. The higher voltage levels are more prevalent where there are high horsepower loads or where platform drilling rigs are not set up to generate their own power. Very large platforms are sometimes designed with primary power system voltages of up to 13.8 kV.

OFFSHORE OPERATIONS i a-45

The primary source of generated power usually consists of one or more brushless synchronous generators with either diesel-engine or turbine prime movers. Turbines may be gas- or diesel-fueled or they may be dual fueled. Dual-fueled turbines generally are set up to run on diesel initially and then to cut over to produced gas as production builds to a stable supply. Turbines offer the advantages of lighter weight and the opportunity for the use of waste heat in oil production processing. Turbine-generator packages, however, are very costly, require more maintenance, and in some cases, may require more deck space than internal-combustion-engine-driven generators of the same capacity. In addition, primarily because of the somewhat limited selection of turbine sizes available, it can be more difficult to match turbine generator packages to the electrical system loads.

The second source of offshore power, submarine cable, offers an efficient means of electrification if adequate sources are located nearby. Depending on the loads and the distances involved, cable-system voltages can range between 480 and 35,000 V. Submarine-power-cable technology is well established. Power cables generally include one or more pairs of small-gauge wires dedicated to telecommunications or telemetry and also are protected with torque-balanced double layers of galvanized armor wires that may be plastic-coated for additional protection. Power cables may be buried, depending on local conditions or regulations. Tubular risers are required at each platform served by submarine cables to protect the cables as they rise from the seabed to the platform deck.

The riser generally should be filled with a corrosion- inhibiting fluid to preserve the long-term integrity of the armor wires. When the submarine cables originate on land, they must be buried and suitably protected through the surf zone. Mechanical protection is frequently provided by additional armor or by the installation of heavy- wall pipe through this area.

Secondary/Back-up Power

Essential loads on offshore facilities must remain ener- gized even when the main power source fails. These loads usually fall in the category of navigation lights (ship and aircraft), foghorns, communications, emergency lighting, and possibly some selected hotel loads. In some instances where marine regulatory bodies have jurisdiction, the emergency system may have to be expanded to serve other loads related to the safety of personnel on board.

High-volume oil and gas facilities are frequently designed to shut in completely in the event of power failure, in which case process or shipping loads do not have to be considered in determining backup power loads. Some smaller facilities, such as wellhead platforms, are designed to continue production operations without any electrification other than lighting, navigation lights, and foghorns.

Emergency power generally is derived from onboard generators, storage batteries, or both. The generators usually are small (less than 500 kW) and diesel-engine driven. Diesel engines that are used in emergency service normally are equipped with automatic starters. These engines also are equipped with cooling water and lube- oil heaters to ensure that they will start reliably and instantaneously (within 10 seconds) on loss of primary power.

Fig. 18.48-Typical electrical one-line diagram with ups and emergency generator.

Offshore facilities with electronic instrumentation or computer-based monitoring or telemetry systems generally need uninterruptable power supplies (UPS’s) to provide reliable, clean power for these loads. A UPS is essential on platforms where some loads are supplied from onboard, solid-state, silicon-controlled rectifiers (SCR’s) because of the high level of electrical noise injected into the power system by the SCR’s.

The emergency bus typically is tied into the normal power bus through an automatic transfer switch. The emergency bus receives power during normal operation from the main bus. On failure of the primary power, the transfer switch opens the tie between the two buses to isolate the emergency bus and its loads from the remainder of the platform loads. Additional controls automatically start the emergency generator to energize the emergency bus. Normally, no provisions are made for operating the emergency generator in parallel with the main power supply. Interlock circuits should be provided, however, to permit testing the emergency generator offline without energizing the emergency bus.

Typical one-line diagrams for a UPS and an emergency generator bus tie are shown in Fig. 18.48.

Distribution System

Offshore power-distribution systems and associated equipment do not differ substantially from land-based operations. Depending on load sizes, distribution voltages will normally range from 120 to 2,400 V. Distribution systems with 4,160 V and higher are rare except on very


large platforms with high horsepower loads. As with land- based systems, accepted good practice and system capacity determine the maximum allowable horsepower for individual loads and the appropriate supply voltages.

Motor-control centers and switchgear usually are installed inside pressurized or purged enclosures or modules. The pressurizing or purging frequently is required to ensure a nonhazardous operating environment for the electrical equipment so that air-break switching equipment can be used safely. Code requirements for purging are prescribed in the Natl. Electrical Code I9 and Natl. Fire Protection Assn. Bull. NFPA 496. *’ As indicated in the references, interlocks usually are provided either to sound an alarm or to shut down all supplies when a loss of pres- surization or purge occurs. Depending on the particular operations involved, a simple alarm may be permissible if shutdown of the process could create a greater hazard than continuing operation during a short-term loss of pres- surization or purge protection. Each installation is site- specific and must be considered on its own merits.

System transformers may be installed either indoors or outdoors. Both single- and double-ended line-ups are used offshore, depending on the nature of the operation. Double-ended 100%~capacity supplies with bus tie breakers are frequently used for increased reliability and con- tinuity of service in the event of equipment failures.

Hazardous Areas

The possibility of fire or explosion because of the ignition of leaking gas or liquids is a concern in all producing facilities. The concern is even greater offshore because of the concentration of personnel and facilities in a relatively confined area where fire can be difficult to extin- guish and platform evacuation can be a complex problem. The requirements for classifying areas according to their degree of hazard and for selecting equipment for the various areas are covered in API RP 500B*” ; Section 500 of the Natl. Electrical Code19; and USCG Regulations 46 CFR, Chapter I, Subchapter I-A, Part 108 and Sub- chapter J, “Electrical Engineering,” Parts 110-l 13. *’ Certifying agencies outside the U.S. have some different rules and categories, but in general, they are no more strict than the U.S. rules for the same situation.

Wiring Methods and Equipment Enclosures

Several wiring methods are applicable to offshore installations, and opinions differ as to whether conduit-and-wire or cable-and-cable-tray systems are best. Each has advantages and disadvantages. If the installation complies with the appropriate procedures as outlined in the Natl. Elec- trical Code I9 and API RP 14F,20 each method is safe and reliable. Rigid steel conduit and wire systems provide the maximum mechanical integrity, but the conduit and fittings should be coated with plastic (PVC) to eliminate corrosion effectively. PVC-coated fittings and accessories are readily available in most locations. Copper-free aluminum conduit systems have been used successfully in lieu of galvanized steel. Aluminum systems require special wrapping of the conduit wherever steel support clamps are used to prevent setting up corrosion cells where dissimilar metals would come in contact. Overall, conduit systems provide excellent protection for wiring systems, but they have the disadvantage of being bulky and relatively expensive to install, maintain, and modify.

Type MC (metal-clad) cable, which has a corrugated aluminum sheath and an overall PVC jacket, is another preferred system that is frequently used offshore. Ar- mored shipboard cable and type TC tray cable are viable alternatives to MC cable in many instances and are suitable as long as the installation complies with the Natl. Elec- tric Code. I9 Cable systems are more flexible, quicker to install, and less subject to corrosion than conduit offshore, but they are more subject to mechanical damage.

As with onshore facilities, all electrical equipment installed in locations classified as Class 1, Div. 1 must be explosion-proof. Requirements for equipment in Class I, Div. 2 areas are slightly less stringent as long as no arc- ing contacts are exposed. Electrical equipment in nonhazardous areas generally is chosen for its applicability to the situation. Switches, for instance, generally are provided with explosion-proof enclosures for mechanical protection and durability even if the contacts are enclosed. General-purpose enclosures are normally used in protected areas. Fiberglass enclosures are seldom used in open areas because of their susceptibility to mechanical impact damage.

Explosion-proof motors are required in Div. 1 areas. Fractional horsepower motors in Divs. 1 or 2 areas must be explosion-proof. Otherwise, both Div. 2 and unclas- sified areas permit the use of totally enclosed fan-cooled (TEFC), totally enclosed nonventilated (TENV), or en- capsulated, open drip-proof motors. Choice of enclosure must be based on exposure and service. Motors operating at 2,400 V or higher should be equipped with integral heaters or low-voltage winding heating systems if they are in exposed locations to ensure that the integrity of the insulation resistance is maintained during periods of nonuse. As warm windings cool in the relatively high- humidity offshore atmosphere, moisture is pulled into the windings, giving rise to a high chance of an internal short circuit. Motors of 500 hp or more that have been shut down for an extended period should always be checked with an insulation tester before they are restarted, regardless of whether they have been heated in the interim.

General Instrumentation

The primary guide for the instrumentation of offshore installations is API RP 14C. 24 In conjunction with the applicable U.S. OCS Orders, 2s,26 the API guide provides an excellent reference for guidelines in the design of monitoring and control systems for offshore-production facilities. Most offshore installations have local control panels for packaged equipment, such as gas compressors, low-temperature gas separation (LTS) units, and electrical generators. Beyond this, the choice of individual local display/controllers or centralized control rooms with field transmitters and local displays depends on the nature of the facility, the complexity of the operation, economics, and preference of the operator. Most large, modern offshore facilities are planned with centralized controls. The one design principle that must be followed regardless of the overall design philosophy is that all systems should be designed to be fail-safe so that loss of a signal represents an alarm or shutdown condition.

Even when a centralized control room has been provided, continuous remote, closed-loop (analog) control is seldom used. Remote, closed-loop controls usually are


considered only when the process includes complex separation processes, sulfur removal and handling, or gas processing. Most oil/gas handling processes are sufticient- ly simple and straightforward that local control loops are satisfactory.

Centralized controls can be based on conventional instrument and relay control panels, but advances in electronics have increased the use of programmable controllers and microprocessor-based instrument systems. These advanced system designs have multiplexed monitoring and control systems that frequently offer significant savings in wiring costs and space for a complex installation. They are more compact and much more flexible than conventional control panels.

Offshore instrumentation is very similar to onshore instrumentation. Process controls frequently include a combination of pneumatic, hydraulic, electric, and electronic instrumentation. Process variables that must be monitored and/or controlled include level, pressure, temperature, flow, oil/water interface, and gas/oil interface. In most offshore installations, the considerations in designing instrument systems to accommodate these variables are identical to those for their land-based equivalent. Foam, gas cutting, sand, wax, and HzS are typical considerations. Radios are used extensively for offshore communications, and radio interference can present some unique problems. Some electronic-based sensors are sensitive to radio- frequency interference and may give false alarms or shutdown signals when high-powered radios are keyed nearby.

Most operators prefer to separate emergency shutdown circuits from alarm and control circuits. Typical examples are high- and low-level shutdowns on tanks or separators, high/low pilots on manifolds, and gas and fire detectors.

Regardless of the level of complexity of an instrument and control system, a conscientious, well-disciplined, and well-documented program of regular testing and maintenance is essential. Offshore instrument systems are not overly complicated, and given adequate maintenance and correct initial installation, every instrument should perform its intended function reliably over the life of the facility.

Safety Systems

Separation of process-related shutdowns from instrument- control loops was mentioned earlier. Process shutdowns generally involve process-related variables that have exceeded preset limits. Other key safety-related systems should be kept separate. These systems are covered in various U.S. Coast Guard Regulations and OCS Orders. 26 Examples of these systems are combustible-gas detectors, poisonous-gas sensors, fire detectors, surface-controlled subsurface safety valves (SCSSV’s), surface safety valves (SSV’s), and emergency shutdowns (ESD’s). Proper design and installation of these systems is perhaps the single most critical aspect of offshore instrumentation and control systems. (Safety shut-in systems are covered in more detail in Chap. 3).

Combustible-gas detection systems are required in most offshore operations. They are intended as early warming devices to alert operators to potentially hazardous conditions where none normally exist. Modern systems have catalytic sensing heads whose electrical characteristics depend on the concentration of hydrocarbon gases surround-

ing the sensor. Units are calibrated and have known variations in their electrical output that are based on the gas concentration in the area of the sensor. Gas sensors normally are connected to a central monitoring panel equipped with individual sensor readouts calibrated in percent of lower explosive limit (LEL). Each readout has at least two alarm outputs. Normally, one alarm is set for about 20% LEL and the other for 60% LEL. Some operations/regulations require automatic shut-in at the higher level. Although gas detection technology has improved over the years, malfunctioning caused by poisoning of the sensing heads by contaminants in the air and loss of circulation caused by dirt accumulations on the sensors continues to be a problem.

H 2 S gas detectors are essential where sour crude and sour gas is handled or produced because of the extreme toxicity of Hz S. Sensors continue to improve, but reliability and maintenance are continuing problems.

Because of their vulnerability, gas detectors must be installed where they are protected from water spray, drilling mud, and other contaminants; yet they must be in areas where they can adequately monitor the environment. They generally are installed in areas where leaks or accumulations might be expected under abnormal conditions- above gas compressors, in a wellhead or manifold area, over drilling mud pits, or in dead-air spaces-or in areas where a gas build-up could be catastrophic, such as in ventilation system inlets.

Fire-detection systems generally are based on the use of ultraviolet (UV) sensors or fusible plugs. The operating principle of UV sensors is that their sensitivity to the UV radiations from flame provides an alarm output in the presence of UV radiations from open flames. Unfortunate- ly, they also are somewhat sensitive to direct and reflected UV radiation from welding arcs. Because of their extreme sensitivity, most UV fire systems include a brief time delay to minimize false triggering of a fire alarm. The layout of UV sensors at a site is important. Consider- able care must be taken in laying out a coverage that con- siders viewing angles, range, and sensitivity.

Infrared (IR) fire sensors were tried on early offshore platforms, but they had many operating problems. IR sensors are seldom used today, primarily because of diffcul- ties with their calibration and reliability.

Fusible plug fields that consist of pressurized stainless steel or plastic tubing, heat-sensitive solder plugs, and pneumatically held pilot shutdown valves are popular and reliable systems for fire detection. With this system, tubing is run through and around various critical areas of an offshore facility. The tubing runs are segregated by process area or some other criterion. Multiple solder plugs are included within each field. If a fire occurs in that field,

the solder plug melts, depressurizing the field and tripping a shutdown valve. The major problem with plug

fields is maintenance and accidental shutdown because of leaks.

Overall, a judicious combination of strategically placed

UV sensors and fusible plugs forms the optimum fire detection and automatic shut-in system. With such a SYS-

tern, it is possible to shut in production facilities, to blow down pressurized vessels, and to activate the appropriate

fire suppression system simultaneously and immediately, thus minimizing fire danger.


SCSSV’s should form an integral part of every offshore production system. These valves are installed in the production tubing below the mudline. They are hydraulically actuated and held open during normal operation by pressurized hydraulic fluid in their individual control lines. They are designed to be fail-safe in that they close when a loss of hydraulic pressure occurs. On modem platforms, the SCSSV hydraulic system is generally a separate, centralized, hydraulic power unit dedicated solely to their control. Surface power units and their associated alarms and controls are available as specialty packaged units. Their design and component selections have been developed over the years to the point where it is not cost- effective to try to design alternative units.

Platform wells also are equipped with surface safety valves (SSV’s) between the wellhead and the production manifold. The actuators on these valves are designed to be fail-safe closed and can be actuated either hydraulically or pneumatically by the automatic safety shutdown system and/or the manual emergency shutdown.

SCSSV’s normally are actuated only in extreme emergency to preserve their integrity. Repairs are expensive. SSV’s, on the other hand, usually are activated in almost all platform or process shut-ins. They are simpler devices that are less expensive to repair, more rugged, and more accessible.

Nearly all platforms are equipped with automatically controlled riser shutoff valves on pipelines and flowlines feeding or leaving the platform. The intent of these valves is to allow isolation of the platform from any outside source of flammable fuel in the event of a platform accident. In the case of subsea wells, it frequently is desirable to shut in the wells by simply blocking the flowline rather than operating the subsea valves unnecessarily. The flowline riser valves commonly are operated first in the event of a subsea well shut-in and last on startup to avoid cutting out seafloor valves by closing or opening them unnecessarily against a flowing stream. Riser valves are far more accessible and maintainable than the subsea tree valves.

The final element in any safety system should be the manual ESD. These controls can be either electrically operated solenoids or pneumatically or hydraulically held pilot valves that control various shutdown control circuits around the facility. ESD stations usually are located on boat landings, on helidecks, in process areas, and in control rooms.

Control of Subsea Production Facilities

The mechanics of subsea production systems, such as wells and manifold centers, were covered under Produc- tion Facilities. This section presents various operating philosophies and discusses methods of providing remote control of subsea equipment.

Subsea controls should be as simple and straightforward as possible and still meet requirements for operational considerations and the physical layout of the field. System reliability, maintainability, control response times, and the need for feedback of tubing or annulus pressure and valve position to the operator are some of the most important factors that must be considered. Physical aspects of the oilfield-water depth, lateral offset of the subsea equipment from its associated platform, anticipated sea or ice conditions, and the potential for well damage caused

by shut-ins-and the complexity of the subsea facility determine the optimal design of the control system.

Subsea facilities require two separate sets of controls. Production controls provide day-to-day operational control of the subsea equipment. Initial completion and subsequent workover of subsea wells require controls designed specifically for installation/maintenance functions. Completion/workover systems generally provide control over more functions than the associated production system. Production controls normally exclude control over hydraulic connectors, test ports, and vertical access valves.

Reliability/Maintainability

Reliability of a control system can be considered to be the probability that the control will not malfunction in such a manner as to preclude performing an intended function. Reliability usually is quoted over some specified period, such as the intended operating life of a particular project or the planned time between scheduled maintenance. For any system, the probability that the system will fail to perform its intended functions over a given time period is a function of the design, quality of the system components, built-in redundancy, and the quality of manufacture. Regardless of how well a system is designed and built and how short an operating period is considered, there will be a finite probability that it may malfunction during that period.

The likelihood of a failure can be minimized, but it can never be completely eliminated. Consequently, rather than going to great lengths and expense to design a system that “can’t fail, ” it may be more cost-effective to set realistic reliability goals and to concentrate on designing the equipment for maximum maintainability. This approach is aimed at minimizing the effect of a problem instead of trying to avoid the inevitable and generally results in a sound design that maximizes ease of retrieval and rein- stallation. How best to implement increased reliability and maintainability should be determined by the incremental cost involved and the benefits to be derived.

Redundancy

A system fault generally results in a well shut-in, and recovery of the control equipment for repair and reinstal- lation is then required before the well can be returned to production. The ultimate cost of the repairs is reflected in the expense of mobilization for repair, the repair, and subsequent demobilization.

One of the most economic and effective means of increasing system reliability is to include active backup for weaker elements within the basic subsea control module or, in the extreme, to provide a completely redundant module.

This approach may be cost-effective in situations where field conditions could impose severe economic penalties or pose a safety hazard in the event of a control system malfunction. The value of deferred/lost production and the possibility of premature well work necessitated by the shut-in also must be taken into account in evaluating the economics of providing redundancy.

Redundancy does not eliminate the ultimate need for repairs, but it may permit postponing them to take advantage of favorable weather windows, contracting rate trends, or vessel availabilities, all of which can work to


substantially reduce repair costs. In addition, built-in redundancy allows operations personnel to schedule the work on the basis of convenience without incurring unnecessary production losses or potential well damage.

Although potentially economical, redundancy is not free. The decision on the extent to which it should be included in a design must be based on a careful examina- tion of the cost impact on the project and the potential benefits.

Operational Considerations

Many operational factors and operating philosophies must be weighed when the control system for a specific installation is selected. Among the items that should be considered are response-time requirements; potential need for diver or remote-operated vehicle intervention; requirements for feedback regarding subsea valve operation, wellhead pressures, and temperatures; single-well or multiwell completions; use of subsea manifolds, commin- gling, and application of subsea chokes; type of control fluid; and the type of control-system design.

Response time for opening a tree valve usually is not critical if it is not unreasonably long, particularly with low flow velocities and clean production. Closing times are of considerable concern, however, and closing response may determine the type of control system to be employed. Excessively long closing times are neither desirable nor necessary. From the standpoint of pollution potential in the event of a flowline rupture, closing response times are less critical for water-injection wells than clean-gas wells, which, in turn, are less critical than oil wells.

All other factors being equal, a system that provides discrete control over each subsea valve and allows the operator to verify or to infer valve operations would always be recommended over one that does not. Some type of feedback on valve position is almost mandatory for safe operation of subsea facilities. If there is any concern about wellhead temperature or pressure because of flowline materials or other considerations, the ability to sense this information and to transmit it to the surface can be an overriding consideration in the selection of a control system.

Control Fluids

High-pressure control fluid is the means of converting a control command into subsea valve operation in both all- hydraulic and electrohydraulic control systems. Military- grade, low-viscosity, conventional oil-based hydraulic oil and highly water-based fluids are the two types of fluids in subsea control systems. Oil-based fluids provide the best system performance from the standpoints of lubricity, component wear, internal leakage, corrosion protection, and ultimately, system reliability. Oil can be used only in closed systems, however, because it cannot be discharged into the ocean when a control loop is vented to deactivate a control. Closed systems imply higher costs and, over long distances, slower response times. Oil-based systems can be particularly troublesome in cold climates. Water-based fluids, on the other hand, are inexpensive, are biodegradable (so they can be discharged into the environment), exhibit very low viscosity, and provide the fastest response times.

Unfortunately, water-based fluids also have certain inherent deficiencies. They exhibit lower lubricity, en-

courage higher leak rates, display lower corrosion inhibition, and are subject to biofouling from bacterial growth. All of these shortcomings can be overcome with proper component selection, design, and operating practices. Water-based fluids are used in most subsea and drilling control systems.

The single most important factor with any hydraulic control system, regardless of the type of fluid, is fluid cleanliness. Failure to keep the hydraulic system clean virtually guarantees an early malfunction.

Umbilicals

Platform control of subsea facilities requires a control umbilical between the platform and the subsea facility. The umbilical may consist of multiple hydraulic lines in a common jacket, a composite of hydraulic hoses, power wires, and communication pairs in a common jacket, or separate hydraulic and electrical bundles. The makeup of the umbilical depends entirely on the nature of the control system and the field conditions.

Hydraulic bundles can be fabricated with either stainless steel tubes or elastomeric hoses. Because of the importance of quick response and because hose expansion is a major determining factor in response time, manufacturers have upgraded hoses to the point that their expansion characteristics approach that of steel tubing. Hose and tubing display different installation and operational characteristics, however, and the decision on umbilical design must consider the unique aspects of the materials and specific project requirements.

Protection of control lines laid on the ocean floor is always a concern. One of the principal means of protecting control lines that are not buried is to armor them with galvanized wires applied in two separate, contrahelically wound layers. Because of the cost of armoring the large diameters encountered in most hose bundles, however, they are seldom armored. As armoring costs are reduced with improved technology, armoring of hose bundles may become more common. Hose bundles that are armored usually have a polyethylene bedding jacket under the armor and may have a thin covering over the armor. Some unarmored designs include a small-diameter wire rope molded integrally with the jacket to provide tensile strength and additional weighting. An alternative hose- bundle design makes use of a thick outer urethane jacket in lieu of armor. The high density of urethane provides the necessary negative buoyancy, and it provides excellent abrasion resistance.

A viable alternative to hose bundles uses stainless steel tubes laid side-by-side in a flat configuration. Relatively large-diameter wirelines (ropes) normally are placed on the outside of a flat bundle to provide mechanical protection and resistance to kinking.

The generally accepted practice is to armor subsea electric control cable for mechanical protection, weighting, and tensile strength. Even though it means a substantial reduction in cost, few, if any, unarmored cables have been installed subsea.

Alternative Subsea Control System

There are currently two primary approaches to the control of subsea equipment-all hydraulic and a hybrid of electric and hydraulic (electrohydraulic). Each offers a number of variations with unique advantages, disadvantages, and associated costs that must be considered and


SUPPLY

CONTROLLER

SUBSEATREE VALVE

AND ACTUATOR

Fig. 18.49-Direct hydraulic subsea control.

evaluated before a system design is finalized. Usually, one design will prove superior to the others for a given situation. Applicability of a given design depends on factors such as water depth, operating environment, offset distances, equipment to be controlled, operating requirements, operating philosophies, reservoir characteristics, and field economics.

The following sections include brief discussions of several basic system designs and their primary advantages and disadvantages. When the final selection is made, trade-offs must be made between simplicity, response time, operability, and costs on a site/project-specific basis.

Direct Hydraulic Control

Direct hydraulic control (Fig. 18.49) is the most straightforward design approach. It uses a single three-way surface control valve-a single, relatively large-diameter, dedicated high-pressure control line between the surface control valve and the subsea tree hydraulic-valve actuator and the valve actuator. When the surface control valve is operated, high-pressure fluid is introduced into the control hose, causing the subsea valve actuator to open the tree valve. When the surface valve is deactivated, the fluid that opened the subsea valve is returned to the surface fluid reservoir. The advantages of this system are simplicity, discrete remote control over each subsea function,

inherent feedback on subsea operations, and minimum cost of the basic control hardware. The disadvantages are slowest response time for a given control-line size, highest control-line costs, and potential problems with corrosion and biofouling of the subsea actuator when a water-based control fluid is used because the fluid is never renewed. System costs are directly proportional to the number of functions controlled and the distance between the control point and the subsea device. This type of system is seldom used with producing wells at offset distances of more than 10,000 ft, or with injection wells at more than 15,ooO ft, because of the cost and response-time considerations.

Response time of the direct hydraulic controls for the valve-closing (shut-in) operation over a given distance with a given line size can be improved significantly by installing a subsea dump valve on each tree-valve actuator. This approach also permits the introduction of new control fluid into the control lines because the fluid in the actuators is vented to the ocean whenever a valve is closed. This provides a gradual renewal of control fluid with each valve operation.

The disadvantage of dump valves is that they introduce additional flow-rate/direction-sensitive hydraulic-control elements that require very clean control fluid. This increases subsea hardware cost and markedly reduces the overall system reliability.

COMMON H P SUPPLY

/ PILOT VALVE

J (1 PER FUNCTION)

1

I

SUBSEA TREE VALVE

k------i AND ACTUATOR

I

I

Fig. 18.50-Discrete-piloted hydraulic subsea control


Fig. l&51--Sequential-piloted hydraulic subsea control.

Discrete-Piloted Hydraulic

Discrete-piloted control (Fig. 18.50) has a single three- way control valve at the surface for each subsea function, a corresponding subsea pilot control valve, and single small-diameter dedicated signal line between the two control valves. A single common high-pressure line provides hydraulic supply from the surface to the tree. When the subsea pilot valve is actuated, it switches high-pressure fluid from the supply line to the subsea valve actuator. When the pilot valve is deactuated, it vents the actuator fluid subsea rather than returning it to the surface.

The advantages of this design are its relative simplicity, discrete remote control over each subsea function, inherent inferred feedback on subsea operations, relatively low-cost hardware, and relatively fast response compared to the direct hydraulic design. The disadvantages are that cost and response time are both proportional to distance just as they are with direct hydraulics, but the effects are significantly less. The pilot lines are dead-ended, but cor-

SlNGLE WIRE PAIR REGARDLESS OF NUMBER OF SIGNALS OR AMOUNT OR TYPE OF INFORMATION

rosion is not a problem because pilot valves usually are made of stainless alloys. Biofouling, however, can still be a problem.

Sequential-Piloted Hydraulic

Sequential control is similar to discrete-piloted control in that it also has subsea pilot valves that direct high-pressure hydraulic fluid to tree-valve actuators (Fig. 18.5 1). How- ever, subsea pressure-sensitive pilot valves that are manifolded to a common signal line are used in this design. Rather than discrete control over each individual pilot valve, they are switched in groups according to the signal pressure and the pilot-valve set points. The valves are interconnected so that supply pressure is applied to the subsea tree actuators in a predetermined sequence in response to changes in signal pressure. Up to six combinations of valve operations can be implemented reliably with a single control line and a single supply line. The advantage of this system is its relatively lower cost for

r 1 ATMOS POD

COMMON ELECTRICAL POWER SUPPLY

ACCUMULATOR

1 ATMOS OR OIL FILLED AMBIENT COMP POD

-I---- SOLENOID VALVE \

COMMON H P HYDRAULIC SUPPLY Y

SUBSEATREE VALVE WITH

HYDRAULIC ACTUATOR

Fig. l&52-Multiplexed control with hydraulic valve actuators.

1 B-52 PETROLEUM ENGINEERING HANDBOOK

hardware and control lines. It has a somewhat slower response than a discrete-piloted hydraulic system, and it is less reliable because it is more complex. The really significant disadvantages, however, are that independent valve control is impossible, there is no ready means of confirming valve operations, and operating sequences must be predetermined before manufacture and installation. Further, the sequence approach offers no means for independent valve control for performing operational tests, troubleshooting or diagnosis, or for changing flow paths through a subsea system in the event of a malfunction.

Multiplexed Electrohydraulic Control

This design (Fig. 18.52) has a single high-pressure hydraulic supply line and an electric cable that, in its simplest form, consists of one pair of signalling wires and one pair of power wires. The limited number of electrical conductors permits the use of inductive couplers rather than the more conventional pin-and-socket connectors. Pin connectors historically have been troublesome in subsea production applications, while inductive couplers have proved to be relatively foolproof. Inductive couplers are easily and reliably remotely mated and unmated subsea, which provides maximum flexibility in hardware layout. Encoding and decoding multiplex logic are provided at the surface and subsea to enable transmission of coded valve-open and valve-close commands and associated status feedback over one pair of single wires. The multiplexed valve commands trigger electric solenoid pilot valves that activate hydraulic-switching valves to direct high-pressure fluid to the tree-valve actuators. The advantages of this system are (1) fastest possible response independent of distance, (2) simplicity of control lines, and (3) total flexibility. With some designs, a totally hydraulic backup system can be included with the multiplexed system at little increase in cost to provide some measure of temporary operability in the event of an electrical failure. The obvious disadvantages of this design are its increased complexity and higher costs.

With a multiplexed electrohydraulic system, a recir- culating (closed) subsea hydraulic power unit is also feasible in the control housing. This approach offers the possibility of eliminating the hydraulic supply line if system leakage can be controlled to a very low level. If this design approach is used, some provision still must be made for fluid makeup to account for long-term leakage. The primary advantage is that it allows the use of conventional oil-based hydraulic fluids, thus ensuring maximum life and reliability. The obvious disadvantages are the substantial increase in system complexity and conse- quent overall lower system reliability.

2. The Tt4~nolog~ oj’Qfj~hore Drilling, Coi,~pl~tiotc md Pmductim. ETA Offshore Seminars, Inc.. The Petroleum Publishing Co.. Tulsa (1976).

3. Silcox, W.H.: “Floating Drilling: The First 30 Years--Part I and Part 2,” J. Per. Tech. (Jan. and Feb. 1983).

4. Burke, B.G.: “Downtime Evaluation for Operations from Float- ing Vessels in Waves,” Proc., 1977 Society of Naval Architects and Marine Engineers Second Ship Tech. and Research Symposi- um. New York City, 279-301.

5. Sheffield, R.: “Floating Drilling: Equipment and Its Use,” Prrw tica/ Drilling Technology. Gulf Publishing Co., Houston (1980) 2.

6. “Recommended Practice for Blowout Prevention Equipment Sys- terns,” second edition, API RP 53. API, Dallas (Jan. 1984).

7. “The Analysis of Spread Mooring Systems for Floating Drilling Units,” latest edition, API RP 2P. API, Dallas.

8. “Design and Operation of Marine Drilling Riser Systems.” latest edition, API RP 24, API, Dallas.

9. “Recommended Practice for Care and Use of Marine Drilling Risers,” latest edition. API RP 2K. API. Dallas.

IO. “Recommended Practices for Safe Drilling of Wells Contaimng Hydrogen Sulfide.” latest edition, API RP 49, API, Dallas.

I I. “Strengthening for Navigation in Ice.” Rulrsfor Buikditlg trnd Clms- ing Steel Vessels, latest edition. American Bureau of Shipping, New York City.

12. Bardgette, J.J. and Irick, J.T.: “Construction of the Hondo Plat- form in 850 Feet of Water,” paper OTC 2959 presented at the 1977 Offshore Technology Conference, Houston. May 2-S.

13. Kinra, R.K. and Marshall, P.W.: “Fatigue Analysis of the Cognac Platform,” paper OTC 3378 presented at the 1979 Offshore Tech- nology Confeience, Houston. April 30-May 3.

14. Tan&ill, C.A., Isenhower, W.M., and Engle. D.D.: “Cerveza-A Project Overview of a Deep-water Platf&m,” paper OTC 4185 presented at the 1982 Offshore Technology Conference, Houston. May 3-6.

15. Goodman, M.A.: Handbook of Arciic Well Completions. Gulf Pub- lishing Co., Houston (1978).

16. “U.S. Arctic Oil and Gas.” Natl. Petroleum Council, Washing- ton, DC (Dec. 1981).

17. “API Bulletin on Planning, Designing, and Constructing Fixed Off- shore Structures in Ice Environments,” latest edition, Bull. 2N. API, Dallas.

18. Willits, K.L. and Lindsey, W.C.: “Well Completions in the Prudhoe Bay Field,” Pet. Eng. Id. (Feb. 1976) 48-56.

19. Narl. Elecrrical Code, latest edition, Natl. Fire Protection Assn.. Quincy, MA.

20. “Recommended Practice for Design and Installation of Electrical Systems for Offshore Production Platforms,” latest edition, API RP 14F, API, Dallas.

21. USCG Regulation 46 CFR Shipping, Chap. I, Subchapter I-A, “MO- bile Offshore Drilling Units,” Subchapter J, “Electrical Engineer- ing,” U.S. Government Printing Office, Washington. DC. Parts 108 and 110-13.

22. “Recommended Practice for Classification of Areas for Electrical Installations at Drilling Rigs and Production Facilities on Land and on Marine Fixed and Mobile Platforms,” latest edition, API RP 5OOB, API. Dallas.

23. “Standards for Purged and Pressurized Enclosures for Electrical Equipment in Hazardous (Classified) Locations,” Bull. 496, Natl. Fire Protection Assn., Quincy, MA.

24. “Recommended Practice for Analysis, Design, Installation and Tcst- ing of Basic Surface Safety Systems on Offshore Production Plat- forms,” latest edition, API RP 14C, API, Dallas.

25. LJSCG Regulation 33 CFR, Chap. I, Subchapter N. “Outer Con-

References tinental Shelf Activities,” U.S. Ciovernmknt Printing Office, Washington. DC, Parts 140 through 147.

I. Rintoul, B.: “Drilling From The Steel Island,” Pcrcific Oi/ 26. USCG Regulation 30 CFR, Part 250, “Oil and Gas and Sulfur Op- World/Annual, Petroleum Publishers, Brea. CA (Jan. 1980) 73, erations in the Outer Continental Shelf.” Minerals Management 18-28. Service, U.S. Dept. of Interior, Washington, D.C.

Chapter 19

Crude Oil Emulsions H. Vernon Smith, Meridian Corp.

Kenneth E. Arnold, Paragon Engineermg Scr~icm Inc.

Introduction Much of the oil produced worldwide is accompanied by water in an emulsion that requires treating. Even in those fields where there is essentially no initial water production. water cuts may increase in time to the point where it is necessary to treat the emulsion. Water content of the untreated oil may vary from a fraction of I % to over 90%.

To prevent increased transportation costs, water treatment and disposal costs, and deterioration of equipment, purchasers of crude oil limit the basic sediment and water (BSSCW) content of the oil they purchase. Limits vary depending on local conditions, practices. and contractual agreements and typically range from 0.2 to 3.0%. BY&W is usually predominantly water but may contain solids. The solids contained in the BS&W come from the producing formation and consist of sand. silt, mud, scale. and precipitates of dissolved solids. These troublesome solids vary widely from producing field to field. zone to zone, and well to well.

Purchasers may also limit the salt content of the oil. Removing water from the stream decreases the salt content. Salt content along with BS&W are the two impor- tsnt crude purchasing requirements.

When water forms a stable emulsion with crude oil and cannot be removed in conventional storage tanks. emulsion-treating methods must be used. The methods. procedures, equipment, and systems generally used in treating crude oil emulsions are considered in this chapter, Space limitation does not permit the rigorous trcat- ment of crude oil emulsions. Many topics and sub-topics exist on which entire chapters can be written. This chap- tcr contains an abbreviated discussion of only a few of the most important and pertinent considerations of crude oil emulsions. More detailed and diversified discussions on crude oil emulsions can be found in the General Refer- ences at the end of the chapter.

Theories of Emulsions Definition of an Emulsion An emulsion is a heterogeneous liquid system consisting of two immiscible liquids with one of the liquids intimately dispersed in the form of droplets in the second liquid. An emulsion is distinguished from a simple dispersion of one liquid in another by the fact that, in an emulsion, the probability of coalescence of droplets on contact with one another is greatly reduced because of the presence of an emulsifier, which inhibits coalescence. Such inhibition is not present in a dispersion.

The stability of the emulsion is controlled by the type and amount of surface-active agents and/or finely divided solids. which commonly act as emulsifying agents or emulsifiers. As shown in Fig. 19.1, these emulsifying agents form interfacial films around the droplets of the dispersed phase and create a barrier that slows down or prevents coalescence of the droplets.

The matrix of an emulsion is called the external or continuous phase. The portion of the emulsion that is in the form of small droplets is called the internal, dispersed, or discontinuous phase. The emulsions considered in this chapter consist of crude oil and water or brine produced with it.

In most emulsions of crude oil and water, the water is finely dispersed in the oil. The spherical form of the water globules is a result of interfacial tension (IFT). which com- pels them to present the smallest possible surface area to the oil. This is a water-in-oil emulsion and is referred to as a “normal” emulsion. The oil can be dispersed in the water to form an oil-in-water emulsion, which is referred to as an “inverse” or “reverse” emulsion. A typical reverse emulsion is shown in Fig. 19.2.

Emulsions are sometimes interrelated in a more complex form. The emulsion may be either water-in-oil or oil-in-water to begin with, but additional agitation may


Fig. 19.1—Photomicrograph of water-in-oil emulsion. Observethe riqid-appearing film or skin that retardscoalescence.

cause it to become multistage. If it is a water-in-oil emul-sion initially, a water-in-oil-in-water emulsion can beformed if a small volume of the original water-in-oil emul-sion is enveloped in a film of water. It is also possibleto form multistage emulsions in an oil continuous phaseas shown in Figs. 19.3 and 19.4. This alternating external-phase/internal-phase/external-phase arrangement has beenknown to exist in eight stages. Multistage emulsions usual-ly add appreciably to the problem of separating the emul-sion into oil and water. The more violent the agitation,the more likely multistage emulsions are to form.

How Crude Oil Emulsions FormThe three conditions necessary for the formation of anemulsion are (1) the two liquids forming the emulsionmust be immiscible, (2) there must be sufficient agitationto disperse one liquid as droplets in the other, and (3) theremust be an emulsifying agent present. Crude oil and waterare immiscible. If gently poured into the same container,they will quickly separate. If the oil and water are vio-lently agitated, small drops of water will be dispersed inthe continuous oil phase and small drops of oil will bedispersed in the continuous water phase. If left un-disturbed, the oil and water will quickly separate intolayers of oil and water. If any emulsion is formed, it willbe between the oil above and the water below.

When considering crude oil emulsions, we are usuallyconcerned with water-in-oil emulsions because most emul-sions are this type. Oil-in-water emulsions are encoun-tered in some heavy oil production, however, such as that

Fig. 19.2—Photomicrograph of reverse emulsion. Uniformlysized oil particles are about 10 µm in diameter andare dispersed in the continuous water phase.

found in areas of Canada, California, Venezuela, andother areas. Oil-in-water emulsions are generally resolvedin the same way as water-in-oil emulsions, except elec-trostatic treaters cannot be used on oil-in-water emulsions.

The agitation necessary to form an emulsion may re-sult from any one or a combination of several sources:(1) the bottomhole pump, (2) flow through the tubing,wellhead, manifold, or flowlines, (3) the surface trans-fer pump, or (4) pressure drop through chokes, valves,or other surface equipment. The greater the amount ofagitation, the smaller the droplets of water dispersed inthe oil. Figs. 19.5 through 19.9 show common crude oilemulsions that demonstrate the range of droplet sizesnormally encountered. Studies of water-in-oil emulsionshave shown that water droplets are of widely varyingsizes, ranging from less than 1 to about 1,000 µm. Emul-sions that have smaller droplets of water are usually morestable and difficult to treat than those that have largerdroplets.

Crude oils vary widely in their emulsifying tendencies.Some may form very stable emulsions that are difficultto separate, while others may not emulsify or may forma loose emulsion that will separate quickly. The presence,amount, and nature of an emulsifying agent determineswhether an emulsion will be formed and the stability ofthat emulsion. If the crude oil and water contain no emul-sifying agent, the oil and water may form a dispersionthat will separate quickly because of rapid coalescenceof the dispersed droplets. On the other hand, if an emul-sifying agent is present in the crude oil, a very stable emul-sion can be formed.

CRUDE OIL EMULSIONS 19-3

Fig. 19.3—Photomicrograph of oil-in-water-in-oil emulsion. Oil Fig. 19.4—Photomicrograph of multiple-stage emulsion fromdroplets are shown dispersed in water droplets that Rocky Mountain field. The dispersed water phaseare dispersed in the continuous oil phase. contains small oil particles.

If an emulsion is not treated, a certain amount of waterwill separate from the oil by natural coalescence andsettling because of the difference in density of oil andwater. Unless some form of treatment is used to accom-plish complete separation, however, there probably willbe a small percentage of water left in the oil even afterextended settling. The water that remains in the oil willbe in minute droplets that have extremely slow settlingvelocities. They will be widely dispersed so that there willbe little chance for them to collide, coalesce into largerdroplets, and settle.

The amount of water that emulsifies with crude oil inmost production systems may vary from less than 1 tomore than 60% in rare cases. The most common rangeof emulsified water in light crude oil-i.e., oil above20° API-is from 5 to 20 vol%. The most common rangeof emulsified water in crude oil heavier than 20° API isfrom 10. to 35%.

Emulsifying AgentsEmulsifying agents are surface-active compounds that at-tach to the water-drop surface and lower the oil/water IFT.When energy is added to the mixture by agitation, thedispersed-phase droplets are broken into smaller droplets.The lower the IFT, the smaller the energy input requiredfor emulsification-i.e., with a given amount of agitation,smaller droplets will form.

There are many theories on the nature of emulsifyingagents in crude oil emulsions. Some emulsifiers arethought to be asphaltic in nature. They are barely soluble

Fig. 19.5—Photomicrograph of loose emulsion from westernKansas containing about 30% emulsified water in theform of droplets ranging in diameter from about 60µm downward.


Fig. 19.6—Photomicrograph of water-in-oil emulsion with dis-persed particles of water ranging in size from about250 to about 1 µm.

Fig. 19.7—Photomicrograph of relatively tight water-in-oil emul-sion. Largest water droplets are about 60 µm,medium droplets are about 40 µm, and the smallestones are about 1 to 20 µm.

Fig. 19.8—Photomicrograph of tight emulsion with the dispersedwater particles varying in size from 1 to 20 µm.

in oil and are strongly attracted to the water. They comeout of solution and attach themselves to the droplets ofwater as these droplets are dispersed in the oil. They formthick films that surround the water droplets and preventthe surfaces of the water droplets from contacting, thuspreventing coalescence when the droplets collide.

Oil-wet solids-such as sand, silt, shale particles, crys-tallized paraffin, iron, zinc, aluminum sulfate, calciumcarbonate, iron sulfide, and similar materials-that col-lect at the oil/water interface can act as emulsifiers. Fig.19.10 shows some of these solids removed from a crudeoil emulsion. These substances usually originate in theoil formation but can be formed as the result of an in-effective corrosion-inhibition program.

Many emulsions are prepared for commercial use. Anemulsion of kerosene and water is used for spraying fruittrees; soap is used as the emulsifying agent. Eggs supplythe emulsifying agent used in the preparation of mayon-naise from vegetable oil and vinegar: These are very stableemulsions.

Most but not all crude oil emulsions are dynamic andtransitory. The interfacial energy per unit of area in pe-troleum emulsions is rather high compared with familiarindustrial emulsions. They are therefore thermodynami-cally unstable in the sense that if the dispersed watercoalesced and separated, the total free energy woulddecrease. Only the presence of an emulsifier film in-troduces an energy barrier that prevents the “breaking”or separation process from proceeding.

The characteristics of an emulsion change continuallyfrom the time of formation to the instant of complete reso-lution. This occurs because there are numerous types of

CRUDE OIL EMULSIONS

Fig. 19.9—Photomicrograph of tight emulsion from HuntingtonBeach, CA; water content 20%, with the averagewater droplet diameter less than 5 µm.

adsorbable materials in a given oil. Also, the adsorptionrate of the emulsion and permanence of location at theinterface may vary as the fluid flows through the proc-ess. Furthermore, the emulsion characteristics are changedas the liquid is subjected to changes in temperature, pres-sure, and degree of agitation.

Prevention of EmulsionsIf all water can be excluded from the oil as it is producedand/or if all agitation of well fluids can be prevented, noemulsion will form. Exclusion of water in some wells isdifficult or impossible, and the prevention of agitation isalmost impossible. Therefore, production of emulsionfrom many wells must be expected. In some instances,however, emulsification is increased by poor operatingpractices.

Operating practices that include the production of ex-cess water as a result of poor cementing or reservoirmanagement can increase emulsion-treating problems. Inaddition, a process design that subjects the oil/water mix-ture to excess turbulence can result in greater treatingproblems. Unnecessary turbulence can be caused by over-pumping and poor maintenance of plunger and valves inrod-pumped wells, use of more gas-lift gas than is needed,and pumping the fluid where gravity flow could be used.Some operators use progressive cavity pumps as opposedto reciprocating, gear, or centrifugal pumps to minimizeturbulence. Others have found that some centrifugalpumps can actually cause coalescence if they are installed

Fig. 19.10—Photomicrograph showing a collection of inorganicsolids removed from an emulsion by filtering andwashing. These solids include calcite, silica, ironcompounds, obsidian, and black carbonaceous ma-terials.

in the process without a downstream throttling valve.Wherever possible, pressure drop through chokes andcontrol valves should be minimized before oil/water sepa-ration.

Color of EmulsionsThe color of a crude oil emulsion can vary widely, de-pending on the oil and water content of the emulsion andthe characteristics of the oil and water. The most com-mon color of emulsions is a dark reddish brown. How-ever, any color from light green or yellow to grey or blackmay be found. “Brightness” is an indicator of the pres-ence of an emulsion. Oil-free water and water-free oil areclear and bright. Emulsions are murky and opaque be-cause of reflection and scattering/of light at the oil/waterinterfaces of the dispersed phase. The greater the totalinterfacial area between the oil and water, the lighter thecolor of the emulsion. That is, an emulsion containingmany small droplets of water will tend to be lighter thanone containing an equal volume of water in larger dropletsbecause the latter has less total interfacial surface area.

Stability of EmulsionsGenerally, crude oils with low API gravity (high density)will form a more stable and higher-percentage volume ofemulsion than will oils of high API gravity (low density).Asphaltic-based oils have a tendency to emulsify morereadily than paraffin-based oils. High-viscosity crude oil


will usually form a more stable emulsion than low- viscosity oil. Emulsions of high-viscosity crude oil usually are very stable and difficult to treat because the viscosity of the oil hinders or prevents movement of the dispersed water droplets and thus retards their coalescence. In addition, high-viscosity/high-density oils usually contain more emulsifiers than lighter oils.

Effect of Emulsion on Viscosity of Fluids

Emulsions are always more viscous than the clean oil contained in the emulsion. The ratio of the viscosity of an emulsion to the viscosity of the clean crude oil in oilfield emulsions depends on the shear rate to which it has been subjected. The authors have found that for many emulsions and the shear rates normally encountered in piping systems, this shear rate can be approximated by the following equation if no other data are available.

~,/~~~=1+2.5f+l4.lJ”, . . . . . . . . . (1)

where

cc e = viscosity of emulsion.

PO = viscosity of clean oil, and

f = fraction of the dispersed phase.

Sampling and Analyzing Crude Oil Emulsions Purchasers of crude oil have established certain specifications that must be met before they will accept oil from a producer. These specifications limit the amount of BS&W in the oil. The limitations are usually strict, and if the amount of ES&W in the oil exceeds the specified limit. the oil may not be accepted by the purchaser. The seller and buyer must agree on the procedure for sampling and analyzing the oil to provide consistent and mutu- ally acceptable data.

The performance of emulsion-treating units or systems can be observed and studied by the practice of regularly and periodically withdrawing and analyzing samples of the contents at multiple levels in the vessels or multiple points in the systems. This is particularly beneficial in treating emulsions involving viscous oils. Samples of emulsions should be representative of the liquid from which they are taken. Emulsification should not occur when the sample is extracted. Samples obtained at the wellhead. manifold, or oil and gas separator may show a high percentage of emulsion, but the oil and water in the system may actually not be emulsified. This indicates that emulsification occurred because of the turbulence created as the sample was removed from the pressure zone to the sample container.

It is possible to take a sample from a pressure zone without further emulsification of the liquids if the velocity of the discharging liquid is controlled. One method is to use a piece of small-diameter tubing approximately 10

the line to the container is absorbed by flow through the tubing. Flow through the tubing, however, can cause either coalescence or additional emulsification.

Another method of withdrawing a representative sample of emulsion is to use a sample container initially filled with water. The sample container is equipped with valves at the top and bottom with the top valve connected to the point from which the sample is to be extracted. The top valve of the container is opened first and the container pressured from the line. The valve at the bottom of the container is then opened and the water discharged into the atmosphere as the sample enters the container. There will be no emulsification in the container because there is no pressure drop between the source and sample container to cause turbulence. Once the sample is taken, pressure can be bled off through a third valve with little effect on the sample.

Small centrifuges are used to determine BS&W content of crude oil. The centrifuges may be driven by hand or electric motor. A small measured volume of sample is diluted with solvent and placed in graduated glass containers. These are then inserted into the centrifuge and rotated at high speed for a few minutes. Separation of the oil, water, and solids is accomplished by centrifugal force. The percentages of each constituent can be read directly from the graduated containers in which the sample is cen- trifuged. The speed used in these small centrifuges varies from 2,000 to 4,000 revimin.

Methods of taking and analyzing samples of crude oil for custody transfer are included in the API Mur~uul of Petroleum Measurement Standards. Also see Chap. 17.

Methods Used in Treating Crude Oil Emulsions Three basic steps usually are required to separate a crude- oil/water emulsion into bulk phases of oil and water.

Step l-Destabilization. An emulsion is destabilized by counteracting the stabilizing effect of the emulsifier. The tough skin or film surrounding the dispersed water droplets must be weakened and broken. This is usually accomplished by adding heat and/or a properly selected, interfacially active chemical compound to the emulsion.

Step 2-Coalescence. After the films encasing the dispersed droplets are broken, the dispersed droplets must coalesce into drops large enough to settle out of the continuous phase of oil. Fig. 19.11 shows a small droplet of water breaking through a destabilized emulsion film to coalesce with the bigger drop. This usually is accomplished by imposing a period of moderate agitation or by subjecting the destabilized emulsion to an alternating electric field. This will increase the dispersed droplets contacting rate. Thus coalescence will increase, resulting in larger droplets.

Step 3-Gravity Separation. A quiet period of settling must be provided to allow the coalesced drops to settle out of the oil because of the difference in density between the water and oil. This is accomplished by providing a

to 15 ft.long. One end of the tubing is connected to a sufficient residence time and a favorable flow pattern in bleeder valve on the line or vessel from which the sam- a tank or vessel that will allow the coalesced drops of ple is to be extracted, and the other end is connected to water to separate from the oil. the sample container. The bleeder valve should be opened Another way of stating the general emulsion-treating fully and the sample allowed to flow through the small- procedure is that to resolve a crude-oil/water emulsion diameter tubing into the container. The pressure drop from into bulk oil and water three things must be done:

CRUDE OIL EMULSIONS

Fig. 19.11 -A waler-in-oil emulsion with the film or skin surrounding the water droplet in the process 01 rupturing.

(I) increase the probability of coalescence of dispersed water droplets on contact, (2) make the rate of contact of dispersed water droplets high without creating high shear forces, and then (3) allow the liquids to settle quietly so that they can separate into bulk phases of oil and water. All the incidental variables, such as selection of proper chemical, rate of chemical injection, treating temperature and pressure, oil and emulsion viscosity, flow rate, vessel design, vessel size, and fluid levels, are controlled to execute these three steps in the quickest and most economical manner.

An emulsion-treating unit or system will use one or more of the methods in Table 19. I to aid in destabiliz- ing, coalescence, and/or settling. Each of these treating methods that can be used to resolve an emulsion is discussed separately.

Heating

The use of heat in treating crude oil emulsions has four basic benefits.

I. Heat reduces the viscosity of the oil, resulting in a greater force during collision of the water droplets. Also, the reduced oil viscosity allows the water droplets to settle more rapidly through the less viscous oil, Fig. 19.12 can be used to estimate crude oil viscosity/temperature relationships. Viscosities vary widely from one crude to another. The curves should be used only in the absence of specific data. If the viscosity of the crude is known at two temperatures, the viscosity at other temperatures can be approximated by a straight line. If the viscosity is known at one temperature, it can be approximated at

19-7

TABLE 19.1-METHODS TO AID DESTABILIZATION, COALESCENCE, AND/OR SETTLING

Destabilization Chemical Heating

Coalescence Agitation Coalescing plates Electric field Water washing Filtering Fibrous packing Heating Retention time Centrifugation

Gravity Separation Gravity settling Heating Centrifugation

other temperatures by drawing a straight line parallel to the others. If the viscosity is unknown at any temperature, the lines on the chart may be used. API Spec. l2L recommends that crude be heated so that its viscosity is below 150 SSV (about 50 cSt) for treating.

2. Heat increases the droplets’ molecular movement. This aids in coalescence through increased collision frequency of the dispersed-phase droplets.

3. Heat may deactivate the emulsifier (e.g., dissolving paraffin crystals) or it can enhance the action of treating chemicals, causing the chemical to work faster and more thoroughly to break the film surrounding the droplets of the dispersed phase of the emulsion.

4. Heat may also increase the difference in density between the oil and the water, thus accelerating settling. In general, at temperatures below 180”F, the addition of heat will increase the difference in density. Most light oils are treated below 180°F; thus the effect of heat on gravity is beneficial. For heavy crudes (below 20”API). which normally are treated above 180”F, heat may have a negative effect on difference in density. In special cases, increased heat may cause the density of water to be less than that of oil. This effect is shown in Fig. 19.13.

Heating well fluids is expensive. Adding heat can cause a significant loss of the lower-boiling-point hydrocarbons (light ends). This results in “shrinkage” of the oil, or loss of volume. Because the light ends are boiled off, the remaining liquid has a lower API gravity and thus may have a lower value. Figs. 19. I4 and 19.15 illustrate typical gravity and volume losses for 33”API crude vs. temperature. The molecules leaving the oil phase may be vented or compressed and sold with the gas. Even if they are sold with the gas, there probably will be a net loss in income.

The gas liberated when crude oil is treated may also create a problem in the treating equipment if the equipment is not properly designed. In vertical emulsion treaters and gun barrels, some gas may rise through the coalescing section. The liberated gas can create enough turbulence and disturbance to inhibit coalescence. Perhaps more important, the small gas bubbles have an attraction for surface-active material and hence for the water droplets; thus they have a tendency to keep the water droplets from settling and even may cause them to be discharged with the oil.

19-8

Fig. 19.12-Approximate viscosity/temperature relationships for crude oil

1 0 *oo 200 300

Temperature. D F

Water

Crude C

0 100 200 300 Temperature. D F

Fig. 19.13--Relationship of specific gravity with temperature for three crude oils.

Fuel is required to provide heat, and the cost of fuel must be considered. If the oil is much above ambient temperature when discharged from the treating unit, it can be flowed through a heat exchanger with the incoming well fluid to transfer the heat to the cooler incoming well fluid. This will minimize evaporation losses and reduce fuel cost. It will also increase the vapor pressure of the crude, however, which may be limited by contract.

If properly applied, heating an emulsion can have great beneficial effect on water separation. The most economical emulsion treating may be obtained by use of less heat and a little more chemical, agitation, and/or settling space.

In some geographic areas, emulsion heating requirements vary in accordance with daily and/or seasonal atmospheric temperatures. Emulsions are usually more difficult to treat when it is cool-at night, during a rain, or in winter months when the atmospheric temperatures are lowest. Treatment, especially heating, may not be required in warmer summer months. Where the treating problem is seasonal, some emulsions can be resolved suc- cesstilly by addition of more chemical demulsifiers during winter months. Study is required to determine the proper economic balance of heat and chemicals.

Crude oil emulsions with similar viscosity ranges do not always require the same type of treating equipment or the same treating temperature. Emulsions produced


Fig. 19.14-API gravity loss vs. temperature for crude oil.

TYPICAL 33’ API

GRAVITY LOSS

TEMPERATURE, ‘F

from different wells on the same lease or from the same formation in the same field may require different treating temperatures. For this reason, it is recommended that low treating temperatures be tested so that the lowest practical treating temperature for each emulsion and treating unit or system can be determined by trial.

The heat input and thus the fuel required for treating depends on the temperature rise, amount of water in the oil, and the flow rate. It requires about twice as much

energy to heat a given volume of water as it does to heat the same volume of oil. For this reason, it is beneficial to separate free water from the emulsion to be treated. Often this is accomplished in a separate free-water knockout vessel upstream of the point where heat is added. Sometimes it is accomplished in a separate section of the same vessel.

The required heat input for an insulated vessel (heat loss is assumed to be 10% of heat input) can be approximated from

Q= 16AT(OSy,y, +q,,,yb,,), (2)

where

Q = heat input, Btu/hr.

AT = increase in temperature, “F,

90 = oil flow rate, B/D,

9 1, = water flow rate, B/D,

Yo = specific gravity of oil, and

Yn = specific gravity of water.

Chemical Demulsifiers

Certain chemical compounds are widely used to destabilize and to assist in coalescence of crude oil emulsions. These are referred to as dehydration chemicals or demulsifiers. This treatment method is popular because the

50 70 80 110 130 150

TEMPERATURE, l F

Fig. 19.15-Percent loss by volume vs. temperature for crude oil.

chemicals are easily applied to the emulsion, usually are reasonable in cost, and usually minimize the amount of heat and settling time required.

The chemical counteracts the emulsifying agent, allowing the dispersed droplets of the emulsion to coalesce into larger drops and settle out of the matrix. For demulsifiers to work, they must (1) be injected into the emulsion, (2) intimately mix with the emulsion and migrate to all of the protective films surrounding all of the dispersed droplets, and (3) displace or nullify the effect of the emulsifying agent at the interface. A period of continued moderate agitation of the treated emulsion to produce contact between and coalescence of the dispersed droplets and a quiet settling period must exist to allow separation of the oil and water.

Four actions are required of a chemical demulsifier. Strong attraction to the oil/warer interjace. The demul-

sifier must have ability to migrate rapidly through the oil phase to reach the droplet interface where it must counteract the emulsifying agent.

Flocculation. The demulsifier must have an attraction for water droplets with a similar charge and bring them together. In this way, large clusters of water droplets gather, which look like bunches of fish eggs under a microscope.

Coalescence. After flocculation, the emulsifier film is still continuous. If the emulsifier is weak, the flocculation force may be enough to cause coalescence. This is not true in most cases, and the demulsifier must therefore neutralize the emulsifier and promote rupture of the droplet interface film. This allows coalescence to occur. With the emulsion in a flocculated condition, the film rupture results in growth of water drop size.

Solids Wetting. Iron sulfides, clays, and drilling muds can be made water-wet, causing them to leave the interface and be diffused into the water droplets. Paraffins and


asphaltenes can be dissolved or altered by the demulsitier to make their films less viscous, or they can be made oil- wet so that they will be dispersed in the oil.

The demulsifier selection should be made with all functions of the treating system in mind. If the process is a settling tank, a relatively slow-acting demulsifier can be applied with good results. On the other hand, if the system is an electrostatic process, where some of the flocculation and coalescing action is accomplished by the electric field, there is need for a quick-acting demulsifier. Time for demulsifier action in a vertical emulsion treater normally will be somewhere between that of a settling tank and that of an electrostatic treater.

As field conditions change and/or the treating process is modified, the chemical requirements may change. Seasonal changes may cause paraffin-induced emulsion problems. Well workovers may change solids content, which may alter emulsion stability. So no matter how satisfactory a demulsifier is, it cannot be assumed that it will always be satisfactory over the life of the field.

While the first commercial emulsion-treating chemical was a solution of soap, present-day chemicals are based on highly sophisticated materials. Chemical emulsion breakers are complex organic compounds with surface- active characteristics. The active properties may be derived from any one or a combination of nonionic, cation- ic, and anionic materials. Within each of these types, compositions are used that will confer various degrees of hydrophobeihydrophile balance to the chemical as desired. The active components are highly viscous and sometimes even solids. It is necessary to use a carrier that will make handling easier: this carrier is almost without exception an organic solvent. Solvent systems are designed to make emulsion breakers compatible with the crude oil system in which they are used. It is also necessary to omit materials that will interfere with refining processes, such as those that will poison catalysts. Therefore, no organic chlorides, bromides, iodides, fluorides, or compounds of arsenic or lead are used in the manufacture of most emulsion-treating chemicals.

There is no simple designation of specific chemicals to treat specific emulsions. There are, however, certain common demulsifier types that tend to produce a consistent reaction in many water-in-oil emulsions. Some of the demulsifier types are as follows.

Pol~~lvcc~l esters are characterized by quick brightening of emulsjons. but frequently tend toward slow water drop and sludging; they are subject to overtreating problems.

Lo~r,-lnolrculrrr-~~‘~~i~~~t resin derivatives tend toward rapid water drop and fair to good overall demulsification properties; they show some tendency toward overtreatment in high-API-gravity emulsions.

High-molecular-weight resin derivatives generally have a strong wetting tendency and fair brightening and water- drop characteristics; they are always used in combination with other materials.

Sulfonates exhibit fair to good wetting and water-drop performance, some ability to brighten oil, and very little tendency to overtreat, particularly in high-gravity emulsions.

Polymerized oils and esters produce specific characteristics on particular emulsions; they are generally poor for widespread application and are always used in combination with other materials.

Alkanolamine condensates promote water drop in some emulsions and may produce some brightening; they are blended with other materials for overall good performance.

Oxyalkylatedphenols are predominantly wetting agents with fair to poor demulsification properties; they are used in blending to improve demulsifier performance.

Polyamine derivatives produce good brightening characteristics and are good blending agents; they are relatively poor in other respects.

There are many specific variations within each of these broad categories. Most demulsifiers used in breaking crude oil emulsions are blends of the above and other compounds. The components selected for a given demulsifier are chosen to provide the necessary actions to achieve complete emulsion treatment. The number of different surface-active materials that can act as emulsifiers in crude oil is large. The possible combinations of these emulsifying agents is almost infinite. Therefore, the number of demulsifiers and their combinations must likewise be numerous to treat the emulsions. The type and composition of the crude oil in the emulsion being treated has more influence on how a certain chemical demulsifier will perform than does the specific category of components included in the treating chemical. For example, a low-molecular-weight resin used in treating an emulsion of 35”API oil may exhibit rapid water drop, but that same chemical, when used in treating an emulsion of I5”API oil, may not cause rapid water drop. This illustrates that demulsifying chemicals must be compounded for each particular emulsion.

Each treating system must be tested and checked to ensure that the chemicals used for treating the water for disposal do not conflict with chemicals used for treating the oil emulsion. One chemical must not react with the other to cause problems, such as stabilizing the oil in the water. Compatibility of the two chemicals must be tested by bottle tests and then by field tests in the actual treating system. Also, compatibility tests should be performed for any other chemicals added to the produced fluids.

Selection of the optimum chemical to use usually starts with bottle tests. A representative sample of fluid is taken and transferred into several test bottles. Several demulsifying chemicals are added to the test bottles in various amounts to determine which chemical will best break the emulsion. Additional tests are made to determine the optimum ratio of chemical to fluid. Several series of tests may be necessary at various ratios and temperatures before a selection can be made. Many factors-such as the color and appearance of the oil, clarity of the water, interface quality, required operating temperature, settling time, and BS&W content-are observed during these bottle tests.

Bottle tests can be made with the samples of emulsion taken at the wellhead. anywhere in the flowline, at the manifold, or at the entrance to the treating system or tank. Well-equipped mobile laboratories are available, so this type of work can be done in the field. These mobile units should be operated by trained technicians who can minimize testing and optimize selection of chemical demulsifiers.

After the bottle tests are made and the best two or three chemicals have been selected, they should be field tested in the treating system to verify that the best chemicals have been selected. Tests should be made in the treating sys-


CHEMICAL

%-IN. COUPLING

DOUGHNUTMADEOF

DOUGHNUT. AREA OF 6 HOLES TO BE LESS THAN CSA OF %-IN. PIPE AREA %-IN. PIPE=O.19635 SQ IN

HOLE A TO BE DRILLED

8 HOLES FOR CHEMICAL D=0.177 IN.

DRILLED ON UPSTREAM USE %-IN. HOLE (0.156) FACE OF DOUGHNUT HOLES ON FAR SIDE

Fig. 19.16-Chemical distributor for flowlines 10 in. and larger

tern at various concentrations, operating temperatures, settling times, degrees of mixing, etc., before the final selection is made on the basis of performance and cost. The optimum chemical is one that will provide the clearest, cleanest separation of water from oil at the lowest temperature in the shortest time at the lowest cost per barrel treated and that will not interfere with subsequent deoiling of the water.

The required concentration of demulsifying chemical may be as high as 8 gal/l,000 bbl (about 200 ppm) or as low as 1 ga1/5,000 bbl (about 5 .O ppm). This is a range of 40 to 1. The most common range of chemical injection is between 10 and 60 ppm.

Application of heat to an emulsion after a demulsifier has been mixed with it increases effectiveness of the chemical by reducing the viscosity of the emulsion and facilitat- ing more intimate mixing of chemical with emulsion. Chemical reaction at the oil/water interface takes place at a more rapid rate at higher temperatures.

The point of injection of demulsifier chemical into the emulsion is important. The chemical should be injected into the emulsion and mixed with it so that it is evenly and intimately distributed throughout the emulsion when it is heated, coalesced, and settled in the treating system. The demulsifying chemical should be injected in a continuous stream, with the chemical volume directly proportional to the emulsion volume. Certain demulsifiers should

not be present in the emulsion during excessively prolonged agitation because the beneficial effect of the demulsifier may be spent or counteracted by the agitation and re-emulsification may occur.

Turbulence accelerates the diffusion of the demulsifier throughout the emulsion and increases the number and intensity of impacts between water droplets. Turbulence must be prolonged for a sufficient time to permit the chemical to reach the interface between the oil and all the dispersed water droplets, but the intensity and duration of the turbulence must be controlled so that it will not cause further emulsification. Turbulence is the dynamic factor that disperses the water in the oil and is a prerequisite

to emulsion formation. A moderate level of controlled turbulence, however, causes the dispersed droplets to collide and coalesce. Usually, this turbulence is provided by normal flow in surface lines, manifolds, and separators and by flow through the emulsion-treating unit or system.

One way of assisting in dispersing the chemical throughout the entire volume of emulsion is to mix a small volume of chemical with a diluenr and then to inject and mix the diluted chemical with the emulsion. The larger volume of the mixture may make it possible for the chemical to be more uniformly and intimately mixed with the emulsion.

When flow rates are low (less than 3 ft/sec) or when laminar flow is encountered, the injection of chemical into a coupling welded in the side of the pipe is not recommended. In such cases, an injection quill (which injects the chemical in the stream at a location removed from the wall), a chemical distributor (Fig. 19.16), and/or a kinetic mixer (Fig. 19.17) is recommended. The kinetic mixer consists of a series of staggered, helically convolut- ed vanes that use the velocity of the fluid to accomplish mixing.

When a tank of wet oil (oil containing more than the permissible amount of water) accumulates, the tank contents can be treated by adding a small proportion of demulsifier, agitating or circulating the tank contents, and then allowing time for the water to settle in the tank. Trailer- mounted units that include a heater, circulating pump, and chemical injector are sometimes used for this method of tank treating. This batch-treatment method normally is used as an emergency measure.

Excessive amounts of treating chemical can result in increased stability of the water-in-oil emulsion or of the oil-in-water emulsion in the produced water, increase the stability or the volume of the interfacial emulsion and/or sludge, or waste money equal to the cost of the excessive volume of chemical over the optimum volume. Also, the cost of handling and injecting the excessive amount of chemical must be considered along with the purchase cost of the chemical. Insufficient treating chemical can fail to


Fig. 19.17—Kinetic (static) mixer for mixing chemical demulsifi-er with emulsion.

break the emulsion; allow a quick buildup of excessiveamounts of emulsion and/or sludge; and result in a needfor excessive heat to break the emulsion, a need for ex-cessive settling time to resolve the emulsion, reduced ca-pacity of the treating equipment, excessive waterremaining in the crude oil causing accumulation of un-salable oil and the resultant cost of retreating the crude,or more difficulty in removing oil from the producedwater.

AgitationAgitation or turbulence is necessary to form a crude oilemulsion. When turbulence is controlled, however, it canassist in resolving the emulsion. Agitation causes in-creased collisions of dispersed particles of water and in-creases the probability that they will coalesce and settlefrom the emulsion. Caution should be exercised to pre-vent excessive agitation that will result in further emul-sification instead of resolving the emulsion. If theturbulence is kept to moderate Reynolds numbers of50,000 to 100,000, good coalescing conditions usuallyshould be achieved.

4WATER IN

S . P . P A C K T A N K

I N S T A L L A T I O N

I:=13

WATER OUT

A T M O S . VENT

OIL OUT

F R E E - F L O W C O A L E S C E N C E

Fig. 19.18—The S. P. Pack’” grows a larger drop size on theinlet separator of a gravity settler.

The flow of emulsions at moderate Reynolds numbersthrough long pipelines has been shown to cause coales-cence and develop droplets that exceed 1,000 µm in di-ameter. The length of the pipeline required to obtaincoalescence can be dramatically decreased by using a de-fined flow path as in the special flow coalescing deviceshown in Fig. 19.18.

The demulsification process may be assisted by the useof baffle plates placed inside the treating vessel. Properlydesigned and located baffle plates can evenly distributeemulsion in a vessel and cause gentle agitation that maybring about collisions of dispersed water particles to aidin coalescing the droplets. Excessive baffling should beavoided because it can cause excessive turbulence, whichmay result in increased emulsification and impede water-droplet settling. Special baffling in the form of perforat-ed plates properly placed inside treating vessels affordssurfaces upon which water droplets in the emulsion maycoalesce. As the emulsion flows through the perforations,slight agitation in the form of eddy currents is created,causing coalescence. If the perforations are too small,however, shearing of the water droplets can occur, re-sulting in a tighter emulsion.

CRUDE OIL EMULSIONS

Other designs of baffle plates provide coalescing sur-faces for the water droplets, as shown in Fig. 19.19. Flowthrough the plates is laminar, but directional changes en-able the water droplets to contact the plates and coalesce.Such a device may plug easily and become inoperablequickly.

Electrostatic CoalescingThe small water drops dispersed in the crude oil can becoalesced by subjecting the water-in-oil emulsion to ahigh-voltage electrical field. When a nonconductive liq-uid (oil) containing a dispersed conductive liquid (water)is subjected to an electrostatic field, the conductive parti-cles or droplets are caused to combine by one of threephysical phenomena.

1. The water droplets become polarized and tend toalign themselves with the lines of electric force. In sodoing, the positive and negative poles of the droplets arebrought adjacent to each other. Electrical attraction bringsthe droplets together and causes them to coalesce.

2. The water droplets are attracted to an electrode be-cause of an induced electric charge. In an AC field, be-‘cause of inertia, small droplets vibrate a greater distancethan larger droplets, promoting coalescence. In a DCfield, the droplets tend to collect on the electrodes, form-ing larger and larger droplets until eventually they settleby gravity.

3. The electric field tends to distort and thus to weakenthe film of emulsifier surrounding the water droplets.Water droplets dispersed in oil subjected to a sinusoidalalternating-current field will be elongated along the linesof force as voltage rises during the first half-cycle. Asthey are relaxed during the low-voltage portion, the sur-face tension pulls the droplets back toward sphericalshape. The same effect is obtained in the next half of thealternating cycle. The weakened film is thus more easilybroken when droplets collide.

Whatever the actual mechanism, the electrical fieldcauses the droplets to move about rapidly in random direc-tions, which increases the chances of collision with otherdroplets. When droplets collide with the proper velocity,coalescence occurs. The greater the voltage gradient, thegreater the forces causing coalescence. Experimental datashow, however, that at some voltage gradient, the waterdroplet can be pulled apart and a tighter emulsion can re-sult. For this reason, electrostatic treaters normally areequipped with a mechanism for adjusting the voltage gra-dient in the field.

If the quantity of water in the oil is large there is a ten-dency for the formation of a chain of charged water par-ticles, which may form links between the two electrodes,causing short-circuiting. This is referred to as “chaining”and has been observed in emulsions containing 4% or lesswater. The short-circuit releases a burst of electrical ener-gy that immediately causes this chain of water particlesto become steam. The resulting explosions sound like pop-ping popcorn. If chaining occurs, the voltage gradient istoo large (i.e., the electrical grids of the electrostatic treat-er are too close together or the voltage is too high) forthe amount of water being handled. Small amounts of gasbreaking out of solution may also create sufficient turbu-lence and impede the coalescing process.

Water-WashingIn some emulsion-treating vessels, separation of liquids

Fig. 19.19—Performax TM plate pack, a special coalescingmedium for crude oil emulsions.

and vapors takes place in the inlet diverter, flume, or gasboot located at the top of the vessel. The liquid flows bygravity to the bottom of the vessel through a large con-ductor pipe or conduit. A spreader plate on the lower endof the conduit spreads the emulsion into many smallstreams or rivulets that move upward through the water,accomplishing a water-wash. After the emulsion haspassed through the water-wash, it flows to the upper por-tion of the vessel, where the coalesced water droplets settleout of the oil by gravity separation.

If an emulsion is flowed through an excess of the inter-nal phase of the emulsion, the droplets of the internalphase will tend to coalesce with the excess of the internalphase and thus be removed from the continuous phase.This is the principle on which a water-wash operates. Thewater-wash is more beneficial if the emulsion has beendestabilized by addition of a demulsifier and if the wateris heated. The effectiveness of a water-wash greatly de-pends on the ability of the spreader plate or distributorto divide the emulsion into small streams or rivulets andto cause the emulsion to be in maximum intimate contactwith the water bath so that the small drops of water cancoalesce with the water.

If an-emulsion-treating system or unit uses a water-wash, it can be charged with water to facilitate initial op-eration. Water from the emulsion to be treated should beused if available. If it is not available, extraneous watermay be used.

19-14

Filtering

A filtering material with the proper size of pore spaces and the proper ratio of pore spaces to total area can be used to filter out the dispersed water droplets of a crude oil emulsion by preferentially wetting the filtering material with oil and keeping it submerged in oil. When used in this manner, the pack is correctly called a “filter” because it filters out the liquid that it prevents from passing through.

When excelsior is used as a filter in an emulsion treater, it is immersed in oil above the oil/water interface level, Excelsior is preferentially wetted by water because of the high affinity the cellulose fibers have for water. If the excelsior is initially wetted by oil, however. the dispersed water droplets in the oil will not normally take possession of the excelsior fibers because the fibers are saturated with oil. If the water droplets do take possession of the excelsior fibers, possession will occur at a slow rate and penetration of the pack by the water will be only partially complete.

Excelsior is wood that is cut into small shreds or fibers. Observed under a microscope, the surfaces of each strand of excelsior bristle with tiny sharp barbs. When emulsion flows through an excelsior pack, these rough surfaces cause distortion of the film surrounding the water droplets, thereby encouraging adherence of the droplets to the strands of excelsior. This results in coalescence of the water droplets into drops large enough to settle out of the oil. Excelsior should be made from pitch-free woods. such as aspen, cottonwood, or poplar. Pine excelsior is not recommended for crude oil emulsion-treating purposes. Excelsior should be used at less than 180°F treating temperature. Higher temperature will delignify and deteriorate the excelsior. It will also make it difficult to remove from the vessel.

Glass wool and other porous materials have been used as filtering material. Glass wool. when the fibers are properly sized and compacted, can serve as a filtering material for filtering water droplets out of a crude oil emulsion. If the glass wool is coated with silicone, its filtering effect will be enhanced because the silicone-coated fibers will be more wettable by oil than untreated glass wool fibers. Glass wool is not widely used for filtering because of its initial expense and its fouling problems.

Porous materials, both plastics and metals, are available that will filter dispersed water droplets from a crude oil emulsion. These porous materials are not widely used because of the difficulty of obtaining and maintaining the proper size pores and because they easily foul and become inoperable.

Treating crude oil emulsions by filtering is not widely used because of the difficulty in obtaining and maintaining the desired filtering effect and because the filtering material is easily plugged by foreign material.

Fibrous Packing

Fibrous coalescing packs are not commonly used in oil treating. They are mentioned for completeness and to differentiate between filtering and coalescence. A coalescing pack is a section or compartment in an emulsion- treating tank or vessel that is packed with a material that is wetted by the water, causing the water to coalesce into larger drops. Separation of two emulsified liquids by use


of a coalescing pack operates on the principle that two immiscible liquids with different surface tensions cannot simultaneously take possession of a given surface. The coalescing pack is wetted with or submerged in water. When the dispersed droplets of water come in contact with the water-wet coalescing material. the water droplets coalesce and adhere to the coalescing surfaces. Oil will pass through the pore spaces of the coalescing material. Separation of the two liquids in a coalescing pack is not caused by filtering but by the greater affinity of the water- wet coalescing material for the water droplets.

The film of oil containing the emulsifying agent surrounding the dispersed water particles must be broken before these droplets will adhere to a coalescing medium. The film is broken with the aid of demulsifying chemicals and/or heat and by repeated contact between the water particles and the surface of the coalescing materials as the emulsion flows through the pack. When this film has been broken, the water particles adhere to the surface of the coalescing material until they combine into drops large enough to settle out of the oil.

Glass wool can be used as coalescing material in emulsion-treating vessels. It will not deteriorate like wood excelsior and will prolong the service life of the coalescing pack. Glass wool fouls rather easily and may cause channeling. Woven wire mesh can also be used but tends to be more expensive than glass wool.

Gravity Settling

Gravity settling is the oldest, simplest, and most widely used method of treating crude oil emulsions. The difference in density of the oil and water causes the water to settle through and out of the oil. Because the water droplets are heavier than the volume of oil they displace, they have a downward gravitational force exerted on them. This force is resisted by a drag force caused by their downward movement through the oil. When the two forces are equal, a constant velocity is reached that can be computed from Stokes’ law as

I .78 x 10 -6(A~CIM.)d2 )> = ) . . . . . . . . .

PL, (3)

where v=

d=

AY o,,‘ =

downward velocity of the water droplet

relative to the oil, ft/sec,

diameter of the water droplet, pm. difference in specific gravity between the

oil and water, and

CL, - dynamic viscosity of the oil, cp.

Several conclusions can be drawn from this equation. 1. The larger the size of a water droplet, the greater

its downward velocity-i.e., the bigger the droplet size, the less time it takes for the droplet to settle to the bottom of the vessel, and thus the easier it is to treat the oil.

2. The greater the difference in density between the water droplet and the oil, the greater the downward velocity-i.e., the lighter the oil, the easier it is to treat the oil. If the oil gravity were lO”AP1 and the water fresh, the settling velocity would be zero because there is no gravity difference.


3. The higher the temperature, the lower the viscosity of the oil, and thus the greater the downward velocity of the water droplets-i .e., it is easier to treat the oil at high temperatures than at low temperatures (assuming a small effect on gravity difference because of increased temperature).

Gravity settling alone can be used to treat only loose, unstable emulsions. When other treating methods destabilize the emulsion and create coalescence, which increases water droplet size, however, gravity settling provides separation of water from oil.

Retention Time

In a gravity settler, such as an oil-treating tank or the coalescing section of an oil-treating vessel, coalescence will occur. Because of the small forces at work, however, the rate of contact between water droplets is small and coalescence seldom occurs immediately when two droplets collide. Thus the process of coalescence, although it will occur with time, follows a steep exponential curve where successive doubling of retention time results in small incremental increases in droplet size.

The addition of retention time alone, after some small amount necessary for initial coalescence. may not significantly affect the size of the water droplet that must be separated by gravity to meet the desired oil quality. A taller tank will increase the retention time but will not decrease the upward velocity of the oil or may not significantly increase the size of the water drop that must be separated from the oil. Thus the additional retention time gained by the taller tank may not materially affect the water content of the outlet oil.

A larger-diameter tank will increase the retention time. More important, it will slow the upward velocity of the oil and thus allow smaller droplets of water to settle out by gravity. In this case, it may not have been the increase in retention time that improved the oil quality but rather the reduction in flow velocity, which decreased the size of the water droplets that can be separated from the oil by gravity.

Centrifugation

Because of the difference in density between oil and water. centrifugal force can be used to break an emulsion and separate it into oil and water. Small centrifuges are used to determine the BS&W content of crude oil emulsion samples. A few centrifuges have been installed in the oil field to process emulsions. They have not been widely used for treating emulsions, however, because of high initial cost, high operating cost. low capacity, and a tendency to foul.

Distillation

Distillation can be used to remove water from crude oil emulsions. The water, along with lighter oil fractions, can be distilled by heating and then separated by appropriate means. The lighter oil fractions are usually returned to the crude oil.

The only current use of distillation is in the “flash system” used in 15”API and lower oil. These systems use the excess heat in the oil received from the treater or treating system and convert this sensible heat to latent heat at or near atmospheric pressure. The flashed steam is condensed in a surface condenser in the incoming cooler

stream of raw crude. thus scavenging the excess heat that would ordinarily be wasted. Fig. 19.20 shows a typical flash distillation system for dehydrating emulsions of heavy viscous crude oils.

The disadvantage of distillation is that it is expensive and that all the dissolved and suspended solids contained in the water are left in the oil when the water is removed by evaporation.

Emulsion-Treating Equipment and Systems The design of equipment or a system for treating crude oil emulsions and the sizing of each piece of equipment for a specific application requires experience and engineering judgment. It would be ideal if a procedure existed that would permit the engineer to infer from measured properties of the emulsion the most economical treating process, taking into account treating temperature, chemical usage, and physical size of treating equipment. Un- fortunately, such a procedure is not available and the engineer must rely on experience and empirical data from other wells or fields in the area and on laboratory experiments.

For example, the economic balance between the amount of chemical and heat to use to destabilize the emulsion and aid in coalescence is difficult to predict. Almost all emulsion-treating systems use demulsifying chemicals. In most instances, the lower the treating temperature, the greater the amount of chemical required to treat the emulsion. In many areas of west Texas and the Gulf of Mexico, some operators do not add heat to treat the relatively light crudes that are produced. Other operators under the same conditions add heat when treating similar crudes to minimize chemical cost and the size of the emulsion- treating equipment.

Another example is the economic balance that must be considered between those factors that promote coalescence (chemicals, water wash, heat, coalescing plates, etc.) and the size of the treating vessels. The larger the size of the treating vessel, the smaller the size of water droplets that can be separated from the emulsion. Thus the use of coalescing aids may reduce the size of the equipment by increasing the size of the water droplet that must be separated from the oil to meet the required quality. The savings in vessel costs must be balanced against the increased capital and operating cost (e.g., fuel and increased maintenance because of plugging) of the coalescing aids.

Bottle tests in the laboratory provide a means for estimating ranges of treating temperature and retention time for design purposes. Unfortunately, these tests are static in nature and do not model closely the dynamic effects of water droplet dispersion and coalescence that occur in the actual equipment because of flow through control valves, pipes, inlet diverters, baffles, and water-wash sections. Bottle tests, however, can be useful in estimating treating parameters such as temperature, demulsifier volume, settling time, etc.

When evaluating empirical data from similar wells or fields, the designer should recognize that the temperature at which an emulsion is treated may not be as critical as the viscosity of the crude at that temperature. The design of an oil-treating system can be assisted by observing an existing system, knowing the viscosity of the crude at treating temperature, and calculating from the flow gem ometry and Stokes’ law the minimum size water droplet


;c ‘FLASH

150 to 18OOF Production Inlet 15 to 30% cut TEMPERATURE OF SHELL SIDE

APPROXIMATELY 20°F

r TREATER AT 280°F

1 CONDENSATE

TOWER

Ll

220°F STORAGE

1 I

-I 1 j-+ WATER

+ TREATED CRUDE AT 2% CUT

Fig. 19.20-Typical flash distillation system for dehydrating emulsions of heavy viscous crude oils

that can be settled from the crude. A treating system can then be designed that will heat the emulsion to the temperature required to obtain the same viscosity that exists in the sample field, and then any one of the pieces of equipment or combinations thereof described in the next section can be selected and sized so that all water droplets larger than the calculated minimum diameter can be separated from the oil.

Because of the uncertainties in attempting to scale up from laboratory data or to infer designs from empirical data from similar wells or fields, a new treating system should be designed with either larger equipment or more heat input capacity than the engineer calculates to be necessary. The amount of “overdesign” to be built into the treating system depends on an assessment of the cost of the extra capacity balanced against the risk of not being able to treat the design throughput.

Description of Equipment Used in Treating Crude Oil Emulsions The characteristics of the emulsion to be treated should be understood before a treating system is selected. Several different types of equipment or systems may satisfactorily resolve an emulsion, but one particular type of equipment or system may be superior to others because of basic considerations in design, operation, initial cost, maintenance

cost, operating cost, and performance. Effort should be made to select the minimum number of pieces of equipment or the simplest design for each treating system to optimize initial and operating costs.

The combination of the various emulsion-treating methods that will provide the lowest use of chemical, lowest treating temperature, lowest loss of light hydrocarbons, lowest overall treating cost, and the best performance should be used. Experience and empirical data may guide the buyer to the optimum combination of treating methods, but field testing will be required to confirm the selection.

The following discussion describes various emulsion- treating equipment and systems. Each piece of treating equipment and each treating system affords a wide selection of the type, configuration, size, component selection, component design, and usage. Additional treating equipment can usually be added to each unit or system until the desired treating results are obtained. The design and selection of all the components of the treating system should be made at the time of initial purchase and installation. Because of the modular design of most systems, however, if the selected equipment does not perform as desired or if operating conditions change, additional features can usually be added or operating procedures altered to obtain the desired results.


WE13 NIPPLE TO MAINTAIN GAS “CUSHION” IR VESSEL

Fig. 19.21-Typical vertical FWKO

Free Water Knockouts

Where large quantities of water are produced, it usually is desirable to separate the free water before attempting to treat the emulsion. When oil and water are agitated with moderate intensity and then allowed to settle for a period of time, three distinct phases normally will form: a layer of essentially clean oil at the top with a small amount of water dispersed in the oil in very small droplets, relatively clean water (free water) at the bottom with a small amount of dispersed oil in very small droplets, and an emulsion phase in between. With time, the amount of emulsion will approach zero as coalescence occurs.

The free water is the water that separates in 3 to 10 minutes. It may contain small droplets of dispersed oil that may require treament before disposal. Equipment to do this is discussed in Chap. 15.

Free-water knockouts (FWKO’s) are designed as either horizontal or vertical pressure vessels. Fig. 19.21 is a schematic of a vertical FWKO. and Fig. 19.22 shows a horizontal FWKO. The fluid enters the vessel and flows against an inlet diverter. This sudden change in momentum causes an initial separation of liquid and gas, which will prevent the gas from disturbing the settling section of the vessel. In some designs, the separating section contains a downcomer that directs the liquid flow below the oil/water interface to aid in water-washing the emulsion.

The liquid-collecting section of the vessel provides sufficient time for the oil and emulsion to form a layer of oil at the top, while the free water settles to the bottom. When there is appreciable gas in the inlet stream, a three-phase

separator can be used as an FWKO. See Chap. 12 for a description of both vertical and horizontal three-phase separators.

Sometimes a cone-bottom vertical three-phase separator is used. This design is used if sand production is anticipated to be a major problem. The cone is normally at an angle to the horizontal of between 45 and 60”. If a cone is used. it can be the bottom head of the vessel, or for structural reasons, it can be installed internally in the vessel. In such a case. a gas-equalizing line must be installed to ensure that the vapor behind the cone is always in pressure equilibrium with the interior of the vessel. Water jets can be used to dislodge and flush the sand from the vessel.

Oil and water are usually separated more quickly and completely in an FWKO when the liquid travels through the vessel in a horizontal rather than a vertical direction. Horizontal flow permits a less restricted downward movement of the water droplets. If the emulsion flows vertically upward, the water must move downward through an upward-moving stream; therefore, the downward movement of water is retarded by upward movement of the oil and emulsion.

19-18

L.L.C. ;VITH WEIGHTED FLOAT TO SIP II: L?IL AND EMJLSION AND FLOAT ONLY I:; FREE VJATZR.


'wEIf? NIPPLE TO MAINTAIN OIL AND GAS GAS "CUSHION" IN VESSF3.a

PERFORATED WAVE I~PINGENENT

BREAKER OUTLET

FLUID IMLST

Fig. 19.22-Typical horizontal FWKO.

It is possible to add a heating tube to an FWKO, as shown in Fig. 19.23, or to add heat upstream of the FWKO. In such cases, even though the vessel may be called an FWKO, it is performing the function of an emulsion treater.

Many configurations are possible for providing baffles and maintaining levels in an FWKO. A good design will provide the functions described previously, i.e., degassing, water-washing, and providing sufficient retention time and correct flow pattern so that free water will be removed from the emulsion.

When the free water is removed, it may or may not be necessary to treat the oil further. In many fields producing light oil, a well-designed FWKO with ample settling time and with a reasonable chemical-treating program can provide pipeline-quality oil. Most often, however, further emulsion treating is required downstream of the FWKO.

Storage Tanks

Oil generally should be water-free before it is flowed into lease storage tanks. If there is only a small percentage of water in the oil and/or if the water and oil are loosely emulsified, however, it may be practical to allow the water to settle to the bottom of the oil storage tank and to draw off the water before oil shipment. This practice is not generally recommended or followed. but for small volumes of free or loosely emulsified water on small leases or for low-volume marginal wells, it may be a practical and economical procedure.

When a storage tank is used for dehydration, the oil is flowed into the tank and allowed to settle. When the tank is full of liquid, flow into the tank is stopped or

switched to another tank and the tank is allowed to remain idle while water settles out of the oil. After the water has been separated from the oil by settling, water is drained from the bottom of the tank and the oil is gauged, sampled, and pumped or drained to a truck or pipeline. No water-wash is used in conjunction with the standard storage tank. If there is a water-wash, its shallowness and the absence of a proper spreader causes it to be of little or no benefit.

Settling Tanks

Various names are given to settling tanks used to treat oil. Some of the most common are gunbarrels, wash tanks, and dehydration tanks. The design of these tanks differs in detail from field to field and company to company. All contain all or most of the basic elements shown in Fig. 19.24.

The emulsion enters a gas separation chamber or gas boot where a momentum change causes separation of gas. Gas boots can be as simple as the piece of pipe shown in Fig. 19.24, or they can contain more elaborate nozzles, packing, or baffles to help separate the gas. If there is much gas in the well stream, it is usually preferable to use a two- or three-phase separator upstream of the settling tank. In this case, the gas boot must separate only the gas that is liberated as the pressure decreases during flow from the separator to the settling tank.

A downcomer directs the emulsion below the oil/water interface to the water-wash section. On most large tanks, a spreader is used to distribute the flow over the entire cross section of the tank. This minimizes short-circuiting. The more the upward-flowing emulsion spreads out and

CRUDE OIL EMULSIONS

BAFFLE

19-19

BAFFLE (SIDE PERFORATED~

LEVEL CONTROLLER

, BAFFLE (SOLID?

Fig. 19.23--Schematic view of FWKO with heating element in each end

approaches plug (or uniform) flow, the slower its average upward velocity and the smaller the water droplets that will settle out of the emulsion.

There are many types of spreader designs. Spreaders can be made by cutting slots in plate, use of angle iron, or holes in pipe. By causing the emulsion stream to separate into many small streams, the spreader causes a more intimate contact with the water to help promote coalescence in the water-wash section. This is shown in Figs. 19.25 and 19.26. Most spreaders contain small holes or slots to divide the oil and emulsion into small streams. Large holes (3 to 4 in. in diameter) will not be nearly as effective in dividing the stream as small holes (‘/8 to 1 in. in diameter). In designing a spreader, however, it is important that the fluid is not agitated to the point where shearing of the water droplets in the emulsion takes place, causing the emulsion to become harder to separate. In addition, small holes can be more easily plugged with solids and are difficult to clean. Free-flow coalescing devices, such as S.P. Packs TM (Fig. 19.18), can be installed on the downcomer/spreader to promote coalescence and to minimize shearing of the water droplets by the spreader.

As the emulsion rises above the oil/water interface, water droplets settle from the oil countercurrent to the flow of the oil by gravity. Because there may be very little coalescence above the oil/water interface, increasing the height of the oil-settling section above some minimum to aid in spreading out the flow may not materially affect the oil outlet quality.

ADJUSTABLE IN:fPRPF”CE

GAS OUTLET

GAS SEPARATING G&S ECUALIZING

WELL PRODuCTtON

OIL SETTLING

Fig. 19.24-Typical settling tank with internal downcomer and emulsion spreader.

1 Q-20 PETROLEUM ENGINEERING HANDBOOK

rOtL SETTLING SECTION

NOTE VELOCITY OF WELL FLUID THRU -THE HOLES IN THE DISTRIBUTOR

INTO THE WATER WASH SHOULD NOT EXCEED I 0 FT PER SECOND (!I

Fig. 19.25-Proper design of well fluid inlet distributor for wash or gunbarrel tank showing use of small holes in dls- tributor.

Fig. 19.26~Improper design of well fluid inlet for wash or gunbarrel tank.

Sometimes oil collectors similar in design to oil spreaders are used to aid in establishing plug flow. The oil collector must not allow vortexing and should collect oil from the top of the tank in such a way that horizontal movement of the oil will be minimized.

Some tanks discharge the water through a water collector designed to cause the flow of water to approach plug flow conditions more nearly. The water outlet collector must prevent vortexing of the water and must minimize horizontal movement of the water. The water outlet collector should be located near the tank bottom. There must be enough vertical distance between it and the inlet spreader to allow sufficient clarification of the water, and it should be at least 6 to 12 in. above the tank bottom to allow for accumulation of sand.

Some tanks have elaborate sand-jetting and drain systems that may or may not be part of the water-collector system. It may be difficult to make these drains operate satisfactorily because the water flow to each drain must be on the order of 3 ft/sec to suspend the sand. Sand drains may lengthen the amount of time between tank cleanings, but the additional cost of sand drains in tanks may not be warranted.

In Fig. 19.24, the oil/water interface is established by an external adjustable weir sometimes called a water leg. The height of the interface is determined by the difference in height of the oil outlet and weir and the fluid properties. It may be calculated from

where hM.d = height of water-draw-off overflow nipple in

weir box above tank bottom, ft,

h 01, = height of clean oil outlet above tank

bottom, ft, h M'W = desired height of water-wash in tank above

tank bottom, ft,

Yo = specific gravity of oil, and

Y II = specific gravity of water.

Water legs are used successfully for emulsions where the gravity is above 20”API and there is sufficient difference in gravity between the oil and water. Marginal performance is obtained on oil between 15 and 20”API. Below lS”API, water legs normally are not used.

It is also common to control the oil/water interface with internal weirs or with an interface liquid-level controller and a water-dump valve. In heavy oils, electronic probes are most often used to sense the interface and operate a water-dump valve. In lighter oils, floats that sink in the oil and float in the water are more common.

Not all settling tanks contain all the sections and design details described previously. The choice depends on

the overall process selected for the facility, emulsion properties, flow rates, and desired effluent qualities. While Fig. 19.24 is representative of the majority of settling tanks currently in use, other tanks have a different flow pattern. A series of parallel vertical baffles from the bottom of the vertical tank to above the oil level, as shown in Fig. 19.27, cause the flow of the emulsion to be


horizontal rather than vertical. With this type of flow path, the water droplets fall at right angles to the oil flow, rather than countercurrent to the oil flow. Some settling tank designs employ a vortex or swirling motion at the inlet of the tank to aid in coalescence and settling and to minimize short-circuiting. Many settling tanks employ heat to aid in the treatment process. Heat can be added to the liquid by an indirect heater, a direct heater, or any type of heat exchanger.

A direct fired heater, sometimes referred to as a “jug*’ heater, is one in which the fluid to be heated comes in direct contact with the immersion-type heating tube or element of the heater. Direct fired heaters are generally used

to heat low-pressure noncorrosive liquids. These units normally are constructed so that the heating tube can be removed for cleaning, repair, or replacement.

An indirect fired heater is one in which the fluid passes through pipe coils or tubes immersed in a bath of water, oil, salt, or other heat-transfer medium that, in turn, is heated by an immersion-type heating tube similar to the one used in the direct fired heater. The contents of the bath of an indirect fired heater are caused to circulate by thermosiphonic currents. The immersion-type heating tube heats the bath, which heats the fluid flowing through coils immersed in the bath. When water is used as the bath, water free of impurities will prolong the life of the heater and prevent fouling of the surface of the heating tube and coils.

Indirect fired heaters are less likely to catch on fire than direct fired heaters and generally are used to heat corrosive or high-pressure fluids. They usually are constructed so that the heating tube and pipe coil are individually removable for cleaning and replacement. They tend to be more expensive than direct fired heaters.

Heat exchangers normally are used where waste heat is recovered from an engine, turbine, or other process stream or where fired heaters are prohibited. In complex facilities, especially offshore, a central heat-transfer system recovering waste heat and supplying it through heat exchangers to all process heat demands is sometimes more economical and may be the only way to meet established safety regulations.

Advantages of heating the entire stream of emulsion before it enters the settling tank are as follows.

1. After the fluid is heated, it flows through piping and into the flume pipe or gas boot of the gunbarrel tank. This moderate agitation of the heated fluid can assist in coalescence of water droplets.

2. The emulsion is heated before it reaches the gunbarrel. which aids in removing gas from the oil in the gas boot. This helps maintain quiescence in the settling portion of the gunbarrel.

3. The heater and gunbarrel can be sized independently, which allows flexibility in sizing the system.

4. Water-wash volume in the gunbarrel can be adjusted over a wide range, providing additional flexibility.

5. Continuous flow of fresh fluids through the heater tends to prevent coking and scaling and helps keep the heating surface clean, which will prolong the heater life.

Heat can also be supplied to the system by circulating the water in the water-wash section to a heater and back to the tank. The hot-water-wash section warms the incoming emulsion. A thermosiphon caused by density differences of the hot and the cold water can be used as the

WATER OUTLET [BELOW)

Fig. 19.27-Plan view of vertical tank with horizontal flow settling pattern.

driving force for the circulation if the heat source is not far from the tank. The water also may be pumped to the heater and circulated back through the flume, as shown in Fig. 19.28. In this system, the settling space in the gun-

barrel may be disturbed by gas released from the oil when it comes in contact with the hot water. It has two advantages. First, oil will not be overheated because it never comes in contact with the heating element in the heater but is heated by the water bath in the gunbarrel. This minimizes vapor losses from the oil and tends to maintain maximum oil gravity. It also minimizes coking and scaling. Second, this system is as safe from fire hazards as a system involving a fired vessel can be because only water flows through the heater. There is no oil or gas in the fired vessel.

Settling or gunbarrel tanks can also be heated directly with a fire tube, as shown in Fig. 19.29, or with internal heat exchangers, using steam or other heat media. Heat exchangers can be either pipe coils or plate-type heating elements.

Plate-type heating elements are usually 18 to 32 in. by about 5 to 8 ft. These usually are preferred over pipe coils because the heat-transfer coefficient is 10 to 20% higher for the plate-type heating elements when immersed in oil than for a corresponding area of pipe. Further, the gentle agitation brought about by the convection flow of the oil up the surface of the plate-type element assists in coalescence. Plate-type heating elements are available with a wide range of pressure ratings. They can be purchased for steam service or hot-water service, but the same unit should not be used for both because the construction of the cells is different for the two types of heating media.

Pipe coils are popular because of the local availability of materials. The cost is normally slighter higher than for plate-type exchangers, however, especially in larger installations.

When settling tanks are heated directly, they operate in much the same manner as vertical or horizontal emulsion treaters.


t

GAS

WELL FLUID INLET

- a-

WA+ER LEVEL -AAA--

I I I I I I

1 GUN BARREL

AUXILIARY WELL INLET

Fig. 19.29-Heater and gunbarrel in forced circulation method of heating.

P

I GAS VAPOR

AS OUTLET

,-- DEGASSER BOOT

e-INLET

:

OIL OUTLET

WATER OUTLET

Fig. 19.29-Heated gunbarrel emulsion treater.

WATER OUTLET

Vertical Emulsion Treaters

The most commonly used one-well lease emulsion treater is the vertical unit. A typical design is shown in Fig. 19.30. Flow enters near the top of the treater into a gas separation section. This section must have adequate dimensions to separate the gas from the liquid. If the treater is located downstream of a separator, this section can be very small. The gas separation section should have an inlet diverter and a mist extractor.

The liquid flows through a downcomer to the bottom portion of the treater, which serves as an FWKO and water-wash section. If the treater is located downstream of an FWKO, the bottom section can be very small. If the total wellstream is to be treated, this section should be sized for suffkient retention time to allow the free water to settle out. This will minimize the amount of fuel gas needed to heat the liquid rising through the heating section.

The oil and emulsion flows upward around the tire tubes to a coalescing section, where sufficient retention time is provided to allow the small water droplets to coalesce and to settle to the water section. Treated oil flows out the oil outlet. Any gas flashed from the oil because of heating flows through the equalizing line to the gas space above. The oil level is maintained by pneumatic or lever- operated dump valves. The oil/water interface level is controlled by an interface controller or an adjustable external water leg.

It is necessary to prevent steam from being formed on the fire tubes. This can be done by employing the “40” rule”-i.e., the operating pressure is kept equal to the

CRUDE OIL EMULSIONS

pressure of saturated steam at a temperature equal to the operating temperature plus 40°F. This is desirable because the normal full-load temperature difference between the fire tube wall and the surrounding oil is approximately 30°F in most treaters. Allowing 10°F for safety, the 40” rule will prevent flashing of steam on the wall of the heating tube.

Baffles and spreader plates may be placed in the coalescing section of the treater above the fire tubes. Many treaters were originally equipped with excelsior or “hay” packs. In most applications these may not be needed, but a manway may be provided in case one may need to be added in the field.

Although Fig. 19.30 shows a treater with a fire tube, it is also possible to use an internal heat exchanger to provide the required heat or to heat the emulsion before it enters the treater. For safety reasons, some offshore operators prefer a heat-transfer fluid and a pipe or plate heat exchanger inside the treater rather than a fire tube.

Horizontal Emulsion Treaters

For most multiwell leases, horizontal treaters normally are preferred. Fig. 19.31 shows a typical design of a horizontal treater. Flow enters the front section of the treater where gas is flashed. The liquid flows downward to near the oil/water interface where the liquid is water- washed and the free water is separated. Oil and emulsion rises past the fire tubes and flows into an oil surge chamber The oil/water interface in the inlet section of the vessel is controlled by an interface-level controller, which operates a dump valve for the free water.

The oil and emulsion flows through a spreader into the back or coalescing section of the vessel, which is fluid- packed. The spreader distributes the flow evenly throughout the length of this section. Treated oil is collected at the top through a collection device used to maintain uniform vertical flow of the oil. Coalescing water droplets fall countercurrent to the rising oil. The oil/water interface level is maintained by a level controller and dump valve for this section of the vessel.

A level control in the oil surge chamber operates a dump valve on the oil outlet line regulating the flow of oil out the top of the vessel and maintaining a liquid-packed condition in the coalescing section. Gas pressure on the oil in the surge section allows the coalescing section to be liquid-packed.

The inlet section must be sized to handle separation of the free water and heating of the oil. The coalescing section must be sized to provide adequate retention time for coalescence to take place and to allow the coalescing water droplets to settle downward countercurrent to the upward flow of the oil.

Fig. 19.32 shows another design of a horizontal emulsion treater with a different flow pattern that minimizes vertical flow of the emulsion. Oil, water, and gas enter the top of the treater at the left side (facing the burners) and travel toward the front and downward. Gas remains at the top, and oil and water are heated as required. Some heat is applied to the water in this section, but because this section has its own temperature controller, it can be regulated up or down for optimum performance.

The cross section in Fig. 19.32 shows that the emulsion flows under a longitudinal baffle and through a large slot in the partition plate near the front of the treater at

19-23

Fig. 19.30-Schematic view of typical vertical emulsion treater.

the bottom of the fire tube where it is water-washed. In the right compartment of Section AA, the oil and emulsion flow longitudinally up across the fire tube at about a 10 to 1.5” incline from horizontal.

The heating and settling section separation baffle blocks the passage of foam at the top and blocks emulsion at the bottom. Heated oil travels through a slot in a partition that is about at the centerline of the top fire tube. Free water is allowed to travel under the baffle. As emulsion accumulates at the interface, it rises to touch the tire tube, which is only 6 in. above the interface. The fire tube then tends to heat and eliminate the emulsion pad to maintain a uniform emulsion-pad thickness.

Channeling, skimming, and stratifying are all reduced by the application of louvered baffles, which are made of a stainless steel sheet punched with a louvered pattern that ranges from 15 to 60% open area. The baffles are solid at the top to prevent foaming or skimming and extend down near but do not touch the water. All the emulsion goes through the louvered openings, which provide a slight impedance to flow to develop even flow distribution and aid in coalescence.

Oil level in the treater is maintained by a weir and an oil box. Water level in the treater is critical; thus a weir is placed approximately 5 ft from the rear head seam, and the oil/water interface level upstream of this weir is maintained by the weir. Adjustment of the water-level controller, which is located downstream of this weir, has no effect on the water level in the main treater body upstream of the weir.

Because the emulsion flow path in this design is essentially horizontal, the water particles are not opposed by the velocity of upflowing oil as in a treater with a vertical flow pattern. This is especially important in heavy crudes where the differential specific gravity between oil and water is small and the settling velocity is low.


GAS EQUALIZER

EMULSION MIST EXTRACTOR INLET

\ GAS OUT

\” f, _- ,-a-OIL OUT

------- ----

- COLLECTOR

--------

+ WATER _r’ ~------

---- WATER - --I

I FIRETUBE FREE WATER

DEFLECTOR OUT

AROUND FIRETUBE

FRONT SECTION 1

ocLH%

I SPREADER WATER

OUT

COALESCING SECTION

Fig. 19.31-Typical horizontal emulsion treater with vertical flow

FIRE TUBE GAS DEMISTER AND COALESCER

/

OIL WEIR 30 DIFFUSION BAF

OIL CONTROLLER

OIL OUTLET

WATER OUTLET

BURNER EATING AND SETTLING

SECTION SEPARATION BAFFLE DAM CONTROLLER ELEVATION

HEATING CHAMBER 1 y- HEATING CHAMBER 2

FOR OIL PASSAGE SETTLING SECTION

CENTER BAFFLE - VESSEL

Fig. 19.32-Horizontal emulsion treater with horizontal flow.


GAS EQUALIZER

EMULSION MIST EXTRACTOR INLET

\ GAS OUT \ I ,--OIL OUT

---------__

c COLLECTOR

FIRETUBE

------

t WATER

e

-- -,- - WATER - - --I

I SPREADER WATER

FREE ‘WATER OUT OUT

DEFLECTOR AROUND FIRETUBE

.-----

Leff

1

i

Di

Fig. 19.33-Typical horizontal electrostatic emulsion treater with vertical flow

Other flow patterns are available if different baffle designs are used in horizontal treaters. The two described previously are examples to show the concepts that are most generally applied. Other methods of heating the emulsion can be used if it is desirable to eliminate the fire tubes.

Electrostatic Coalescing Treaters

Electrostatic treating can be used in either vertical or horizontal emulsion treaters by including electrical grids in the settling or coalescing sections. Figs. 19.33 and 19.34 show how grids can be installed in the horizontal treaters shown in Figs. 19.31 and 19.32.

Two grids of electrodes typically are installed in electrostatic emulsion treaters. One is wired to a source of electric current and the other is grounded. The emulsion flows between these electrodes, which are charged with a very high voltage. The electrodes are installed in the vessel to provide a final stage of coalescence to the emulsion after it has already been treated to near pipeline quality. In the design of Fig. 19.33, the upflowing oil passes the “hot” electric grid, which is usually steel or stainless steel rods or bars spaced 4 to 6 in. apart. This grid is stationary and hangs from multiple electric insulators. AC current is wired to this grid from an external single- phase transformer. The “cold” electric grid is mounted directly above the hot grid and is adjustable from 2.5 to 12 in. from the hot grid. The normal operating spacing between the two grids is usually 4 to 6 in.

Coalescing takes place between the oil/water interface and the hot grid, as well as between and above the grids. The oil continues vertically to the outlet collector pipe with

small calibrated holes in the top of the pipe to ensure uniform distribution. The electric section has no oil/gas interface. All gas must be removed in the heating section.

The electric coalescing function in the treater shown in Fig. 19.34 is similar to the horizontal grid unit of Fig. 19.33 except that the vertical grids provide the advantage of the horizontal flow pattern for the emulsion and improve the performance of this unit.

In addition to the safety controls normally found on emulsion treaters with fireboxes, there are also low-liquid- level safety switches on the electric treater to avoid the possibility of the electric power being applied when the high-voltage grid is surrounded with gas instead of liquid.

The greater the voltage gradient, the greater the forces causing coalescence. Experimental data show, however, that at some voltage gradient the water droplets can be pulled apart and a tighter emulsion can result. For this reason, electrostatic treaters are normally equipped with a mechanism for adjusting the voltage gradient in the field so that the optimum can be obtained.

The voltage gradient can be changed (1) by selecting optional transformer voltage taps. (2) by adjusting the voltage gradient by raising or lowering the oil/water inter face in units using horizontal grids-the water level is. in effect, a grounded electrode against which most of the coalescing takes place; or (3) by adjusting the hot or cold grid location to change the voltage gradient.

The transformer is normally an 18.000- to 20,000-V secondary, single-phase, oil-tilled. 100%reactance-type transformer. It is mounted on the top, side, or end of the treater with a short, high-voltage conduit connected to an


Fig. 19.34-Typical horizontal electrostatic emulsion treater with vertical electric grids for horizontal flow of fluids

appropriate entrance probe assembly. The high-voltage line is entirely submerged in transformer oil, which is normally a highly refined hydrocarbon that has been vacuum dried and contains no moisture.

The application of electrostatic treaters should be limited to “polishing” of oil to avoid chaining and short-circuiting of the grids. They are particularly effective in reducing the water content of oil to very low levels (less than 0.5 to 1 .O%). Electrostatic coalescence may also aid in reducing heat and/or chemicals required to treat crude oil to a specific quality.

Desalting Crude Oil

Most produced water contains salts, which may cause problems in production and refining processes when the solids precipitate to form scale on heaters, plug exchangers, etc. This can cause accelerated corrosion in piping and equipment.

In almost all cases, the salt content of crude oil consists of salt dissolved in small droplets of water that are dispersed in the crude. In some instances, the produced oil can contain crystalline salt, which forms because of changes in pressure and temperature as the fluid flows up the wellbore and through the production equipment. Crystalline salt will flow out with the water and is not of importance in desalting operations.

The salinity of produced brine varies widely, but most produced water falls in the range of 15,000 to 130,000 ppm of equivalent NaCl. Crude oil containing only 1 .O% water with a 15,000 ppm salt content will have 55 lbm salt/l ,000 bbl of water-free crude. The chemical composition of these salts varies, but the major portion is nearly always NaCl with lesser amounts of calcium and

magnesium chloride. Because of the operational problems associated with salts, most refineries buy crude at a salt content of 10 to 20 lbm/l,OOO bbl, then desalt the oil to 1 to 5 lbm/l,OOO bbl before charging to crude stills.

The purpose of a desalting system is to reduce the salt content of the treated oil to acceptable levels. In cases where the salinity of the produced brine is not too great, salt content can be reduced by merely ensuring a low fraction of water in the oil. In this case, the terms desalting and emulsion treating are identical, and the concepts and equipment described previously can be used.

The required maximum concentration of water in oil to meet a known salt specification can be derived from

c,, =0.35c,,y,“f&., . . . . . . . . . . . . (5)

where C,, = salt content of the oil, lbm/l,OOO bbl, c SM’ = concentration of salt in produced water,

Pw

Y w = specific gravity of produced water, and

fW = volume fraction of water in crude oil.

If the produced brine has a high salt concentration, it

may not be possible to treat the oil to a low enough water content (less than 0.2 to 0.25% is difficult to guarantee). In such a case, desalting implies the mixing of low-salt- content water with the emulsion before treating, as shown in Fig. 19.35. This lowers the effective value of C,, in Eq. 5. If a single-stage desalting system will require too much dilution water, then a two-stage system, such as that shown in Fig. 19.36, is used.

CRUDE OIL EMULSIONS

MIXER

19-27

OIL -a

OIL TREATER ) CLEAN STREAM OIL

DILUTION WATER WATER TO DISPOSAL

Fig. 19.35-Single-stagedesalting with dilution water injection.

Although it is possible to desalt with most of the emulsion-treating equipment discussed previously, most desalting systems use electrostatic treaters to obtain the lowest possible water content in the oil and thus minimize the amount of dilution water needed.

One of the most important parts of desalting systems is the method and efficiency of the method of mixing the wash water with the crude. The smaller the diameter of the wash-water drops dispersed in the oil, the greater the possibility of their coming in contact and coalescing with entrained saltwater droplets.

Excessive agitation when mixing the wash water with the crude oil can result in emulsions that are too tight (stable) to resolve easily. Therefore, the amount of mixing provided should be adjustable to zero. This requirement tends to make pumps and level-control valves poor choices for mixing. The most commonly used mixing system consists of some type of special mixing tee or static mixer followed by a globe-type mixing valve.

Mixing efficiency in a desalting system refers to the fraction of wash water that actually mixes with the produced water. The remainder of the water, in effect, bypasses the desalting stage and is disposed of as free water. A mixing efficiency of 70 to 85% can be considered a reasonable range of attainment. Part of the energy for mixing is obtained from the pressure of the wash water, which should enter the mixer at approximately 25 psi above the pressure in the vessel.

MIXER 1

Reverse Emulsions Most emulsions are the water-in-oil type; they occur much more frequently than the oil-in-water type. Oil-in-water (reverse) emulsions are most likely to be produced where the WOR is high, the dissolved solids content of the water is low, the water is slightly alkaline, and the oil has a naphthenic base. The oil content of these emulsions may vary from as low as a few ppm to 40%. They may vary in consistency from watery thin to a moderately heavy cream.

The produced water from some leases and ballast water from some oil tankers contain sufficient oil to be or have the characteristics of an oil-in-water emulsion Such water is usually treated with chemicals formulated for water treating and with equipment described in Chap. 15.

Reverse emulsions may not require much, if any, heat. Because the external phase is water, the viscosity is quite low at ambient temperature. The chemicals used to treat reverse emulsions are usually some type of surface-active compounds that will neutralize the charges on the oil particles and allow them to coalesce during the gentle agitation that should follow introduction and mixing of the chemical with the emulsion. Overtreatment of this type of emulsion with chemical can result in stabilization rather than breaking of the emulsion.

Empirical data and experience are required to design equipment and/or systems for resolving reverse emulsions.

MIXER 2

WATER DILiTION

D&AL WATER

OIL STREAM -a OIL TREATER 7 OIL TREATER T

CLEAN OIL

RECiLE PUMP

Fig. 19.36-Two-stage desalting using second-stage recycle.


Treating Emulsions Produced From EOR Projects

Standard emulsion-treating procedures, equipment, and systems used during primary and secondary oil production may not be adequate to treat the emulsions encountered in EOR projects. EOR methods of oil production- such as in-situ combustion and steam, COz, caustic. polymer, and micellar (surfactantipolymer) floods-may result in the production of emulsions that may not respond to treatment normally used in primary and secondary oil production operations.

The treatment of the emulsions from EOR projects is usually handled independent of the primary and secondary emulsions from the same fields. Emulsion-treating procedures, equipment, and systems have been and are continuing to be developed for use in these EOR projects.

Clarification of Water Produced with Emulsions

Even though a normal (water-in-oil) emulsion exists in the oil production system, when produced water is separated from crude oil, the water usually contains small quantities of oil. The oil has been divided into small particles and dispersed in the water by agitation and turbulence caused by flow in the formation; into the wellbore; through the bottomhole pump, standing valve, traveling valve, and tubing; reciprocation of sucker rods; flow through the wellhead choke, flowline, manifold, oil and gas separator, and treating system; and by surface transfer pumps.

These small particles of oil will be suspended in the water and held there by mechanical. chemical. and electrical forces. The amount of oil contained in the untreated produced water in most systems will vary from an average low of about 5.0 ppm to an average high of about 2,000 ppm. In some water systems, oil contents as high as 20.000 ppm (2.0%) have been observed.

The oil particles in the untreated produced water will usually vary in size from 1 to about 1,000 pm, with most of the oil particles ranging between 5 and 50 pm in diameter.

Nine methods can be used to remove oil from the produced water: chemical, heat, gravity settling (skim tanks, API separators, etc.), coalescence (plate, pipe/free flow), tilted plate (corrugated) interceptors, flotation. flocculation. filtering, and combinations of the above. Refer to Chap. 15 for a discussion of the details of deoiling the produced water.

Operational Considerations for Emulsion-Treating Equipment Burners and Fire Tubes

The design of burners and fire tubes is of importance because of the high cost of fuel and the operating problems that can occur when they malfunction. The burner should be designed to provide a flame that does not impinge on the walls of the fire tube, but that is almost as long as and concentric with the fire tube. If the flame touches the fire-tube wall. hot spots can develop, which can lead to premature failure.

Burners should not be allowed to cycle off and on frc- quently because thermal stresses caused by temperature reversals can damage the firebox. The combustion con-

trols should be accessible and designed so that the operator can easily adjust the air and gas to achieve optimum tlame pattern and peak combustion efficiency.

A reliable pilot burner is required. Many operators and some regulatory agencies require burner safety shutdown valves that will shut off fuel to the burner in case of pilot failure. Unless specifically requested by the purchaser, most small emulsion treaters normally will not include this feature.

API RP 14C, “Analysis, Design, Installation and Test- ing of Basic Surface Safety Systems for Offshore Produc- tion Platforms,” contains a basic description of recommended safety devices needed for fired- and exhaust-heated units. Consideration should be given to installing these devices on onshore, as well as offshore, fired treaters. They include process high-temperature shutdown. burner exhaust high-temperature shutdown, low- flow devices and check valves for heat exchangers, high- and low-pressure shutdown sensors, pressure-relief valves, flame arresters, fan motor starter interlocks on forced draft burners, etc.

Every gas-fired crude oil heating unit should be provided with fuel gas from which liquids have been “scrubbed.” In large facilities, this can be accomplished with a central fuel-gas scrubber or filter providing fuel gas to all fired units. Many small facilities are equipped with individual fuel-gas scrubber vessels for each fired unit. These fuel-gas scrubbers are typically 8 to 12 in. in diameter and 2 to 4 ft tall, and contain a float-operated shutoff valve. If liquid enters the fuel-gas scrubber. the float will close a valve and stop gas flow to the burners of the heating unit. This will prevent oil from entering the combustion chamber and possibly prevent a fire.

Most fire tubes that transfer heat to crude oil or emulsion are sized to transfer 7,500 Btuihr-sq ft. although some manufacturers use heat-transfer rates as high as 10,000 Btuihr-sq ft. Fire tubes that transfer heat to the water- wash section of a treater, as in a vertical treater. are sized for 10,000 Btuihr-sq ft, although some manufacturers use heat-transfer rates as high as 15,000 Btu/hr-sq ft. These higher rates are not recommended because they can be overly optimistic and thus may undersize the required fire tube area.

The temperature controller. fuel control valve, pilot burner, main burner, combustion safety controls. and fuel- gas scrubber for controlling and cleaning the fuel gas for fired treating vessels should be inspected and cleaned periodically as required. A schedule of preventive maintenance is recommended for this equipment.

Deposits of soot, carbon, sulfur, and other solids, if any, should be removed from the combustion space periodically to prevent reduction in heating capacity and loss of combustion efficiency. On oil-fired units, the following items should be inspected and maintained periodically: combustion controls, burner nozzles, combustion refrac- tory, air/fuel control linkages. oil pump, oil preheater, pressure and temperature gauges, and 02 and/or CO2 analyzers.

Cleaning Vessels

Crude oil emulsions may contain mud, silt, sand, salts, asphalt. paraffin, and other impurities produced in conjunction with crude oil and accompanying water. In most


instances, these impurities are present in small quantities and add little to the treating problem. However, the treating problem may be made difficult and expensive because of the presence of one or more of these impurities in appreciable quantity. Special equipment and techniques may be required to handle these materials.

It is good practice to equip all treating vessels with cleanout openings and/or washout connections so that the vessels can be drained and cleaned periodically. Larger vessels should be equipped with manways to facilitate cleaning them. Steam cleaning may be required periodically. Acidizing may be required to remove calcium car- bonate or similar deposits that cannot be removed by hot water or by steam cleaning.

One of the most likely causes of difficulty in operating fired emulsion-treating vessels is the deposition of solids on heating tubes and nearby surfaces. It is desirable to prevent such deposits, but if they cannot be prevented, these surfaces should be cleaned periodically. The deposits insulate the heating tube, reducing heating capacity and efficiency. Also, these materials may cause accelerated corrosion.

Of the salts commonly found in oilfield waters, the chlorides, sulfates, and bicarbonates of sodium, calcium, and magnesium are predominant. The most prevalent of the chlorides is NaCl. Calcium and magnesium chloride are next in quantity. These salts can be found in practically all water associated with crude oil. Salts are seldom found in the crude oil, but if they are present, they are mechanically suspended and not dissolved in it.

Emulsion-heating equipment is particularly susceptible to scaling and coking. These processes of deposition are not distinctly separate but may occur simultaneously. Also, one may hasten the other.

Calcium and magnesium carbonates and calcium and strontium sulfates are readily precipitated on heating surfaces in emulsion-treating equipment by decomposition of their bicarbonates and the resultant reduced solubility in the water carrying them. These materials will be deposited in pipes, tubes, fittings, and the inside surface of treating vessels. Maximum deposition will occur at the hottest surfaces, such as on heating coils and fire tubes.

Scale deposition also may occur when pressure on the fluid is reduced. This is the result of release of CO* from the bicarbonates in salt water to form insoluble salts that tend to adhere to surfaces of equipment containing the fluid.

Coke is not generally a primary fouling material. When deposits of salt, scale or any other fouling material build up, however, coking begins as soon as the insulating effect of the fouling material causes the skin temperature of the heating surface (heating tube or element) to reach 600 to 650°F. At this temperature, coke begins to form, which further aggravates fouling and reduces heat transfer. Once coking starts, a burnout of the lirebox may follow quickly.

In areas where fluids cause considerable scaling or coking, the amount of such deposits can be reduced to a minimum by decreasing the treating temperature or by use of chemical inhibitors, properly designed spreader plates, and favorable fluid velocities through the equipment. Ar- ranging the internals of the equipment so that all surfaces are as smooth and continuous as possible will also reduce such deposits. The operator should periodically inspect the equipment internally and clean the surfaces as required

if trouble-free operation is to be obtained over a long period of time. It is impossible to eliminate the deposition of solids entirely in emulsion-treating equipment, but it can be minimized.

Removing Sand and Other Settled Solids

Sand and silt may be produced with many crude oils. They may settle out in the vessel and be difficult to remove. It is common to shut down and drain the vessel periodically for cleaning. Sand can be removed from the unit with rakes and shovels or with a vacuum truck. The use of “sand pans,” automated water jets, and drain systems can eliminate or minimize the problem of sand and silt in emulsion-treating vessels, but it is very difficult to eliminate sand buildup in large-diameter tanks.

Sand pan is the name given to a special perforated or slotted box or enclosure located in the bottom portion of a vessel or tank. Sand pans are designed to cover the area of the vessel that the flow of discharging water will clean. Often they are designed to work in conjunction with a set of water jets. The sand pans for horizontal vessels usually consist of elongated, inverted V-shaped troughs that are located parallel with and on the bottom of the vessels and that straddle the vertical centerline of the vessel. In the design in Fig. 19.37, the sand pans have sides that make an angle of 60” with the horizontal. The bottom edges of the sloping sides are serrated with 2-in. V-shaped slots and are welded to the interior of the shell of the treater. Most sand pans used in horizontal vessels are 5 ft long; a 60-ft-long horizontal vessel will typically have 11 sand pans and 11 sand-dump valves.

Sand pans, without a water jet system, have satisfactorily removed sand from most horizontal vessels up to 6 or 8 ft in diameter. Horizontal vessels larger than 4 ft in diameter should be equipped with a water jet system in addition to sand pans to keep sand cleaned from the vessel. Typical sand pans with a water jet system are illustrated in Fig. 19.37.

In vertical vessels, the sand pan may be a flanged and dished head approximately one-third the diameter of the vessel in which it is concentrically located. The sand pan is usually serrated around the periphery where it is welded to the bottom head of the vertical vessel concentric with the water outlet.

Water jets usually are designed to flow approximately 3.0 galimin of water through each jet with a differential pressure of 30 psi. Standard jets are available for this service that have a 60” flat fan jet pattern. The jets are usually spaced on 12- or 16-in. centers. The water jet header is U-shaped so that the vessel is cleaned on both sides of the sand pan simultaneously. The water jets can be programmed for all the jets to flow at the same time, or they can be controlled by operation of the water jets and the sand dump valves in sequential cycles.

One problem in removing sand from vessels is that very few, if any, water-discharge control valves can withstand the abuse of sand-cutting during the water-discharge period for the long term. The partial answer to this problem is to arrange the instrumentation to open and close the water-discharge control valve on clean sand-free water and to use special slurry-type valves.

The most sophisticated sand-removal systems use programmable logic controllers. This solves the problem of selection of the proper time intervals between dumps


Fig. 19.37-Sand pans and water jet system in a horizontal vessel.

and automatically controls the length of the water/sand discharge. The timing must be coordinated with the water jet system and the normal water-dump controller. A properly designed sand-removal system with proper water jetting and water/sand dumping can operate for many years without the need for a shutdown to clean out the sand or to repair or replace the dump valve.

Most emulsion-treating systems that handle large volumes of sand should not rely on hand or non- programmed operation for removal of the sand. If the operator fails to activate the dump valve often enough, the sand will cover the sand pans and plug or partially plug the water outlet, and the drains will become inoperative. With sand pans in the treater but without a programmer, large volumes of sand will usually cause trouble by plugging or partially plugging the water outlets and/or by cutting or wearing the drain valve.

Because both the amount and type of sand vary greatly, the length and frequency of the water-jetting and dumping cycles must vary to suit local conditions. Most of the coarser sand will settle out in the inlet end of the treater; the fine sand will settle out near the outlet end of the treater. It may be necessary to cycle the water jets and drain valves near the inlet end of the treater three to four times more frequently than those near the outlet end of the unit. Many timers are set for 30 minutes between jetting and dumping cycles and for 20- to 60-second jetting and dumping periods.

Interfacial Buildup

Interfacial buildup, sometimes referred to as sludge, is material that may collect at or near the oil/water interface of emulsion-treating tanks and vessels. Interfacial

buildup may contain paraffin, asphaltenes, bitumen, water, sand, silt, salt, carbonates, oxides, sulfides, and other impurities mixed with the emulsion. It can be rem moved from the vessel through a drain installed at the interface. The most common procedure, however, is to close the water dump valve and float it out to a bad-oil tank for further processing or disposal. Interfacial buildup can also be discharged with the water by opening the water-drain valve.

Corrosion

Emulsion-treating equipment that handles corrosive fluids should periodically be inspected internally to determine whether remedial work is required. Extreme cases of corrosion may require a reduction in the working pressure of the vessel or repair or replacement of vessel and piping. Periodic ultrasonic tests can measure the wall thickness of vessels and piping to detect the existence and extent of corrosion.

Corrosion of emulsion-treating equipment is usually mitigated or controlled by a combination of the following.

Exclusion of Oxygen. Corrosion rates in most oilfield applications can be kept low if O2 is excluded from the system. Care must be taken in the process design to install and maintain gas blankets on all tanks in the process and to exclude rainwater from the system. Recycled water from sump systems and storage tanks is a prime source for 02 entry into the process.

Corrosion Inhibitors. An inhibitor is a material that, when added in small amounts to an environment potentially corrosive to a metal or alloy, effectively reduces


the corrosion rate by diminishing the tendency of the metal Level Controllers and Gauges or alloy to react with the solution. Inhibitors, in the form of liquid solutions or compounds, can be injected into the

A wide selection of liquid-level controllers is available

flow stream in the flowline, manifold. or production sys- for liquid/gas control and for oil/water interface control

tern to inhibit corrosion that would occur otherwise. in light crude oil (above 20”APl) systems. For interfacial controllers in light crude oils. floats that sink in the

Cathodic Protection. Sacrificial galvanic anodes are commonly used for cathodic protection. They are made of a metal that will provide sacrificial protection to the steel vessel because of its relative position in the galvanic series. Most galvanic anodes used in emulsion treaters are about 3 to 6 in. in diameter and about 3 to 4 ft long. Mul- tiple anodes usually are installed in each vessel. The anodes usually are sized to last from 10 to 20 years. They are considered expendable and are always installed in the vessel through a flange or quick coupling so that they can be easily replaced when expended.

The galvanic anodes must be installed so that they are immersed in the water, which serves as an electrolyte. They will not protect the treater if they are immersed in the oil. The anodes must “see” all metal surfaces that are to be protected-i.e., there must be no obstructions between the anodes and the surface they are to protect. Each anode should be located as near the center of the compartment or area they are to protect as practical.

An impressed electric-current cathodic protection sys- tern can also be used to inhibit corrosion. It is a direct electric current supplied by a device using a power source external to the electrode system. The DC current can be obtained by rectifying AC current. When resistivity ol the electrolyte ranges from 25 to 300 Q.crn and above. consideration should be given to the use of an impressed current system. Impressed current systems are difficult and costly to maintain. however. and usually require skilled technicians.

Electrical current density requirements for cathodic protection of emulsion-treating vessels usually range from 5 to 40 mA/sq ft of bare water-immersed steel.

Internal Coating. Emulsion-treating vessels can be coated internally for protection from corrosion. It is important that the internal surfaces and the welds of the vessels be properly prepared to receive the coating. A coating system must be selected that will withstand the physical and chemical environment to which it will be exposed. Coating specifications. application techniques. and final inspcc- tion are very important considerations. Most coating systems will contain some holidays (breaks in the coating) or may be damaged in shipping or installation. Therefore. coating alone should not be relied on to prevent corrosion.

Steel tanks can be galvanized or lined in the field with fiberglass or other coatings. Some operators use fiberglass tanks in their emulsion-treating process, while others feel that this represents an unnecessary fire hazard.

Special Metallurgy. In particularly severe environments, such as where large quantities of CO: are cxpccted and where O1 cannot be practically eliminated from the system. it is possible to minimize corrosion by using stainless steel vessels or an internal stainless cladding in carbon steel vessels. In most low-pressure applications, stainless vessels are less expensive than clad vessels. It may also bc cheaper to use a stainless vessel than one that is internally coated because of the labor required to prepare and to coat the internal surfaces of the vessel.

oil but float in the Later normally are used.

For heavy crude oils, electronic interface controIIers have been very successful. These operate on the princi- plc of the differences between oil and water electrical con- ductivity, electrical capacitance, or radio frequency. The most common type is called “capacitance probes”; they USC the dielectric strength of the fluid in which they arc immersed.

Standard gauge glasses (reflex or transparent) are used on 20”API crude oil and higher. Reflex gauges normally are used on liquid/g% levels and transparent gauges for oil/water levels. Armored gauge glasses normally are used on pressure vcsscls and tubular gauge glasses on tanks. Some operators use the tubular gauge glasses on low pressure treating equipment. Tubular gauge glasses nor mally are furnished on standard low-pressure vessels unless armored gauges are specified.

For API gravities below 20”APl, gauge glasses are not recommended. particularly for interfacial service. because they are difficult or impossible to read. In lieu of gauge glasses. a system of sample valves is used with the sample lines all piped to a single point just above a sample box. Generally, the lines are insulated to keep them warm. A nameplate is clamped on each sample valve to dcsig- nate the elevation it represents in the treater. The sample box is fitted with a drain line piped to the sump.

Water-in-Oil Detectors (BS&W Monitors)

Several companies manufacture devices for detecting and measuring the water content of crude oil. They are commonly called BS&W monitors. BS&W monitors are typically analog instruments that measure dielectric strength and are specifically designed for determination of the water content of crude oil containing a low percent of water. They do not operate satisfactorily on streams containing free water. The unit provides a water reading car responding to the water content of the oil. It can be made to alarm, record. and control if the detected percentage exceeds the field-selectable preset limit of BS&W content.

Special Safety Features for Electrostatic Treaters

Because of the high voltage and the associated potential hazard to personnel that can result from entering a drained vessel with the grids activated, electrostatic treaters require a positive shutdown switch for the high-voltage transformer. This disconnects the transformer if the fluid lcvcl falls below a predetermined level in the treating vessel. Some manufacturers install an internal grounding device inside the treater that grounds the hot grid if fluid is not present.

Changing Excelsior Packs

Excelsior used in treating emulsions may have a serviceable lift ranging from just a few days to several years. The best grade excelsior should be used because cost of the excelsior is small compared to the expense of removing and replacing it. In some fields. the excelsior must be chopped into blocks with an ax and removed in chunks.


Serviceable life of the excelsior may be extended by periodically washing it with hot water. The water level can be allowed to build up above the top of the excelsior pack and the water heated to about 2 10°F. The temperature should be kept this high for about I hour. The water should be drained quickly from the vessel while it is hot. The hot water may carry most of the foreign material with

it. If the application of heat only partially cleans the excelsior, a second heating may clean it further. Care should be taken to use water that will not deposit solids in the treater while it is being heated.

Economics of Treating Crude Oil Emulsions The object of operating oil-producing properties is to deliver consistently the maximum volume of highest-API- gravity oil to the pipeline at the lowest possible cost. Emulsions should be prevented wherever feasible and treated at the lowest cost where they cannot be prevented.

Implementation of the following directives can minimize the occurrence of emulsions and the costs of treating.

1. Eliminate production of water with oil where possible and practical.

2. Minimize the investment in emulsion-treating equipment by studying the treating problem and the selection and use of appropriate treating methods, equipment, and procedures. The emulsion treating system should be as small as possible, yet capable of adequately handling treating requirements on the lease. Treating systems may be initially oversized to allow for future development. lease expansion, or increased water production. Such antici- pation of future needs should be considered when treating equipment is purchased; however, needlessly oversized systems involve unnecessary expense and accomplish nothing that properly sized systems will not accomplish.

3. Minimize the loss of oil with water and salvage oil from interfacial sludge and tank bottoms where feasible. Oil may be discharged with the water as it flows from FWKO’s, emulsion treaters, gunbarrels, or other treating vessels. The fraction of oil is low and the oil is usually dispersed in small droplets. Sometimes this oil is pumped

along with the water to disposal wells or delivered to operators of water disposal companies without recovery or without credit being received by the lease. This oil loss can be minimized by maintaining proper operating variables with adequately sized and maintained vessels and controls and by properly designed water treating systems.

4. Minimize chemical treating costs by use of the most appropriate chemical demulsifier compound(s), the optimum quantity of chemical, the proper location and method of injection of chemical, the proper means of intimately mixing chemical with emulsion, and the proper use of heat. Treating chemicals are not recoverable and constitute a continuing expense. Some crude oil can be adequately treated by chemical injection used in conjunction with coalescing and/or settling without heat. However, some emulsions require an increased temperature during the coalescing and settling period. A proper balance of chemical and heat aids in providing the most economical and efficient treating system. The chemicals must be intimately mixed with emulsion so that a minimum amount of chemical will provide maximum benefit. Chemicals may be wasted by being injected into the oil in large slugs and not intimately mixed with the emulsions.

5. Ensure that chemicals added to the produced fluids are compatible. Some corrosion inhibitors can cause emulsions or affect the action of oil-treating chemicals. Chem- icals used in produced-water-treating systems may be recycled to the oil-treating system with the skimmed oil and cause emulsion-treating problems.

6. Conserve gravity and volume of oil by using optimum treating temperature, cooling oil before discharging to storage, discharging vent gases from treating vessels through cooler oil in stock tanks. maintaining slight gas pressure on treating system and storage tanks, and using vapor recovery equipment on vessels and tanks. Crude oil emulsions should be resolved at the lowest effective temperature. Excessive heat drives condensible vapors from the oil, and they are discharged with the gas. Loss of these light ends lowers the API gravity of the oil and simultaneously reduces the oil volume. A further disadvantage of overheating is the increased maintenance rc- quired on treating systems caused by hot spots. salt deposition, scaling, and increased rate of corrosion, especially of the fireboxes.

7. Use all treating equipment to the best advantage. By careful observations, supervision, and record keeping, emulsion-treating equipment can be used to maximum efficiency, Transfer of equipment and alterations or additions to the treating system may be made to use existing equipment better. Constant testing and vigilance are required to obtain maximum results.

8. Practice preventive maintenance to minimize ir- retrievable loss of oil production because of downtime for equipment repairs. The more complex the treating system, the greater will be the possibility of mechanical failure. Oversized and overly complex systems have a greater failure frequency than more appropriately designed, simpler, and more compact systems.

9. Exchange information on treating methods and results among company personnel, with other operators, engineering firms, vendors, and chemical-treating companies. Experience can be gained and shared by personnel responsible for handling treating problems, which will result in lower treating costs.

Cost records are important in oil emulsion-treating operations. To achieve optimum operation of emulsion- treating equipment at minimum cost, proper records must be kept of operating temperatures, pressures, fuel and/or power consumption, chemical usage, performance, etc.

Such records should be kept on a daily, weekly, or monthly basis, reviewed regularly, and kept available for supervisory personnel.

Cost records make it possible to predict whether an existing system should be modified or replaced. Justifica- tion for modifying or replacing an existing system will depend on the efficiency of the system, and this can be determined best from accurate and reliable cost and performance records. Cost records on existing methods or systems will assist in determining the type, kind, and size of treating systems for new leases. Treating cost records should make it possible to determine current operating costs, probable installation costs for new systems, and probable future operating costs of similar systems.

Each operator should determine what is to be charged to emulsion-treating costs. Listed in Table 19.2 is an out-

line of items that may be considered part of the data base for a cost-accounting system. Some of these items will


not be applicable to all treating systems, and some operators may elect to group some of the categories. A comprehensive general listing is presented for those who wish to consider all items of cost. Special systems and conditions may require additional cost items.

It is necessary to consider all factors listed in Table 19.2 if complete and accurate treating cost records are to be compiled. This information should be obtained and recorded on a continuous daily, weekly, or monthly basis, and it must be accurate and concise if cost records are to be of maximum value.

Investment costs must include the initial cost of all equipment used, including the cost of transporting it to the location, of erecting and installing the equipment, and of readying the system for operation. Such items as pipe and pipe fittings, valves, grade work, foundations, and fencing should be included.

The labor cost should include supervisory personnel, cost of company gang and contract gang, and other labor required to obtain, install, and put the treating system into operation. Operating cost should be kept separate from maintenance cost and should include such items as supervisory labor, operating labor, chemicals, fuel, and miscellaneous supplies. Maintenance cost should include cost of maintaining and repairing all treating equipment. This should include such items as cleaning. repairing, painting, and other similar items.

The Overall System Performance section should include an accurate record of the volume of oil treated and the volume of water separated, treated, and handled. This part of the record should also include reference to troubles experienced with the system and a commentary on day-today performance of the unit or system.

Key Equations in Metric Units

Q=53.09aT(0.5y,,y,,+y,,.y,,.), . . (2)

where Q is in watts. ilT is in Kelvin. y,, is in m’ld. y,, is dimensionless, (I,,. is in m j/d. and yII. is dimensionless.

5.43x 10 -‘o(Ay(,\,)d2 ,‘Z . (3)

CL il

where 1’ is in m/s.

AY l,il is dimensionless. d is in pm. and p,, is in Pa.s.

TABLE 19.2-COSTS OF EMULSION TREATING

Equipment Investment Costs

Material Treating equipment and facilities Auxiliary equipment, controls, and accessories

Labor to install equipment Company labor Contract labor

Other expenses Surface rights Special services Other

Operatmg Costs

Material Chemicals Chemical injection equipment Diluents and solvents Testing apparatus Depreciation allowances Other

Labor Supervisory Operator or pumper Steaming or cleaning crew Gang Contract Other

Other expenses Equipment rental Fuel (gas or liquid) Electric current Transportation Laboratory expenses Other

Maintenance Costs

Material Maintenance Replacements and additions Repalrs

Labor Supervisory Pumper or operator Mechanic Painting Gang Contract Other

Other Expenses Equipment rental Transportation Other

Overall System Performance

Volume of oil treated Water content of treated oil Volume of water produced Volume of water treated Volume of oil salvaged from water treatment Remarks or observations

Bmdxd.. P.L. and Bcshlcr. D.U.: “Cold Trealing of Oilfield Einul- \ionr,” presented ;~t the Southwehlern Petroleun~ Short Courw. Dcpr. ot Purroleu~n Englncerinf. Tcx;r\ Tech U.. Luhboch. April 1975


Bechcr. P : Priwiples of’ Emulsion Twh~toio~y, Reinhold Pubhhhing Corp.. New York City (1955).

Besaler. D.U.’ “Demulsification of Enhanced Oil Recovery Produced Fluids,” Tretolite Div.-Petrolite Corp., St. Louis, MO (March 23. 1983).

Blair. C.M.. “Handling the Emulsion Problem in the 011 Fields.” Magna Corp.. Santa Fe Sprmgs, CA (Dec. 6, 1971).

Blair, C.M.: “lnterfdclal Films Affecting the Stability of Petroleum Emulsions,” Chemi.w,v and Indu.stryv (1960) 538-44.

Breuking Emdsions by Chemical Technology-Thuorws of Emu/;rmn Erwkin#, Technology Series CTS-V3, Nalco Chemical Co., Sugarland. TX (1983).

Coppel. C.P.: “Factors Causing Emulsion Upsets in Surface Facllitles Following Acid Stimulation,” JPT (Sept. 1975) 1060-66.

“Corrosion Control.” Corrmion. L.L. Shreir (ed.), John Wiley and Sons Inc.. New York City (1963) 2, 18.12.

Cwwsion Cwzrwl iu Prrroleum F’n~~wtion. NACE TPC5. Item 5 I 103. Natl. Assn. of Corrosion Engineers, Houston (1979).

Corrtnio/~ Inhrhirors. Item 5 1073. Natl. Assn. of Corrosion Engineers, Houston (1973).

CO, Corrosion in Oil und Gas Production: S&wed Pup~rs, Abstrcrcts, and Refprewes. Item 51 120. Natl. Asan. of Corrosion Engineers. Houston (1984).

“Demulsification.” Tretolite Div.-Petrollte Corp.. St. Louis. MO (1975).

“Design, Installation. Operation. and Maintenance of Internal Cathod- ic Protection Systems in Oil Treating Vessels,” NACE Standard RP-05-75, Natl. Assn. ofcorrosion Engineers, Houston (Oct. 1975).

Fontaine, E.T.: “Oilfield Brine Vessels-Cathodic Protection for Brine Handling Equipment.” Murrriuls Prorecrion (March 1967) 6, No. 3, 41-44.

H, S Cormsiorz in Oil und Gas Production: A Cmpilarion of Ciasr~c Puperr, Item 51113, Natl. Assn. of Corrnslon Engmeers. Houston (1981).

hwoduciion IO Oiljrkl Wufer Twhnology, Item 52 140. Natl. As&n. of Corrosion Engineers, Houston (1979).

Jones. T.J . Neuatadter, E.L.. and Whittingham. K.P.: “Water-in-Crude Oil Emulsion Stability and Emulsion Destabilization by Chemical Demulsltiers.” J. C&z. Per. Twh. (April-June 1979) 17, No. 2,

lot-08.

Mansurov. R.I. ef rrl.: “Sravnitel’Nye lapytaniya Elektrodegldratorov Trekh Konstruktsll (Comparatwe Tests of Three Differently Deslgned Electrodehydrators).” Nefi Kho: (Dec. 1976) No. 12. 50-53.

McCiaflin, G. el ul.: “The Replacement of Hydrocarbon Diluent With Surfactant and Water for the ProductIon of Heavy, Viscous Crude Oil,” JPT (Oct. 1982) 2258-64.

McGhee. E.: “Una Sola Planta Deshidratara 150,000 BPD,” Pew&u Interameritnno (Aug. 1965) 42-46.

Mennon, V.B. and Wassan, D.T.: “Demulsifications,” Encyclopedia ofEwl.sion Technology, P. Becher (ed.), Marcel Dekker, New York City (1984) 2.

Moilliet, J.L., Collie, B., and Black, W.: Surface Activity, the Physi- cal Chemistry, Technical Applications. and Chemical Constitution of S~ntheti~Surfare-Artive Agents, D. Van Nostrand Co. Inc., Prince- ton, NJ (1960).

“Nalco Announces New Emulsion Breaker High Temperature/Pressure Heater Treater Simulator,” Visparch (Sept. 1984) 3, No. 2.

“New Mechanical Coalescing Medium is Used in Treaters,” Oil and Gas J. (Jan. 23, 1984) 82-83.

Petrov, A.A. and Shtof, I.K.: “Investigation of Structure of Crude Oil Emulsion Stabilizers by Means of Infrared Spectroscopy,” Chrmi- cal Technology Fuels Oils (July-Aug. 1974) 10, No. 7-8, 654-57.

“Practices and Methods of Preventing and Treating Crude Oil Emul- sions,” Bull. 417, U.S. Bureau of Mines, Superintendent of Docu- ments. Washington, D.C. (1939).

“Recommended Practice for Analysis, Design. Installation. and Test- mg of Basic Surface Safety Systems on Offshore Production Plat- forms,” latest edition, API RP 14C, API. Dallas.

“Recommended Practice for Analysis of Oil-Field Waters.” latest edition, API RP 45, API, Dallas.

“Recommended Practice for Design and Installation of Offshore Pro- duction Platform Piping Systems,” latest edition, API RP l4E. API, Dallas.

“Specification for Indirect-Type Oil Field Heaters.” latest edition. API Spec. 12K. API. Dallas.

“Specification for Vertical and Horizontal Emulsion Treaters.” latest edition. API Spec. l2L. API. Dallas.

“Standard for Welding Pipelines and Related Facilities.” latest edition. API Std. 1104, API, Dallas.

Stockwell, A.. Graham. D.E.. and Cairns, R.J.: “Crude Oil Emulsion Dehydration Studies,” paper presented at the 1980 Oceanology lntl.. Brighton, England, March 2-7, available from BPS Exhibitions Ltd., London, England.

Twaiing Oil Field Emulsions, third edition, Petroleum Extension Serv- ice. U. of Texas, Austin (1974).

“Tretolite Chemical Demulsifiers for Petroleum Producers,” &l//, , Tretolite Div.-Petrolire Corp., St. Louis, MO (Sept. 1978).

Wassan. D.T. PI a/.: “Observations on the Coalescence Behavior of Oil Droplets and Emulsion Stability in Enhanced Oil Recovery.” SPE/ (Dec. 1978) 409-17.

Tanker. K.J.: “Radio-Wave Interface Detector Measures Low Concew trations of Oil in Water, Controls Dumping.” Oil and Gas J. (Jan. 30. 1984) 150-52.

Chapter 20

Gas Properties and Correlations Robert S. Metcalfe, Amoco ProductIon Co.*

Molecular Weight Molecules of a particular chemical species are composed of groups of atoms that always combine according to a specific formula. The chemical formula and the international atomic weight table provide us with a scale for determining the weight ratios of all atoms combined in any molecule. The molecular weight, M, of a molecule is simply the sum of all the atomic weights of its constituent atoms. It follows, then, that the number of molecules in a given mass of material is inversely proportional to its molecular weight. Therefore, when masses of different materials have the same ratio as their molecular weights, the number of molecules present is equal. For instance, 2 lbm hydrogen contains the same number of molecules as 16 lbm methane. For this reason, it is convenient to define the term “lbm mol” as a weight of the material in pounds equal to its molecular weight. (Similarly, a “g mol” is its weight in grams.) One lbm mol of any compound, therefore, represents a fixed number of molecules.

Ideal Gas The kinetic theory of gases postulates that a gas is com-

posed of a large number of very small discrete particles. These particles can be shown to be identified with molecules. For an ideal gas, the volume of these particles is assumed to be so small that it is negligible compared with the total volume occupied by the gas. It is assumed also that these particles or molecules have neither attractive nor repulsive forces between them. The average energy of the particles or molecules can be shown to be a function of temperature only. Thus, the kinetic energy, EL, is independent of molecule type or size. Since kinetic energy is related to mass and velocity by

Ek = 5/2mv’,

it follows that small molecules (less mass) must travel faster than large molecules (more mass) when both are

‘Author of the ormmal ChaDter on ths ~ODC in the 1962 edllion was Charles F Wemaua.

at the same temperature. Molecules are considered to be

moving about in all directions in a random manner as a result of frequent collisions with one another and with the walls of the containing vessel. The collisions with the walls create the pressure exerted by the gas. Thus, as the volume occupied by the gas is decreased, the collisions of the particles with the walls are more frequent, and an increase in pressure results. It is a statement of Boyle’s law that this increase in pressure is inversely proportional to the change in volume at constant temperature.

"I P2 -=- “2 PI

where p is the absolute pressure and V is the volume. Further, if the temperature is increased, the velocity

of the molecules and, therefore, the energy with which they strike the walls of the containing vessel will be increased, resulting in a rise in pressure. To maintain the pressure constant while heating a gas, the volume must be increased in proportion to the change in absolute temperature. This is a statement of Charles’s law,

“I TI -=-

“2 T2

where T is the absolute temperature and p is constant. From a historical viewpoint, it is interesting to note that

the observations of Boyle and Charles in no small degree led to the establishment of the kinetic theory of gases, rather than vice versa.

It follows from this discussion that, at zero degrees absolute, the kinetic energy of an ideal gas, as well as its volume and pressure, would be zero. This agrees with the definition of absolute zero, which is the temperature at which all the molecules present have zero kinetic energy.

20-Z PETROLEUM ENGINEERING HANDBOOK

TABLE 20.1-VALUES OF THE GAS CONSTANT, R, IN pV= RT FOR 1 MOLE OF IDEAL GAS

Temperature Pressure Units Units

K -

Volume Energy Units Units R/g mol - - -

cm3

: L L

calories absolute joules

international joules - - - - -

1.9872 8.3144 8.3130

82.057 0.082054

62.361 0.08314 0.08478

Rllbm mol

R R R R R R

Ft K K K

- -

atm atm

mm Hg bar

kg/cm 3

- - -

atm in. Hg

mm Hg lbmlsq in., abs. lbmlsq ft, abs.

atm mm Hg

- Btu (IT) 1.986 - hp-hr 0.0007805 - kw-hr 0.0005819

cu ft - 0.7302 cu ft - 21.85 cu ft - 555.0 cu ft - 10.73 cu ft ft-lbm 1545.0 cu ft - 1.314 cu ft - 998.9 - - 1.988

Because the kinetic energy of a molecule is dependent only on temperature, and not on size or type of molecule, equal molecular quantities of different gases at the same pressure and temperature would occupy equal volumes. The volume occupied by an ideal gas, therefore, depends on three things: temperature, pressure, and number of molecules (moles) present. It does not depend on the type of molecule present. The ideal-gas law, which is actually a combination of Boyle’s and Charles’s laws, is a statement of this fact:

PV=nRT, . . . . . . . . . . . . . . . . . (1)

t T1 < T2 <T3 <Tc XT,

Fig. 20.1-Typical pressure volume diagram for pure component.

where

P= v= n= R= T=

pressure, volume, number of moles, gas-law constant, and absolute temperature.

The gas-law constant, R, is a proportionality constant dependent only on the units of p, V, n, and T. Table 20.1 presents different values of R for the various units of these parameters.

Critical Temperature and Pressure Typical PVT relationships for a pure fluid are illustrated in Fig. 20.1. The curve segment B-C-D defines the limits of vapor/liquid coexistence, B-C being the bubblepoint curve of the liquid and C-D, the dewpoint curve of the vapor. Any combination of temperature, pressure, and volume above that line segment indicates that the fluid exists in a single phase. At low temperatures and pressures, the properties of equilibrium vapors and liquids are extremely different-e.g., the density of a gas is low while that of a liquid is relatively high. As the pressure and temperature are increased along the coexistence curves, liquid density, viscosity, etc. generally decrease while vapor density, viscosity, etc. generally increase. Thus, the difference in physical properties of the coexisting phases decreases. These changes continue as the temperature and pressure are raised until a point is reached where the properties of the equilibrium vapor and liquid become equal. The temperature, pressure, and volume at this point are called the “critical” values for that species. Location C on Fig. 20.1 is the critical point. The critical temperature and pressure are unique values for each species and are useful in correlating physical properties. Critical constants for some of the commonly occurring hydrocarbons and other components of natural gas can be found in Table 20.2.

GAS PROPERTIES AND CORRELATIONS 20-3

TABLE 20.2-SOME PHYSICAL CONSTANTS OF HYDROCARBONS

Number Compound

1

2

3 4

z

7

a

9

10 11

12

13

14

15

16 17

la

19 20 21

22

23

24

25 26 27

28

29

30 31

:z 34

35

36

37

38

39

40 41

42

43

44

45

46 47

48

49

50 51

52

53

54 55

56

57

58

59

60 61

62

63

64

methane

ethane

propane n-butane

rsobutane n-pentane

rsopentane

neopentane n-hexane

2-methylpentane

3-methylpentane neohexane

2,3-dimethylbutane

n-heptane 2-methvlhexane

3-methylhexane 3-ethylpentane

2,2-dimethylpentane

2,4-dimethylpentane

33.dimethylpentane triptane

n-octane

dirsobutyl

isooctane

n-nonane n-decane

cyclopentane

meihylcyclopentane

cyclohexane

methylcyclohexane ethylene

propene

I-butene

cis-2.butene

trans-Pbutene

isobutene

1-pentene

1.2.butadiene

1,3-butadiene

isoprene acetylene

benzene

toluene

ethylbenzene

o-xylene m-xylene

p-xylene

styrene

Isopropylbenzene

methyl alcohol ethyl alcohol

carbon monoxide carbon dioxide

hydrogen sulfide sulfur dioxrde

ammonra

air

hydrogen

oxygen

nrtrogen chlonne

water

hehum

hydrogen chlonde

Formula

Molecular

Weight

16043

30.070

44097 58124

56124 72151

72151

72.151

86 178

86 178 86.178 86178

86.178

100.205 100205

100.205

100.205

100205

100.205 100205

100205

114232 114232

114.232

120259 142286

70 135 84 162

84.162

98.189 28054

42.081

56.108 56.108

56.108

56.108 70.135

54.092 54.092

68.119

26.038 78.114

92.141

106.168 106.168

106168 106168

104152

120.195

32.042 46.069

28010 44010

34076 64059

17031 28.964

2.016

31.999

28.013 70.906

la.015

4.003 36.461

Vapor Pressure

(lOOoF, psia)

(5000)

(800) 188.0 51.54

72.39 15.575 20.4444

36.66 4.960

6.767 6.103

9.859

7.406

1.620 2.2719

2.131

2.013

3.494

3293 2.774

3.375

0.537 1.1017

1.709

0.1796 0.0609

9.914

4503

3.266

16093

227.6

62.10

45.95

49.94

63.64 19.117

36 5

59.4

16.68

3.225

1.033

0.376

0.263

0.325 0.3424

0.238

0.188

4.63 2.125

387 I 8546

211.9

-

154.9

0.9495

906.3

Critical Constants

Pressure

bsial

667.8

707.6

616.3 550.7

529.1

488.6

490.4

464.0

436.9

436.6 453.1 4469

453.5

396.8 396.5

408.1 419.3

402.2

397.0 427.1

428.4

360.6

360.6

372.5

331 8 3044

6530 549 0

590.9

503.6 731 1

6672

583.5 612 1

587 1

580 0

591.8

(653.0) 628.0

(55a 4)

a904 7104

595 5 5234

541 6

5129

5092

580.0

465.4 1.174.4

9253

5075 1,071 0

1,306 0 1,145 0

1,636 0 5469

188 I

7369

493 0 1.1184

3,207 9 32.99

205.1

Temperature

(OF)

- 116.68

90.1

206.01 305.62

274.96

385.6

369.03

321.08

453.6

43574 448.2 420.04

4400

512.7

494.89

503.67 513.36

477.12

475.84

505.74

496.33

564.10 530.31

519.33 610.54 651.6

461.6

499.24

536.6

570.15 48.56

197.06

295.48 324.37

311.86 292 55

376.93

(340.0)

305.0

(412.0)

95.32 552.22

605.57

651.29 674.92

651.02 649.54

706.0

676.3

463.08

465.39

-220.4 87.67

212.6

315.6

270.4 -221.4

- 399.9 -181.2

-232.7 291.0

705.5

-450.308 124.8

Volume

(cu ftllbm)

0.0988

0.0788

0.0737 0.0703

0.0724 0.0674

0.0679

0.0673

0.06887

0.0682 0.0682

0.0668

0.0665

0.0690

0.0673 0.0646 0.0665

0.0665

0.0668

0.0682 0.0636

0.0690 0.0676

0.0657 0.0684

0.0679

0.0594 0.0607

0.0589

0.0601 0.0748

0.0689

0.0686

0.0668

0.0679 0.0682

0.0676

(0.0649) 0.0655

(0.0850) 0.0695

0.0525

0.0549 00565

0.0557

0.0567 0.0570

0.0541

0.0572

0.0589

0.0580 0.0532

0.0342

0.046 0.0306

0.0681

0.0517

0.5164 0.0367

0.0516 0.0280

0.0509

0 230

0.0356

Gas Densrty (60°F. 14.696 psia)

Calculated as Ideal Gas’

(cu 11 gas/gal liquid)

59.1. 37.48"

36.49' t 31.80"

30.65" 27.67

27.38

26.16"

24.38

24.16

24.56 24.02

24.47

21.73

21.56

21.64 22.19

21.41

21.39

22.03 21.93

19.58 19.33

19.26 17.81

16.32

33.85

28.33

29.45

24.92

39.25”

33.91" 35.36"

34.40"

33.86" 29.13"

38.4"

36.69' *

31.67

35.82

29.94

25.97

26.36

25.88 25.80

27.68

22.80

78.61 54.36

-

59.78' *

73.07 69.01

114.71 -

- 63.53

175.6

74.88


Specific Gravity (Relative Density) The specific gravity of a gas, y, is the ratio of the density of the gas at a given pressure and temperature to the density of air at the same pressure and temperature. The ideal- gas laws can bc used to show that the specific gravity (ratio of densities)* is also equal to the ratio of the molecular weights, When the ideal-gas assumptions are not valid (high pressures or most real gases), this will not always be true. By convention, specific gravities of all gases at all pressures are defined as the ratio of the molecular weight of the gas to that of air (28.966).

Mole Fraction and Apparent Molecular Weight of Gas Mixtures The analysis of a gas mixture can be expressed in terms of a mole fraction, y;, of each component, which is the ratio of the number of moles of a given component to the total number of moles present. Analyses also can be expressed in terms of the volume, weight, or pressure fraction of each component present. Under limited sets of conditions, where gaseous mixtures conform reasonably well with the ideal-gas laws, the mole fraction can be shown to be equal to the volume fraction but not to the weight fraction. The apparent molecular weight of a gas mixture is equal to the sum of the mole fraction times the molecular weight of each component.

Specific Gravity of Gas Mixtures The specific gravity (yR) of a gas mixture is the ratio of the density of the gas mixture to that of air. It is measured easily at the wellhead in the field and, therefore, is used as an indication of the composition of the gas. As mentioned earlier, the specific gravity of gas is proportional to its molecular weight (M,) if it is measured at low pressures where gas behavior approaches ideality. Once again, by convention, the specific gravity is defined as the mole weight of the gas mixture divided by 28.966. Specific gravity also has been used to correlate other physical properties of natural gases. To do this, it is necessary to assume that the analyses of gases vary regularly with their gravities. Since this assumption is only an approximation and is known to do poorly for gases with an appreciable nonhydrocarbon content, it should be used only in the absence of a complete analysis or of correlations based on a complete analysis of the gas.

Dalton’s Law The partial pressure of a gas in a mixture of gases is defined as the pressure that the gas would exert if it alone were present at the same temperature and volume as the mixture. Dalton’s law states that the sum of the partial pressures of the gases in a mixture is equal to the total pressure of the mixture. This law can be shown to be true if the ideal-gas laws apply.

Amagat’s Law The partial volume of a gas in a mixture of gases is defined as that volume which the gas would occupy if it alone were present at the same temperature and pressure as the

mixture of the gases. If the ideal-gas laws hold, then Amagat’s law, that the sum of the partial volumes is equal to the total volume, also must be true.

Real Gases At low pressures and relatively high temperatures, the volume of most gases is so large that the volume of the molecules themselves may be neglected. Also, the distance between molecules is so great that the presence of even fairly strong attractive or repulsive forces is not sufticient to affect the behavior in the gas state. However, as the pressure is increased, the total volume occupied by the gas becomes small enough that the volume of the molecules themselves is appreciable and must be considered. Also, under these conditions, the distance between the molecules is decreased to the point where the attractive or repulsive forces between the molecules become important. This behavior negates the assumptions required for ideal-gas behavior, and serious errors are observed when comparing experimental volumes to those calculated using the ideal-gas law. Consequently, a real-gas law was formulated (in terms of a correction to the ideal-gas law) by use of a proportionality term called the compressibility factor, z. The real-gas law is thus

pV=znRT. . . . . (2)

Tables of compressibility factors are available for most pure gases as functions of temperature and pressure. Com- pressibility factors for mixtures (or unknown pure compounds) are measured easily in a Burnett’ apparatus or a variable-volume PVT equilibrium cell. Excellent correlations are also available for the calculation of compressibility factors as discussed in the section on equations of state (EOS’s). For this reason, compressibility factors are no longer routinely measured on dry gas mixtures or most of the leaner wet gases. Rich gas condensate systems require other equilibrium studies, and compressibility factors can be obtained routinely from these data. A knowledge of the compressibility factor means that the density, p, is also known from the relationship

PM P=-,

ZRT

because V=(IIM)/P. where M is the molecular weight. Many times it is more convenient to report compressi-

bilities than densities because the range in z is usually small-e.g., between 0.3 and 2.0.

Principle of Corresponding States The principle of corresponding states has been useful in correlating the properties of gases. This principle was dc- veloped because observers noticed that the behavior of pure gases was qualitatively similar when compared (on p-V plots, for instance) even though the quantitative values of p and V were very dissimilar. The idea was advanced that the properties of substances could be correlated if they were all compared at “corresponding” values of T and p, which could be referenced easily. In the application


of the principle of corresponding states to a single- component gas, the critical state of the gas is used as the reference point. The following terms are used.

P~=~, 7.,=$, and V,=I, PC c V,

where p,. = reduced pressure, T, = reduced temperature, V, = reduced volume, PC = critical pressure, T,. = critical temperature, and V, = critical volume.

Compressibility factors of many pure compounds are available as functions of pressure in most handbooks deal- ing with gas properties (e.g., Katz et al. *). While the principle of corresponding states is not entirely rigorous, its application has been used widely in the determination of gas volumes for engineering purposes. It also has application in the estimation of gas viscosities.

In application of the principle of corresponding states to a mixture of gases, the true critical temperature and pressure for the gases cannot be used because the paraffn- ic hydrocarbon series does not strictly follow the principle as stated above. “Pseudocritical” temperature and pressure are defined for use in place of the true critical temperature and pressure to determine the compressibility factor for a mixture. The pseudocritical temperature and pseudocritical pressure normally are defined as the molal average critical temperature and pressure of the mixture components, Thus

Ppc =CYiPci

and

Tpc = Cyi Tci 3

where

PPC = pseudocritical pressure of the gas mixture, T PC = pseudocritical temperature of the gas

mixture, pci = critical pressure of Component i in the gas

mixture, Tci = critical temperature of Component i in the

gas mixture, and yi = mole fraction of Component i in the gas

mixture.

These relations are known as “Kay’s rule” after W.B. Kay, who first suggested their use.

The pseudocritical pressure and temperature are then used to determine the pseudoreduced conditions:

P p/W=----,

PPC

where pPr is the pseudoreduced pressure, and

T

T,,,=k. P’

PSEUDOREDUCEDPRESSURE


Fig. 20.2~Compressibility factor for natural gases (from Ref.

3).

where Tpr is the pseudoreduced temperature. These reduced conditions are used to determine the compressibility factor, z, from Fig. 20.2, which was developed by Standing and Katz3 from data collected on methane and natural gases. The data used to develop Fig. 20.2 ranged up to 8,200 psia and 250°F. Compressibility factors of high-pressure natural gases (10,000 to 20,000 psia) may be obtained from Fig. 20.2A, which was developed by Katz ef al. * Figs. 20.2B and 20.2C may be used for low- pressure applications after Brown et al. 4

Fig. 20.3 presents a correlation developed by Brown et al. 4 between the pseudocritical temperatures and pseudocritical pressures of naturally occurring systems with their specific gravities. Values from this chart then can be used to determine the compressibility factor of a gas whose complete analysis is not known but should be used with caution since many different compositions can result in similar gravities. It should be used only when small amounts of nonhydrocarbons are present.

Figs. 20.2A through 20.2C do not consider the presence of large quantities of nonhydrocarbons such as nitrogen, carbon dioxide, and hydrogen sulfide. However, it has been shown that nitrogen does not pose a problem for the calculation of compressibilities, and Wichert and Aziz5 have proposed corrections for the pseudocritical constants for natural gases with significant concentrations of carbon dioxide and hydrogen sulfide. Their procedure involves calculation of corrected pseudocritical constants for mixtures. The corrections are defined as follows.

7;;< =TP’.-t _. _. _. (3)


and

Pbc = PpC GC

Tpc+Y~~s(l-YH,Sk '

where

PSE”cQRED”CEo PRESSURE

Flg. 20.2A-Compressibility factor for natural gases at pressures of 10,000 to 20,000 psia (from Ref. 2).

0.90 I I I I 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07

PSEUDOREDUCED PRESSURE PSEUDOREDUCEDPRESSURE

+lS(~H~s’.~-yH~s~.‘), . . . . . . . . . . . . . .(4)

Tbc = corrected pseudocritical temperature,

P;, = corrected pseudocritical pressure,

Yco, = mole fraction of CO2 in mixture, and yH,s = mole fraction of Hz S in mixture.

The correction factor, E, has been plotted against hydrogen sulfide and carbon dioxide concentrations in Fig. 20.4 for convenience. This correction is reported to reproduce compressibility factors with less than 1% error.

Equations of State An EOS seeks to describe specific PVT relationships of fluids mathematically. There are hundreds of these equations ranging from those for a specific pure compound to generalized forms that claim to relate the properties of multicomponent mixtures. Naturally, there is a large range of complexity from the simple ideal-gas law to

z-

1111111111”““’

hi i

0.6

I I I I I I’ I\1 ‘\, \ \-

Fig. 20.2B-Compressibility factors for natural gases near atmospheric pressures.

Fig. 20.2C-Compressibility factors for natural gases at low reduced pressures.


modern equations with 15 or more universal constants plus adjustable parameters. Historically, use of these equations has been limited to applications by researchers having large computing facilities. Recently, however, operating engineers have been provided with the same computing tools previously reserved only for researchers and special projects. The use of EOS’s, therefore, has become relatively common. Some applications, such as calculation of compressibility factors, are possible on hand-held programmable calculators. The modern engineer should not forget the use of EOS’s when the need arises for calculation or estimation of fluid properties.

Van der Waals’ Equation Van der Waals6 added terms to the ideal-gas law in an attempt to take into account forces between molecules as well as volume of the molecules themselves. His equation becomes

. ._ ____.

where VM is the molar volume and a and b are constants characteristic of the gas.

The term b is a constant to correct for the volume occupied by the molecules themselves. The term a/Vi is a correction factor to account for the attraction between molecules as a function of the average distance between them (which is related to the molar volume). When an EOS such as the van der Waals equation is applied to mixtures, either special constants for a and b must be developed for each mixture or constants for each gas in the mixture must be included in the equation along with adjustments for the interaction between unlike gases. The latter is the more common approach.

t i i i i i i i I

MS QMV1l-Y (AIR I 1) PERCENT H,S

Fig. 20.3-Pseudocritical properties of natural gases.

Van der Waals’ law extends the range of pressures and temperatures for describing gas behavior beyond that of the ideal-gas law. However, it has two disadvantages in actual application. The correction factors are inadequate at very high pressures and it is not always easy to obtain the mixture coefficients and interaction constants. In addition, this two-parameter formulation does not really treat the attractive and repulsive forces correctly. Despite these criticisms, modifications of the van der Waals equation have been used successfully in industry for many years.

Redlich and Kwong7 developed the first major extension of the two-parameter EOS when they proposed their own form and showed how they related the a and b terms toR,p,, and T,. Other researchers since have modified the original Redlich and Kwong equation to improve its accuracy and generality further. Most notable of the modifications are those of Soave,’ Zudkevitch and Joffe, 9 and Peng and Robinson. lo Some companies have their own versions, such as the one published by Yar- borough. ’ ’

The most common equations of state in use today and the computer programs available are the following.

1. The Starling-Hon I2 extension of the Benedict- Webb-Rubin I3 EOS:

p=RTpM+ B,RTM,-Cf5-5 T= T3 1-4

+ (bRT-a-++a(a+dj$

2

+ 2E (1 +y&) exp(-yp$), . . . . . ( > T*

PSEUDOCRITICAL TEMPERATURE ADJUSTbENT FACTOA, E. “F

Fig. 20.4-Pseudocritical temperature adjustment factor, 5, “F.

20-a PETROLEUM ENGINEERING HANDBOOK

0.024 I

# / .\ I

i 0.018

B

E O.OlS} , U'] I/ I

0.008

0.006

0.004 [ I I I I I I r;n inn im 3nn 3m m vin Ann

TEMPERATURE, "F

Fig. ZOS-Viscosity of pure compounds at 14.7 psia.

where A ,, , B,,, C, D,,, E,, a. b, c, d, 01, and y are empirical constants, and pM equals n/V,++ (subscript M refers to molar values). This equation usu ,lly is called the “BWRS” and is available from Exxon C ,rp.

2. The Peng-Robinson lo EOS (Equipha eTM):

RT a(T) -- ‘= I’,-b VM(V,,,+b)+b(VM-b)’ ““”

. (7)

where a and b are constants characteristic of the fluid, a(T) is a functional relationship, and V, is the molar volume. It is available from the Gas Processors Suppli- ers Assn. (GPSA).

3. The Soave’ modification of the Redlich-Kwong’ EOS:

RT a(T) -- p= V~.lb vM(V~+b), . . (8)

where a(T) is a functional relationship. It, too, is available from the GPA.

The first equation, BWRS, is an empirical form using 11 constants. The values of these constants have been determined fiti properties measured on many different fluids. It is ext :mely accurate in the prediction of most thermodynamic properties. Eqs. 7 and 8 are variations of the original equation proposed by van der Waals ?.?d as such are not as accurate as the BWRC for calculation of pure component properties or properties of mixtures of light hydrocarbons. Both the Peng-Robinson and the

Soave RK EOS’s are more reliable for phase equilibrium calculations or for calculation of properties of gas condensate systems. One cannot assess their accuracy directly because it is dependent on how well the constants represent the specific components.

The Redlich-Kwong EOS and its extension are cubic in compressibility factor. J.J. Martin14 proposed a generalized cubic equation that, through suitable adjustment of parameters, can be used to obtain any other cubic including those that have been proposed after his work was published. All cubic equations have limitations in their ability to represent behavior at near-critical conditions. They are incorrect in the prediction of the critical compressibility factor and/or the shape of the critical isotherm. They can be manipulated by additional terms to circum- vent this problem but errors then appear in some other region of pressure-temperature-composition space. In general, however, EOS’s can be used routinely to calculate gas properties for both hydrocarbon and nonhydrocarbon systems and their mixtures.

One particularly useful application of EOS’s in gas property estimations is the direct calculation of the compressibility factor, z. As noted previously, the principle of corresponding states can be used to obtain compressibilities with reasonable accuracy. However, one can solve an EOS directly for z quite readily. The most reliable methods for typical natural gases are those of Robinson and Jacoby I5 and Hall and Yarborough. I6 Robinson and Jacoby proposed the following equations.

RT a -_ P= vM-b fiv,(v,+b) ? . . (9)

ai=cr,+PjT, . . . . . . . . . . . . . . . . . . . . . . . . . . ..(lO)

and

bi=yi+hiT, . . . . . . (II)

where 01, 0, y, and 6 are constants for Substance i, and for mixtures

a~; %[Kjjai+(l-Kij)aj],

a f7t =CiC;YjYiay,

and

b,,, =C,y,bi,

where KY is a constant for each binary pair when used for mixtures.

Their equations are another modification of the Redlich- Kwong equation designed specifically for the region of temperatures from 70 to 250°F and pressures below 1,500 psia. It is untested t’Jr gas mixtures containing large amounts of Cd+ material. Within these stated limits, it should be expected to calculate 8~. mpressibility factors with less than 2% error.

The Hall-Yarborough equation is

(1 +x+x2 -X”)-,4X+&c z= (1 + ) . . (12)


I I 1 I 1 i : ! ! : ; ’ 1 “ao.S 1.0 12 1.4 1.6 1.8 2.0 2.2 2.4 26 2.3 3.0 3.2

I

MOLEWLAR WEIQHT PSEUDOREDUCED TEMPERATURE

Fig. 20.6-Viscosity of gases at 14.7 psia. Fig. 20.7-Viscosity ratio vs. pseudoreduced temperature.

where L = compressibility factor,

A = (14.76t-9.76t2 +4.5&‘), B = (90.7t-242.2t2 +42.4t3), C = 1.18+2.82t,

xi = bpM14, b = 0.245(RT,/p,) exp[-1.2(1-t)*], and t = TJT.

It is designed specifically to fit the Standing-Katz charts and provides excellent results for multicomponent systems. Hall and Yarborough also include the correction factors proposed by Wichert and Aziz for systems with high concentrations of nonhydrocarbons. The method has been programmed for hand calculators by Ajitsaria. I7 Note that the equations contain both z and PM, making the solution trial and error and not well suited for use

without a computer or calculator algorithm.

Viscosity Viscosity is an important property in determining resistance to flow during production and marketing of gas. Generally, the viscosity of a gas increases with increasing pressure, except at very low pressures where it becomes more or less independent of the pressure. At low pressures, the viscosity of a gas, unlike that for liquids, increases as the temperature is raised. This is caused by the increasing activity of the molecules as temperature increases. Viscosity of a fluid is obtained by determining the force per unit area necessary to shear two parallel planes with a standard spacing and velocity difference. The standard unit of viscosity is the poise, which is defined as 1 dyne-s/cm2 [6.9 lbf-seclsq in.]. However, the common unit is the centipoise (0.01 poise). Carr et al. ‘* used the data of many researchers to produce Fig. 20.5, which presents viscosities as a function of temperature at atmospheric pressure for a number of pure compounds.

Viscosity Correlations Viscosities can be estimated both by the principle of corresponding states and by a residual viscosity function based on reduced density. Carr et al., I8 using the the-

ories of transport processes, correlated viscosities of pure gases and gas mixtures against molecular weight and temperature. Fig. 20.6 presents their correlation for viscosities at atmospheric pressure. Fig. 20.7 permits estimation of a pressure correction for gas viscosities by corresponding-states techniques. The ratio of the viscosity at some elevated pressure to the viscosity from Fig. 20.6 is plotted against pseudoreduced temperature and pressure. Viscosi- ties calculated from this correlation should be expected to have less than 2% error.

The residual viscosity function (P--C(*) also has been used to correlate gas viscosities with even better success

than the corresponding-states technique described previously. (I* is a correlating parameter obtained from Fig. 20.8.) Thodos et al. 19,*o have shown that the residual viscosity function can be well correlated against density, thereby making it a useful tool for both gas and liquid viscosities. The Thodos method requires two steps, as does the technique of Carr et al. First p* must be estimated, then the effect of pressure can be calculated from another correlation. The correlation for CL* is shown in Fig. 20.8, and the effects of pressure can be estimated

PSEUDO REOUCEO TEMPERATURE

Fig. 20.6-Thodos viscosity correlation


Fig. 20.9-Thodos viscosity correlation-pressure correction.

from Fig. 20.9. Viscosities calculated using the correlations of Thodos et al. can be expected to have an accuracy on the order of 3%.

To use Figs. 20.8 and 20.9, one mug first calculate the average mole weight of the mixture, M, =CyiMi, and the pseudocritical temperature, pressure, and volume by Kay’s rules (T, in units of Kelvin and V, in cm3/g mol) or Fig. 20.3 if the CT+ concentration is small. Alterna- tively, the correlation of Matthews et al. *’ (Fig. 20.10) may be used to get T,. andp, for C7+ fractions. The following may be used for V, of the C7+ fraction.

(WC,+ = 1.561(Mc,+ IpR) ‘.15,

where MC,+ is the molecular weight of the CT+ fraction, and ~a is the relative density of the CT+ fraction.

Calculation of the pseudocritical density, ppcr and the viscosity parameter, t, are as follows.

ppc =M,/V,, . . . . . . . . . . . . . . (13)

(Mg),~(Ppc)% ) . . I.. . . . . . (14)

where Tpc is the pseudocritical temperature, K, and ppc is the pseudocritical pressure, atm. Caution: This is a correlation and the terms should not be converted to a consistent set of units.

Fig. 20.10-Pseudocritical properties of C,+ fractions

For very quick estimations, Katz’ provides graphs of viscosity vs. temperature (“F) and pressure (psia) for gas gravities ranging from 0.6 to 1 .O. Errors can be expected to be on the order of 4 to 5 % .

If gas density is not known it can be obtained from the compressibility factor through pR =M,pl (z,RT). Com- pressibility factors can be obtained by using the methods discussed above. Reduced conditions then can be calculated making sure ,o and pPc are in the same units. It is possible to use Fig. 20.8 to obtain p* and then obtain t from p*=(p*l) 14. The final step is to obtain (p-p*)[ from Fig. 20.9 and solve for p with P=/~*+[(P-P*)[]/[.

Within the limitations of each correlation, that of Carr et al. may have a slight advantage. That of Thodos et al. is a more general relationship and can be used for both gases and liquids, making it the preferred method for phase equilibrium calculations or for the near-critical region.

Natural Gasoline Content of Gas In the handling and evaluating of gas, determination of natural gasoline or liquefiable content is important. This can be accomplished because the liquid volumes of the heavier components in natural gasoline are essentially additive. The required number of cubic feet of gas to form, by condensation, 1 gal of various materials is shown in Table 20.2 under the heading “cu ft gas/gal liquid.” Mole fraction, or cubic feet of any component per cubic foot of mixture, divided by the cubic feet of gas per gallon of liquid gives the total gallons of liquid that each component could contribute to the natural gasoline per cubic

GAS PROPERTIES AND CORRELATIONS 20-l 1

foot of gas mixture. If only a part of the component under consideration is to be recovered as liquid, a suitable correction must be made. Using the principle of additive volumes, the sum of contributions of each component can be assumed to give the recoverable gasoline content per cubic foot. Use of this procedure can lead to errors of about 10% if relatively large amounts of aromatic and/or naphthenic compounds are present.

Formation Volume Factor The gas FVF, B,, is defined as barrels of reservoir gas contained in 1 scf. It is sometimes erroneously reported as the reciprocal of this definition. In either case, it is a way of relating reservoir PV to produced surface volumes. The definition of B, assumes that no liquids will condense as the reservoir gas is brought to standard conditions (60°F and 1 atm [288 K and 100 kPa]). This may be an invalid assumption for gas condensates but is probably acceptable for most wet gases.

The real-gas law, pV=:nRT, can be used to convert measurements at standard conditions to reservoir conditions. If the above assumption holds, then

where the subscript rc refers to reservoir conditions and SC to standard conditions. Since, by definition,

temperature divided by the actual volume. It can be written in differential form as

i av CR=-- - .

( > v ap T

If a gas is ideal it can easily be shown that cg = l/p. As we have already discussed, however, reservoir gases and most surface gases do not follow the ideal-gas law. Consequently, this result should only be considered as an order-of-magnitude approximation.

When the real-gas law, pV=znRT, is differentiated to calculate c,, the result is

1 i a7. CR=---

( > - . . . . . . . . . . . . . .

P z aP r (17)

If z’s are known as function of pressure, it can be evaluated over a small range as

However, Trube** has correlated a term called pseudoreduced compressibility against pseudoreduced pressure to eliminate the need for these evaluations. His definition of pseudoreduced compressibility is

VU cpr=cK Xp,,‘., . . . . . (18)

B = 5.61458 s

v.>,. .

it follows that

TI.J r<, B,s =0.005035- . (16)

P,-c~z\c~

when T is in “R and p in psia, or

T,.,.z j.<. B, =0.34722-

P/G.W

when T is in K and p in kPa. Many times it is assumed that z,,, = 1 .O, but this is not

necessarily true. If greater precision is desired, Fig. 20.2B or 20.2C can be used to determine z for the gas at standard conditions. For rough engineering calculations, this extra precision may not be required.

Coefficient of Isothermal Compressibility Reservoir engineering equations that deal with system compressibility require a gas compressibility term. This is not the gas compressibility factor, z, but the coefficient of isothermal compressibility, c~, It is defined as the rate of change of volume with respect to pressure at constant

and is nondimensional. The correlating work of Trube is presented in Figs. 20.11 and 20.12. A knowledge of pseudoreduced temperature and pseudoreduced pressure is required to obtain the pseudoreduced compressibility. The coefficient of isothermal compressibility then can be calculated directly from this relationship. Trube does not give any estimates of the accuracy of his correlation, but a method based on pseudoreduced properties should be at least as accurate as the z-factor correlations on which it is based because the coefficient of compressibility is a slope rather than an absolute number.

Vapor Pressure At a given temperature, the vapor pressure of a pure compound is the pressure at which vapor and liquid coexist at equilibrium. The term “vapor pressure” should be used only in conjunction with pure compounds and is usually considered as a liquid (rather than gas) property. For a pure compound, there is only one vapor pressure at any temperature. A plot of these pressures for various temperatures is shown in Fig. 20.13 for n-butane. The temperature at which the vapor pressure is equal to 1 atm (14.696 psia or 101.32 kPa) is known as the normal boiling point.

The Clapeyron equation gives a rigorous quantitative relationship between vapor pressure and temperature:

dp, Lt. dT =TAv, . . . . ..I................... (19)


2 3 4 5 6 7 6 9 10

PSEUDOREOUCEDPRESS”RE

Fig. 20.1 l-Reduced compressibility coefficients for low pseudoreduced pressures and fixed pseudoreduced temperatures

0.07

0.06


Fig. 20.12~Reduced compressibility coefficients for moderate pseudoreduced pressures and fixed pseudoreduced temperatures.

500

400

300

200

100

0 100 200 300

TEMPERATURE ‘F

Fig. 20.13-Vapor pressure of n-butane

where pv = vapor pressure,

T = absolute temperature, AV = increase in volume while vaporizing 1

mole, and L, = molal latent heat of vaporization.

Assuming ideal-gas behavior of the vapor and neglect- ing the liquid volume, the Clapeyron equation can be simplified over a small temperature range to give the approximation

d In pv L, -=-

dT RT2’

which is known as the Clausius-Clapeyron equation. This equation suggests that a plot of logarithm of vapor

pressure against the reciprocal of the absolute temperature would approximate a straight line. Such a plot is useful in interpolating and extrapolating data over short ranges. However, the shape of this relationship for real substances is not a straight line but rather S-shaped. There- fore, the use of the Clausius-Clapeyron equation is not recommended when other methods are available.

Cox Chart COXES further improved the method of estimating vapor pressure by plotting the logarithm of vapor pressure against an arbitrary temperature scale. The vapor- pressure/temperature plot forms a straight line, at least for the reference compound, and usually for most of the materials related to the reference compound. This is especially true for petroleum hydrocarbons. A Cox chart using water as a reference material is shown in Fig. 20.14. In addition to forming nearly straight lines, compounds of the same family appear to converge on a single point.

GAS PROPERTIES AND CORRELATIONS 20-l 3

VAPOR PRESSURE, PSIA

Fig. 20.14-Cox chart for normal paraffin hydrocarbons.

Thus, it is necessary to know only vapor pressure at one temperature to estimate the position of the vapor-pressure line. This approach is very handy and can be much better than the previous method. Its accuracy is dependent to a large degree on the readability of the chart.

Calingeart and Davis Equation The Cox chart was fit with a three-parameter function by Calingeart and Davis. x Their equation is

B lnp,,=A-- T-c’ . . . .

where A and E are empirical constants, and, for compounds boiling between 32 and 212”F, C is a constant with a value of 43 when T is in K, and C is a constant with a value of 77.4 when T is in “R.

This equation generally is known as the Antoine25 equation because he proposed one of very similar nature that used 13 K for the constant C. Knowledge of the vapor pressure at two temperatures will fix A and B and permit approximations of vapor pressures at other temperatures. Generally, the Antoine approach can be expected to have less than 2% error and is the preferred approach if the vapor pressure is expected to be less than 1,500 mm Hg [200 kPa] and if the constants are available.

Lee-Kesler Vapor pressures also can be calculated by corresponding- states principles. The most common expansions of the Clapeyron equation lead to a two-parameter expression. Pitzer extended the expansion to contain three parameters:

In pvr=fo( T,)+wf’( T,), . . .(21)

where pvr is the reduced vapor pressure (vapor pressure/critical pressure), f” andf’ are functions of reduced temperature, and w is the acentric factor.

Lee and Kesler26 have expressed f’ and f’ in analytical forms:

f” =5.92714-(6.09648/T,)

- 1.28862 In T, +O. 169347( T,-)6 . .(22)

and

f’ =15.2518-(15.6875/T,)- 13.4721 In T,

+0.43577(T,)6, . . . . . . . . . . . . . . . . . . . . . . . ..(23)

which can be solved easily by high-speed computer or a hand-held calculator. Lee-Kesler is the preferred method of calculation but should be used only for nonpolar liquids.

The advent of computers and calculators makes use of approximations and charts much less advantageous than they were in the 1960’s. Values of acentric factors can be found in Ref. 27, which also presents many other available vapor-pressure correlations and calculation techniques with comments about their advantages and limitations.

Example Problems Example Problem 1. Calculate relative density (specific gravity) of the following natural gas. All compositions are in mole percent.

Cl 83.19 C2 8.48 c3 4.37 iC4 0.76 nC4 1.68 iC5 0.57 nC5 0.32 c6 0.63 Total 100.00


TABLE 20.3-DATA FOR EXAMPLE PROBLEM 1 TABLE 20.4-DATA FOR EXAMPLE PROBLEM 2

I Y,

-0.8319 3

2 0.0848 0.0437

iC, 0.0076

nC4 0.0168 iC, 0.0057

nC5 0.0032 C6 0.0063

Total 1 .oooo

‘From Table 20 2

M,’ Y,M, 16.04 13.344 30.07 2.550 44.10 1.927 58.12 0.442 58.12 0.976 72.15 0.411 72.15 0.231 86.18 0.543

20.424

Methane Ethane Propane i-butane n-butane i-pentane n-pentane Hexanes

Mole Fraction (O:‘) ~ ~ 0.8319 343 0.0848 550 0.0437 666 0.0076 735 0.0168 766 0.0057 829 0.0032 846 0.0063 914

(p%*) M;

xii-- 16.04 708 30.07 616 44.09 529 58.12 551 58.12 490 72.15 489 72.15 437 86.17

1 .oooo

‘From Table 20 2

Solutioion. First calculate the apparent mole weight from

information in Table 20.3.

ti, =Cy;M; ~20.424.

and

M* 20.424 - =0.705

‘“=z= 28.966

Then From Fig. 20.3 we obtain YI: = M,/M,=CviMi128.966

TPc =392”R. = 20.424128.966

535

= 0.705, Tp,=---1.36,

392

where M, is the molecular weight of air=28.966 pPc. =663 psia,

Example Problem 2. Calculate actual density of the same mixture at 1,525 psia and 75°F.

Solution.

PM, PK = -

z,RT’

p = 1,525 psia,

1,525 ~ =2.30, pv= (-63

and

zR =0.712.

M, = 20.424,

Conclusion. Composition and gas gravity methods yield identical results for this hydrocarbon gas at surface processing conditions. Then,

1,525 x 20.424

Psi = 0.712 x 10.73~535 R = 10.73 psiaxcu ft

“R Xlbm mol (from Table 20. l),

T = 75”F+460=535”R, and =7.62 lbm/cu ft=O.122 g/cm3.

zK must be obtained from Fig. 20.2. Example Problem 3. Calculate the z factor for the reservoir fluid in Table 20.5 at 307°F and 6,098 psia. For the

Calculate zg from known composition or gas gravity in C 7 + fraction: Table 20.4. From the known gas composition we obtain y = 0.825(40”API),

T,,,. =Ey;Tci =393.8”R, Mh’ = 119, and

535 the experimental zR = 0.998.

T/Jr= -=11.36, 393.8 Solution. From the known gas composition we obtain

(Fig. 20.2) ~~~.=~y;p~i=662.6 psia,

Tp, = C>‘iT,.i=487”R,

1,525 = - =2.30, and

767

Ppr 662.6 T/w = ==1.58,

zR =0.712.

From gas gravity we obtain

iis =CyjMj =20.424

ppc = CJ’;p,i=822 psia,

6,098 PP = -=7.42. and

824

ZR = 0.962 (-4% error).


TABLE 20.5-DATA FOR EXAMPLE PROBLEM 3

Mole Fraction &J (:!a) & ~~

Nitrogen 0.1186 226 493 28.02 Methane 0.3636 343 660 16.04 Carbon Dioxide 0.0849 546 1071 44.01 Ethane 0.0629 550 708 30.07 Hydrogen sulfide 0.2419 673 1306 34.08 Propane 0.0261 666 616 44.09 I-butane 0.0123 735 529 58.12 n-butane 0.0154 766 551 50.12 kpentane 0.0051 I329 490 72.15 n-pentane 0.0052 846 489 72.15 Hexanes 0.0067 914 437 66.17 Heptanes plus 0.0373 1,116* 453’ 119.00

Total 1 .oooo

‘Otdamed from Ffg 20 10

From gas gravity we obtain

M,? = EyiMi=31.87, and

31.87 YK = MS/M, = ~=l.loo.

28.966

T I’( = 524”R,

767 T,j, = 524 = 1.464,

P[K = 652 psia,

6,098 Ppr = __ =9.3.5, and

652

zx = 1.087 (9% error).

By including corrections to calculated criticals with Wichert and Aziz’s chart we obtain

c = 31.2 (Fig. 20.4),

T;,. = 487-31=456”R,

Pi, = (822)(456)/[487+(0.2419)(1-0.2419)(31.2)],

= 762 psia.

767 T,,, = 456 = I .68,

6,098 P/Jr = ~ =8.00, and

762

zfi = 1.010 (1% error).

Example Problem 4. Calculate the viscosity at 150°F and 2,012 psia for the gas of the composition shown in Table 20.6.

TABLE 20.6-DATA FOR EXAMPLE PROBLEM 4

Mole Fraction

Nitrogen 0.156 Methane 0.739 Ethane 0.061 Propane 0.034 i-butane 0.002 n-butane 0.006

Total 1.000

M, Molecular T,

Weight (OR) (p?a)

---5&z---- 22a 492 16.04 343 666 30.07 550 706 44.09 666 616 58.12 735 529 56.12 765 551

Solution by the Carr-Kobayashi-Burrows Method.

Tpc = CyiT,.i =350”R,

460+ 150 Tpr = =1.74,

350

ppc = Cyip,i=639 psia.

2,012 PPr = -=3.15,

639

M, = CyiMi=l9.98, and

19.98 “fh’ = -=0.690.

28.966

Viscosity at 150”F, 1 atm (Fig. 20.6) = 0.0116 cp Correction for N2 (Fig. 20.6) = +0.0013 cp

Viscosity, r.i r = 0.0129 cp

Viscosity ratio, h/h, (Fig. 20.7) = 1.32 Viscosity, ~=(1.32)(0.0129) = 0.0170 cp

Solution by the Thodos Method.

Vllc =CyiV,.i=lO4.5 cm3/g mol.

Viscosity parameter,

(350/1.8)x

= (19.98)“(639/14.7)” =0.0435.

Pseudocritical density,

~-0.1912 g/cm”.

Viscosity factor, p*l (Fig. 20.8)=55x 10m5,

/~*=55xlO-s/0.0435=0.0126 cp.


Density, From Fig. 20.2, zrc = 1.095, zSr =0.998 (probably could have assumed 1 .O), and

zg = 0.876 (Fig. 20.2),

MgP (19.98)(2,012) pg=-= =7.017 lbm/cu ft

z,RT (0.876)(10.73)(610)

Td, 0.005035 x704.6x 1.095 B, =0.005035- =

Pd& 6,000x0.998

=0.00065 RB/scf. =O. 112 g/cm3, and

p,,=O.l12/0.1912=0.58.

Viscosity factor, (p--*)4= 18.9~ lo-’ (from Fig. 20.9).

Viscosity, ~=~*+(~--~*)[I{

=O.Ol26+l8.9x1O-5/O.O435

=0.0169 cp

Results. Carr et al. =0.0170 cp, Thodos et al. =00.0169 cp, and experimental=0.0172 cp.

Conclusion. Excellent results are obtained from either correlation for viscosity of a natural gas.

Example Problem 5. A new discovery in the Lower Tus- caloosa formation produces a gas consisting of 96 % C t and 4% C?. There is no liquid production at the surface. Reservoir conditions are 6,000 psia and 245°F. Calcu- late the gas formation volume factor and the coefficient of isothermal compressibility.

Solution. The pseudocritical pressure and temperature of the mixture are

T,,,.=O.96~343=329.3 +0,04x550= 22.0

=351.3”R.

and

p,,.=O.96~668=641.3 +0.04x708= 28.3

=669.6 psia.

The reduced quantities at reservoir (subscript TC) and surface (subscript SC) conditions are

245 +459.6 VP,),,. =

704.6

351.3 =~=2.00,

352.24

60+459.6 (T,,).,,. = 351 .3 = 1.48,

6,000 (p,r),.=669.6=8.96,

and

From Fig. 20.11, by using Tp, =2.00 and ppr =8.96, we obtain:

Cpr =0.074,

CPr =cp XPpc,

and

Cpr 0.074 CR=-= - =0.0001105

ppr 669.6

=llO.5XlO-6 psi-‘.

By using a computer to calculate the numerical derivative of z with respect to pressure. we get cg = 107.4X lop6 psi-t, which indicates Trube’s correlation to be in error by about 3%.

Example Problem 6. The vapor pressure of pure hexane as a function of temperature is 54.04 kPa at 50°C and 188.76 kPa at 90°C. Estimate the vapor pressure of hexane at lOO”C, using all the methods outlined in the text.

Solution. Clausius-Clapeyron, The Clausius-Clapeyron equation can be solved graphically by plotting log of vapor pressure vs. reciprocal absolute temperature and extrapolating. It also can be solved by slopes.

T, = 50°C [581.67”R],

l/T, = 0.001719,

T2 = 90°C [617.67”R], l/T1 = 0.001619,

p,, at TI = 54.04 kPa=7.8374 psia, logp,, = 0.89417,

p,. at T2 = 105.37 kPa= 15.2826 psia, log p,, = 1.18420,

A log p,, = -0.29003, l/T1 - l/T2 = 0.0001, and

slope =

=

Alogp,. 4

i-’ = >

-0.29003

TI T2 0.0001

-2900.3.

Solving log p v = -2900.3(l/T)+b for h yields

14.7 ~Pp,L,.=~=0.022.

b = 5.87977, T3 = 100”C=671.67”R, and

l/T, = 0.001489.

GAS PROPERTIES AND CORRELATIONS

Solving for pr at 100°C:

log Pl, = -2900.3(0.001489)+5.87977

= 1.56122, and

p,, = 36.4102 psia r251.04 kPa].

However, if you know that the vapor pressure at 70°C is 105.37 kPa, you can use the 70 to 90°C temperature differential to calculate the slopes and ultimately will calculate p,,=35.81 psia=246.7 kPa.

Cox Chart. 2g From Fig. 20.14, the vapor pressure at 100°C can be approximated between 35 and 36 psia. A larger chart is required for more precise readings.

The Calingeart and Davis or Antoine Equation. This can be used by obtaining the Antoine constants from Reid et al. ?’ For n-hexane, with temperature in K, these constants are A= 15.8366, B=2697.55, and C= -48.78. Then,

B lnp,. = A--

T-tC

2697.55 = 15.8366- =3.60223, and

373 -48.78

pY = 36.68 psia [252.73 kPa].

Lee-Kesler. The use of the Lee-Kesler equation requires pr, T,., and w for n-hexane. These can be obtained from Table 20.2.

pc, = 436.9 psia [29.7 atm] T,. = 453.7”F or 913.3”R or 507.4 K, and w = 0.3007.

For lOO”C,

T,. = 0.7351, (T,)6 = 0.15782, In T, = -0.30775,

f’=5.92714-(6.0964810.7351)+1.28862(0.30775)

+O. 169347(0.15782),

and

f’=l5.2518-(15.687510.7351)+13.4721(0.30775)

+0.43577(0.15782).

Therefore,

f o =5.92714-8.29340+0.39657+0.02673

= - 1.94296,

20-17

f’ =15.2518-21.34063+4.14604+0.06877

z-1.87402,

In pvr= -1.94296+0.3007(-1.87402)

= -2.50648,

P VI =p”=O.O816, PL

and

p,=O.O816~29.7=2.4235 atm=35.62 psia

=245.59 kPa.

Experimental Value. 35.69 psia=245.90 kPa. Conclusions. Lee-Kesler gives the best answer, but the

Clausius-Clapeyron method can be even more accurate if the extrapolation is short.

Nomenclature a=

a; = aij = a, =

a(T) = A=

4, = b=

bi = b, =

B= B, =

Bo = c=

constant characteristic of the fluid constant for Substance i mixture parameter Parameter a characteristic functional relationship empirical constant empirical constant constant characteristic of the fluid constant characteristic for Substance i Parameter b for mixture empirical constant gas FVF

cg =

empirical constant

empirical constant coefficient of isothermal

c= compressibility

constant with value of 43 when temperature is in K, and a value of 77.4 when temperature is in “R

d= empirical constant

Do = empirical constant Ek = kinetic energy

E,, = empirical constant

fO,f' = functions of reduced temperature K, = constant for each binary pair when

L,. = m= M=

M, =

MC,+ =

M, =

P’ Pc =

used for mixtures molar latent heat of vaporization mass molecular weight molecular weight of air molecular weight of CT+ fraction

average mole weight of gas mixture absolute pressure critical pressure

20-l 8

PC, = critical pressure of Component i in References gas mixture

Ppc = pseudocritical pressure of gas mixture

P’pc = corrected pseudocritical pressure

Pr = reduced pressure

Pm = pressure at reservoir conditions

Pw = pressure at standard conditions

PI, = vapor pressure

Pw = reduced vapor pressure (vapor

R= t=

pressure/critical pressure) absolute temperature ratio of critical to absolute

temperature T,. =

T,.i = critical temperature critical temperature of Component i in

8

9

gas mixture T,,<. =

T, = T,,. = 7-w. =

v= v=

v,. = (Vc)c. =

v, = v, =

v,,. = v.,, = AV =

corrected pseudocritical temperature reduced temperature temperature at reservoir conditions temperature at standard conditions velocity volume

10

II

critical volume critical volume of CT+ fraction molar volume

12

13

reduced volume volume at reservoir conditions

volume at standard conditions increase in volume while vaporizing

1 mole x; = mole fraction of Component i in

liquid

14

15

16.

17.

YCO! = mole fraction of CO* in mixture

L’H,S = mole fraction of H 2 S in mixture ?‘; = mole fraction of Component i in gas

18.

mixture compressibility factor

compressibility factor at reservoir

19.

20.

conditions

z.\, = compressibility factor at standard 21.

conditions a; =

Pi =

Yg =

Yi =

6j =

E=

CL=

lJ *=

c;=

PM =

P/X =

PR =

CO=

empirical constant for Substance i empirical constant for Substance i specific gravity for gas empirical constant for Substance i empirical constant for Substance i correction factor viscosity correlating parameter viscosity parameter molar density pseudocritical density relative density of CT+ fraction acentric factor

22.

23.

24.

25. 26.

21.

28.

29.


Burnett, E-S.: “Compressibility Determinations Without Volume Measurements,” .I. Applied Mechanics (1936) 3, Al36-40. Katz, D.L. et al.: Hmdbook of Nalural Gas Engineering, McGraw- Hill Book Co. Inc., New York City (1959). Standing, M.B. and Katz, D.L.: “Density of Natural Gases,” Truns., AIME (1942) 146, 140-44. Brown, G.G. ef al. : “Natural Gasoline and the Volatile Hydrocar- bons,” Natural Gas Assn. of America, Tulsa (1948). Wichert, E. and Aziz, K.: “Compressibility Factor for Sour Natural Gases,” Cdn. J. Chem. Eng. (1972) 49, 269-75. vanderWaals,J.D.:Proc.,Acad. Sci.Amsterdam(1901)3.515. Redlich, 0. and Kwong, J.N.S.: “On the ThermodynamicsofSo- lutions. V-An Equation of State. Fugacities of Gaseous Solutions,” Chem. Reviews (1949) 44. 233-44. Soave. Cl.: “Equilibrium Constants from a Modified Redlich-Kwong Equation of State,” Chem. Eng. Sci. (1912) 27, 1197-1203. Zudkevitch, D. and Joffe, J.: “Correlations and Predictions of Vapor-Liquid Equilibrium with the Redlich-Kwong Equation of State.” AfChEJ. (1970) 16, 112-19. Peng. O.Y. and Robinson, D.B.: “A New Two-Constant Equa- bon of State,” Ind. und Eng. Chetn. Fundamcntu/s (1976) 15. No. 1, 59-64. Yarborough, L.: “Application of a Generalized Equation of State to Petroleum Reservoir Fluids,” Equations of Store in En@xvr- ing and Rewurch, K.C. Chao and R.L. Robinson (eds.). Advances in Chemistry Series No. 182, ACS, Washington, DC (1979) 385-439. Starling, K.E. and Han, M.S.: “Thermo Data Refined for LPG.” Hqdrot~urbon Processing (May 1972) 129-33. Benedict, M., Webb, G.B., and Rubin. L.C.: “An Empirical Equa- tmn for Thermodynamic Properties of Light Hydrocarbons and Their Mixtures,” Chem. Eng. frog. (1951) 47.419: and J. Chem. Phy. (1940) 8. 334; (1942) 10, 747. Martin, J.J.: “Cubic Equations of State-Which?” hd Eng. Chrm. Fundmwnta/~ (May 1979), 18, No. 2, 81-97. Robinson, R.L. Jr. and Jacoby. R.H.: “Better Compressibility Fac- tors,” Hydrocarbon Processmg (April 1965) 14 l-45. Hall. K.R. and Yarborough, L.: “A New Equation of State for Z-Factor Calculations.” Oil and Gas J. (June 16. 1971) 71. 82-91. Ajitsaria. N.K.: “Hand Held Calculator Programs Determine Natur- al Gas Physical Properties,” Oil and Cm J. (June 6. 1983) 81. 69-72. Carr. N.L., Kobayashi. R., and Burrows. D.B.: “Viscosity of Hydrocarbon Gases Under Pressure.” Trans.. AIME (1954) 201. 264-78. Stiel. L.I. and Thodos. G.: “The Viscosity of Non-Polar Gases at Normal Pressures,” AIChE J. (1961) 7, 61 l-20. Jossi. J.A.. Stiel. L.I.. and Thodos, Cl.: “The Viscosity of Pure Substances in the Dense Gaseous and Liquid Phases,” AlChE J. (1962) 8, 59-70. Matthews, T.A., Roland, C.H.. and Katz. D.L.: “High Pressure Gas Measurement,” Proc., Natural Gas Assn. of America (1942) 41-51. Trube, A.S.: “Compressibility of Natural Gases,” Trans., AIME (1957) 210. 355-57. Cox, E.R.: “Pressure Temperature Chart for Hydrocarbon Vapors,” Ind. Eng. Chem. (1923) 15. 592-98. Calingeart, G. and Davis, D.S.: “Pressure Temperature Charts Ex- tended Ranges,” lad. Eq. Chem. (1925) 17, 1287~1300. Antoine, C.: Chetn. Reviews (1888) 107, 836-50. Lee, B.I. and Kesler, M.G.: “A Generalized Thermodynamic Car relation Based on Three-Parameter Corresponding States,” AfChE J. (May 1975) 21, 510-27. Reid. R.C., Prausnitz, J.M., and Sherwood. T.K.: T/w Propertitcv of Guse.s andLiquids, 3rd ed.. McGraw-Hill Book Co. Inc., New York City (1977). Perry. J.H. and Chilton. C.H.: Chenzrtzrl Enwineer’.~ Handbook. tifth.edition. McGraw-Hill Book Co. Inc., New York City (1975). GPSA Enginrering Databook, ninth edition. fifth revision. Gas Processors Suppliers Assn., Tulsa (198 1)

Documents

yyifuuyf