Upload
others
View
5
Download
0
Embed Size (px)
Citation preview
Louisiana State UniversityLSU Digital Commons
LSU Historical Dissertations and Theses Graduate School
1996
Improved Method for Selecting Kick ToleranceDuring Deepwater Drilling Operations.Shiniti OharaLouisiana State University and Agricultural & Mechanical College
Follow this and additional works at: https://digitalcommons.lsu.edu/gradschool_disstheses
This Dissertation is brought to you for free and open access by the Graduate School at LSU Digital Commons. It has been accepted for inclusion inLSU Historical Dissertations and Theses by an authorized administrator of LSU Digital Commons. For more information, please [email protected].
Recommended CitationOhara, Shiniti, "Improved Method for Selecting Kick Tolerance During Deepwater Drilling Operations." (1996). LSU HistoricalDissertations and Theses. 6159.https://digitalcommons.lsu.edu/gradschool_disstheses/6159
INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI
films the text directly from the original or copy submitted. Thus, some
thesis and dissertation copies are in typewriter free, while others may be
from any type o f computer printer.
The quality a f this reproduction is dependent upon the quality of the
copy submitted. Broken or indistinct print, colored or poor quality
illustrations and photographs, print bleedthrough, substandard margins,
and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete
manuscript and there are missing pages, these will be noted. Also, if
unauthorized copyright material had to be removed, a note will indicate
the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand comer and
continuing from left to right in equal sections with small overlaps. Each
original is also photographed in one exposure and is included in reduced
form at the back of the book.
Photographs included in the original manuscript have been reproduced
xerographically in this copy. Higher quality 6” x 9” black and white
photographic prints are available for any photographs or illustrations
appearing in this copy for an additional charge. Contact LJMI directly to
order.
UMIA Bell & Howell Information Company
300 North Zeeb Road, Ann Arbor MI 48106-1346 USA 313/761-4700 800/521-0600
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
IMPROVED METHOD FOR SELECTING KICK TOLERANCE DURING DEEPWATER DRILLING OPERATIONS
A Dissertation
Submitted to the Graduate Faculty of the Louisiana State University and
Agricultural and Mechanical College in partial fulfillment of the
requirements for the degree of Doctor of Philosophy
in
The Department of Petroleum Engineering
byShiniti Ohara
B.S., Univcrsidadc Estadual de Campinas, Brazil, 1979 M.S., Univcrsidadc Estadual de Campinas, Brazil, 1989
May 1996
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
UMI Num ber: 96 2 8 3 1 3
UMI Microform 9628313 Copyright 1996, by UMI Company. AH rights reserved.
This microform edition is protected against unauthorized copying under Title 17, United States Code.
UMI300 North Zeeb Road Ann Arbor, MI 48103
Reproduced with permission of the copyright owner. Further reproduction prohibited w ithout permission.
ACKNOWLEDGMENTS
The author is deeply indebted to Dr. Adam “Ted” Bourgoyne, Campanile
Charities Professor o f Offshore Mining and Petroleum Engineering, under whose
valuable guidance, supervision, and encouragement this work was accomplished. Sincere
appreciation is extended to Dr. Andrzej Wojtanowicz, Dr. Julius P. Langlinais. Dr.
Ronald F. Malone, Dr. William J. Bernard, and Dr. Zaki Bassiouni for serving on the
dissertation committee.
The author extends his deepest thanks to Petroleo Brasileiro S.A. (Petrobras) for
providing him the financial support to attend the doctoral program at LSU. Furthermore,
this research was financed by Petrobras through funds from its Technological
Development Program on Deepwater Production System — PROCAP 2000. Sincere
appreciation is due to Dr. Edson Y. Nakagawa, Mr. Saulo Linhares, Mr. Andre Barcelos.
and Mr. Djalma R. de Souza of Petrobras for their firm support throughout this project.
Special thanks are also due to Mr. O. Allen Kelly and Mr. Richard Duncan of
PERTTL--LSU for their attention, ideas, and support during the experimental phase of
this project Acknowledgments are extended also to Mr. Jason Duhe, Mr. Bryant
LaPoint, Mr. Eddy Walls, Jr., and Mr. Ben Bienvenu for their invaluable help during the
experimental work. Mrs. Jeanette Wooden is thanked for her kind words of
encouragement. The assistance of Mrs. Brenda Macon for proofreading the manuscript is
gratefully acknowledged.
The author thanks the following persons for donating or lending equipment and
material that were crucial to accomplish the experimental work: Mr. Mike Strout. Mr.
ii
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
Richard Kriege, and Mr. Nicholas Lavolpicella. Geophisical Research Corp.; Mr. Rickie
Menard, Francis Drilling Fluids, Ltd.; Mr. Donald LeJeune, Drillogic, Inc.; Mr. Gene Lee
and Mr. Ed Taupier, Daniel Flow Products, Inc.; Mr. Eurphy Lantier, Production
Wireline Services, Inc.; Mr. George Murphy, SWACO; Mr. James Pontiff, Halliburton
Energy Services; and Mr. Huey Bertrand, Cameron.
Finally, the author dedicates this work to his mother Julia for her constant moral
support, to his wife, Neuza. and to his children. Sara and Fabio for their support and love.
hi
Reproduced with permission of the copyright owner. Further reproduction prohibited w ithout permission.
TABLE OF CONTENTS
ACKNOWLEDGMENTS................................................................................................... ii
LIST OF TABLES............................................................................................................ vi
LIST OF FIGURES.......................................................................................................... vi
NOMENCLATURE....................................................................................................... xi
ABSTRACT...................................................................................................................... xvi
CHAPTER
1. INTRODUCTION............................................................................................. 1
2. LITERATURE REVIEW................................................................................. 102.1 Kick Tolerance.......................................................................................... 102.2 Kick Simulators......................................................................................... 172.3 Two-Phase Flow Through Annular Section............................................ 20
3. CIRCULATING KICK TOLERANCE MODEL............................................ 283.1 Wellborc Model......................................................................................... 28
3.1.1 Continuity Equations...................................................................... 293.1.2 Momentum Balance Equation....................................................... 293.1.3 liquations of State........................................................................... 31
3.2 Gas Reservoir Model................................................................................ 323.3 Choke Line Model..................................................................................... 353.4 Upward Gas Rise Velocity M odel............................................................ 353.5 Solution of the Differential Equations.................................................... 363.6 Simplification of the Differential Equations System.............................. 39
4. COMPUTER PROGRAM AND RESULTS OF NUMERICALSIMULATIONS................................................................................................ 424.1 Computer Program..................................................................................... 424.2 Results from a Typical Deep Water Drilling Experience........................ 444.3 Comparison with Commercial Kick Simulator....................................... 474.4 Selecting Kick Tolerance........................................................................... 53
4.4.1 Selecting Kick Tolerance for Well Design.................................... 544.4.2 Selecting Kick Tolerance while Drilling....................................... 55
5. EXPERIMENTAL WORK.............................................................................. 575.1 Description of a Full-Scale Well: LSU No. 2 ......................................... 575.2 Methodology of Experimentation............................................................. 595.3 Instrumentation of the Well...................................................................... 62
iv
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
5.4 Methodology Used to Measure Gas Rise Velocities................................ 65
6. EXPERIMENTAL RESULTS.......................................................................... 686.1 A Typical Experiment................................................................................. 686.2 Zuber - Findlay P lo t.................................................................................... 74
7. CONCLUSIONS AND RECOMMENDATIONS........................................... 777.1 Conclusions.................................................................................................. 777.2 Recommendations for Future Work............................................................ 78
7.2.1 Gas Distribution Profile.................................................................... 787.2.1.1 The Triangular Gas Distribution Profile............................. 797.2.1.2 Triangular Gas Distribution Velocities............................... 83
7.2.2 Improvement in the Gas Flow Out Measurements.......................... 857.2.3 Modification for Inclined W ell......................................................... 857.2.4 Instrumentation of a Real W ell......................................................... 85
REFERENCES.................................................................................................................. 86
APPENDIX A. SHUT-IN KICK TOLERANCE............................................................ 92A.l Maximum Shut in Casing Pressure (SICP)............................................... 92A.2 Kick Tolerance............................................................................................ 93A.3 Safety Factor and Surge Gradient............................................................. 95
APPENDIX B. INPUT DATA FOR KICK TOLERANCE PROGRAM..................... 97B.l Input Data for a Typical Deepwater W ell................................................. 97B.2 Input Data for the Well RJS - 4 57 ............................................................ 100B.3 Input Data for the Well CES - 112.......................................................... 103
APPENDIX C. GAS DISTRIBUTION PROFILE......................................................... 106
VITA.................................................................................................................................. 152
v
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
LIST OF TABLES
2.1 Effect of trip and safety margin on the kick tolerance.................................... 11
2.2 Safety factors used in the Beaufort Sea ........................................................... 12
2.3 Kick tolerance, alternate levels and procedures for drilling in theBeaufort S ea ..................................................................................................... 13
4.1 Typical casing setting used for a deep-water well in Campos Basin 45
5.1 Test matrix for water and natural gas experiments........................................ 60
5.2 Test matrix for mud and natural gas experiments with gas injectedthrough tubing.................................................................................................. 60
5.3 Test matrix for mud and natural gas with sensors 1,200 ft apart................. 61
5.4 Test matrix for mud and natural gas with sensors 100 ft apart.................... 61
5.5 Drilling fluid properties utilized in the experiments...................................... 62
5.6 Gas flow out calculations for gas measurements system.............................. 65
7.1 Front and center velocities for different circulation...................................... 83
A. 1 Safety factors used in the Beaufort Sea.......................................................... 96
C. 1 Test matrix for mud and natural gas experiments with gas injectedthrough tubing.................................................................................................. 106
C.2 Test matrix for mud and natural gas with sensors 1,200 ft apart................. 106
C.3 Test matrix for mud and natural gas with sensors 100 ft apart.................... 107
C.4 Drilling fluid properties used in the experiments........................................... 107
vi
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
LIST OF FIGURES
1.1 Deep-water drilling world records.................................................................... 2
1.2 Deep-water drilling activities for water depth greater than 400 m .................. 3
1.3 World record for subsea completions............................................................... 4
1.4 Kick Tolerance for deep-water.......................................................................... 7
2.1 Effect o f depth and kick volume on kick tolerance.......................................... 11
2.2 Pump-rate requirements and equipment lim its................................................. 16
2.3 Zuber-Findlay plot for flow-loop experiments................................................. 24
3.1 Finite difference scheme for a cell.................................................................... 37
3.2 Flowchart of the complete program.................................................................. 40
3.3 Flowchart of the simplified program................................................................ 41
4.1 A typical well design for deep water drilling in Campos Basin...................... 46
4.2 Kick tolerance, casing pressure, and fracture pressure at casing depthfor a typical deep-water w ell............................................................................ 48
4.3 Pit volume, bottom hole pressure, and drill pipe pressure for a typicaldeep-water w ell................................................................................................. 49
4.4 Gas flow rate and gas leading edge depth for a typical deep-water w ell 50
4.5 Kick tolerance for the well RJS - 457 ............................................................... 51
4.6 Kick tolerance for the well CES - 112.............................................................. 53
5.1 LSU No. 2 well completion schematic............................................................. 58
5.2 Gas flow out measurements system.................................................................. 64
5.3 Downhole pressure sensors disposition............................................................ 66
6.1 Example of downhole pressure data................................................................. 69
6.2 Example of gas flow rates and drill pipe pressure............................................ 70
vii
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
6.3 Example o f casing pressure, pit volume, and pump speed................................ 71
6.4 Differential pressure between on-line and casing and between downholeand pressure recorders......................................................................................... 72
6.5 Differential pressure between top and on-line sensors and between bottomhole pressure and bottom sensor....................................................................... 73
6.6 Zuber - Findlay plot of experimental data ......................................................... 75
6.7 Zuber - Findlay plot of the present and previous flow loops experiments .... 76
7.1 Gas fraction between sensors for different depths and times forexperiment M 1..................................................................................................... 80
7.2 Gas fraction profile as a function of depth for various tim es............................ 81
7.3 Proposed triangular gas distribution profile....................................................... 82
7.4 Gas velocity profile for various liquids velocities............................................. 84
C.l Gas fraction for different depths and times for experiment Ml ........................ 108
C.2 Gas fraction as a function of depth for experiment Ml ..................................... 109
C.3 Gas fraction for different depths and times for experiment M6...................... 110
C.4 Gas fraction as a function of depth for experiment M 6.................................... 111
C.5 Gas fraction for different depths and times for experiment M 7...................... 112
C.6 Gas fraction as a function of depth for experiment M 7....................................113
C.7 Gas fraction for different depths and times for experiment M8........................ 114
C.8 Gas fraction as a function of depth for experiment M 8.................................. 115
C.9 Gas fraction for different depths and times for experiment M 9..................... 116
C.10 Gas fraction as a function of depth for experiment M 9................................... 117
C.l 1 Gas fraction for different depths and times for experiment M il ..................... 118
C.l 2 Gas fraction as a function of depth for experiment M il .................................. 119
viii
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
C. 13 Gas fraction for different depths and times for experiment M 1 2 ................... 120
C.14 Gas fraction as a function of depth for experiment M12.................................. 121
C. 15 Gas fraction for different depths and times for experiment M 1 3 ................... 122
C. 16 Gas fraction as a function of depth for experiment Ml 3................................. 123
C.17 Gas fraction for different depths and times for experiment M14.................... 124
C. 18 Gas fraction as a function o f depth for experiment M 14................................ 125
C. 19 Gas fraction for different depths and times for experiment M 1 5 .................. 126
C.20 Gas fraction as a function o f depth for experiment M15................................. 127
C.21 Gas fraction for different depths and times for experiment M l6 .................. 128
C.22 Gas fraction as a function of depth for experiment M l 6 ............................... 129
C.23 Gas fraction for different depths and times for experiment M 17 ................. 130
C.24 Gas fraction as a function of depth for experiment Ml 7 ............................... 131
C.25 Gas fraction for different depths and times for experiment M 1 8 .................. 132
C.26 Gas fraction as a function of depth for experiment M 18............................... 133
C.27 Gas fraction for different depths and times for experiment M l9 .................. 134
C.28 Gas fraction as a function of depth for experiment M l9................................ 135
C.29 Gas fraction for different depths and times for experiment M 20.................. 136
C.30 Gas fraction as a function of depth for experiment M 20............................. 137
C.31 Gas fraction for different depths and times for experiment M 21.................. 138
C.32 Gas fraction as a function of depth for experiment M 21................................ 139
C.33 Gas fraction for different depths and times for experiment M 22 .................. 140
C.34 Gas fraction as a function of depth for experiment M 22............................... 141
ix
Reproduced w ith permission o f the copyright owner. Further reproduction prohibited w ithout permission.
C.3 5 Gas fraction for different depths and times for experiment M 23................ 142
C.36 Gas fraction as a function of depth for experiment M 23............................. 143
C.3 7 Gas fraction for different depths and times for experiment M 24................ 144
C.38 Gas fraction as a function of depth for experiment M 24.............................. 145
C.39 Gas fraction for different depths and times for experiment M 27.................. 146
C.40 Gas fraction as a function of depth for experiment M 27................................ 147
C.41 Gas fraction for different depths and times for experiment M 28.................. 148
C.42 Gas fraction as a function of depth for experiment M 28................................ 149
C.43 Gas fraction for different depths and limes lor experiment M 29................... 150
C.44 Gas fraction as a function of depth for experiment M 29............................... 151
x
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
NOMENCLATURE
Roman Letters
an cross sectional area o f annulus
cr formation compressibility
CK ~ gas compressibility
c„ = gas distribution factor
total compressibility
d = distance
D = turbulence or non-Darcy factor
true vertical depth of hole
Dr true vertical depth of weakest formation
3“ II fracture depth or casing depth
4 ,= diameter of hole
d = diameter of drill pipe
dpjdt = shut-in pressure rise rate (choke pressure)
e = ratio of the two-phase slip to no-slip friction factor
f f = no-slip Fanning friction factor
two-phase flow friction factor
g = gravitational acceleration
Xc = conversion factor
h = permeable zone thickness
H = liquid holdup
Reproduced w ith permission of the copyright owner. Further reproduction prohibited w ithout permission.
h,e = gas leading edge height
h,pf = height o f two phase flow
J = productivity index
k = permeability
K = kick tolerance
K = circulating kick tolerance
Lk = true vertical length of influx
M = gas molecular weight
m(p) = real gas pseudopressure
p = pressure
ph = bottom pressure
phh = bottom-hole pressure
P, = choke pressure
/ ’/, = dimensionless pressure
Pf = formation pore pressure
Psc = pressure at surface condition (standard condition)
P s = safety factor pressure
p, = top pressure
O ~ gas flow rate
qt. = average filtrate loss rate to formation
= gas flow rate
q = gas flow rate at standard conditions
X ll
Reproduced w ith permission o f the copyright owner. Further reproduction prohibited w ithout permission.
qi = liquid flow rate
R = gas constant
ROP = rate of penetration
rw = wellbore radius
S = skin effect
iSICPmaj. = maximum shut-in casing pressure
Swi = initial water saturation
t = time
T= temperature
tD = dimensionless time
Tsc = temperature at surface condition (standard condition)
vcenter = volume centered gas velocity
Vj= fluid loss volume
vfront= gas front velocity
vg = mean gas velocity
vgs = superficial gas velocity
Vk~ influx volume
V/ = liquid velocity
vh= superficial liquid velocity
vm = mixture or homogeneous velocity
Vm = mud volume
vm£ry= front mixture velocity
xiii
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
vmixt = tail mixture velocity
Vol = volume
vs= slip velocity
vtaii= g3̂ tail velocity
Vw = wellbore volume
Xk = influx compressibility
Xm = mud compressibility
Xw = wellbore elasticity
YP = yield point
z = gas compressibility factor
Greek letters
a = gas fraction
Pg = velocity coefficient
X = no-slip liquid holdup or input liquid content
<}> = formation porosity
p. = viscosity
p.y= formation fluid viscosity
pg= gas viscosity
p = equivalent mud density
py = fracture gradient expressed in equivalent mud density
pg = density of gas
p̂ . = density of fluid influx
xiv
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
Pz = density o f Liquid
pm = density o f drilling fluid or mud
pnj = two-phase no-slip density
pp = pore pressure expressed in equivalent mud density
psf= safety factor expressed in equivalent mud density
pjg = surge pressure expressed in equivalent mud density
pt = trip margin expressed in equivalent mud density
dp/ dz = gradient pressure
(dp/ dz)elev = gradient pressure due to elevation
(dp/ dz)fric = gradient pressure due to friction
Subscripts
an = annulus
bh = bottom hole
Dp = differential pressure
frac - fracture
max= maximum
min = minimum
res = reservoir
sc = standard conditions
stab= stabilized
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
ABSTRACT
One of the most critical aspects in the design of oil and gas wells is the selection
o f the depths at which steel casing is set. As the length of open borehole increases, the
risk o f formation fracturing during drilling operations increases. Formation fracture often
leads to an underground blowout that can be very expensive to control. Because of the
special problems involved in drilling deepwater well, accurately measuring the risk of
formation fracture is essential. A calculated parameter called “kick tolerance” is often
used to measure this risk.
In this study, improved computer software specifically designed for computing
kick tolerance for wells drilled in deep waters was developed. During well design, the
software can be used to confirm previously calculated casing setting depths. The software
can also be used during drilling to estimate the fracture risk of the weakest exposed
formation if a kick was taken and circulated. If an unacceptable fracture risk is indicated,
drilling can be interrupted and the casing string can be set earlier. The developed
computer program has been proven to be fast, reliable, and suitable for available rig site
computers. The accuracy achieved was similar to that obtained using commercially
available well control simulators that are much more time consuming to run. The
availability of this simulator may result in safer drilling operations and improved
capability for drilling in deeper water depths.
Experiments were performed using a drilling fluid and natural gas in a 6,000 ft
research well to verify and improve previously published empirical correlations for gas
rise velocities. An empirical correlation relating the gas velocity to the sum of the average
xvi
Reproduced w ith permission o f the copyright owner. Further reproduction prohibited w ithout permission.
mixture velocity and the relative slip velocity was determined using both the available
data from previous flow-loop experiments and data from the present experiments. This
correlation was used in the new computer software. The experimental data may also
allow additional improvement to be made in the accuracy of the kick tolerance
calculation in the future. Investigation of a triangular gas distribution profile along the
path of upward gas migration is proposed as a future area o f study.
XVll
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
CHAPTER 1
INTRODUCTION
Geologists have long believed that significant hydrocarbon accumulations exist at
deep-water locations. However, these locations have been left unexplored until recently
because they were not considered to be economically and technically viable. Deep-water
exploration and development concepts have changed over the years. For example, in the
early sixties the exploration and development of offshore hydrocarbons were restricted to
46 m (150 ft) by the physical and economic limitations of bottom-supported drilling and
producing rigs. The major concern was to overcome this 46 m limit, which was
considered a deep-water location at the time.
During the oil crisis o f 1973, the oil price jumped from $2.00 to $11.00 per barrel.
A second oil price shock occurred in 1979 when the oil price reached $30.00 per barrel.
Motivated by the improved economics of oil exploration that was brought about by these
oil crises, the oil industry began searching for hydrocarbons in deeper water, as can be
seen in Figure 1.1. Deepwater technology has advanced from moored semi-submersibles
to today’s advanced dynamic positioned (DP) vessels. Today, wells drilled in water
depths over 400 m (1,312 ft) are considered to be deep-water wells. This depth
corresponds to the maximum depth that a human being can dive using saturation
techniques. Beyond this depth ROVs (remotely operated vehicles) are used to service the
well heads on the sea floor, including those of ultra-deep-water wells, which are
considered to be wells in water depths of 1,000 m (3,281 ft) or more. The current world
record for deep-water drilling is held by Shell for a well at the Mississippi Canyon 657 #2
drilled in 1988 in a water depth of 2328.1 m (7,638 ft).
1
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
1
Year68 70 72 74 76 78 80 82 84 86 88 90
iI
S 500 -I
1000 - -
3 1500
2000 -
2500
Figure 1.1 Deep-water drilling world records
Brazil is now one of the most active countries in deep-water drilling and
producing. The deep-water drilling program using dynamically-positioned (DP) units
began in 1985, when nine wells were drilled in water depths of more than 400 m (1,312
ft). In 1992, 51 deep-water wells were drilled, as shown in Figure 1.2.
The deep-water drilling activities in Brazil were intensified by the discovery of a
giant field, Albacora, in September of 1984 with the wildcat well l-RJS-297. Albacora
field, located in Campos Basin (Southeast Brazil) in water depths ranging from 293 m
(755 ft) to 1,900 m (6,234 ft), has an estimated oil-in-place volume of 4.4 billion barrels
2over an area of 235 km' (90 mi').
Marlim, another giant field, was discovered in 1985 when the well 1-RJS-219A
was drilled at a water depth of 853 m (2,797 ft). Marlim field is also located in Campos
Basin in water depths ranging from 600 m (1,967 ft) to 1,050 m (3,445 ft). The total
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
reserves (recoverable oil volume) for this field are estimated to be 1.5 billion barrels of
2 2oil (6.6 million of oil-in-place) over an area of 132 km (51 mi ).
90 ---------------------------------------------
80-!-! □ World (excluding Brazil) B Brazil
| 70 -h
i 60 t03 j
1 50 to 40 -U0
| 30 T1 iz 2 0 |
10 -
\ /
\ A
70 72 74 76 78 80 82 84 86 88 90 92 94YearSource: Petrobras E&P
Figure 1.2 Deep-water drilling activities for water depth greater than 400 m
Brazilian offshore exploration was not limited to the Albacora and Marlim fields.
Prospecting in the Campos Basin soon pinpointed the Barracuda, Bijupira and Salema
fields with reserves of 106,43, and 13 million barrels of oil, respectively, in water depths
ranging from 400 m (1,312 ft) to 1,000 m (3,281 ft). In addition, other deep-water
prospects are currently being drilled outside the Campos Basin and may also reveal new
deep-water fields.
The new challenge after these discoveries was overcoming the technological
barriers involved in producing these deep-water fields. Offshore production using fixed
platforms in Brazil started in the shallow water of the Sergipe/Alagoas Basin (Northeast
Reproduced w ith permission of the copyright owner. Further reproduction prohibited w ithout permission.
4
Brazil) in 1968. Nine years later, floating production system (FPS) technology was used
for the first time in Brazil to bring Enchova field on stream. The simplicity o f the FPS
reduced the lead-time needed to bring this field into production. The next step was the
application of FPS for field development using subsea completion techniques. The first
subsea completion in Brazil was performed in 1979 in a water depth of 189 m (629 ft).
Since then the world water depth record for subsea completion has been repeatedly
broken over a short period of time, culminating in the current subsea world record of
1,027 m (3,370 ft) established in May 1994 with the completion of well 3-MRL-4 in
Marlim field, as shown in Figure 1.3. This world record may be broken again in 1997 in
the Mensa field, where the water depth is 1,646 m ( World Oil, July 1995). Currently, of
all subsea trees installed worldwide, one third have been installed in Brazil.
Year
78 80 82 84 86 88 90 92 94 96 98
j f 1000: ♦ Petrobras
o Placid
gS hell (expected in 1997)
' 1500 -
2000
Figure 1.3 World record for subsea completions
Deepwater drilling and production are now a reality. However, deep water
drilling poses special problems, such as low fracture gradients, high pressure loss in
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
5
choke lines, overbalanced drilling due to a riser safety margin, and emergency riser
disconnection problems. As operators search for hydrocarbons in deeper waters, key
factors for successfully drilling deep-water wells are to have (a) a detailed well design
and drilling plan and (b) a close control while drilling to avoid kicks, loss of circulation,
and a possible underground blowout, which can be especially costly. Therefore, special
care must be used when planning and drilling these wells. The kick tolerance concept is a
powerful tool that can be used during well design, along with the pore pressure and
fracture gradients, to determine depths at which casing should be set. In addition, kick
tolerance can be used during drilling to estimate the fracture risk of the weakest exposed
formation. If a kick is taken and circulated, break down of this formation could lead to an
underground blowout. This parameter can be used to stop the drilling and run the casing
string and to regulate drilling activities by governmental regulatory agencies, such as the
US Mineral Management Service.
Even though kick tolerance has been used in the drilling industry, the concept has
been controversial (Redman, 1991). Much confusion can be credited to the original
definition: “a difference between formation pressure and mud weight in use (expressed
as mud weight equivalents) against which the well could be safely shut in without
breaking down the weakest formation.” According to Redman, much confusion is also
credited to the term “zero gain,” which is either misunderstood or omitted entirely.
Another accepted definition is “kick tolerance is the maximum increase in mud weight
allowed by the pressure integrity test of the casing shoe with no influx (zero gain) in the
wellbore.” Often the zero pit gain condition is omitted. For example, with a pressure
Reproduced w ith permission o f the copyright owner. Further reproduction prohibited w ithout permission.
integrity test result of 1.68 gr/cm3 (14 lb/gal) at the casing shoe and a mud density of 1.20
gr/cm3 (10 lb/gal), many may consider that they are secure because they have a kick
tolerance o f 0.48 gr/cm3 (4 lb/gal). This is only true if no influx (zero pit gain) occurs, but
generally a kick is detected by the pit gain (increase of volume in the mud pits). As a
result, kick tolerance decreases as kick volume and depth increase.
Kick tolerance is calculated assuming that natural gas (worst case) is the kick
fluid. Another extremely important assumption is the maximum pit gain that would be
expected before the blowout preventers are closed. The maximum pit gain used in the
calculation is critical and must be appropriate for field operating practices,
instrumentation, and rig crew training. Shut-in kick tolerance applies to well conditions
when the well is shut in. Circulating kick tolerance applies to the most severe conditions
expected during the well control operations to remove the kick fluids from the well. The
circulating kick tolerance can easily be calculated as a simple model which assumes that
the influx of gas enters as a slug and remains a slug during the circulation. This simple
model, although easy to calculate, is very conservative if compared with a modem kick
simulator, as shown in Figure 1.4.
However, calculation of kick tolerance using an existing commercial kick
simulator can be very time consuming. For example, it took almost one full day to
calculate the five points used to draw the upper curve in Figure 1.4. Furthermore, existing
kick simulators are known to fail in many deep-water drilling situations (Negrao, 1995).
Although time consuming, using a kick simulator rather than a simplified “slug” model
to calculate kick tolerance in this well saved around SI00,000 in drilling costs. Thus, the
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
development of a more realistic, reliable, and much faster kick simulator dedicated to
calculate kick tolerance for use in well planning and while drilling at a deep-water
location has motivated the present research.
11.3 ,
11.2Kick S im ula to r
11.0Sim plified Model10.9V5 CJ
=* 10.8c>I 10.7
10.6 -
After Nakagawa and Lage, 1994
30 400 10 20 50 60Pit gain (bbl)
Figure 1.4 Kick Tolerance for deep-water
The concept of kick tolerance is more complex in deep water drilling because
dynamic positioned drilling ships (DPDS) are used, and normally a riser safety margin is
applied to avoid a potential loss of hydrostatic pressure due to an emergency
disconnection and BOP failure. Depending on water depth, leak-off test results, and pore
pressure, the riser margin cannot always be applied because of the risk of formation
fracture. The kick tolerance value can be near zero or even negative in this case without
implying a dangerous situation. Another important factor in deep-water is the high
pressure loss, which was considered in the proposed kick simulator, in the long subsea
flow lines.
A computer model is proposed in this research to calculate circulating kick
tolerance. The model is based on: a) mass-balance equations (continuity equations) for
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
8
the mud and gas; b) a momentum-balance equation for the gas-mud mixture; c) equations
of state for mud and gas; and d) a correlation relating the gas rise velocity in the annulus
to the average mixture velocity plus the relative slip velocity between mud and gas.
Developing a more accurate circulating kick tolerance calculation procedure
requires the determination of a correlation for the gas rise velocity in the annulus. Many
studies have been performed in this area using flow loops or a real well and using mud,
Xantham gum, or water as a liquid phase and air, nitrogen, or argon gas as a gas phase.
These studies were used to develop empirical methods for computing gas slip velocity
and gas concentrations in well control operations. The empirical gas slip correlation used
in the new circulating kick tolerance simulator is based on this previous work.
Despite these previous studies, no single experiment was made with the
combination o f real well conditions, drilling fluids, and natural gas. The previous studies
concentrated on the bubble front velocity, but how the shape of the gas fraction
distribution profile will change with time during the gas migration is still unknown. Also,
since the velocity of the gas behind the two-phase interface, or tail velocity, is low, its
volume along the well can be considerable. Consequently, these unknowns have
motivated the present experimental works to determine these velocities and distribution
profiles.
In summary, a kick simulator that is dedicated to calculate kick tolerance for deep
water drilling has been developed. The developed software has been proven to be fast,
reliable, and suitable for available rig site computers. Experiments were performed in a
6,000 ft research well, using a drilling fluid and natural gas, to verify and improve
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
9
previous published empirical correlation for gas rise velocities. An empirical correlation
relating the gas velocity to the average mixture velocity plus the relative slip velocity was
determined using the available data from previous flow-loop experiments combined with
data from the present experimental work. The experimental data may also allow
additional improvement to be made in accuracy of the kick tolerance calculation in the
future. Investigation of a triangular gas distribution profile along the path of upward gas
migration is proposed as a future area o f study.
Reproduced w ith permission o f the copyright owner. Further reproduction prohibited w ithout permission.
CHAPTER 2
LITERATURE REVIEW
2.1 Kick Tolerance
Shut-in kick tolerance is defined as the difference between mud weight in use
and formation pressure (expressed as mud weight equivalents) against which the well
could be safely shut-in without breaking down the weakest formation. Circulating shut-in
kick tolerance applies to the most severe conditions expected during the well control
operations that will allow the removal of the kick fluids from the well. The shut-in kick
tolerance can be defined as:
k , = p , - p . = | K p ( - p . ) - ^ - A ( p . - p ( ) - p , (21)
Appendix A presents a derivation of Equation 2.1.
Pilkington and Niehaus (1975) compared the effects of safety margin, trip margin,
and fluid influx on the kick tolerance using the formula:
V (p / _ Ptf — Pm)P/ Pm(^* — Lk )+ p* LkK- -------- D— — + — — 5 ---------------- P - - P .
" u >> (2.2)
The data used for comparison is:
Casing set at: 4,000 ft
Drilling at: 12,000 ft
Fracture gradient: 13.5 lb/gal (at 4,000 ft)
Mud weight: 10.0 lb/gal
Pilkington and Niehaus comparison is summarized in Tabic 2.1.
10
Reproduced w ith permission of the copyright owner. Further reproduction prohibited w ithout permission.
11
Table 2.1 Effect of trip and safety margin on the kick tolerance
Ignoring trip margin and fluid influx
Accounting for trip margin with no influx
Accounting for trip margin and influx
(42 bbl)safety margin safetv maruin safety margin
with (0.5 lb/gal)
without with(0.5 lb/gal)
without with (0.5 lb/gal)
without
1.2 1.3 1.1 1.1 0.7 0.8
Pilkington and Niehaus showed that the kick tolerance decreases with depth and
with increased kick volume, as shown in Figure 2.1.
4000
5000
6000
~ 7000 a£ 8000 ST“ 9000
10000
11000
120000 1 2 3 4
Kick tolerance (lbm/gal)
Figure 2.1 Effect of depth and kick volume on kick tolerance
Pilkington and Niehaus concluded that the fracture gradient gives operators a
false sense of security. In addition, the fracture pressure, and not a mud weight
equivalent, at the shoe is the critical factor in well planning because the fracture pressure
------------------ 1■
■1
1---------------0•
A A
----------------A 1
■VI A
V M
• ▲ «
■ •H A
4 *
wm■ • i
A 1A
0 (A fierP i Ik ington and
■
w m
• MNiehauj . 1975)
AH I03bblinflux
0 42bblinflux
▲ 8.5bblinflux
♦ Obblinflux
■as m
B J+■B V
■ •M A
JKr
*
■ ■ •
B r
W ' ' L
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
12
(at the casing shoe) minus the hydrostatic pressure o f the mud column (at the casing
shoe) is the maximum surface pressure that can be tolerated.
A practical form of kick tolerance was used to drill in the Canadian Beaufort Sea,
which has high abnormal pressure and an unconsolidated formation, as well as
permafrost, gas hydrates, and plastic shale (Wilkie and Bernard, 1981). The problems
associated with this location made the optimum setting of the casing string critical. As a
result, the safety factor was redefined as a function of depth and expressed in pressure.
Moreover, a surge gradient factor was introduced into the calculation of kick tolerance. A
surge gradient was created on restarting the mud pumps, after the well was shut in, to
read the drill pipe and casing pressures. The formula used to drill in the Beaufort Sea is:
( ^ 0 , - 1 0 1 . 9 4 ^ ) p . ( L , + Df )
A A (2.3)
where the surge gradient is defined as:
5.33x103Y P D f 101.94
Pn ( < w , ) " ' V
Their proposed safety factors (Psf ) are shown in Table 2.2.
Table 2.2 Safety factors used in the Beaufort Sea
(2.4)
Below Casing Safety Factor(mm) (inches) kPa psi406 16 225 33340 13 3/8 345 50244 9 5/8 690 100
In addition, Wilkie and Bernard proposed procedures to be used while drilling, as shown
in Table 2.3.
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
13
Table 2 3 Kick tolerance, alternate levels and procedures for drilling in the BeaufortSea
Level 1 Level 2 Level 3Kick Volume Vk = 4 nr* (25 bbl) Vk = 2.8 m3 (17.5 bbl) Vk=1.6m 3(10bbl)Kt Greater than zero Greater than zero Greater than zero1. General safety
(a) BOP drills Weekly (each crew) Weekly (each crew) Each tour
(b) Dog house safety meeting
As required Each tour Each tour; written instruction
(c) Drilling rate By cuttings in hole (3.c) By cuttings in hole (See 3.c) < 9 meter per hour(d) Tripping speeds (casing and open hole)
Calculate for each trip based on swab/surge
Calculate for each trip based on swab/surge
Calculate for each trip based on swab/surge
(e) Barite plug preparation
Pilot test; review procedures; measure chemicals
Pilot test; review procedures; measure chemicals
Prepare mix water; line out cement unit
(f) Weather/ice conditions
Normal forecasts Favorable forecast 24 hr Favorable forecast 48 hr
2. Kick detection(a) Active pit volume
Normal Reduced Minimum
(b) PVT (while circulating)
Sensitivity +/- 1.6 m3 Sensitivity +/-1.1 m3 Sensitivity +/- 0.6 m* Man on pits continuously
(c) On drilling breaks
Flow check Flow check Shut in well
(d) Hole fill procedures
Follow normal hole fill/trip record procedures
Follow normal hole fill/trip record procedures
Supervisors check procedures and records during trips
(e) Mud weight Check every 1 hour* Check every 30 min* Check every 15 min*
(f)Communications
Normal Open from mudlogger to floor
Open from mudlogger to floor
* If mud weight out drops more than 36 kg/m water cutting
, flow check and check mud properties for possible
3. Pressuredetection
(a) General procedures
Observe normal** indicators; report significant trends
Observe all indicators; report significant trends
Observe all indicators; report all trends
(b) Gas units i) Calibrate daily i) Calibrate on each tour i) Calibrate every 4 hours
ii) Run degasser if necessary
ii) Run degasser to check response
ii) Run degasser
iii) Observe and report trends
iii) Limit max gas units iii) Limit max gas units
(c) Cuttings in hole Less than 30 m Less than 18 m Less than 9 m(d) Wireline logs At casing point Approx. every 762 m or as
required for overpressure confirmation (wellsite team recommendation)
Approximately every 305 m
table cont.
Reproduced w ith permission of the copyright owner. Further reproduction prohibited w ithout permission.
14
Level 1 Level 2 Level 3Kick Volume Vk = 4 m ' (25 bbl) Vk = 2.8 m'’ (17.5 bbl) Vk =1.6 n f (10 bbl)(e)Dummyconnections
As required As required As required; every 5 m if increasing pore pressures are indicated
Kick tolerance Greater than zero Greater than zero Greater than zero4. O ther measures
(a) On tripping Flow check after first 5 stands, at sl.oe and before pulling collars into BOP stack.
Flow check every 5 stands, at shoe and before pulling collars into BOP stack.
Consider increasing mud weight for tripping.Flow check every 5 stands, at shoe and before pulling collars into BOP stack.
(b) Short trip (dummy trip)
As dictated by hole conditions
As dictated by hole conditions
Make 5 stands short trip and circulate bottoms up before tripping out of hole.
** Mud gas units, penetration rate, ‘d ’ - exponent or equivalent
Chenevert (1983) presented a microcomputer program to calculate the kick
tolerance. The formula used in this program is similar to Equation 2.1 but without the
safety factor and trip margin gradients.
During the well control process, calculating the pressure of the influx fluid when
it reaches the casing shoe is desirable. Using the "driller's method," which employs the
existing mud weight to remove the influx from the well, Redmann (1991) proposed an
iterative process to calculate the top of the influx pressure and the volume of influx.
Therefore, he defined the circulating kick tolerance as:
K ic ~ P /~ P « i (2.5)
where equivalent-mud density (p ) is calculated by the iterative process.
Reproduced with permission of the copyright owner. Further reproduction prohibited w ithout permission.
15
Although an underground blowout is highly undesirable, for a hole section with a
known but manageable underground flow potential, necessary unconventional well
control contingency plans can be developed. Wessel and Tarr (1991) reported a new
strategy to optimize well costs by managing the well-control risks better than an arbitrary
minimum kick tolerance value. A direct tradeoff exists between kick tolerance and well
cost: Specifying a higher kick tolerance than necessary' can increase the well cost because
additional casing strings will be required. Specifying lower kick tolerance can lead to
costly well-control incidents.
Wessel and Tarr first simplified the productivity index (./) as a function of only
the product o f the permeability (k ) and the permeable zone thickness (/i) multiplied by a
constant (y).
J = Y kh (2.6)
By estimating the kh value for a potential gas zone to be drilled, one can determine
whether an underground gas flow can be controlled with the available rig equipment or
whether an additional pumping unit or a relief well will be required. Furthermore,
neglecting the two-phase-flow liquid hold up and any friction pressure loss in the annu
lus, the kill-mud density and pump rate combination required to kill the underground
flow is dependent on the volume of kill-mud available, as shown in Figure 2.2.
Leach and Wand (1992) reported the use of a kick simulator to generate well
control procedures and kick tolerance calculations during the planning stage for a deep
high pressure well in the Norwegian North Sea. Recently Nakagawa and Lage (1994)
reported cases o f exploratory deep water drilling on the Brazilian coast.
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
16
One open hole annular volume of mud required for kill
Pumping capacity with additional high pressure pum ps
Additional kill options with high pressure f umps
Effective kill options with rig pur ip;
Infinite volume of mud required
Pumping capacity with rig pumps
(After Wessel and Tarr. 1991) KILL MUD WEIGHT
Figure 2.2 Pump-rate requirements and equipment limits
Nakagawa and Lage reported that the kick tolerance was considered to be a
crucial aspect and was calculated both before (during the well design) and during
drilling. One of the cases, a well located in 1,214 m (3,983 ft) of water with 508 mm (20
in) casing set at 1,590 m (5,217 ft) had a low fracture gradient of 1.38 gr/cm’ (11.5
Ibm/gal) and indicated some problems. Wellbore stability problems led to an increase in
mud density to 1.31 gr/cm’ (10.9 lbm/gal) while drilling at 2,000 m (6,562 ft). At this
point the possibility of safely reaching the final depth of 2,340 m (7,678 ft) before
running the 340 mm (13 3/8 in) casing was in doubt. Controlling a 4.8 m’ (30 bbl) gas
kick from a formation with 1.32 gr/cm’ (11.0 lbm/gal) pore pressure was considered to be
possible, and the simulation showed that drilling ahead as planned was also possible.
However, the maximum allowed surface pressure was 1,586 kPa (230 psi), and the
pressure loss through the choke line was 1,241 kPa (180 psi) for 34 m’/hour (150 gpm) of
mud flow rate. Those close pressures could result in some problems during the well
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
17
control procedure. It was decided that, if a kick occurred, the pump flow rate to circulate
the kick would be reduced to 22.7 m3/h (100 gpm), and the kick would be circulated by
choke and kill line in a parallel arrangement. Fortunately a kick did not occur, and the
casing was set as programmed.
2.2 Kick Simulators
Many old computer models for gas kick simulations were limited by the
assumption of an arbitrary distribution of the gas in the wellbore. The most common
assumption is that the gas enters as a slug and remains as a continuous gas slug through
the annulus to the surface.
One of the first mathematical models using this assumption was presented by
LeBlanc and Lewis (1968). Also assumed were that the annular frictional pressure loss is
negligible, the gas is insoluble in the mud, the annular capacity well is uniform, and the
gas travels at the same velocity as the mud.
Mackenzie (1974) also developed a mathematical model to predict the annular
pressure profile. He broke the circulation o f a kick into six zones. Three of those zones
are due to influx of: a) water and mud; b) oil and mud; and c) gas and mud. The other
three zones are occupied only by the mud. The computer program that was developed
changes the volume of each of the zones as the kick circulated due to gas slip and to fluid
expansion. His model corresponds well with data from a kick generated in the Louisiana
State University (LSU) training well.
A transient model was proposed by Hoberock and Stanbery (1981). They used
equations of motion that describe the pressure and flow in a rigid, vertical fluid
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
18
transmission line of a constant cross-sectional area. They adjusted the two-phase flow
properties that reflected average fluid properties to allow it to treat two-phase flow as a
single fluid flow. Also, the two-phase region was assumed to be distributed uniformly by
volume with the mud but to change with time due to gas expansion and elongation of the
region.
Later, Nickens (1985) proposed a dynamic computer model complete with
equations, assumptions, computational strategy, boundary conditions, and suggestions on
timesteps. His model is based upon mass-balance equations for the mud and gas, a
momentum-balance equation for the gas-mud mixture, an empirical correlation relating
the gas velocity to the average mixture velocity plus the relative slip velocity between
mud and gas, and equations of state for mud and gas. Chapter III presents a more detailed
description of this model.
Using the Nickens’ model, Podio and Yang (1986) proposed a well control
simulator for personal computers. The main difference between the two models is related
to the solution method of differential equations. While Nickens’ model uses a fixed space
grid, Podio and Yang’s model uses a moving boundary solution. Other differences are the
calculation of influx rate, slip velocity, and friction factor. The benefit o f using Podio and
Yang’s model is that it facilitates simulation of multiple kicks taken in the same well.
Negrao and Maidla (1989) developed a mathematical model to predict the
pressure variation in the choke line and the annular section of the well during a well
control in deep-water. They used the model to select the flow rate for kick control.
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
19
Element, et al. (1989) presented a complete overview of well kick computer
simulator codes. They compared and contrasted the existing computer models with
respect to the differing modeling capabilities, solution methods, numerical
approximations, and the description o f physical effects using either physical models or
correlations.
Kato (1989) developed a two-phase model with the assumption that no
coalescence nor breakage of bubbles occurs with initial input o f bubble sizes.
Santos (1991) proposed a mathematical model for well control operations in
horizontal wells. He modified his previous work for a vertical well (Santos, 1989), based
on the Nickens’ model, to use in horizontal wells.
Vefring, et al (1991) presented a kick simulator for use on a workstation with a
Unix operating system and X-Window system installed. They reported that many selected
downhole parameters could be plotted graphically on the screen as a function of time and
space. The mathematical model is composed of the conservation of mass (mud, free gas,
dissolved gas, and formation oil), conservation of total momentum, and functional
relationships o f mud density, gas density, free gas velocity, gas influx, rate o f gas
dissolution, and frictional pressure loss. They used a finite difference method to solve the
system of equation with a simple front tracking technique.
Miska, et al. (1991) presented a computer simulation of the reverse circulation
well control procedure for gas kick. Their model assumes a steady-state flow of all fluid
in the well, slip velocity of zero, and gas flowing as a continuous slug.
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
20
Santos, et al. (1991) analyzed the dynamic pressures developed on the marine
riser and diverter line during gas removal from a riser-diverter system. They studied the
“riser blowout” that can cause collapse o f the riser and the risk o f fire. They developed a
riser model that was coupled with an existing diverter model (Santos, 1989) to simulate
numerically the gas removal from the riser-diverter system.
The dynamic two-phase model OLGA (Bendlksen et al, 1991), originally
developed for two-phase oil and gas flow in pipelines, has been modified to use in well
control. For example, Rygg and Gilhuus (1990) and Rygg, et al. (1992) describe the use
of the two-phase model OLGA during the kill planning phase of a 1989 underground
blowout in the North Sea.
Schofmann and Economides (1991) compared kick control in ultra deep wells
with shallow wells. The basic pressure equations used are similar to the equations used
by LeBlanc and Lewis (1968).
Two commercial kick simulators are available now: the RF kick simulator from
Rogaland (Rommetveit and Vefring, 1991) and the R-model from an association of
Schulumberger Cambridge Research, BP International, and Sunbury, and supported by
the United Kingdom Department of Energy (Tarvin and Walton, 1991; White and
Walton, 1990).
2.3 Two-Phase Flow Through Annular Section
The development of a reliable kick simulator also requires an accurate model of
gas-mud mixture flow as it moves upward in the wellbore. The Department of Petroleum
Engineering at Louisiana State University has been conducting projects in well control
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
21
(Bourgoyne, 1982) at the Petroleum Engineering Research and Technology Transfer
Laboratory (PERTTL) for more than a decade. Most o f the experiments cited in this
section were performed at this well facility.
Rader, et al. (1975) verified that the assumption of gas flowing as a continuous
slug and with the same velocity as the liquid did not work well when applied in a 1,828
m (6,000 fit) LSU research well. They observed lower gas velocity and lower casing
pressure than expected during a well control operation.
After evaluating kick control methods, Mathews (1980) and Mathews and
Bourgoyne (1983) reported the occurrence of bubble fragmentation. Mathews observed
that the bubble fragmentation is smaller in viscous fluids and less intense using the
dynamic volumetric method.
Caetano (1986) studied two-phase flow in a flow loop using both air-water and
air-kerosene flows. He defined flow pattern maps for concentric and fully eccentric
geometry. He concluded that eccentricity affects both the friction factor and the transition
from bubble to slug flow. Furthermore, he proposed models for liquid hold up and
pressure gradients for each flow pattern based on Taitel’s (1980) equations.
Motivated by the need for a better knowledge of the bubble fragmentation
process, Casariego (1987) and Bourgoyne and Casariego (1988) made theoretical and
experimental studies of gas kicks in vertical wells. Their model closely predicted the
measured casing pressure with data from a 1,828 m (6,000 ft) LSU research well.
Rommetveit and Olsen (1989) used an inclined (maximum of 63°) research well
to perform gas kick experiments using nitrogen and argon gas with oil-base mud. They
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
used nine surface sensors to monitor the pump strokes, mud return flow rate, pit level,
choke position, choke pressure, gas injection rate, choke line fluid density, standpipe
pressure, and gas injection pressure. In addition, they used one hardwire sensor and four
downhole memory tools to log the pressure and temperature. Based on the differential
pressure among the sensors in the well, and between the choke pressure and the sensors
in the well they concluded that: the gas starts to dissolve immediately as it enters the
wellbore; the bubble flow regime prevails in the two-phase section; the gas bubbles rise
and dissolve; the initial gas-oil ratio (GOR) in the experiment was higher than the
saturated GOR: gas bubbles rise and distribute over a longer section of the well; the gas
dissolution is governed by convective diffusion; and the mud does not become saturated
with gas immediately. They also observed some pulsations on the return flow, and their
explanation was that gas bubbles first coalesce and form a slug of gas, which rises
quickly and expands. After this a new dissolution process takes place in the upper part of
the annulus.
Continuing the well control research at LSU, Nakagawa (1990) and Nakagawa
and Bourgoyne (1992) performed an experimental study in a fully eccentric flow loop at
different inclinations to determine the gas fraction and gas velocity during the gas kick.
They presented a simplified model for the gas-rise velocity eliminating the bubble size
and shape for the calculation. Following this study, Mendes (1992) and Wang (1993)
continued Nakagawa’s experiments with low-er superficial gas and liquid velocities that
were not covered in previous experiments.
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
23
Using a flow loop, Johnson and White (1990) performed some experiments to
examine gas rise velocities during kicks. They used water and Xanthan gum as the liquid
phase and air as the gas phase. They concluded that, in drilling fluids, the bubbles rise
faster than in water despite the increased viscosity. They explained that these surprising
results are due to the change in the flow regime, with large slug-type bubbles forming at
lower void fractions. Furthermore, their results show that a gas bubble will rise faster
than any previously published correlation would predict. One of their results, for vertical
flow, is shown in a Zuber-Findlay (1965) plot along with Nakagawa’s. Mendes’, and
Wang's data in Figure 2.3. We can observe from this figure that Johnson's and
Nakagawa's data are similar and can be fitted in a Zuber-Findlay correlation for the mean
velocity of gas (vc).
Hovland and Rommetveit (1992) experimented with gas kicks in the same well
used by Rommetveit and Olsen. In these experiments, the authors used oil and water-
based mud. Nitrogen and argon were injected to simulate the gas kick. They varied mud
type, mud density, gas concentration, mud flow rates, and gas injection depth in their
experiments. They concluded that, in a high concentration gas kick, the gas rises faster
than in low and medium concentration. The gas rise velocity correlations obtained from
v (2.7)
where the superficial mixture velocity (vm) is defined as:
V* +<//( 2 .8 )V
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
24
these experiments are not significantly dependent on gas void fraction, mud density,
inclination, mud rheology, and surface tension.
4.0
ZUBER-FINDLA Y PLOT Inclination = 0 degree
3.5
3.0
2.5
« 2.0*❖
1.5
Johnson and White (1990)
1.0 Nakagawa (1990)
Mendes (1992)
0.5Wang (1993)
0.00.0 0.5 1.0 1.5 2.0 2.5 3.0
Mixture velocity (m/s)
Figure 23 Zuber-Findlav plot for flow-loop experiments
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
25
Hovland and Rommetveit presented one Zuber-Findlay plot, but the graphic was
normalized (divided by the maximum value). As a result, the experimental data cannot be
compared with previous work.
Using the same flow loop used by Johnson and White, Johnson and Cooper
(1993) investigated the effects o f deviation and geometry on the gas migration velocity.
For vertical orientation they conclude that the flow in the pipe and annulus are almost the
same. They conclude that the gas distribution coefficient iC0) is the same while the gas
slip velocity (v,) is slightly larger in the annular geometry. In deviated flows, C'0 is larger
for the annulus and v. is larger for the pipe. Up to a deviation of 45°, v, remains almost
constant. They also conclude that, even in a stagnant mud, the gas normally migrates at a
velocity over 0.5 m/s (5,900 ft/hr), almost six times the conventional field model of 0.085
m/s (1,000 ft/hr). The conventional field model considers only the hydrostatic effect of
gas migration as;
where dpc dt is shut-in pressure rise rate. They used an equation developed by Johnson
and Taruin, 1993 ( in Johnson and Cooper, 1993) to calculate the shut-in pressure rise
rate (dpc dt):
Equation 2.9 considers the mud and wellbore compressibility and fluid loss into
formation.
(2.9)
dPc X y iiP mg Vs - < I edt x kvt + x ^ + x mvm
(2.10)
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
26
Lage, et al.(1994) reported gas kick experiments performed in a 1,310 m (4,298
ft) vertical training well. The well has a 400 mm (13 3/8 in) casing set at 1,310 m and
cemented up to surface. Inside this casing, a second 178 mm (7 in) casing was placed to
simulate the wellbore. A 48 mm (1.9 in) tubing string was used to inject air at the bottom
of the 178 mm casing, and it was placed in the annulus o f400 mm and 178 mm casings.
In this same annulus, an additional 48 mm tubing string was placed at 800 m (2,625 ft)
to simulate the casing shoe and circulation losses. Inside the 178 mm casing, a drillstring
composed of 121 mm (4 3/4 in) drill collars and 89 mm (3 1/2 in) drill pipe was run. A
special sensor sub was made to accommodate the pressure sensor. Four sensors were
placed at 302 m (991 ft), 600 m (1,968 ft), 877 m (2,877 ft), and 1,267 m (4,157 ft). Air
and water were used in four tests. They measured three velocities: bubble front, volume
centered, and bubble tail using data from the pressure sensors. They state that, if no gas is
present, the differential pressure is equal to hydrostatic pressure between two sensors. In
addition, they measured the bubble front velocity dividing the distance between two
upper sensors by the time elapsed between the beginning of differential pressure decrease
in the two upper sensors and two lower sensors. Next, they also measured the volume
centered velocity, but they assumed that the center of the largest gas volume (when the
differential pressure is minimum) is at the middle point o f two sensors. They assumed
that the air expansion and concentration changes are negligible as the air rises from the
center of two the lower sensors to the center o f the pair above. Therefore, the volume
centered velocity could be measured by dividing the distance between the lower and
upper pairs of sensors by the time elapsed between the minimum differential pressure
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
27
between these two sets o f sensors. Lage, et al. measured the tail velocity considering the
distance and differential pressure stabilization between two sensors. They observed no
significant difference among the velocities for open or shut-in well conditions. They
obtained an average bubble front velocity of 0.26 m/s (3,070 ft/hr), an average tail
velocity o f 0.09 m/s (1,063 fit/hr), and an average volume centered velocity of 0.08 m/s
(944 ft/hr) to 0.15 m/s (1,772 ft/hr). In addition, they derived an interactive equation for
the pressure build-up (choke pressure) prediction that fitted very well with experimental
data:
1 ln- K - K
X" K + vr -Pc
(2 . 11)
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
CHAPTERS
CIRCULATING KICK TOLERANCE MODEL
Deep-water drilling has intrinsic problems, such as low fracture gradients, high
pressure loss in subsea lines, overbalanced drilling due to a riser safety margin, generally
high permeability formations, and emergency riser disconnection problems. As a result,
key factors to successfully drilling deep-water wells are, first, a detailed well design and
drilling plan; and, second a close control while drilling to avoid kicks, loss o f circulation
and underground blowouts.
Therefore, circulating kick tolerance can be used during the well design, along
with the pore pressure and fracture gradients, to determine depths at which to set casing
strings. It can also be used while drilling to estimate the fracture risk of the weakest
exposed formation if a kick is taken and circulated. Based on this analysis, a decision to
stop the drilling and run the casing string may be made if the results show a dangerous
fracture risk.
A mathematical model of a kick simulator dedicated to calculating the circulating
kick tolerance is presented here. The proposed model is divided into submodels: a
wellbore model, gas reservoir model, choke line model, and upward gas rise velocity
model.
3.1 Wellbore Model
The wellbore unloading model includes the upward two-phase flow inside the
annulus (well/drillstring, casing/drillstring, and riser/drillstring). This model is based on
the model proposed by Nickens (1985). A similar approach was used by Santos (1989)
28
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
29
and Negrao (1994), but Santos’ program is restricted to only two annular sections. The
proposed model can theoretically handle any number of different annular sections, for
practical purposes the number was limited to 15 different annular sections in this study.
The model is based on: a) mass-balance equations (continuity equations) for the
mud and gas; b) a momentum-balance equation for the gas-mud mixture; c) equations of
state for mud and gas; and d) a correlation relating the gas velocity to the average
mixture velocity plus the relative slip velocity between mud and gas.
3.1.1 Continuity Equations
The continuity equations are founded on the principle of mass conservation.
Under unsteady two-phase flow conditions, the liquid phase continuity equation is given
by.
3.1.2 Momentum Balance Equation
The momentum balance equation is based on Newton's second law of motion,
which states that the summation of all forces acting on a system is equal to the rate of
i ^ V>H) Q (3.1)
where liquid holdup H is defined as:
volume of liquid in an annular segment volume of annular segment
(3.2)
and for the gas phase is given by:
(3.3)
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
30
change of momentum of that system. For two-phase flow the momentum balance
equation is given by:
The friction term, or frictional pressure gradient, is calculated using Beggs and
Brill’s (1973) correlation modified to account for the non-Newtonian characteristic of
drilling fluids. The Beggs and Brill correlation was adopted for this study because this
correlation can be used in inclined flow (directional drilling) or even in horizontal
drilling. Although the present study does not account for inclined wells, it can be
extended in the future.
The two-phase flow friction factor/ , is given by:
where the no-slip friction factor/*- is obtained from a Fanning diagram (Craft et al, 1962).
The no-slip friction factor used by Beggs and Brill is for a smooth pipe curve on a Moody
diagram. The ratio of the two-phase slip to no-slip friction e' is calculated as:
where (cp / ct) is the gradient pressure.
The elevation term or hydrostatic pressure gradient is given by:
I p = / f - (3.6)
s = (3.7)\ +0.01853 In X,
H' H ‘
A
-0.0523+ 3.182 In -0.8725 In
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
31
where the no-slip liquid holdup or input liquid content X is defined as:
(3 .8 )
If X I H 2 is greater than 1.2 or less than 1.0 then the exponent s is calculated from:
3.1.3 Equations of State
In deep-water drilling, only water-based mud is used because of environmental
pollution problems that could result from an emergency disconnection of the riser. As a
result, the drilling fluid can be considered incompressible for the well depth range of
interest. Hoberock, et al. (1982) studied the effect of this assumption and showed that an
error of order of hundreds of psi are possible in deep abnormally pressured wells. In the
case of deep-well drilling, or if oil-based mud is used, the effects of temperature and
pressure should be considered (Ekwere, et al. 1990). The reduction in bottom hole
pressure for well depths up to 4,572 m (15,000 ft) is not significant, and the mud density
(3.9)
The frictional term is calculated from:
(3.10)
where the mixture velocity vm is defined as:
v„ = v ,// + vjr(l - H ) (3.11)
and the two-phase no-slip density is defined as:
A,< = p,X + p g( l - X) (3.12)
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
32
can be considered as incompressible for conditions encountered today in deepwater
drilling operations. Therefore, the density o f mud is given by:
p, = constant (3.13)
As for the gas density, a real gas equation of state is given by:
= pM_ (3.14)* :RT
3.2 Gas Reservoir Model
Since little is initially known during well design or while drilling, about the
properties of the gas reservoir, a detailed reservoir model is not usually justified.
However, in the proposed simulator, two reservoir models can be chosen by the user: the
Thomas, et al. (1982) or the Al-Hussainy, et al. (1966) models.
In 1982 Thomas, et al. (in Element, et al., 1989) introduced the use of the
following equation:
nkhTi t . =
where: PD = ~ ln ( 'D +0.809) (3.16)
and <„ = ■■■ kt— (3.17)4>Pf cf r J
The approximate solution of the diffusivity equation (Equation 3.15) requires the
assumption of a constant gas flow rate. Since, during a gas kick, the bottom hole pressure
and fluid flow rate vary, this assumption is not true. Nickens made a slight modification
to Equation 3.15. He divided the gas formation into axial segments of thickness h, equal
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
33
to the rate of penetration (ROP) multiplied by the time step in use. Then he considered
that each segment flows independently o f the other. As a result, the total gas-influx rate is
then:
where N(t) is the number of segments at time t. This modification to the flow equation
removes the approximation that gas flow is axially symmetric within the exposed gas.
Implicit also is that the reservoir extends to infinity. In most kick control
situations, this assumption is acceptable because the gas flow time is short, and the
reservoir boundary is not reached. In contrast, simulation of small pockets of gas is not
allowed but should not occur in a serious underground blowout.
Santos used the Al-Hussainy, et al. (1966) equations, modified to account for
changes in flow rate and pressure, in his studies of diverter operations. The wellbore
pressure in an infinite gas reservoir produced at a constant flow rate, including skin and
the turbulence or non-Darcy effects, has the following expression:
(3.18)
0.367p j ~ L 5V
where the real gas pseudo pressure is defined as:
= 0[log(2.245/i)) + 0.87(5 + DO)] (3.19)
(3.20)
and the dimensionless time by:
kt (3.21)D
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
34
where the total compressibility may be approximated as:
(3-22)
The turbulence or non-Darcy factor is calculated with:
P-Mp-k£> = 0.159 z (3.23)
RpghrKTx
and the velocity coefficient for consolidated sandstone as:
P g = ~ ^ J k (3'24)
Since the bottom hole pressure and gas flow rate vaiy with time, the solution for
the wellbore pressure can be found by applying the principle of superposition for
different flow rates in the right hand side o f Equation 3.19, which becomes:
1 ^ ) - ^ ) ] ^ _ g ( g . , ) log[2245(,„- , o .,)]+0.87a(-S-+/X?„)
(3.25)
After algebraic manipulation of Equation 3.25, the flow rate can be obtained from the
solution of the following quadratic equation:
^ 0.87S + log(2245(,0 - O L . B - A - Q ^ Iog(2.245(,c - O )& ---------------o 5 td -------------- a + ----------------- MTO------------------= 0 (" '26)
where:
0367p J T
and??-i
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
35
The flow rate for each time step in the computer model is calculated through
Equation 3.26 using the bottom hole pressure in that time step.
The reservoir height or thickness (A) in Equation 3.27, necessary to calculate gas
flow rate, is considered variable with time as a function of the rate o f penetration while
drilling:
ht = R O P it ' - t , ^ ) (3.29)
A'</)
h = Y JK (3-30)i=0
3.3 Choke Line Model
A choke line is employed to carry fluids to the surface after the subsea blowout
preventers (BOP) are closed. The long and narrow (usually 3 inches) choke line in deep-
water leads to high velocity and consequently high pressure loss.
Elfaghi (1982) performed experiments using a full-scale model at LSU, consisting
of 914 m (3,000 ft) of 60 mm (2 3/8 in.) subsurface choke line. For single-phase mud
flow, both the Bingham plastic and the power law non-Newtonian models provided
acceptable comparisons with the observed data. For two-phase flow through the choke
line, the Hagedom and Brown (1965) and Beggs and Brill (1973) correlations provided
acceptable comparisons with the observed data. As a result, the Beggs and Brill
correlation was selected for this work for two-phase flow conditions. In addition, the
power law model is used for the single phase mud flow.
3.4 Upward Gas Rise Velocity Model
An empirical correlation relating the gas velocity to the average mixture velocity-
plus the relative slip velocity was determined using both the available data from flow
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
36
loop experiments and data from an experimental work with a 6,000 ft well using drilling
fluid and natural gas. Chapter V presents a more detailed description of the experimental
work and the proposed correlation, which is as follows:
vg = 1.425vm + 0.2125 (3.31)
3.5 Solution of the Differential Equations
The solution of the differential equations in Section 3.1 is achieved using the
numerical method of finite difference. This method was also used by Nickens and
Santos. Many techniques can be used to solve the differential equations by the finite
difference method. In the proposed model, a centered in distance and backward in time
with a fixed space grid technique is used because it is a stable method that does not
present a convergence problem. The flow path is divided into a finite number of cells.
Figure 3.1 shows a cell for two different time steps.
The finite difference formulation for the continuity equation in the space
derivative is approximated by:
(3.32)dU = Ue - U s ct Ar
and the time derivative by:
cU UA- U 3 .. Ub +Ui - U 2 - U \ a i 2 a t
where U is a function of r and t. Substituting these approximations into Equations 3.1
and 3.3, the finite difference formulation for the continuity equation becomes for liquid:
( v ; A # ) 6 - ( v ; P | f l ) s [ { P i H \ + { p , H \ - ( p , H ) 2
Ar 2 A/
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
37
z + Az
Az
FINITE CELL
AT TIME
t - At
FLOW
( 2 ) KNOWN
©Q KNOWN
4depth
:+ Az
FINITE CELL
AT TIME
t
© UNKNOWN
©( ? ) KNOWN
time FLOW
and for gas:
Figure 3.1 Finite difference scheme for a cell
hp,(i-g>L-[v>»a-/oLAr
■ +
[/> ,(!- tt)]t + [p /1 - tf)]s - [ / , .( ! - H ) \ - [p, (1-/J)],2 A/
(3.35)
= 0
The finite difference formulation for the momentum balance equation in the time
derivative is the same as Equation 3.33, but the spatial derivative becomes:
oU U6 +U2 - U 5- U l& 2Ar
(3.36)
and substituting Equation 3.36 into Equation 3.4 gives:
+[v.v,(i- #)], -[v.v.o- *oH”.v,(i- *)],+ +{v,V,* )6 *Wp,h\ -(v ,V ,» )5 *)]4 +
+[v> ,r(I - H ) \ - [vf p s(l - ff)]2 - [vsp ,(1 - //)], + ( v ,p ,H \ +(y,P,H)s - (3.37)
♦ ( £ ) , + ( £ )
(A £l + ( > 1 + fA p)'.A rJ, VA: ) , \A:J< \ A z) .
fiic
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
38
The calculation of the flow properties at point 6 in Figure 3.1 from the known
properties at points 1, 2, and 5 requires an iterative procedure. Points 1 and 2 represent
the flow properties in the previous time step (t - At) at the lower and upper boundaries,
respectively. Points 5 and 6 represent the same points as 1 and 2, but at the time step t.
Points 3 and 4 represent arithmetic averaging at the center of the cell at the t - A t and t
time steps, respectively. The procedure to calculate the two-phase flow properties at
point 6 in a cell is:
1) Assume an initial liquid hold-up at point 6.
2) Calculate the liquid velocity using Equation 3.34 at point 6.
3) Calculate the gas velocity’ using the empirical correlation (Equation 3.31) at point 6.
4) Calculate the gas density’ using Equation 3.35 at point 6.
5) Calculate the pressure using Equation 3.14 at point 6.
6) Use the flow properties, determined in steps 1 through 4, in the finite difference
approximation for the mixture momentum balance equation (Equation 3.37) and
solve for the pressure at point 6.
7) Compare the pressures calculated in steps 5 and 6. If the difference between them is
less than an arbitrary value, stop the procedure. Otherwise assume another liquid
velocity and repeat the process until it converges.
The discretization procedure is only applied to the two-phase region. A single cell
exists the first time, two cells for the second time, and so on. The process for each time
step starts in the bottom cell and ends in the uppermost cell that coincides with the two-
phase flow leading edge. With this procedure, the pressure at any given time step and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
39
position can be determined. The flowchart for calculating the bottom hole pressure is
presented in Figure 3.2.
3.6 Simplification of the Differential Equations System
The procedure explained in Section 3.5 can be simplified with great benefit in the
computation time; it can be 10 to 20 times faster depending on the number of cells. The
simplification was made in the calculation of the liquid hold-up, which is calculated
directly from Equation 3.31. The simplified procedure is:
1) Calculate the superficial liquid and gas velocities.
2) Calculate the liquid hold-up and gas velocity directly from Equation 3.31.
3) Calculate the gas density using Equation 3.35 at point 6.
4) Calculate the pressure using Equation 3.37 at
5) point 6.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
40
starti=l
i>N
time step= i *
Calculate pressure @ two-phase leading edge
STOP
Assume: Bottom Hole Pressure Pbh i=i+l
Calculate: vsl and vsg
I Pa - Pc I<sl
i=i
Assume: HI
Calculate: vl and vg
Calculate pg
Calculate: Pressure eq. 3.14
Pa
Calculate: Pressure eq. 3.37
Pb
I Pa - Pb< e
j= j+ l
Figure 3.2 Flowchart of the complete program
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
41
starti=l
i >N
time step= i
Calculate pressure ffi. two-phase leading edge
STOP
Assume: Bottom Hole Pressure Pbh i=i+l
Calculate: vsl and vsg
1 Pb - Pc I< e !
Calculate: vl, vg, and HI
Calculate: Pressure eq. 3.37
Pb
Calculate pg < ►
Figure 3.3 Flowchart of the simplified program
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER4
C O M PU T E R PR O G R A M AND R ESULTS OF N U M E R IC A L SIM U L A T IO N S
The procedures to calculate the kick tolerance, as explained in Chapter III, have
been implemented as a computer program, and are described in this chapter. In addition,
the developed software was first applied in a typical deep water well design and was then
compared with results from a commercial kick simulator for real drilling problem cases.
Furthermore, selection of the kick tolerance during well design and while drilling is
proposed here.
4.1 C om puter Program
A computer program was written in FORTRAN applying the theory previously
described. During the development phase of the program, the goal was to produce a
program that is fast, reliable, and suitable for available rig site computers.
The program includes four major scenarios:
1. Taking the kick while drilling (the gas enters into the well, mixing with pumped
drilling fluid).
2. Detecting the kick, stopping the mud pump, and making a flow check (the reservoir
produces gas).
3. Closing the well and observing the shut-in-drill-pipe-pressure (SIDPP) and the shut-
in-casing-pressure (SICP) (the reservoir still produces gas until the bottom hole
pressure equalizes with the formation pressure).
4. Circulating the kick out and keeping the bottom hole pressure constant (Driller’s
method).
42
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
43
The program can theoretically handle any number of different annular sections,
but for practical purposes it was limited to 15 different annular sections. Unlike most
programs that require the annular sections to be input, the new program calculates the
annular sections using the wellbore, casing, and drillstring configurations data.
The two-phase flow region is divided in cells, a necessary step to solve the system
of equations by the finite difference technique. The user can control the size or volume of
the cell. Normally a cell volume of one or two barrels was found to be adequate, but with
a cell volume of ten barrels the program will run faster. A direct tradeoff exists among
the cell size, computing time, and accuracy. A greater cell size allows the program to run
faster with some loss in the accuracy and sometimes with instability of the system.
Internally, the program uses one half of the cell volume input by the user in the
drill collar-wellbore annulus because the cell height can be high. On the other hand, the
cell volume in the large riser and drill-pipe annulus is fourfold because the height can be
small. Initially, a full model was developed, as shown in the flowchart of Figure 3.2.
Because this approach deals with two iterations (one for liquid hold-up and another for
bottom hole pressure), it is more suitable for use on a main frame computer. Using a
typical deep-water well with a cell of one barrel in volume, pit gain o f 30 barrels, and a
Pentium 90 MHz computer, the running time was about 50 minutes. Therefore, this
model has been found to be slow and not adequate for use at a rig site.
In order to achieve a faster run time, the model was simplified by calculating the
liquid hold-up directly from Equation 3.31, as shown in the simplified program flowchart
of Figure 3.3. Consequently, under the same conditions as the previous model, this
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
44
simplified model calculated the circulating kick tolerance in 5 minutes instead of the
previous 50 minutes. The accuracy o f the simplified model was found to be acceptable
and will be discussed later.
In all simulations, starting the pump was the most critical part of the overall well
control procedure. During the pump start-up, the smallest kick tolerance value was
achieved. This fact also was reported by Bourgovne, et al. (1978) in their studies of well
control procedures for deepwater drilling. The high pressure that developed at the
weakest exposed depth is due to a high choke line friction, which increases with water
depth. To minimize the frictional pressure loss, a slower pump kill speed or the use of the
kill and choke line in a parallel arrangement should be adopted. Unfortunately, a third
option, of using a larger diameter choke line, cannot be applied easily because choke line
diameter is a characteristic of the rig. Thus, the user can specify the start-up pump rate
and determine whether or not a parallel flow arrangement, using the choke and kill line,
will be used.
4.2 Results from a Typical Deep W ater Drilling Experience
A typical well design applied to drill deep water wells in Campos Basin, in
Southeast Brazil, is used here to simulate a kick and to calculate circulating kick
tolerance. The typical casing design up to the surface casing is shown in Table 4.1. The
structural casing is used because of the weakness of the soil at the sea floor that cause the
temporary guide base to sink. The structural casing was placed only by jetting in the past,
but today the structural casing is lowered with a single guide drilling base system (BUP
system) developed by Petrobras, w'hich is an improvement in the guidelines drilling
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
system. Temporary and permanent guide bases were replaced by the BUP system which
has these advantages:
a. A single guide base;
b. Reusable and mechanically retrievable guide funnel;
c. Allows guideline or guidelineless operations;
d. Cutting returns at mudline elevation, which prevents BUP burial by cuttings;
e. Minimum rig up time; and
f. Low *r costs.
Table 4.1 Typical casing setting used for a deep-water well in Campos Basin
CASING DIAMETER
(in)
THICKNESS
(in)
LENGTH BELLOW
SEA LEVEL (m)
GRADE
STRUCTURAL 42 12 BCONDUCTOR 30 1 1.2*- 1 60 X - 52
SURFACE 20
00*rti*
400 K - 55* f ir s t / w y ; jo in ts only
The selection of weight, grade, and coupling of casings were based on the high
loading conditions in the well head The forces involved in the casing head are
consequences of:
a) Currents at the sea bottom of about 2 knots acting on the riser and blow out preventer
(BOP) stack;
b) Watch circle of the dynamic positioned drill ship;
c) Weight of the BOP stack;
d) Inclination of the well head, which creates a momentum (mostly due to a inclination
of the sea floor); and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
46
e) Mud inside the riser.
Calculations presented by Falcao et al (1985) show that, at the well head level,
the acting forces are about 240 tonf (529 Klbf) horizontal, 10 tonf (22 Klbf) vertical, and
130 tonf.m (940 Klbf.ft) momentum. About 10 m (33 ft) below the sea floor, the
maximum shear force of about 30 tonf (66 Klbf) occurs. Also the maximum momentum
of 608 ton.m (1,338 Klbf.ft) occurs at 4.5m (15 fit) below the sea floor with a shear force
of 10.5 tonf (23 Klbf). A typical casing setting profile used in Campos Basin is shown in
Figure 4.1.
DYNAM IC POSITIONED D R ILL SH IP
v — -i- 1—• J
W a te r dep th = 1OOO m
Asea f lo o r i—
^ BOP
ST R U C T U R A L C A SIN G £ « " (12 m)
CONDUCTOR 3 0 " (60 m )
SURFACE “ *-20" (400 m )
-I I- 13 3 /8"(1 5 0 0 m )
F ra c tu re G ra d ie n t =12 ppg"* ^ 9 5 /8 " (2800 m )
M ud W eigh t =9.8 ppg
G A S IN FLU X 1' [ N 9 i P o re P ressu re= 1 0 .5 ppg
T V D =3500 m
Figure 4.1 A typical well design for deep water drilling in Campos Basin
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
47
Low fracture gradients (Holden and Bourgoyne. 1982) are known to occur in deep
water wells. Campos Basin deep water wells are no exception. The results of leak-off
tests at the 20-in casing depth have shown lower values of about 10 to 11 lbm'gal
equivalent mud weight. In the beginning, to overcome this low fracture gradient, many
attempts were made to set the 20-in casing at a depth of 550 m (1805 ft) below the sea
floor, but the casing got stuck many times with little significant gain in the fracture
gradient. Because of this experience, setting the 20-in casing at a depth of 400 m below
the seafloor has become a standard.
The results using the simulator for a typical deep water well design is presented in
Figures 4.2. 4.3, and 4.4. The simulation represents a case using 15 bbl of pit gain, an
equivalent fracture gradient of 12 lbm/gal. mud weight of 9.8 lbm/gal. and a pore
pressure o f 10.5 Ibm gal Complete input data for this simulation is given in Appendix B
4.3 Comparison with Commercial Kick Simulator
Two real drilling cases were analyzed. The first case, a well drilled in Campos
Basin, was simulated by Lage, et al. (1994) using the RF kick simulator developed b\
Rogland to analyze two options of casing design:
a. Option 1
• 30-in casing from 345 m to 420 m
• 20-in casing from 345 m to 790 m
• 16-in casing from 345 m to 2.490 m
• 11 3/4-in casing from 345 m to 4,290 m
• 9 5/8-in liner from 4,190 m to 4,740 m
• 7-in liner from 4,640 m to 5.600 m
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Pres
sure
®
Frac
ture
D
epth
(p
si)
Cas
ing
Pres
sure
(p
si)
48
KICK TOLERANCE
<sotI
Kiel fluids enter the wellOC*
Gas let ding edge atthe casnig shoe depth
0 20 40 60 80 100 120Time (min)
CASING PRESSURE6000
5000
4000
3000B low out p re v e n te r2000
closed1000
0 20 40 60 80 100 120Time (min)
PRES S URE ® FRACTURE DEPTH6000
5000
4000P u m p s ia r te d3000
2000
1000
0 20 40 60 80 100 120Time (min)
Figure 4.2 Kick tolerance, casing pressure, and fracture pressure at casing depth for a typical deep-water well
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
49
PIT VOLUME
.*=- 7000
3000-
5 2000 -
41ua.01s . 1000
100
XS
Kick wafe detected
G as a t su rface
4020 60 SO0 100 120Time (min)
BOTTOM HOLE PRESS I'RE
20 40 60 80
DRILL PIPE PRESSURE
100 120 Time (min)
I20 40 60 80 100 120
Time (min)
Figure 4.3 Pit volume, bottom hole pressure, and drill pipe pressure for a typical deep-water well
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
50
GAS FLOW RATE
80 10040 60 1200 20Time (min)
GAS LEADING EDGE DEPTH
2000
Z 4000- ■ ex5 6000-et
8000-
J lOOOOr
5 12000-
Gps a t surface
100 12040 60 80200Time (min)
Figure 4.4 Gas flow rate and gas leading edge depth for a typical deep-water well
b. Option 2
• 30-in casing from 345 m to 420 m
• 20-in casing from 345 m to 790 m
• 13 3/8-in casing from 345 m to 2,490 m
• 9 5/8-in casing from 345 m to 4,290 m
• 7 5/8-in liner from 4,190 m to 4,740 m
• 5 1/2-in liner from 4,640 m to 5600 m
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
51
A fracture gradient of 2.10 gr/cm3 (17.5 lbm/gal) was expected at the 9 5/8”
casing shoe depth (4,740 m) for option 1 with a mud weight of 1,92 gr/cmJ (16 lbm/gal).
Additional input data from well RJS - 457 that was used for this comparison is given in
Appendix B. Figure 4.5 shows a comparison of results among the proposed model, the
RF kick simulator, and the simplified model for the drilling phase between 4,740 m to
5,600 m.
Well: RJS - 45717.4
Simplified mode!17.2
RF kick simulatorS. 17.0
Proposed simulator16.8
16.6
U 16.4
16.2
16.00 10 20 30 40 50 60 80 90 10070
Pit gain (bbl)
Water depth: 1,132 ft Fracture pressure: 17.5 ppg@ 15,552 ftWell depth : 18,374 ft Casing 9 5/8" @ 15,552 ftM ud weight: 16 ppg Open hole: 8 1/2 "
Figure 4.5 Kick tolerance for the well RJS - 457
The simplified model considers that the kick enters into the well as a slug and
remains as a slug throughout the upward path of the kick circulation. Although this
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
52
simplified approach is simple to calculate, it is very conservative and can result in an
expensive well design. On the other hand, using a commercial kick simulator to calculate
kick tolerance is time consuming, because the simulator is not specifically designed to
calculate the kick tolerance parameter. Figure 4.5 shows the kick tolerance as a function
of the pit gain volume and the pore pressure that might be encountered during the
drilling. When drilling a wildcat well the formation pressure or pore pressure is
frequently an unknown parameter. For this reason, the analysis was made as a function of
pore pressure because this parameter can greatly influence the final decision regarding
which option should be selected.
A second case, a well drilled offshore of Ceara State, Northeast Brazil, was also
simulated by Lage. et al. (1993) using the RK kick simulator. The well CES - 112 had
been drilled in water depth of 1.314 m (4,311 ft) with an 8.5-in bit when the possibility
that a high pressure formation could be encountered was raised, jeopardizing the drilling
operation. The main concern was whether to continue drilling to the depth of the original
well design or to set the casing early. The initial plan was to place the 7-in casing at the
depth o f4,500 m (14,765 ft). After analysis, the decision was made to continue according
to the previous plan. The complete input data from well CES-112 that was used for this
comparison is given in Appendix B. Figure 4.6 shows a comparison of results among the
proposed model, the RF kick simulator, and a simplified model for this well.
The results of kick tolerance calculations using the proposed circulating kick
tolerance model have shown a good agreement when compared with results of a
commercial kick simulator. The advantages of using the proposed model are that the
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
53
program can run in the available rig site computers, is reliable, and is much faster than
the currently available commercial simulators.
Well: CES 11210.4
£ 10.0
Simplified model
RF kick simulator9.8
Proposed simulator
9.6100 1200 40 8020 60
Pit gain (bbl)
Water depth 4,311 ft Fracture pressure: 10.5 ppg @ 12,993 ftWell depth : 14.764 ft Casing 9 5/8" @ 12,993 ftM ud weight: 9.5 ppg Open hole: 8 1/2 "
Figure 4.6 Kick tolerance for the well CES-112
4.4 Selecting Kick Tolerance
Since the shut-in kick tolerance can be calculated using an ordinary calculator,
many drilling plans have a value for the kick tolerance (for a given pit gain) with which
compliance is expected at the well site. An example of the kick tolerance notation is 0.5
lbm/gal' 30 bbl, which should be understood as a kick tolerance o f 0.5 lbm/gal with a pit
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
54
gain of 30 barrels. As drilling proceeds, the kick tolerance is calculated, and if it falls
below 0.5 lbm/gal, the drilling is interrupted.
In contrast, the calculation for the circulating kick tolerance is more complex and
involves a kick simulator. Furthermore, the simulations with a dynamic model have
shown that for deep water wells the worst case occurs when the pump starts to circulate
the kick out of the well, not when the well is shut-in. Therefore, a circulating kick
tolerance should be used, which can be defined as a difference between the maximum
circulating pressure and the fracture pressure, at the w-eakest exposed formation depth,
expressed in equivalent mud weight:
K!c = P-f- ~ Pf™ (4.1)D,*
Since the circulating kick tolerance cannot be calculated as a single equation, a
kick simulator should be used to calculate this value.
4.4.1 Selecting Kick Tolerance for Well Design
After the casing setting depths are determined, the circulating kick tolerance
should be calculated to confirm those depths. Selecting the pit gain is the most important
step in kick tolerance calculations, eclipsing all other unknowns such as reservoir and
mud properties, pore pressure, and temperature. A particular pit gain should be adopted
based on the ability o f the rig crew to detect a kick. Confirming the casing setting depths
using a high selected pit gain can be expensive because additional casing strings may be
necessary. On the other hand, using a small pit gain that the drilling rig crew cannot
detect can be very' costly if a kick or blowout occurs. As a result, three levels of pit gain
are proposed here to be used in well design:
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
55
a. Level 1:30 barrels of pit gain
b. Level 2: 20 barrels of pit gain
c. Level 3:10 barrels of pit gain
If level 3 is adopted, additional procedures should be taken, such as:
a. Use a mud logging unit to monitor the pit gain and to calculate the pore pressure
while drilling.
b. If a drilling break occurs, close the BOP without any flow-check.
c. The hard shut-in should be adopted to avoid further influx.
d. If the pore pressure increases, intermediate well logging is recommended to estimate
the pore pressure from a sonic log.
e. Someone must monitor the pit level constantly.
f. Each person on the rig should be advised about the meaning of level 3. The dog
house safety meeting should cover this before each tour.
The minimum value of the circulating kick tolerance that is obtained when
adopting one of the pit gain levels should be reported in the drilling plan Moreo\er, the
complete input data used to calculate this value should also be reported for comparison
purposes if a circulating kick tolerance must be calculated while drilling.
4.4.2 Selecting Kick Tolerance while Drilling
If a Level 3 pit gain is adopted in the drilling plan, or if the pore pressure
increases, the circulating kick tolerance should be calculated each 30 m (100 ft) drilled or
each time that the mud weight changes. The pit gain to be simulated should be equal to
the adopted pit gain in the drilling plan, or should be one that the crew can detect. Based
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
56
on this calculation, a decision to stop and run the casing or continue the drilling as
planned can be made.
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
CHAPTER 5
EXPERIMENTAL WORK
The experimental procedure for determining the upward gas rise velocity during
well control operations is presented here. This procedure was performed using a full scale
well and natural gas. Despite many studies in this area using flow loops and well, as
described in Chapter II, no experiment that had used a full scale well with natural gas as a
gas phase was found in the literature reviewed. During this project, thirty seven
experiments were performed — 8 with water and 29 with drilling fluid.
5.1 Description of a Full-Scale Well: LSU No. 2
The experiments were carried out in the existing LSU No. 2 w'ell (also known as
the DEA well), as shown in Figure 5.1, located at the Petroleum Engineering Research
and Technology Transfer Laboratory (PERTTL) at Louisiana State University in Baton
Rouge, Louisiana. The drilling and completion o f this well was funded through the
Drilling Engineering Association (DEA Project 7). The LSU well No. 2 is a vertical well
that is 1,793 m (5,884 ft) deep and cased with 244 mm (9 5/8 in) casing. The w’ell is
completed with a 32 mm (1 1/4 in) gas injection line that runs concentrically inside a 89
mm (3 1/2 in) drilling fluid injection line. The well also contains 60 mm (2 3/8 in)
perforated tubing (94 half-inch holes per joint) that serves as a guide for w'ell logging
tools to be run in the annulus without risk of the logging cable wrapping around the drill
string and becoming stuck. The research facility also has these features: a choke manifold
containing four 15,000 psi adjustable drilling chokes; a 250 hp triplex pump; two mud
tanks with a combined capacity of 550 bbl; and a high capacity mud-gas separator.
57
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
58
Loggkig Cable
H *
y \2 3/8” J55 Perforated
Tubing (4.7 Ibf/ft)94 x 1/2" hoies/jont
Pressure On-Line Sensor
Pressure Recorder Sensors ------
SSlffTVD
S822'T\D
Wei Logging Tiixiar
L
II
T
958" Casing
0-3170 f t-53.5 lb/ft 3170-3908 ft -47.0 lb/ft 3908-5553 ft - 43.5 lb/ft 5553-5876 ft - 53.5 lb/ft 5876-5884 ft - 47.0 lb/ft
31/2'J55 EUE Tubing — (93 lb/fit)
1j66"N 80 Tubing (3j02t>/ft)Last joint perforated
40x1/4" holes
Gasnjection Line
Mri njection Line
4" Gas/Fluid Return line
Figure 5.1 LSI' No. 2 well completion schematic
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
59
5.2 Methodology of Experimentation
A drilling fluid with properties matching those used to drill deep-water wells in
the Campos Basin, offshore of Brazil, was used, and it was circulated down the annulus
between the 89 mm (3 1/2 in) and 42 mm (1.66 in) tubing at the desired mud flow rate
with returns taken from the 244 mm (9 5/8 in) casing.
The gas was injected through the 32 mm (1 1/4 in) tubing, or was pumped down
at the desired injection rate through the annulus between the 89 mm (3 1/2 in) and 32 mm
(1 1/4 in) tubing. Before injecting the gas into the well to simulate a kick, the gas was
compressed up to 4,200 psi. This pressurization was accomplished using three 610m
(1,200 ft) storage wells cased with 7-in. 38 lb/ft N-80 and P-l 10 casings connected to a
152.4 mm (6-in) natural gas pipe line that operates at 700 psi pressure. First, one w'ell
annulus was filled with gas from the pipeline, and then the gas was compressed by
pumping mud down the tubing forcing the gas into the annulus of the other well. The
final desired pressure was obtained by alternating the fill-and-compress cycle.
After compressing the gas, it was injected or pumped down until the desired pit
gain was obtained. Following this, the circulation of the gas kick began until all the gas
was out of the well. In most o f the experiments, a back pressure of 150 to 200 psi at the
choke was kept by using an automatic choke (Warren choke). This procedure was used to
avoid a dangerous situation in case a large volume of gas reached the surface. After the
mixture of gas and drilling fluid left the well, it passed through a separator, where the
drilling fluid and gas were separated. The liquid phase returned to the mud pit, and the
gas phase was directed to the flare line, where it was burned. Some parameters were
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
60
varied in the experiments, such as the water or drilling fluid circulation rate, gas pump-
down circulation rate, kick size measured by the pit gain, and position of the downhole
pressure sensors. Table 5.1 shows the test matrix for the water and natural gas
experiments. Tables 5.2 and 5.3 show the test matrix used for drilling fluid and natural
gas experiments. Table 5.4 shows the drilling fluid properties used in the experiments.
Table 5.1 Test matrix for water and natural gas experiments
Experiment Pit gain Pump speed(b bl) (spm)
# 10 20 0 32* 62* 90*W1 10 0W2 10 32W3 10 62W4 10 90W5 20 0W6 20 32W7 20 62W8 20 90
32* spm — vl= 0.64 ft/sec 62* spm — vl= 1.24 ft/sec 90* spm — vl= 1 80 ft/sec
Table 5.2 Test matrix for mud and natural gas experiments with gas injected through tubing
Downhole Pressure Sensors 1200ft Apart
Testpit gain Pump
speedChoke
back pressure Note# <bbl) (spm (psi)
10 20 0 32 62Ml 10 62 170M2 0 choke closed Failure due to valve leakM3 10 32 100 Downhole pressure lostM4 20 62 170 Downhole pressure lostM5 20 0 choke open Downhole pressure lostM6 10 32 170M7 20 0 choke closed
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
61
Table 5.3 Test matrix for mud and natural gas with sensors 1,200 ft apart
Gas Was Pump Down Through 3.5 J55 Eue (93 #/ft) X 1.66” N-80 (3.02 #/ft) Innulus
TestU
Gas pump down speed
(spm)
Pump speed (spm)
Choke back pressure
(psi)Note
62 82* 0 32 62M8 82 0 choke openM9 82 32M10 62 62 170 on line data was lostMil 62 32 170 middle downhole sensor failedM12 62 62 180 middle downhole sensor failedMl? 82 62 170 middle downhole sensor failedM14 82 0 choke closed middle downhole sensor failed
82* spm — v=I 64 ft/sec
Table 5.4 Test matrix for mud and natural gas with sensors 100 ft apart
Test#
Gaspumpdownspeed(spm)
Position of downhole
tools
Pumpspeed(spm)
Viscousfluid
Chokeback
pressure Note
62 82 b* m* 1 t* 0 32 62 yes noM15 82 j B 62 N 0 one sensor failedM16 62 B 62 N 0 one sensor failedM17 82 B 62 N 170 one sensor failedM18 82 M 32 N 170 one sensor failedM19 82 M 62 N 180 one sensor failedM20 82 M 0 N 170 one sensor failedM21 82 M 62 N 170 one sensor failedM22 62 M 32 N 170 one sensor failedM23 62 M 62 N 0 one sensor failedM24 82 T 32 N 200 one sensor failedM25 82 T 0 N choke open one sensor failedM26 82 T 62 N 200 one sensor failedM27 82 M 62 Y choke openM28 82 M 0 Y choke
closedM29 82 M 0 Y choke open
b* = bottom (on line tool @ 5.422 ft) m* = middle (on line tool @ 2.761ft) t* = top (on line tool @ 100 ft)
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
62
Table 5.5 Drilling fluid properties utilized in the experiments
Experiment#
Mudweight(lb/gal)
Marshviscosity
(sec)
Plasticviscositv
(cp)
Yieldpoint
Gel strengthlOsec IOmin
lbf/lOOsq ftMl 9.9 54 15 5M2 9.9 54 15 5
M3-M4 9.9 53 15 5M5 9.9 54 15 5
M6 - M7 9.9 53 15 5M8 9.9 54 14 5
M9-M10 10.0 53 12 5M il -M12 9.8 49 10 2
M13 9.6 62 15 8M14 9.6 58 15 7
M15-M16 9.7 40 12 9 o 10M17-M18-M19 9.6 38 11 6 2 5M20-M21 -M22 9.6 40 12 3 2 9
M23 - M24 9.6 42 12 6 2 15M25 - M26 9.6 43 12 6 2 13M27 - M28 9.7 78 30 15 4 25
M29 9.7 78 30 15 4 25
5.3 Instrumentation of the Well
A data acquisition system from National Instruments was used to acquire and
record data and included:
a) The SCXI-1200: A data acquisition and control module that acquires the signal in the
SCXI-1200, digitizes the conditioned analog signals, and transmits the digital data to
the parallel port o f the PC.
b) The SCXI-1100: A 32-differential channel multiplexer that allows the module to
sample the volt source.
c) The SCSI -1001: A chassis that can house 12 modules.
d) The SCXI-1124: A 6-channel isolated digital to analog converter (DAC) module.
e) The SCXI-1163: A-32 channel isolated digital output module.
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
In addition, a data acquisition and recording program was developed using
LabView for Windows (a graphical programming software) and was used with the data
acquisition system. The following parameters were continuously monitored using the
data acquisition system and program:
a. Drill pipe pressure (psi);
b. Casing pressure (psi);
c. Gas-injection line pressure (psi);
d. Down-hole on-line pressure (psi);
e. Pump speed (spm);
f. Percent of gas-in-mud at the shale shaker (%);
g. Gas flow in (MSCF/hour);
h. Gas flow out through 12-in. line (MSCF/hour);
i. Gas flow out through 4-in. line (MSCF/hour);
j. Gas flow out through 1-in. line (MSCF/hour);
k. Gas flow out through 0.5-in. line (MSCF/hour); and
1. Pit volume (bbl).
In addition to one wired-to-surface downhole pressure sensor, three downhole
pressure recording sensors monitored the pressures developed at desired depth during the
well control experiments. The pressure recorders (model EMR710) used were from
Geophysics Research Corporation (GRC). These recorders acquire and record about
21,000 pressure points. The recording interval could be programmed with a minimum
time of 3.8 seconds. The on-line pressure sensor was connected to the surface through a
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
64
logging cable to the Schulumberger logging unit. The three downhole pressure recorders
were connected to each other and to the on-line sensor with a single strand wireline
(slickline).
A modification was made in the gas-out or vent line gas measuring system. The
previous system had only a 12-in Daniel’s Senior orifice meter installed. A 4-in Junior
orifice meter, a 1-inch honed flow section, and a 0.5-inch honed flow section were added
as shown in Figure 5.2. The measuring system of 12,4, 1, and 0.5-inch configuration was
chosen based on orifice calculations that overlap the expected gas flow rate. The results
of the calculations are presented in Table 5.6.
Orifice(4.625 ’) Valve
1.625
Valve Valvi
OrificeValve 0.5" 0 . 2 "
Figure 5.2 Gas flow out measurements system
The gas could be diverted into different line sizes by opening and closing
pneumatic valves using a switch board. The gas flow out was measured initially with a
12-in line, and as the gas flow rate decreased, the flow was diverted to the 4-in line, and
so on. An accurate gas flow out measurement was important because it could be used to
calculate the gas distribution profile along the well, which was one of the objectives of
this research. Unfortunately, the gas flow out measurements for the different orifices did
not overlap as expected and could not be measured continually. When the gas flow- rate
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
measurement for a given orifice size was minimum or zero, the gas was diverted to the
next smaller line, but sometimes the maximum flow rate for this new orifice size was
exceeded. As a result, the gas had to be diverted back to the original line, and the system
did not measure any gas flow.
The gas flow rate was calculated using the Daniel model 2500 flow computer that
acquires signals for differential pressure, absolute pressure, and temperature.
Table 5.6 Gas flow out calculations for gas measurements system
Linediam eter(inches)
Orificediam eter
Differential pressure
(inches o f water)
Flow ra te (M SCF/hour)
12 4.62567.0 250,0007.28 83,3330 65 25.000
4 1.62563 3 30,00044.0 25.0006.81 10,0000.60 3,000
1 0.500100.0 3,655
65 2,94110.55 1.2180.92 365
0.5 0.200
47.5 40026 30011 200
2.8 1001.8 80
0.69 50
5.4 Methodology Used to M easure Gas Rise Velocities
The velocities of the kick front (leading edge), the peak gas concentration, and
the tail of the two-phase region were calculated through an analysis of the measured
differential pressures. The rationale of this analysis was that if no gas is present between
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
66
two consecutive pressure sensors and mud is not being circulated, the differential
pressure should reflect the hydrostatic pressure between them. Also, when mud is being
circulated, the differential pressure between two sensors should be equal to the sum of
the hydrostatic pressure and the pressure losses between them.
When the gas front reached each sensor, the differential pressure began to
decrease, denoting the arrival o f the bubble front. Thus, the velocity o f the front could be
estimated by dividing the distance between the sensors by the elapsed time between the
first arrival o f the front. Figure 5.3 ilustrates the sensor positioned and the parameters
used for this estimate. The bubble front velocity between sensors 3 and 4 could be
estimated if the distance d3A and the time elapsed between the observed initial decrease
in differential pressure between sensors 2,3 and sensors 3,4 is known by using the
following equation:
frontSAini^p (5.1)
S 1
- i — S 2
t2,3S 3
d3,4S 4 — L .
Figure 5.3 Downhole pressure sensors disposition
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
Similarly, the tail velocity could be calculated as the distance between two
sensors (for example, sensors 2 and 3) divided by the elapsed time to stabilize two
adjacent differential pressures:
In general, when the differential pressure between two sensors is a minimum, the
largest amount of gas is present between the sensors, but the exact position of the peak
concentration is not known. If it is assumed that the peak concentration occurs at the
mid-point between two sensors, then the velocity of peak concentration can be computed
as the distance between two mid points (e.g. at the mid-point between sensors 3 and 4,
and at the mid-point between sensors 2 and 3) divided by the elapsed time between them
when the minimum values of differential pressure were recorded in the two adjacent
well segments.
(^3.4skj6.V ^2.3 s a b ip (5.2)
v. 2 2(5.3)
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
CHAPTER 6
EXPERIMENTAL RESULTS
The results of the experimental work are presented in two forms: first, data
acquired during a typical experiment are shown in graphs as functions of time; second,
the results of the experimental work described in Chapter V are presented here in Zuber-
Findlay plots. In addition, a simplified gas distribution profile is proposed based on
observation of the data.
6.1 A Typical Experiment
The data from each downhole pressure recorder (GRC EMR710) was downloaded
to a file in a PC computer using the parallel port. In addition, all the data from surface
sensors and the on-line downhole pressure sensor were recorded in a file using the data
acquisition system (DAQ) and a special computer program developed for this project
using LabView. Then, the four files (three from EMR710 sensors and one from DAQ)
were combined and adjusted to the same time scale. Since a time delay between the clock
from the EMR710 sensors and the DAQ system was observed, the time was corrected.
This was accomplished by observing a major pressure change at the beginning and end of
each experiment, for example, pump start-up and pump shut-down. Typical data collected
during the experiment are shown in Figures 6.1 through 6.3. These data and graphs are
from experiment M9 in which the gas was pumped down at 82 spm (v=1.64ft/sec), and
the kick was circulated out at 32 spm (v=0.64 ft/sec). Differential pressures used to
calculate gas rise velocities are shown in Figures 6.4 and 6.5.
68
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Down
hole
On-
line
Pres
sure
(p
si)
Down
hole
Pres
sure
(p
si)
69
3000
Bottom sensor
Middle se nsor2000
1500 •Top sensor
10000 20 40 60 80 100 120 140 160 180 200
Time (minutes)
1000
900
800
700
600
500
4000 20 40 60 80 100 120 140 160 180 200
Time (minutes)
Figure 6.1 Example of dow nhole pressure data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
70
600Gas Flow In
G is FlovO ut 1:5 In.500
200 in.c
100
20 80 100 120 140 160 180 2000 40 60
3500
3000
C- 2500uK 2000oclo 1500C.
£1000
500
0
Time (minutes)
Ik 1
mm
- ---
0 20 40 60 80 100 120 140 160 180 200Time (minutes)
Figure 6.2 Example of gas flow rates and drill pipe pressure
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
71
500
400
C.IT 300
m 200’{ftU
100
00 20 40 60 80 100 120 140 160 180 200
Time(minutes)
300 300
250 250iii: volurr
200 200
§ 150“ 150gas jlumped down pump speed increased
£ 100 100
pumf pressure
0 20 40 60 80 100 120 140 160 180 200Time(minutes)
Figure 6.3 Example of casing pressure, pit volume, and pump speed
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
choke cl oss
ed
cho te opei ined
j t------ r
ch ake wi ie opei ined
! ■ > «
nr
. y i. , i
Pump
sp
eed
(spm
)
72
700ONLINE itND CASING
s 600
500
!= 400
3000 20 40 60 80 100 120 140 160 180 200
Time (minutes)
650
\ 6000 k.
1 550!sco
§ 500 d
4500 20 40 60 80 100 120 140 160 180 200
Time (minutes)
Figure 6.4 Differential pressure between on-line and casing and between downhole and pressure recorders
f|o LE PRES S j RE R k o RpDO\\7V
Bottom - Middle s >nsors
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
73
700C P DO1 VNHOLE R IC O R I ER
650
8 600
fc 550
500
-= 450
!= 400
350
3000 20 40 60 80 100 120 140 160 180 200
Time (minutes)
900BOTTOM HOLE PRES AND BOTTOM RECO
SURERDER
800
* 700
t 600
400
3000 20 40 60 80 100 120 140 160 180 200
Time (minutes)
Figure 6.5 Differential pressure between top and on-line sensors and between bottom hole pressure and bottom sensor
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
74
6.2 Zuber - Findlay Plot
The leading edge gas velocities obtained from experimental data are shown in the
Zuber - Findlay plot of Figure 6.6 for different superficial liquid velocities. This data is
plotted with published flow' loop data, as shown in Figures 6.7 and 6.8, in w'hich a fair
agreement can be seen between the present and previous experimental work. The
regression analysis of Figure 6.8 provided the empirical correlation for gas rise velocity
that was used in the kick tolerance computer program:
vg = 1.426vm+0.2125 (6.1)
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
Gas
velo
city
(ni/s
ec)
75
Mixture velocity (ft/sec)0.0 1.0 2.0 3.0
2.0
- 6.01.8
1-1= 0.64 ft, sec1.6-i= 1.24 ft sec
-5.0
1.4
-4.01.2slope = 1.4
1.0
-3.0
0.8
- 2.00.6
0.4
- 1.0
0.2
0.0 hO.O0.0 0.2 0.4 0.6 0.8 1.0 1.2
MixtureveIocity(m/sec)
Figure 6.6 Zuber - Findlay plot of experimental data
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
Gas
velo
city
(ft/s
ec)
76
Mixture Velocity (ft/sec)0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0
Johnson and White (1990)V Nakagawa (1990)G Mendes (1992)□ Wang (1993)♦ Ohara (1995) j
f - 1 2 .0
- 10.0
Vg = 1.42575 Vmix + 0.2125 R*2 =0. 96
0.0 0.5 1.0 1.5 2.0 2.5 3.0Mixture Velocity (m/sec)
Figure 6.7 Zuber - Findlay plot of the present and previous flow loop experiments
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
Gas
V
eloc
ity
(ft/
sec)
CHAPTER 7
CONCLUSIONS AND RECOMMENDATIONS
7.1 Conclusions
Deep-water exploration and development are now a reality, and hydrocarbons
have been produced from water depths as much as 1,027 m (3,370 ft). However, deep
water drilling poses special problems, such as low fracture gradients, high pressure loss in
choke lines, overbalanced drilling due to a riser safety margin, and emergency riser
disconnection problems. Therefore, special care must be used when planning and drilling
these wells. The kick tolerance concept is a powerful tool that can be used during well
design, along with the pore pressure and fracture gradients, to determine depths at which
casing should be set. In addition, kick tolerance can be used during drilling to estimate
the fracture risk of the weakest exposed formation. This parameter can be used to stop the
drilling and run the casing string and to regulate drilling activities by governmental
regulatory agencies.
1. The proposed simplified computer model, which calculates the liquid hold up
directly from the empirical equation of Zuber and Findlay, not only saves computing
time, but has been shown to be accurate when compared with a commercial kick
simulator. Furthermore, the developed computer program is suitable for use with the
available rig site computers.
2. The minimum kick tolerance values for deep water wells in all computer simulations
performed were found to occur at the beginning of the circulation to remove the kick
out of the well. This fact is due to the high pressure loss inside the long choke line.
77
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
78
3. Results of gas rise velocity obtained using flow loop data and the present full scale
well experiments are in a close agreement.
4. Based on the experiments results, a simplified triangular gas distribution profile along
the upward migration of the gas is proposed. The accuracy of circulating kick
tolerance calculations may be further improved with the use of this distribution
profile.
7.2 Recommendations for Future W ork
7.2.1 Gas Distribution Profile
Even though extensive data were collected during experimental work, a fully
study of the gas distribution profile was not possible because of time restraint.
Improvement in the kick simulator may be made through prediction of the gas
distribution profile along the flow path in the annulus. The use of a proposed triangular
gas distribution profile (discussed later) in the circulating kick tolerance simulator is
strongly recommended.
The kick tolerance can be calculated easily if the gas distribution profile along the
upward path of gas migration is known. The differential pressure between two pressure
sensors shows the hydrostatic pressure between them if no gas nor liquid is flowing. If
only liquid phase is flowing then the differential pressure shows the hydrostatic pressure
plus the pressure drop due to friction losses between the sensors. As the gas flows
between two sensors, the differential pressure drops until the gas starts to leave the
interv al between sensors. This fact can be transformed in calculation of gas fraction as a
function of time for a given interval between two sensors by:
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
79
Api (7.1)
where the numerator represents the difference between two pressure sensors and the
denominator represents the initial differential pressure (no gas is present between two
sensors). Figure 7.1 shows calculations for the gas fraction as a function of time from the
experiment for migration with the choke closed.
However, the main interest here is not the gas fraction as a function of time for a
given interval as shown in Figure 7.1, but how the gas fraction profile will van,' along the
well. For a fixed time an average gas fraction can be picked up from the graphs on Figure
7.1 and plotted as a function of depth as shown in Figure 7.2. Appendix C shows more
examples of gas distribution profiles as a function of time and depth.
7.2.1.1 The Triangular Gas Distribution Profile
Observing the Figure 7.2, the gas fraction profile as a function of depth may be
approximated by a triangle. Figure 7.3 shows a scheme of section of the well and the
proposed triangular gas distribution profile. The triangular gas distribution profile is a
function o f the two-phase leading depth:
where hle(Q) = Initial two-phase-flow depth
The length of the base o f the triangle or the two-phase flow interval is given by:
(7.2)
(7.3)
where hlpf{ 0) = Initial two-phase-flow height
Reproduced w ith permission of the copyright owner. Further reproduction prohibited w ithout permission.
80
T ype o f ex p e rim en t: M igration w ith choke closed E xperim ent: M7
s 0.1501 °-10 t 0.05
~ 0.00
: Interval == 0 - 1,000« 1 J U U
i1 J T T1 U l _________
0 50 100 150 200 250 300Tim e(m inutes)
5 0-15 ̂ Interval f 1.000 - 2,
Z 0.10
200 250 300T im e (m inutes)
0. 10 -
0.05-1/3 V.V+.
o.oo-
i Interval = 2,235 - 13.470 ft I! !
] k i i/ V i !
1 ' 1 i i t
50 100 150 200 250 300Tim e( m inutes)- 0.15
cZ 0.10e;
5 0.05
0.000 50 100 150 200 250 300
es
0.15-
0.10-
0.05-
0.00-
Tim e(m inutes)
4 Interval = 4 ,705- ?,822 ft
3
l
Oft
On-line 1,000 ft
Top
2,235 fi
Middle
3,470 ft
Bottom
U 4,705 fi
5,822 ft
0 50 100 150 200 250 300Tim e(m inutes) j TD = 5,884 ft
F igure 7.1 G as frac tio n between sensors fo r d ifferen t dep ths an d tim es fo r experim en t M l
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
Depth
(ft
) De
pth
(ft)
Depth
(ft
)
SI
Experim ent:M 7
01000
2000
3000
I
4 0 0 0 i
5000o cs -a- >o 00 oo o o o o —o' © o d d ©
Gasfraction
0-1000
- 2000 -
oD
4000
5 0 0 0 -™o cs rr o oo o© © © © o —d o ' d d d d
Gasfr;
0-1000-
2000-
oO
4000-
5000-
IIF ■— ,—.
9——,—
O TT \C 00O O C O Od o d d o
Gasfraction
2000 -
3000- oQ
1000- • 2000 -
■3000-
4000-
5000-
■—
>*
o o o o o o Gas fraction
O CN TT so 00 oo o o o o —o o o o o o
Gasfraction
110.03 min
g-3000Q
4000-J
C tN -o- O 00© o © c ©© © © © c
Gasfraction
■ ■ 130.02min
|
yyiio o oo oO O O O O «—O 0 * 0 0 * 0 o’
Gasfraction
40 0 0 -
PTTT TTT1 PT TTT1
lOOOi
■ 2000 -
4 0 0 0 i
5000o CN TT so oo oo o o o o —o o o © o o
O rsj tt O oco o o o oo o © o' o
Gasfraction
Figure 7.2 Gas fraction profile as a function of depth for various times
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
82
In Equation 7.2, one can observe that when hle(t)=0, the two-phase flow has
reached the top of well and starts to leave the well.
The average gas fraction can be defined as:
VoIgJ 0a( t ) =4 » V ( 0
(7.4)
max
wjixfmud Depth
Figure 7.3 Proposed triangular gas distribution profile
If h/r(t) > 0 implies that the two-phase flow has not reached the surface, and the
volume of gas can be defined as:
o)m p (o) T( t ) (7.5)r(0) p{t) T(0)
w'here VoI%a!L{0) is the initial gas volume.
The gas will expand as it migrates upward, but its volume at standard conditions
must be the same. This condition will change when the gas reaches the surface After
this, a mass balance must be applied to calculate the volume of the gas inside the well
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
83
The height of the triangle is determined by equating the area o f a rectangle and
the triangle:
ff(O A ^(0 = ^ ( 0 - M ' ) (7.6)
therefore:
« « ( ' ) = 2a(t ) (7.7)
7.2.1.2 Triangular Gas Distribution Velocities
Each vertex of the triangle will move with a different velocity. The front velocity
( v/i.« ) travel faster than the center velocity ( vcmler ). Also was observed that the
center velocity will travel faster than the tail velocity. Those velocities equations as a
function of depth were determined using the experimental data and are shown Table 7.1
and Figure 7.4.
Table 7.1 Front and center velocities for different circulation
CASE FRONT VELOCITY CENTER VELOCITYMigration with
choke openV_. exp( 1.2~3-3.014E-4*d) vcenter = exp(l.255-4.161E-4*d)
Migration with choke closed
v,_, - exp(l.332-4.83lE-4*d) vce„,cr = exp(1.407-6.382E-4*d)
Circulation with V|S = 0.64 ft'sec
xv.,» = exp(1.613-2.780E-4 *d) vcenter = exp(l.686-2.883E-4*d)
Circulation with vk = 1/24 ft/sec
Vim = exp(l ,767-2.953E-4*d) vcenter = exp(1.772-2.274E-4*d)
The equations of front and center velocities could be determined from
experimental data , as can be seen in Table 7.1, but unfortunately few tail velocities
could be calculated In the proposed model, the tail velocity will be assumed to be equal
to the liquid velocity or equal to vmm in Figure 7.3.
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
84
1. Migration with choke dosed (vsl=0) 0
_ 10 0 0 r 2000
£■3000 a 4000
5000
1w
/f
Fr ant
t rveh icil;
_ 1000r 2000
C inter
0 1 2 3 4 5Gasfrontvelocity(ft/sec)
2. Migration with choke open (vsl=0)0-
0 1 2 3 4 5Gas center velocity (ft/sec)
_ 1000 2000
£•3000 ° 4000
5000
/ ■
i i i•ron
I Vt:IuCi >
0 1 2 3 4Gas front velocity (ft/sec)
3. Circulation with vsl=1.24ft/sec0-
^ 1 0 0 0 -r 2000-
40005000
; HV
i- i
l r
I 1
i'H 111
; 1J
0 1 2 3 4 5 6 7 8 Gas front velocity (ft/sec)
4. Circulation with vsl=0.64 ft/sec
^ 1000
rontv ilm ity
0 1 2 3 4 5 6 Gas front velocity (ft/sec)
0_ 1000 ~ 2000 £-3000 C 4000
5000
i /<,ent
i /tIocity 1
0 1 2 3 4Gas center velocity (ft/sec)
0_ 1000 ~ 2000 £■3000 ° 4000
5000
H/ y p B-|
y
A1L« ni er
- tJ/ ve u<:i I r
....
0 1 2 3 4 5 6 7 8 Gas center velocity (ft/sec)
_ 1000
£-3000 Lenter vcljci
0 1 2 3 4 5 6 Gas center velocity (ft/sec)
Figure 7.4 Gas velocity- profile for various liquids velocities
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
85
The proposed triangular gas distribution profile needs to be implemented
in a kick tolerance model, and the results should be compared with the commercial kick
simulators results. If the results show a good agreement between them, the calculation of
circulating kick tolerance may be performed using a simple hand calculator.
7.2.2 Improvement in the Gas Flow Out Measurements
Since the knowledge of the gas distribution profile along the well can simplify the
circulating kick tolerance calculations, additional experimental work should be done for
different well geometry. Better gas flow out measurements should be sought, or a
gamma-ray density meter at the flowline may be used. Moreover, the downhole pressure
sensors should be installed at least 300 ft apart, or a differential pressure sensor like
gradiomanometer should be utilized to obtain the differential pressures
7.2.3 Modification for Inclined Well
The computer program should be modified to simulate inclined or even horizontal
wells. Data from flow loops experiments performed by Nakagawa. Mendes. and Wang
can be used to obtain the empirical correlation of gas rise velocity for various angles.
7.2.4 Instrumentation of a Real Well
A rig should be fully instrumented with mud logging unit and gas out
measurements to collect data from a kick in deep water drilling. The analysis of the
collected data will improve the present circulating kick tolerance model. Furthermore,
the collected data can be used to verify the accuracy of the proposed kick tolerance
model.
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
REFERENCES
1. Al-Hussainy, R., Ramey H. J., and Crawford, P. B. “The Flow of Real Gases Through Porous Media.” Journal o f Petroleum Technology, May 1966, vol. XVIII No. 5,624 - 636.
2. Beggs, H. D., and Brill, J. P. “A Study of Two-Phase Flow in Inclined Pipes.” Journal o f Petroleum Technology, May 1973, 607 - 617.
3. Bendlksen, K. H., Malnes, D., Moe, R., and Nuland, S. “The Dynamic Two-Fluid Model OLGA: Theory and Application.” SPE Production Engineering, May 1991, 171 - 180.
4. Bourgoyne Jr., A. T., Hise, W. R., and Holden, W. R “Well Control Procedures for Deepwater Drilling Part 3 - Initiation o f Well Control Operations.” Ocean Resources Engineering, December 1978,26 - 37.
5. Bourgoyne Jr., A. T., and Casariego, V. “Generation, Migration, and Transportation of Gas-Contaminated Regions o f Drilling Fluid.” SPE 18020. 63r<* Annual Technical Conference and Exhibition of the Society o f Petroleum Engineers, Houston, TX, October 2 -5 , 1988, 19-28.
6. Caetano Filho, E. Upward Vertical Two-Phase Flow through an Annulus. Ph.D. dissertation, The University o f Tulsa, 1986.
7. Casariego, V. Generation, Migration, and Transportation o f Gas Contamined Regions o f Drilling Fluid. Ph.D. dissertation, Louisiana State University, July 1987.
8. Chenevert, M. E. “Microcomputer program helps determine kick safety factor.” World Oil, December 1983, 62 - 66.
9. Craft, B. C., Holden, W. R., and Graves, E. D. Well Design: Drilling andProduction. Prentice-Hall, Inc., 1962.
10. Ekwere, J. P., Chenevert, M. E., and Zhang, C. “A Model for Predicting the Density of Oil-Based Mud at High Pressures and Temperatures.” SPE Drilling Engineer, June 1990,141 - 148.
11. Element, D. J., Wickens, L. M., and Butland A. T. D. “An Overview of Kicking Computer Models.” International Well Control Symposium/Workshop, Baton Rouge, LA, November 27 - 29, 1989.
12. Elfaghi, F. A Pressure Losses in Subsea Choke Lines During Well ControlOperations. M.S. thesis, Louisiana State University, May 1982.
86
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
87
13. Falcao, H.L.M., Chita, L.C., and Rodrigues, R.S. “Perfura^ao em Aguas Profundas no Brasil.” (in Portuguese) 3° Congresso Brasileiro de Petroleo, Rio de Janeiro, RJ, Brazil, October 5 - 10, 1986.
14. Hagedom, A. R., and Brown, K. E. “Experimental Study of Pressure Gradients Occurring During Continuous Two-Phase Flow in Small Diameter Vertical Conduits.” Journal o f Petroleum Technology’, April 1965.
15. Hoberock, L. L., and Stanbery, S. R. “Pressure Dynamics in Wells During Gas Kicks: Part 1 - Fluid Line Dynamics.” Journal o f Petroleum Technology\ August1981, 1357- 1366.
16. Hoberock, L. L., and Stanbery, S. R. “Pressure Dynamics in Wells During Gas Kick: Part 2 - Component Models and Results " Journal o f Petroleum Technology, August 1981,1367- 1378.
17. Hoberock, L. L., Thomas, D. C., and Nickens. H. V. “Here's how compressibility and temperature affect bottom-hole mud pressure.” Oil and Gas Journal, March 22,1982, 159- 164.
18. Holden, W. R., and Bourgoyne Jr., A. T. “An Experimental Study of Well Control Procedures for Deep Water Drilling Operations." OTC 4353. 14th Annual Offshore Technology Conference, Houston. TX. May 3 -6 . 1982. 635 - 641.
19. Hovland, F., and Rommetveit, R. “Analysis of Gas-Rise Velocities From Full-Scale Kick Experiments." SPE 24580. 67th Annual Technical Conference and Exhibition of Society of Petroleum Engineers, Washington DC, October 4 - 7,1992,331 - 340.
20. Johnson, A. B., and White. D. B. "Gas Rise Velocities During Kicks.” SPE 20431 65th Annual Technical Conference and Exhibition of Society of Petroleum Engineers, New Orleans, LA, September 23 - 26, 1990, 295 - 304.
21. Johnson, A. B., and Cooper, S. “Gas Migration Velocities During Gas Kicks in Deviated Wells.” SPE 26331. 68th Annual Technical Conference and Exhibition of the Society o f Petroleum Engineers, Houston, TX, October 3 -6 ,1993,177 - 185.
22. Johnson, A. B., and Tarvin, J. A. “Field calculations underestimate gas migration velocities.” IADC European Well Control Conference, 1993.
23. Kato, S. “A New Two Phase Flow Model of Kick Control.” International Well Control Symposium,'Workshop. Baton Rouge, LA, November 27 - 29,1989.
24. Lage, A.C.V.M., Nakagawa, E. Y., and Cordovil, A.G.D.P. “Experimental Tests for Gas Kick Migration Analysis." SPE 26953, III SPE-LACPEC, Buenos Aires. Argentina, 1994.
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
88
25. Lage. A.C.V.M., Gonsalves, C.J.C., Nakagawa, E.Y., and Cordovil, A.G.D.P., “Analise do Pogo CES-112, Fase V Atraves do Simulador de Kicks.” (in Portuguese) Comunicagao Tecnica - 034/93, Petrobras/Cenpes, September, 1993.
26. Lage, A.C.V.M., Gonsalves, C.J.C., Nakagawa, E.Y., and Cordovil, A.G.D.P., “Analise do Pogo RJS-457, Fase VI Atraves do Simulador de Kicks.” (in Portuguese) Comunicagao Tecnica - 044/94, Petrobras/Cenpes, January, 1994.
27. Leach, C. P., and Wand, P. A. “Use of a Kick Simulator as a Well Planning Tool.” SPE 24577. 67th Annual Technical Conference and Exhibition, Washington DC, October 4 - 7, 1992.
28. LeBlanc, J. L., and Lewis. R. L. “A Mathematical Model of a Gas Kick.” Journal o f Petroleum Technology. August 1968. 888 - 898.
29. Mackenzie, M. F. Factor Affecting Surface Casing Pressure During Well Control Operations. M.S. thesis. Louisiana State University, August 1974.
30. Mathews, J. L. Upward Migration o f Gas Kicks in a Shut-in Well. M.S. thesis, Louisiana State University’. 1980.
Sl.Matheus, J. L., and Bourgoyne Jr., A. T. “Techniques for Handling Upward Migration of Gas Kicks in a Shut-In Well." IADC/SPE 11376. IADC/SPE Drilling Conference, New Orleans. LA. February 20 - 23, 1983, 159 - 170.
32. Mendes, P. P. M. Two-Phase Flow in Vertical and Inclined Eccentric Annuli. M.S. thesis, Louisiana State University. August 1992.
33. Miska. S.. Beck. F. E.. and Murugappan. B S. “Computer Simulation of the Reverse Circulation Well Control Procedure for Gas Kicks.” SPE/LADC 21966. SPE/IADC Drilling conference, Amsterdam, March 11 - 14, 1991.
34. Nakagawa, E. Y. Gas Kick Behavior During Well Control Operations in Vertical and Slanted Wells. Ph.D. Dissertation, Louisiana State University, December 1990.
35. Nakagawa, E. Y., and Bourgoyne Jr., A. T. “Experimental Study of Gas Slip Velocity' and Liquid Holdup in an Eccentric Annulus.” Multiphase Flow in Wells and Pipelines, ASME - FED vol. 144, 1992, 71- 79.
36. Nakagawa, E. Y., and Lage. A. C. V. M. “Kick and Blowout Control Developments for Deepwater Operations.” IADC/SPE 27497. IADC/SPE Drilling Conference, Dallas. TX. February 15 - 18, 1994.
37. Negrito, A. F., and Maidla E. E. "Optimization of Flow Rate Selection for Kick Control.” SPE 19656. 64th Annual Conference and Exhibition o f the Society' of Petroleum Engineers. San Antonio. TX, October 8 - 11, 1989.
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
89
38. Negrao, A. F. Personal Communication. October, 1995.
39. Nickens, H. V. “A Dynamic Computer Model of a Kicking Well: Part II Predictions and Conclusions.” SPE 14184. 60th Annual Technical Conference and Exhibition o f the Society o f Petroleum Engineers, Las Vegas, NV, September 22 - 25, 1985.
40. Nickens, H. V. “A Dynamic Computer Model of a Kicking Well.” SPE Drilling Engineering, June 1987, 159 - 173.
41. Petrobras/E&P - Personal Communication.
42. Pilkington, P. E., and Niehaus, H. A. “Exploding the myths about kick tolerance." World Oil, June 1975, 59-62.
43. Podio, A. L., and Yang, A. P. “Well Control Simulator for IBM Personal Computer.” IADC/SPE 14737. IADC/SPE Drilling Conference, Dallas. TX, February 10- 12, 1986.
44. Quitzav, R., and Muchtar, J.B. “Drilling Safely at Well Design Limits: A Critical Well Design Case History." IADC/SPE 23930. IADC/SPE Drilling Conference New Orleans, LA, February 18 - 21,1992, 749 - 754.
45. Rader, D. W., Bourgoyne Jr., A. T., and Ward, R. H. “Factors Affecting Bubble- Rise Velocity of Gas Kicks.” Journal o f Petroleum Technology, May 1975, 571 - 584.
46. Redman Jr., K. P. “Understanding Kick Tolerance and Its Significance in Drilling Planning and Execution.” SPE Drilling Engineering, December 1991, 245 - 249.
47. Rommetveit, R., and Olsen, T. L “Gas Kick-Experiments in Oil-Based Drilling Muds in a Full-Scale Inclined Research Well.” 64th Annual Technical Conference and Exhibition of the Society o f Petroleum Engineers, San Antonio, TX, October 8 - 11, 1989,433-446.
48. Rommetveit, R., and Vefring, E. H. “Comparison of Results From an Advanced Gas Kick Simulator With Surface and Downhole Data From Full Scale Gas Kick Experiments in an Inclined Well.” 66th Annual Technical Conference and Exhibition of the Society o f Petroleum Engineers, Dallas, TX, October 6-9 ,1991 .
49. Rygg, O. B., and Gilhuus, T. “Use of a Dynamic Two-Phase Pipe Flow Simulator in Blowout Kill Planning.” SPE 20433. 65th Annual Technical Conference and Exhibition of the Society o f Petroleum Engineering, New Orleans, LA, September 2 3-26 , 1990.
50. Rygg, O. B., Smestad, P., and Wright, J. W. “Dynamic Two-Phase Flow Simulator: A Powerful Tool for Blowout and Relief Well Kill Analysis.” SPE 24578. 67th
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
90
Annual Technical Conference o f the Society of Petroleum Engineers, Washington, DC, October 4 - 7 , 1992.
51. Santos, O. L. A. A Dynamic Model o f Diverter Operations for Handling Shallow Gas Hazards in Oil and Gas Exploratory Drilling. Ph.D. dissertation, Louisiana State University, May 1989.
52. Santos, O. L. A., Lima, H. R. de P., and Bourgoyne Jr., A. T. “An Analysis of Gas Removal From the Marine Riser.” SPE/LADC 21968. SPE/IADC Drilling Conference Amsterdam, March 11 - 14,1991.
53. Santos, O. L. A. “Well Control Operations in Horizontal Wells." SPE Drilling Engineering. June 1991, 111 - 117.
54. Scf offmann. F., and Economides, M. J. “ Controlling Kicks in Ultradeep Wells and Comparison With Shallow Wells.” SPE 22561. 66th Annual Technical Conference and Exhibition of the Society of Petroleum Engineers, Dallas. TX, October 6 - 9 , 1991.
55. Taitel, Y., Bamea, D., and Dukler, A. E. “Modeling Flow Pattern Transitions for Steady Upward Gas-Liquid Flow in Vertical Tubes.” AIChEJ. 26(3), 1980, 345 - 354. '
56. Tarvin, J. A., Walton, I., and Wand, P. “Analysis of a Gas Kick Taken in a Deep Well Drilled With Oil-Based Mud.” SPE 22560. 66th Annual Technical Conference and Exhibition of the Society of Petroleum Engineers, Dallas, TX, October 6 - 9 , 1991.
57. Thomas. D. C., Lea, J. F., and Turek, E. A. “Gas Solubility in Oil-Based Drilling Fluids: Effects on Kick Detection." SPE 11115. 57th Society of Petroleum Engineers Annual Fall Technical Conference and Exhibition, September 26 - 29, 1982.
58. Up-to-date. Published by Petrobras Public Affairs Service, number 3, April 1994.
59. Vefring, E. H., Rommetveit, R., and Borge, E. “An Advanced Kick Simulator Operating in a User-Friendly X-Window System Environment.” SPE 22314. Sixth SPE Petroleum Computer Conference, Dallas, TX, June 17 - 20, 1991.
60. Wang, Y. Gas Slip Velocity through Water and Non-Newtonian Liquids m Vertical and Inclined Eccentric Annult. M.S. thesis, Louisiana State University, December 1993.
61. Wessel, M., and Tarr, B. A. “Underground Flow Well Control: The Key to Drilling Low-Kick-Tolerance Wells Safely and Economically.” SPE Drilling Engineering, December 1991,250 - 256.
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
91
62. White, D. B., and Walton, I. C. “A Computer Model for Kicks in Water-and Oil- Based Muds.” IADC/SPE 19975. IADC/SPE Drilling Conference, Houston, TX, February 27 - March 2, 1990.
63. Wilkie, D. I., and Bernard, W. F. “Dome’s ‘Kick Tolerance’ Formula for Safe Beaufort Sea Drilling.” Ocean Industry, March 1981,33 - 36.
64. Wilkie, D. I., and Bernard, W. F. “Detecting and controlling abnormal pressure.” World Oil, July 1981,129 - 144.
65. World Oil, M y 1995,13.
66. Zuber, N., and Findlay, J. A. “Average Volumetric Concentration In Two-Phase Flow System.” Journal o f Heat Transfer, November 1965,453 - 468.
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
APPENDIX A
SHUT-IN KICK TOLERANCE
Shut-in kick tolerance can be defined as the difference between the formation pore
pressure (expressed in equivalent mud weight) and mud weight that, if a kick occurs, the
well can be shut in without breaking the weakest open hole formation (normally at the
last casing set depth).
K, = Pp - p m (A.l)
where: Kr = kick tolerance [kg/nr]
p p = formation pore pressure [kg/m3]
p„, = mud weight [kg/m3]
A.1 Maximum Shut in Casing Pressure (SICP)
If a kick occurs, the maximum shut-in casing pressure SICPmax that will not
fracture the weakest formation below the last casing set depth can be found as:
Hydrostatic pressure Fracture pressure = SICP.’ + (A.2)
due to a mud column
P f gDf = SICPm3x + p mgDj (A.3)
p j = equivalent density o f fracture [kg/m3]
g = acceleration of gravity [m/s3]
Dj = depth o f weakest formation [m]
SICPmax = maximum shut in casing pressure [Pa]
p m = mud density [kg/m3]
SICpm3x = ( p / — P
then:
or in field units:
(A.4)
92
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
93
or in field units:
SICPmlx = 0.052(p f - pS)P j (A.5)
where: SICPm2x = [psi]; p / ,p m[lbm /gal]; D7[ft]
Since after setting the casing, cementing the casing, and drilling a few feet of
formation, a leak off test (LOT) is made, the p^ can be assumed as the value obtained
from LOT, and Df can be adopted as last casing set depth.
It is not always true that the weakest formation is at the casing set depth because
normally the casing is set at the shale formation. If. for example, a sandstone appears
below the casing set depth, this sandstone should be the weakest point. On the other hand,
if the LOT is made, we know the p ; value, and in most of cases we do not know the
sandstone fracture pressure unless we also do a LOT at the sandstone depth. Therefore, in
m o st of the cases, we assume that the weakest formation is at the last casing set depth
( D, = depth of last casing set depth).
A.2 Kick Tolerance
If a kick occurs, and we assume that the gas enters into the well as slug:
Hydrostatic PressureFormation Pore Pressure
max at the bit depthdue to a mud (A.6)
at the bit depth
s ic p m„ = (p r ~ P „ ) A (A.7)
where Dh = bit depth
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
94
substituting (A.7) in (A.6) becomes:
(p / - P = p pgDb - p mgDb (A.8)
Since the kick tolerance Kt = p p - p m as defined in equation (A.1):
(A.9)
The equation (A.9) is same for field units with K, , p p , p m, p } in [Ibm/gal], and
Dj and D h [ft].
The equation (A.9) is valid only for a “zero pit gain." That is, the kick will be
detected without any increase in the pits, and no fluid will enter the well. However, the
kick is normally detected by the increase in the mud pits due to influx of fluids (water,
oil, or gas) into the well. Therefore, if we consider that the influx fluid will enter as a slug
we will have:
/ Hydrostatic PressureN + due to a influx (A. 10)
fluid column ,
Formation Pore Hydrostatic PressurePressure due to a mud column
V
SICPm3X = p„gDb - [pmg (A - Lk)+ p kgLk (A-l 1)
where: Lk = kick height [m]pk = equivalent kick fluid density [kg/m3]
Applying the concept of equation (A.l), equation (A.l 1) becomes:
(A. 12)
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
95
Equation (A.12) is the same for field units with p p , pm, pf , p k in [lbm/gal];
and Df . Dh , Lk in [ft]
A.3 Safety Factor and Surge Gradient
A practical form of kick tolerance was used to drill in the Canadian Beaufort Sea
which has a high abnormal pressure, unconsolidated formation, presence of permafrost,
gas hydrates, and plastic shale (Wilkie and Bernard, 1981). All of the problems associated
have made the optimum setting of casing string critical.
As a result, the safety factor was re-defined as a function of depth and expressed
in pressure instead of a fixed value expressed in equivalent mud weight. Moreover, a
surge gradient factor was introduced in the calculation of kick tolerance. A surge gradient
is created on restarting the mud pumps after the well was shut in to read the drill pipe and
casing pressures. The surge gradient was defined as:
The shut-in kick tolerance equation considering safety factor and surge gradient is
given by:
Reproduced w ith permission o f the copyright owner. Further reproduction prohibited w ithout permission.
where: p 1R = surge gradient [kg/m3]
P5.33x10 *YPDf 101.94
(4, -< * ,) ' ~ a T
[kg/m3](A.13)
y?= Yield Point [Pa]
dh = diameter of hole [m]
dr = diameter of drill pipe [m]
The proposed safety factors {Psj) are shown in Table A.l
96
i' ( \ Lk ( \k > = « - (P / - P . j - T J - " p*>- P*Sc1'/) u h
where: Ps, = safety factor [Pa]
gk = conversion factor [9.807 kg.m/kgf.sec']
using the conversion factor Equation A. 14 becomes:
f, D, f ^ 101.97x10 Ps{ L k {
' = aT^P/ ~ Pm' ----------Dh—
Table A.l Safety factors used in the Beaufort Sea
BELO W CASING SAFETY FACTOR
(mm) (inches) kPa psi
406 16 225 33
340 13 3/8 345 50
244 9 5/8 690 100
(A. 14)
(A-15)
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
APPENDIX B
INPUT DATA FOR KICK TOLERANCE PROGRAM
Data used to run the kick tolerance program are presented in this appendix.
B .l INPUT DATA FOR A TYPICAL DEEPWATER WELL
f t * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * ** INPUT DATA FOR KICTOL PROGRAM* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *•WELL ID: TYPICAL*• a. Complete table bellow for each section of the WELL having a different• ID and the respective depth (from top to bottom).
WELL ID DEPTH[inches] [ft]
18.8 3281.8.755 9187.8.5 11483.
b. Complete table below starting at top for drillstring OD.lD.and DEPTH
PIPE OD PIPE ID DEPTH[inches] [inches] [ft]
5.0 4.28 105836.25 2.81 11483
-------- -------- ------
c. Enter option: (1) for bit jet diameter in [ /32] (2) for total flow area in [inchesA2]
d. Enter table below with: bit jet diameter (option 1) or TFA (option 2)
JET 1 JET 2 JET 3 JET 4 Total Flow Area [/32] [/32] [/32] [/32] [inchesA2]
14. 14. 14. 0.
e. Enter the following mud properties
MUD VISCOMETER VISCOMETER DENSITY READING READING
@600 rpm @300 rpm
97
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
98
9.8 53. 34.
f. Enter the pipe absolute roughness
.00065
g. Enter the following RESERVOIR data
POROSITY PERMEABILITY THICKNESS RADIUS PORE INITIAL GAS SPECIFICPRESSURE WATER VISCOSITY GAS
SATURAT. DENSITY[d'less] [mD] [ft] [ft] [ppg] [d'less] [cp] [d’less]
.20 350. 66. 8000. 10.5 .2 0.015 0.604
h.Enter the following temperature and pressure data
SURFACE OCEAM BOTTOM STANDART STANDARTTEMP. BOTTOM HOLE TEMP. PRESSURE
TEMP TEMP.[F] [F] [F] [F] [psia]
70. 40. 220. 60. 15.0
* i. Enter the mud flow rate
* MUD REDUCED* FLOW MUD
FLOW* Ispm] [gpm]
500. 100.
j. Enter the volume(s) of pit gain(lf more than one enter data in column)
VOLUMEPIT
GAIN[bbl]
15.
k. Enter the factor that control the size of each cell (factor^ 1 - Ibbl)
1.0
1. Enter the fracture data
SKINFACTOR
[d’less]
0.0
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
FRACTURE DEPTH GRADIENT [ppge] [ft]
12.0 9187.
m. Enter the rate of penetration ROP [ft/hours]
n. Enter the time to close the BOP after the kick was detected [min]
1.0
o. Enter the time between closing the BOP and starting to pump [min]
5.0
p. Enter the inside diameter of:Choke line (in) and Kill line (in)
3.0 3.0
q. Enter option: (1) for circulation through kill line only(2) for circulation through kill line AND choke line (paralell)
1
r. Enter the pressure above the bottom hole pressure to be maintenaine security factor (normally 50-300 psi)
0.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
100
B.2 INPUT DATA FOR THE WELL RJS - 457
INPUT DATA FOR KICTOL PROGRAM
WELL ID: RJS-457
a. Complete table bellow for each section of the WELL having a different ID and the respective depth (from top to bottom).WELL ID DEPTH
[inches] [ft]
18.8 1132.8.535 155528.5 18374
b. Complete table below starting at top for drillstring OD.ID.and length
PIPE OD PIPE ID DEPTH [inches] [inches] [ft]
5.0 4.28 17291.6.5 2.81 18374.
c. Enter option: (1) for bit jet diameter in [ '32] (2) for total flow area in [inches'^]
d. Enter table below with: bit jet diameter (option 1) or TFA (option 2)
JET 1 JET 2 JET 3 JET 4 Total Flow Area[/32] [/32] [/32] [/32] [inchesA2]
17. 17. 17. 0.----- ----- ----- ----- ----------------
e. Enter the following mud properties
MUD VISCOMETER VISCOMETER DENSITY READING READING
@600 rpm @300 rpm [ppg] [d’less] [d’less]
16. 53. 34.
f. Enter the pipe absolute roughness
.00065
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
101
g. Enter the following RESERVOIR data
POROSITY PERMEABILITY THICKNESS RADIUS PORE INITIAL GAS SPECIFICPRESSURE WATER VISCOSITY GAS
SATURAT. DENSITY[d’less] [mD] [ft] [ft] [ppg] [d'less] [cp] [d'less]
.08 50. 66. 8000. 17.0 .2 0.015 0.604
h.Enter the following temperature and pressure data
SURFACE OCEAM BOTTOM STANDART STANDART TEMP. BOTTOM HOLE TEMP.
TEMP TEMP.[F] [F] [F] [F]
70. 50 . 2 9 0 . 60.
i. Enter the mud flow rate
MUD REDUCED FLOW MUD
FLOW [gpm] [gpm]
4 0 0 . 100.
j. Enter the volume(s) of pit gaindf more than one enter data in column)
VOLUMEPIT
GAIN[bbl]
10.
k. Enter the factor that control the size of each cell (factor=i - lbbl)
2.0
1. Enter the fracture data
FRACTURE DEPTH GRADIENT [ppge] [ft]
17.5 15552.
PRESSURE
[psia]
15.0
SKINFACTOR
[d’less]
0.0
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
m. Enter the rate o f penetration ROP [ft/hours]
19.7
n. Enter the time to close the BOP after the kick was detected [min]
0.5
o. Enter the time between closing the BOP and starting to pump [min]
0.5
p. Enter the inside diameter of: Choke line (in) and Kill line (in)
2.5 2.5
q. Enter option: (1) for circulation through kill line only(2) for circulation through kill line AND choke line (paralell)
1
r. Enter the pressure above the bottom hole pressure to be maintenaine security factor (normally 50~200 psi)
0.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
103
B.3 INPUT DATA FOR THE WELL CES - 112
• • • * • * • * * * * * * • * * • • • * « • * * * * * * « * * • * • * * * * * * * « *
* rNPUT DATA FOR KICTOL PROGRAM* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *’ WELL ID: CES-112* a. Complete table bellow for each section of the WELL having a different* ID and the respective depth (from top to bottom).* WELL ID DEPTH* [inches] ♦ _____
[ft]
17.6 4311.8.7 12993.8.5 14764.
b. Complete table below starting at top for drillstring OD,ID,and length
PIPE OD PIPE ID DEPTH [inches] [inches] [ft]
5.0 4.28 13780.6.5 2.81 14764.
c. Enter option: (1) for bit jet diameter in [ /32](2) for total flow area in [inchesA2]
d. Enter table below with: bit jet diameter (option 1) or TFA (option 2)
* JET 1 JET 2 JET 3 JET 4 Total Flow Area* [/32] [/32] [/32] [/32] [inchesA2]
12. 12. 12. 0.
e. Enter the following mud properties
MUD VISCOMETER VISCOMETER DENSITY READING READING
@600 rpm @300 rpm [ppg] [d'less] [d’less]
9.5 46. 29.
f. Enter the pipe absolute roughness
.00065
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
104
g. Enter the following RESERVOIR data
POROSITY PERMEABILITY THICKNESS RADIUS PORE INITIAL GAS SPECIFICPRESSURE WATER VISCOSITY GAS
SATURAT. DENSITY[d’less] [mD] [ft] [ft] [ppg] [d'less] [cp] [d'less]
-10 500. 34.5 8000. 10.2 .2 0.015 0.604
h.Enter the following temperature and pressure data
SURFACE OCEAM BOTTOM STANDART STANDARTTEMP. BOTTOM HOLE TEMP. PRESSURE
TEMP TEMP.[F] [F] [F] [F] [psia]
70. 40. 200. 60. 15.0______ _____ ----—
i. Enter the mud flow rate
MUD REDUCED FLOW MUD
FLOW [gpm] [gpm]
430. 100.
j. Enter the volume(s) o f pit gain(If more than one enter data in column)
VOLUMEPIT
GAIN[bbl]
16.
k. Enter the factor that control the size o f each cell (factor=l ~ lbbl)
1.0
1. Enter the fracture data
FRACTURE DEPTH GRADIENT [ppge] [ft]
10.5 12993.
SKINFACTOR
[d'less]
0.0
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
m. Enter the rate o f penetration ROP [ft/hours]
10.
n. Enter the time to close the BOP after the kick was detected [min]
0.5
o. Enter the time between closing the BOP and starting to pump [min]
0.5
p. Ent :r the inside diameter of:Choke line (in) and Kill line (in)
3.0 3.0
q. Enter option: (1) for circulation through kill line only(2) for circulation through kill line AND choke line (paralell)
1
r. Enter the pressure above the bottom hole pressure to be maintenaine security factor (normally 50-100 psi)
50.
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
APPENDIX C
GAS DISTRIBUTION PROFILE
The gas fraction as a function o f time was calculated from the experimental data
and is shown here in odd numbered figures for a given depth. The gas distribution profile
as a function of depth was derived from the gas fraction and is shown in even numbered
figures. Experimental details are shown in Tables C.l through C.3, and the drilling fluid
properties used are shown in Table C.4.
Table C .l Test matrix for mud and natural gas experiments with gas injected through tubing
Downhole Pressure Sensors 1200 ft Apart
Test#
pit gain
(bbl)
Pumpspeed(spm)
Choke back pressure
(psi)Note
10 20 0 32 6262M l 10 170
M2 0 choke closed Failure due to valve leakM3 10 32 100 Downhole pressure lostM4 20 62 170 Downhole pressure lostM5 20 0 choke open Downhole pressure lostM6 10 32 170M7 20 0 choke closed |
Table C.2 Test matrix for mud and natural gas with sensors 1,200 ft apart
Gas Was Pump Down Through 3.5 J55 Eue (9.3 #/ft) X 1.66” N-80 (3.02 #/ft) Annulus
Test#
Gas pump down speed
(spm)
Pump speed (spm)
Choke back pressure
(psi)Note
62 82* 0 32 62M8 82 0 choke openM9 82 32
M10 62 62 170 on line data was lostM il 62 32 170 middle downhole sensor failedM12 62 62 180 middle downhole sensor failedM13 82 62 170 middle downhole sensor failedM14 82 0 choke closed middle downhole sensor failed82* spm — v=1.64 ft/sec
106
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
107
Table C.3 Test matrix for mud and natural gas with sensors 100 ft apart
Test#
Gaspumpdownspeed(spm)
Position o f downhole
tools
Pumpspeed(spm)
Viscousfluid
Chokeback
pressure Note
62 82 b* m* t* 0 32 62 ves noM15 82 B 62 N 0 one sensor failedM16 62 B 62 N 0 one sensor failedM17 82 B 62 N 170 one sensor failedM18 82 M 32 N 170 one sensor failedM19 82 M 62 N 180 one sensor failedM20 82 M 0 N 170 one sensor failedM21 82 M 62 N 170 one sensor failedM22 62 M 32 N 170 one sensor failedM23 62 M 62 N 0 one sensor failedM24 82 T 32 N 200 one sensor failedM25 82 T 0 N choke open one sensor failedM26 82 T 62 N 200 one sensor failedM27 82 M 62 Y choke openM2 8 82 M 0 Y choke
closedM29 82 M 0 Y choke open
b* = bottom (on line tool @ 5,422 ft); m* = middle (on line tool @ 2,761ft), t* = top (on line tool @ 100 ft)
Table C.4 Drilling fluid properties used in the experiments
Experiment#
Mudweight(lb/gal)
Marshviscosity
(sec)
Plasticviscositv
(cp)
Yieldpoint
Gel strengthlOsec | 10m in
Ibf/lOOsq ftMl 9.9 54 15 5M2 9.9 54 15 5
M3 -M 4 9.9 53 15 5M5 9.9 54 15 5
M6 - M7 9.9 53 15 5M8 9.9 54 14 5
M 9-M 10 10.0 53 12 5M il -M 12 9.8 49 10 2
M13 9.6 62 15 8M14 9.6 58 15 7
M 15 - M 16 9.7 40 12 9 3 10M 1 7 -M 1 8 -M 1 9 9.6 38 11 6 2 5M 20-M 21 -M 22 9.6 40 12 3 2 9
M23 - M24 9.6 42 12 6 2 15M25 - M26 9.6 43 12 6 2 13M27 - M28 9.7 78 30 15 4 25
M29 9.7 78 30 15 4 25
Reproduced w ith permission o f the copyright owner. Further reproduction prohibited w ithout permission.
Gas
fr
actio
n G
as
frac
tion
Gas
fr
actio
n G
as
frac
tion
Gas
fr
acti
onType of experiment: Circulation at 62 spm (vsl=1.24 ft/sec)
Experiment: M l
0.80-0.60-0.40-0.20-0.00-
| Interva = 0 - 1,(100 ftg 1jj I i i
i i f f lW ™ L0 20 40 60 80 100 120 140
0.301
0.20-0.1510.10H0.050.00
T im e (m inutes'-1,000 - 2.238 ft 1LI 1
: / \: / HJlwif I= /
- r - r - 4 1 1 W w p L
0.30-0.25-0.20-0.15-0.10-0.05-
0 20 40 60 80 100 120 140Time (minutes)
- 2,23^ 3 476 ft
i /1 / ViI 1 • ! i i i
0.30-0.25-0.200.1510.1010.050.00
20 40 60 80 100 120 140Time(minutes)
i Interval = 3.476 - 4.714 f t:;: A; / V
/■ , s -----i * * 120 40 60 80
0.301 0.251 0.201 0.151 0.101 0.05 0.00
100 120 140Time(minutes)
: Interval = 4,714 -5 .822 t:
=::
4 .
0 20 40 60 80 100 120 140Time(minutes)
O ft
On-line
1,000 ft
Top
2,238 f
M iddle
3,476 ft
Bottom
4,714 f
5,822 ft
TD = 5,884 ft
Figure C.l Gas fraction for different depths and times for experiment Ml
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
109
Experiment:Ml
51 min
0 1000
« 2000i£ 3000- I " 4000H
50006000-
2000-
0.00 0.05 0.10 0.15 0.20 Gas fraction
q 4000- 5000- 6000-
—H - 55 min
i
?
! “ u
;
i l — — I
I1 -------1------- '
60 min
0 1000
« 2000 5 3000 q 4000
5000 6000
T
0.00 0.05 0.10 0.15 0.20 Gas fraction
0.000.05 0.10 0.15 0.2 Gas fraction
S 2000 £ 3000' q 4000'
5000 6000
— 70 min
■ BlB II
IK----- —■
n
i
i ____
= 3000 o 4000
75 min
0 1000
S2000- 5 3000- Q 4000-
5000 6000
80 min
X 5=
0.00 0.10 0.20 0.30 Gas fraction
0.000.05 0.100.15 0.20 Gas fraction
0.00 0.05 0.100.15 0.2 Gas fraction
—B 85 min —H — 90 min —B 110 min
0 1000
« 2000
1
= 3000 I " 4000-
5000 j 6000
I
T
° 21000-11« 2000-ii= 3000- I" 4000-
5000- 6000-
t0q
1000£;2000H: !
= 3000 q 4000
5000 i6000 i
0.00 0.05 0.10 0.15 0.20 Gas fraction
0.00 0.05 0.10 0.15 0.20 Gas fraction
0.00 0.05 0.10 0.15 0.2 Gas fraction
Figure C.2 Gas fraction as a function of depth for experiment M l
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
Gns
frac
tion
Gas
fr
actio
n (j
flS Tr
actio
n G
as
frac
tion
Gas
fr
acti
on
110
Type of experiment: Circulation at 32 spm (vsl=0.64 ft/sec)Experiment: M6
Interval = 0 -1,000 ft
0.00150 200 250
Time(minutes)
0.3010.25 H0.20 H0.15HO.lOi0.050.00
: 1 _4 1 -1 000 **1 mji krU 1 I: 1 j0 50
0.30-0.25-0.20-0.15-0.10-0.05-=o .oo4-
100 150 200 250Timefminutes)
| In te rv a l = 2 .228 - 3 ,456 f t111
1 - A - I1 ■ ■ ■ ■ 1[— 4-------- r - j — ------- - ----------- --- 1 — 1 l
0.30-q 2S 3 Interval -0l20-0.15- 0.10- 0.05- 0.00-
T
0.301 0.25 i
0 50 100 150 200 250Time(minutes)
0.15i0.10H0.050.00
: Interval = 4.684 - 5,82: ft::::
J
0 50
50 100 150 200 250Time(minutes)
100 150 200 250Time(minutes)
Oft
On-line 1,000 ft
Top 2,228 fi
Middle 3,456 ft
Bottom 4,684 fi
5,822 ft
TD = 5,884 ft
Figure C.3 Gas fraction for different depths and times for experiment M6
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
Dept
h (ft
) De
pth
(ft)
Dept
h (ft
)
111
Experiment: M6
2000-3000-
5000-6000-
— 50 min
1--------■
1
1
1
60 min 70 min
0-p 1000
§2000•5^3000- q 4000-
5000-6000-
0- 1000-*
§2000- £ 3000- q 4000-
5000- 6000-
0.00 0.05 0.10 0.15 Gas fraction
0.00 0.05 0.10 0.15 Gas fraction
0.00 0.05 0.10 0.15 Gas fraction
90 min
1000-:§ 2000-1
f . 3000^1*4000
50006000
100 mm 110 min
S 3000
0.00 0.05 0.10 0.15 0.00 0.05 0.10 0.15 0.00 0.05 0.10 0.15Gas fraction Gas fraction Gas fraction
—B — 120 min —B — 150 min —B — 150 min
o-1000-2000-3000-4000-5000J6000-
t ----- S■j
B-|■
rH B
----------
o-1000-
§2000-
f . 3000'q 4000-
5000- 6000-
0.00 0.05 0.10 0.15 Gas fraction
0.00 0.05 0.10 0.15 Gas fraction
0 1000
§ 2 0 0 0 5 3000 q 4000-
5000-7 6000-
0.00 0.05 0.10 0.15 Gas fraction
Figure C.4 Gas fraction as a function of depth for experiment M6
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
112
Type of experiment: Migration with choke closedExperiment: M7
c 0 .2 0_o
0 .1 5eer4-
0 .1 0V3«5 0 .0 5
O 0 .0 0
: Interval = 1I - 1.000 f t i: 1:
ft u bj l , 1 m
5 0 100 150 200 2 5 0
eeo
0.20- 0 .1 5 - 0.10- 0 .0 5 -
0 J
3 Interval =1 ,000 - 2,228 t1
t
-̂---------------iJJ ijiii latfi
5 0 100 150 2 0 0 2 5 0Time(minutes)
c 0 .20 -g
0 .15 -:ec> 0 .10 -f(fiec 0 .0 5
0 .0 0 J
Interval = 2,228 - 3,45 . ft
5 0 100 150 200 2 5 0
= 0.20-q
* | 0 .1 5 i | 0.10- S 0 .0 5 H
Interval = 3,456-4,68-1 ft
0 5 0 100 150 200 250Time (minutes)
- 0 .2 0 -30 .15 -:
ec£ 0 .10-:(fiec 0 .0 5 -E
o 0 .0 0 J
Interval = 4,684 - 5,82; : ft
5 0 100 150 2 0 0 25 0Time (minutes)
Oft
On-line
1,000 ft
Top
2,228 f
M iddle
3,456 ft
Bottom
4,684 f
5,822 ft
TD = 5,884 ft
Figure C.5 Gas fraction for different depths and times for experiment M7
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
113
Experiment: M7
1000- S2000- :S 3000- q 4000-
5000- 6000-
—B — 50 min
I------------■
1
B
1
60 min
o - p 1000
£ 2 0 0 0 i t5 3000- ( f 4000-
0.00 0.05 0.Gas fraction
5000-f 6000-
10 0.00
£ 2000
70 min
q 4000- 5000- 6000-
0.05 0.10Gas fraction
0.00 0.05 Gas fraction
0.10
1000 S2000-
3000' O 4000'
5000- 6000
— 80 min
1------------■
r ■“ 1■ i ,,
1
01000 +
£ 2 0 0 0-£^3000 q 4000
5000 i 6000
90 min
OH f1000+
£ 2000-j f £ 3000 §4000
5000 6000
100 min
£0.00 0.05 0.10
Gas fraction0.00 0.05 0.10
Gas fraction0.00 0.05 0.10
Gas fraction
110 min 120 min 180 min
1000- £2000- :£ 3000- q 4000-
5000- 6000-
HE-- -- - - - - - - -H I - - - - - - - - - - -
: B ----------
HI -- - - - - - - -
I ——
1
o- 1000-
£2000- £ 3000- 1*4000-
5000- 6000-
I- f —
I _IIF------j |
= £ —
1
0 llo o o i
£ 2 0 0 0 3000o.oq 4000H 5000-! 6000
II
0.00 0.05 Gas fraction
0.10 0.00 0.05 0.10Gas fraction
0.00 0.05 0.10Gas fraction
Figure C.6 Gas fraction as a function of depth for experiment M7
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
114
Type of experiment: Migration with choke openExperiment: M8
e 0 .8 0 1.2 a /rc\ J Interv il = 0 1,000 [ft
U.oU ;
U.4U"(mCSS U.2U ;
° 0 .0 0 j r r-r-r
5 0 7 0 9 0 110 130 150 170 190 2 1 0 2 3 0
•2 0.161ues 0 .1 2ch 0 .0 8wes 0 .0 4o 0 .0 0
0 .2 00 .1 6
e; 0 .1 2£ 0 .0 8«5a 0 .0 4O 0 .0 0
0 .2 0 -o 0 .1 6 -
: Interval #1 000 - 2L228 f t: I I\ U . . I llj:
J f"l 'I . . . . 1 u • I 1 •5 0 7 0 9 0 1 1 0 130 150 170 190 2 1 0 2 3 0
Time(minutes)1 In ten al = 2 .2 2 8 -1 4 5 6 t1
1
j---- - 41 1 • 1 -----/ ' ■ ■ 15 0 7 0 9 0 110 130 150 170 190 2 1 0 230
Time(miRutes)
S 0.12 * 0 .0 8 -
o.oo-
: In ten a l= 3!.456 J 1.684 f |:jI ^ ):
• * * . . . [ . , T ] ‘ • ‘
- 0.20- •2 0 .1 6 - 3 0.12-
* 0 .0 8 -
5 0 7 0 9 0 110 130 150 170 190 2 1 0 2 3 0Time(minutes)
3 0.04-° o.oo-
1 In ten al = 4,6 8 4 -: i,822 i t114
5 0 7 0 9 0 110 130 150 170 190 2 1 0 2 3 0Time(minutes)
O ft
O n-line
1,000 ft
T op
2,228 f
M iddle
3,456 ft
lottom
,684 ft
5,822 ft
TD = 5,884 ft
Figure C.7 Gas fraction for different depths and times for experiment M8
Reproduced w ith permission o f the copyright owner. Further reproduction prohibited w ithout permission.
Dept
h (ft
) De
pth
(ft)
Dept
h (ft
)
115
Experiment: M8
0-1000-2000-3000-4000-5000-6000-
— 50 min
■—n
■
■
■
01000
£2000£ 3000 q 4000
5000^ 6000
o o o o o o o o O — N r r l T < n ' O P ' o ' o o o o o ’ o ' o '
Gas fraction
0100020003000400050006000
—® 130 min
i J i—
.11: 1
m
J■ t
j i ii i
200 min
o-1000-2000-3000-4000-5000-6000-
%
70 min
1000 £2000 £ 3000
4000 5000 6000
—H — 100 min
■■1 R
IT: ■-11
in 11 1o o o o o o o oO — N n ^ ’ i n ' O t 'o o o ' o ' © o ' o ’ o '
Gas fraction
150 min
o o o o o o oO —̂ 04 i/%o o* o ' o ' o ' o* o '
Gas fraction
0- 1000-
£2000- £ 3000- q 4000-
5000- 6000-
0 1000
£2000- ■5̂ 3000- | ’ 4000-
5000- 6000-
170 min
-L
i t i
I
o o o o o o o o o o ' o o o o* o ' o
Gas fraction
o o o o o o o o o — ( S n T u - . o r -o o ' o ' o ' o o ' o ' o '
o o o o o o o cO — M r , T f ' A ' S fo o o ' o o ' o ’ o ’ c
—® 220 min —® 228 min
o- 1000-
£ 2000- •£ 3000- q 4000-
5000- 6000-
o-1000-
£2000 '
O O O O O O O O O — c N m T T * / - . o r - *o o* o* o ' o ’ o ' o ' o '
Gas fraction
io o o o o o o o o ' o ' o ' o ’ o o ' o ' o
Gas fraction
£ 3000 q 4000-i)
5000- 6000-
t
I io o o o o o oo - n n m oo> o o ' o ' o ’ o ' o
Gas fraction
Figure C.8 Gas fraction as a function of depth for experiment M8
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
0L 0
ni'n
i
1 1
1 1
1 r
-0Z, 0
Gas
fr
actio
n G
as
frac
tion
Gas
fr
actio
n G
as
frac
tion
Gas
fr
acti
on
116
Type of experiment: Circulation at 32 spm (vsl=0.64 ft/sec)Experiment: M9
0.60|0.5010.4010.3010.2010.1010.00"
Interv; il = 0 - 1.000
0 20 40 60 800.400.300.200.10
Int< rval =1 000 - 2, 228 ft
i W T|- ." ." P lT V T *
0.40-0.30-0.20-0.10-0.00-
0 20 40 60 80 100 120 140 160Time (minutes)
Interv: il= 2,2 28-3 ,4 56 ft
/ \: V
. . . . . . i t . . . .
0.40- 0.30 H
0 20 40 60 80 100 120 140 160Time (minutes)
0.20 i 0.10 0.00
Interv: il = 3,456 - 4,6 54 ft •
jj / \ l: / K
r • • • ! ' . . 1 1 1 ♦ ‘ . . 4
0.20-0.15-:
20 40 60 80 100 120 140 160Time (minutes)
0.1010.050.00
~ Interval = 4,684 - 5,82: : ft::
!
0 50 100 150 200 250Time (minutes)
O ft
On-line 1,000 ft
Top 2,228 f
Middle 3,456 ft
Bottom [ J 4,684 ffi
5,822 ft
TD = 5,884 ft
Figure C.9 Gas fraction for different depths and times for experiment M9
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
Dept
h (ft
) De
pth
(ft)
Dept
h (ft
)
117
Experiment: M9
20 min 30 min
o- 1000-
3 2000 -
•5̂ 3000- q 4000-
5000- 6000-
1000 3 2000 = 3000 q 4000
5000 6000
— 40 min
i ■
\m 1M .
: p - l i
i i - a
i i0.00 0.20 0.40 0.60
Gas fraction
0.0 0.2 0.4Gas fraction
0.6 0.00 0.20 0.40 0.60 Gas fraction
o-1000-2000-3000-4000-5000-6000-
—® 50 min
1--------|■11 "■
1T -1B
1---- 1H;
1
o- 1000-
3 2 0 0 0 - = 3000- i 4000-
5000- 6000-
— 55 min
■■
I f iI1
ri 1K----- — ■
i i
o- 1000-
3 2000- = 3000- q 4000-
5000- 6000-
60 min
0.00 0.20 0.40 0.60 0.00 0.20 0.40 0.60 0.00 0.20 0.40 0.60Gas fraction Gas fraction Gas fraction
—H — 65 min — 70 min —B - 80 min
o- 1000
3 2 0 0 0 - = 3000- q 4000-
5000 6000
I t
* 01000-
3 2 0 0 0 - = 3000- q 4000-
5000- 6000-
I t
0.00 0.20 0.40 Gas fraction
0.600.00 0.20 0.40 0.60
Gas fraction0.00 0.20 0.40 0.60
Gas fraction
Figure C.10 Gas fraction as a function of depth for experiment M9
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
118
Type of experiment: Circulation at 32 spm (\sl=0.64 ft/sec)Experiment: M il
_ 0.80TT"— : 5 3 Interval = 0-1,0 )0ft•- 0.60 4" n in l n« 0.40 4 i / \g 0.20 q j | ( \£ o.oo I !
0 20 40 6
00o
o V-4 )o i: Time (r
>0 14 ninutes
- 0.601 o Interval =1,000 - 2,225 ft sZ 0.40Fes / \ l« 0.201 j J k^ O.OOT . . ,
20 40 60 80 100 120 140Time (minutes)
- 0.30-qcZ 0.20czt 0.10-:
Interva = 2,225-4.675
0.000 20 40 60 80 100 120 140
Time(minutes)
= 0.20 | 0.15 .2 0.10
o0.050.00
Interval = 4,675 5,822
1 !20 40 60 80 100 120 140
Time (minutes)
O ft
O n-line
1,000 ft
Top 2,225 f
M iddle
3,450 f t
(failed)
Bottom 4,675 f
5,822 ft
..TJQ .=..5*884. f t .
Figure C.l 1 Gas fraction for different depths and times for experiment M l 1
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
Depth
(ft)
De
pth
(ft)
Depth
(ft
)
119
Experiment: M il
1000-112000-jr
3000-i4000-i-
6000-
—B - 30 min
■-------■ 1
i
l1
0 *looo-ii
S 2000i f£ 3000- Q 4000-
s50 min
° Tlooo-p 2000 3000 4000 5000 6000
0.00 0.20 0.40 Gas fraction
70 min
o-1000-2000-300040005000-q6000
f-H I
5000i6000
40 min 45 min
I H i
o-1000H
3 2 0 0 0 - £ 3000- q" 4000-
5000- 6000-
0.00 0.20 0.40 0.60 Gas fraction
0.0 0.2 0.4Gas fraction
0.6 0.00 0.20 0.40 0.6 Gas fraction
01000
« 2000 £ 3000 |"4000-i
5000-1 6000
55 min
" I .
o-p 1000
g 2 0 0 0£ 3000 q 4000
5000-6000-
60 min
0.60 0.00 0.20 0.40 Gas fraction
0.60 0.00 0.20 0.40 Gas fraction
0.60
—B - 80 min —B — 90 min
1000-11- S 2 0 0 0 if £ 3000-; 1*4000-
5000- 6000-
0q1000
s 2000 i f£ 3000- 1*4000-
5000 i 6000
0.00 0.20 0.40 0.60 Gas fraction
0.00 0.20 0.40 0.60 Gas fraction
0.00 0.20 0.40 0.6 Gas fraction
Figure C.12 Gas fraction as a function of depth for experiment M il
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
Gas
fr
actio
n G
as
frac
tion
Gas
fr
actio
n G
as
frac
tion
Type of experiment: Circulation at 62 spm (\sl=1.64 ft/sec)Experiment: M12
0.40 ~ 0.301 0.201 0.10 0.00
: Interval = 0 - 1,(r r,Aj Ai\ / Ij i \ _ .
i ■ ■0
0.4(T0.3CF0.20T0.10"o.otr
20 40 60 80 100 120 140Time(minutes)
'̂ >■<4 ii
nterval = *1,000 - ; ,225 ft
20 40 60 80 100 120 140Time (minutes)
0.40-0.30-0.20-0.10-0.00-
3 Interval = 2,225 - 4,675 rt1
j
f
0 20 40 60 80 100 120 140Time (minutes)
0.200.150.100.050.00
Interval = 4,675 -5 ,822 t
20 40 60 80 100 120 140Time (minutes)
O ft
On-line
1,000 ft
Top
2,225 f
Middle
3,450 ft
(failed)
Bottom
4,675 f
5,822 ft
..TJD.= .5JS84.Xt.
Figure C.13 Gas fraction for different depths and times for experiment M12
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
121
Experiment: M12
20 min
o-100(H
§2000- £ 3000- |"4000-
5000- 6000
25 min
OH 1000H
§2000- ■£^3000- q 4000-
5000- 6000-
0.00 0.20 0.40 0.60 Gas fraction
0.2 0.4 0.6Gas fraction
35 min
<=: 2000
40 min
•£ 3000 O 4000
5000 6000
o- 1000-
§2000- £ 3000- q 4000-
5000- 6000-
0.00 0.20 0.40 0.60 Gas fraction
01000%
§ 2 0 0 0 £ 3000 q 4000
5000 6000
30 min
0.00 0.20 0.40 0.6 Gas fraction
0-11 1000%
§ 2 0 0 0£ 3000- n 4000-
5000i6000
45 min
0.00 0.20 0.40 0.60 Gas fraction
0.00 0.20 0.40 0.60 Gas fraction
— 50 min ' H 55 min 70 min
§ 4000
02 1000%
§2000% £ 3000- § 4000-
5000- 6000-
0%lo o o -ii
§2000-ir
o.Q>Q£ 3000i ^4000-
5000 6000t
0.00 0.20 0.40 0.60 Gas fraction
0.00 0.20 0.40 0.60 Gas fraction
0.00 0.20 0.40 0.6 Gas fraction
Figure C.14 Gas fraction as a function of depth for experiment M12
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
122
Type of experiment: Circulation at 62 spm (\sl=1.64 ft/sec)Experiment: M13
mm0.40-© 0.30-es£ 0.20-crCC o .io -o o.oo-
0.400.30
eeeh 0.20«5ec 0.10o0.00
Interval = 0 - 1,000 ft
20 40 80 100 120 Time(min
interval =1,000 - 2,
80 100 120 Time (minutes)
- 0.40- 3 0.30-S 0 .2 0 -
s 0.10- ° o.oo-
a Interval ■= 2,222-. 1,666 ft
i13i • * 1 i • • • ■ • • 1 • ■ • 1
20 40 60 80 100 120Time (minutes)
s 0.200.15
<h 0.10ifiCS 0.05o 0.00
jj Interval = 4,666 - 5 ,822 ft
iii1 ■ ■ ■ ‘
20 40 60 80 100 120Time(minutes)
0 ft
On-line 1,000 ft
n Top J 2,222 ft
r*i Middle 3,444 ft
(failed)
Bottom 4,666 f
5,822 ft
Figure C.15 Gas fraction for different depths and times for experiment M13
Reproduced w ith permission of the copyright owner. Further reproduction prohibited w ithout permission.
Depth
(ft
) De
pth
(ft)
Depth
(ft
)
123
Experiment: M13
1000-2000-jf3000-i4000 H
6000-
—B - 20 min
I-------■
1
1
30 min
o- 1000 -
S 2 0 0 0 - :5 3000- Jj" 4000-
5000- 6000-
40 min
200030004000
6000-
—H — 45 min
II-------■ ■I
11----- ■
j
60 min
o-1000-2000 -
3000-4000-5000-6000
0-^ 1000
« 2000|[3000 i 4000-
5000-1
0.00 0.20 0.40 0.60 Gas fraction
0.0 0.2 0.4Gas fraction
0.6
0 *1000
S2000- •S 3000- I t 4000-
5000-6000
50 min
o- 1000-
S 2 0 0 0 - ■5 3000- q 4000-
5000- 6000-
u ------0.00 0
G.20 0 as frac
.40 0 tion
j —■— 55 min
1 _ 1L i
1
-!—0.00 0.20 0.40 0.60
Gas fraction0.00 0.20 0.40 0.60
Gas fraction0.00 0.20 0.40 0.60
Gas fraction
— 70 min —® 80 min
£
o-* 1000
S 2 0 0 0 tE 3000^ q 4000
5000 6000
o-1000-lt
S20oo-jr5 3000- Q 4000-
50006000
0.00 0.20 0.40 0.60 Gas fraction
0.00 0.20 0.40 0.60 Gas fraction
0.00 0.20 0.40 0.60 Gas fraction
Figure C.16 Gas fraction as a function of depth for experiment M13
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
124
Type of experiment: Migration with choke dosedExperiment: M14
_ 1.00 -a .2 0.80 i
:0.40 30.20 ^ 0.00
Interval = 0 - 1.000 ft
2 0.60 -
50 100 150 200 250 300 350 400
£ 0.2(
0 Interv: II © o ) - 2.222 ft I0 ju
10i r i
i 1t 1
50 100 150 200 250 300 350 400Time(minutes)
g 0.30 iI 0 .2 0 -jg 0.10 u 0.00
Interval = 2,2 22-4.666 ft | |
!| |
1 i -0 50 100 150 200 250 300 350 400
Time (minutes)
“ 4 Interval =n i t - -{ 4,666 - 5 ,822 ftU.lJi 4
O 3 C5 A i n 1U*lUT
Wa j
CS v.U-> J u 0 .00 ■
20 40 60 80 100 120 Time(minutes)
0 ft
On-line 1,000 ft
Top 2,222 f
Middle 3,444 ft (failed)
Bottom 4,666 f
5,822 ft
..TO.=.5*884.0.
Figure C.17 Gas fraction for different depths and times for experiment M14
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
Depth
(ft)
De
pth
(ft)
Depth
(ft
)
125
Experiment: M14
1000H*2000i f 3000-i4000-i500016000
—® 30 min
1-------■
1 '
1
40 min
o- 1000-
£ 2000 -
■£ 3000- q 4000-
5000 6000
o- 1000 -
£ 2000 -
£ 3000- q 4000-
5000- 6000-
— 50 min
i i i0.00 0.20 0.40 0.60
Gas fraction0.0 0.2 0.4 0.6
Gas fraction0.00 C.20 0.40 0.60
Gas fraction
o-1000 -
2000 -
3000-4000-5000-6000-
- » ~ 60 min
1 B-----] |
I ■
70 min 80 min
o- 1000-
S 2 0 0 0 - £ 3000- I" 4000-
5000- 6000-
leto-
1000-ft- £ 2000 -
£ 3000- I" 4000-
5000-6000-
0.00 0.20 0.40 0.60 Gas fraction
0.00 0.20 0.40 0.60 Gas fraction
0.00 0.20 0.40 0.60 Gas fraction
90 min ~ S — 100 min —B — 200 min
02 1000-j* 20003000-i 4000-i 5000 6000
0-1000*
£ 2000 -
£ 3000- q 4000-
5000-6000
0^1000
£ 2 0 0 0 i f£ 30001 a 40001
50006000
0.00 0.20 0.40 0.60 Gas fraction
0.00 0.20 0.40 0.60 Gas fraction
0.00 0.25 0.50 0.75 l.00 Gas fraction
Figure C.18 Gas fraction as a function of depth for experiment M14
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
Gas
fr
actio
n fr
flct
i011
frac
tion
Gas
fr
acti
on
Type of experiment: Circulation at 62 spm (\sl=1.24 ft/sec)Experiment: M15
0.10
0.05
0.00
Interval = 0 -5 ,4 12 ft
0 5 10 15 20 25 30 35Tirae(minutes)
0.20H
0.10
0.00
Interv: j =5,422
t5
- 5, 522 rt
-I10 15 20 25 30 35
Time(minutes)
0.60-
0.40-i
0.20-i 0.00 J
Interva = 5,52^ -5 ,622
0 10 15 20 25 30 35Time(minutes)
0.200.150.100.050.00
Interva = 5,622 - 5,822 ft
10 15 20 25 30 35Time (minutes)
Oft
On-line
5,422 ft
Top
5,522 ft
M iddle
5.622 ft
JL, Bottom
5,722 ft
(failed)
5,822 ft
TD = 5.884 ft
Figure C.19 Gas fraction for different depths and times for experiment MI5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Dept
h (ft
) De
pth
(ft)
Dept
h (ft
)
127
Experiment: Ml 5
5400-
5500-
5600-
5700-
5800 a
—® 15 min
1
1
1
—B - 16 min —B 17 min
.5500-
■B 5600-Q.o° 5700-
5800-
5400-
> 5500 -
•B 5600-Q.oQ 5700-
5800-0.00 0.20 0.40 0.60
Gas fraction
0.0 0.2 0.4 0.6Gas fraction
0.00 0.20 0.40 0.60 Gas fraction
5400-
5500-
5600-
5700 ■
5800 a
18 min
0.00 0.20 0.40 0.60 Gas fraction
19 min 20 min
5400 5400
5500 5500-e:■S 56005600
5700 t5700 7
580058000.00 0.20 0.40 0.60 0.00 0.20 0.40 0.60
Gas fraction Gas fraction
—B - 21 min —B - 22 min —f i - 23 min
58004
5400-
.5500-
■B 5600-C-o° 5700-
5800
5400-
.5500-
•B 5600-O.oQ 5700-
58000.00 0.20 0.40 0.60
Gas fraction
0.00 0.20 0.40 0.60 Gas fraction
0.00 0.20 0.40 0.60 Gas fraction
Figure C.20 Gas fraction as a function of depth for experiment M15
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Gas
Trac
tion
Gas
frac
tion
128
Type of experiment: Circulation at 62 spm (vsl=1.24 ft/sec)Experiment: M16
0 .0 6
0 .0 4
0.020.00
3 Interval = 0 - 5,422 ft
j N *
/| , . i . . . . . • • • ‘
10 15 20 25 30Time(minutes)
0 .4 0 : Interva] =5,422 - 5. 522 ft0 .3 0 :
0 .2 0 "0 .1 0 :
j
10 15 20 25 30Time (minutes)
c 0.60
“ 0.40Interval := 5,522 1,622 ft
0 5 10 20 2515 30Time (minutes)
s 0 .2 00 .1 5
c;c 0 .1 0(/]C 0 .0 5w 0 .0 0
In terval:= 5,622-i 1,822 ft
10 15 20 25 30Time (minutes)
0 ft
On-line 5,422 ft
Top 5,522 ft
Middle 5.622 ft
Bottom
5,722 ft (failed)
5,822 ft
TD = 5,884 ft
Figure C.21 Gas fraction for different depths and times for experiment M16
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
Dept
h (ft
) De
pth
(ft)
Dept
h (ft
)
129
Experiment: M16
20.7 21 min
54005400
55005500
rS 5600a .<aa 5700
5600
5700
580058000.0 0.2 0.4 0.60.00 0.20 0.40 0.60
22 min
5400
.5500
■S 5600a .oD 5700
05800-
Gas fraction Gas fraction0.00 0.20 0.40 0.60
Gas fraction
5400-
5500-
5600-
5700-
5800-
23 min
0.00 0.20 0.40 0.60 Gas fraction
24 min 25 min
5400 5400
5500cS 5600c.o° 5700
tS 5600
5700
5800 58000.00 0.20 0.40 0.60 0.00 0.20 0.40 0.60
Gas fraction Gas fraction
26 min 27 min 30 min
5400-
5500-
5600-
5700
5800
B~ T
r■—■
5400-
^5500-cS 5600-
C l.o° 5700-
5800
5400-
.5500-I
5 5600*a .oQ 5700
5800-
II- I
0.00 0.20 0.40 0.60 Gas fraction
0.00 0.20 0.40 0.60 Gas fraction
0.00 0.20 0.40 0.60 Gas fraction
Figure C.22 Gas fraction as a function of depth for experiment M16
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
130
Type of experiment: Circulation at 62 spm (vsl=1.24 ft/sec)Experiment: M17
- 0.061
Z 0.041
Iu 0.00-
In terval= 0 - 5,422 rt
I* ,
. A0 20 40 60 80 100 120
Time(minutes)
- 0.40joos* 0.201C/3cz
LOO
Interv; J =5,422 - 5, 522 ft
20 40 60 80 100 120Time(minutes)
- 0.60-q oZ 0.40
« 0.201 u 0.00
Interval = 5,522 - 5, 522 ft
— J 10 20 40 60 80 100 120
Time (minutes)
c 0 .2 0 1 *“ 0 1 5 "
Interval == 5,622 -4,822 ft— U.1S . « A in - :V*1U W A AC 'cc : ° 0 .00 J
10 15 20 25 30Time(minutes)
Oft
On-line 5,422 ft
Top 5,522 ft
Middle 5,622 ft
Bottom 5,722 ft (failed)
5,822 ft
TD = 5,884 ft
Figure C.23 Gas fraction for different depths and times for experiment M17
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
131
Experiment: M17
38 min
54005400
55005500
- 56005 56006 OQ 5700 5700
580058000.00 0.20 0.40 0.60
39 min 40 min
i t
Gas fraction
0.0 0.2 0.4 0.6Gas fraction
5400
5500
■B 5600
5700
58000.00 0.20 0.40 0.60
Gas fraction
■E 5600
41 min
0.00 0.20 0.40 0.60 Gas fraction
42 min 43 min
54005400
5500 5500
£ 5600 ~ 5600a.o° 57005700
580058000.000.200.40 0.600.80 0.00 0.20 0.40 0.60
Gas fraction Gas fraction
— 45 min —H 46 min —B - 49 min
5400
.5500
•S 5600CL.oa
i
£5700 t
5800-0.00 0.20 0.40 0.60
Gas fraction
5400
5500
S 5600
5700
58000.00 0.20 0.40 0.60
5400
.5500 L
5 5600-CL.0>° 5700-
Gas fraction
5800-r0.00 0.20 0.40 0.60
Gas fraction
Figure C.24 Gas fraction as a function of depth for experiment M l 7
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
Type of experiment: Circulation at 32 spm (vsl=0.64 ft/sec)Experiment: M18
Oft= 0.16
— 0.12 1 0.08 «« 0.04' «0.00
Interv al = 0 - i ,422 ftj
/ " I * *
J H] !—— -
P
0 20 40 60 80 100 120 140Time(minute:
- 0.60 Interval 5, 522 ft=5,422 -
25 0.40
0 20 40 60 80 100 120 140Time(minutes)
- O M t T Z —3 3 In te rv a = 5,52: - 5,622 f t
w U.4U JjA" j a 1« U-20 J
« d ^ 0.00 i ------------ 11 '
0 20 40 60 80 100 120 140Time(minutes)
= 0.20- •§ 0.15- 1 0.10-
0.05-0.00-
3 Interva = 5,62: - 5.822 ft1111 1
20 40 60 80 100 120 140Tin
On-line 5,422 ft
Top 5,522 ft
Middle 5,622 ft
Bottom 5,722 ft (failed)
5,822 ft
TD = 5,884 ft
Figure C.25 Gas fraction for different depths and times for experiment M18
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
133
Experiment: M18
18.4 19 min
54005400
5500-
£ 5600
5700mu
580058000.0 0.2 0.4 0.60.00 0.20 0.40 0.60
20 min
5400'
^5500-s£ 5600 £2.° 5700
5800
Gas fraction Gas fraction0.00 0.20 0.40 0.60
Gas fraction
22 min
5400
5500
£ 5600 o. cjQ 5700
58000.00 0.20 0.40 0.60
23 min
5400
5500
-£ 5600
5700
58000.00 0.20 0.40 0.60
5400-
5500'c:£ 5600
CLCJ>° 5700-
5800'
24 min
Gas fraction Gas fraction0.00 0.20 0.40 0.60
Gas fraction
— 25 min — 30 min —B — 40 min
5400-
—. 5500-
£ 5600-c .oQ 5700-
5800-
1
0.00 0.20 0.40 0.60 Gas fraction
5400
5500
£ 5600a.oQ 5700
58000.00 0.20 0.40 0.60
5400-
.5500-
£ 5600-o.oQ 5700-
Gas fraction
5800 t0.00 0.20 0.40 0.60
Gas fraction
Figure C.26 Gas fraction as a function of depth for experiment M18
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
Gas
fr
actio
n ^
as fr
actio
n G
as
frac
tion
Gas
fr
acti
on
134
Type of experiment: Circulation at 62 spm (>sl=1.24 ft/sec)Experiment: M19
0.08-; Interval =
0.04-j
0.00
0-2,761 It
0.30"
0.20'
0.10'
0.00"
0.60-
0.40-i
0.20 i
0.000 20
0.20-0.15-0.10-0.05-0.00-
A Co—0 20 40 60 80 100 120
Time (minutes)
i iInterva = 2,761 - 2,861 ft
j
§1 *
20 40 60 80 100 120Time (minutes)
Interval = 2,861 - 2, 161 ft
—.. . - j 140 60 80 100 120
Time (minutes)
3 Interva = 2,961 - 5,822 ft1
j . . . 1 . ‘ ‘ 1 . •0 20 40 60 80 100 120 140
Time (minutes)
Oft
On-line 2,761 ft
Top 2,861 ft
Middle 2,961 ft
Bottom
3,061 ft (failed)
5,822 ft
TD = 5.884 ft
Figure C.27 Gas fraction for different depths and times for experiment M19
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
Depth
(ft)
De
pth
(ft)
Depth
(ft
)
135
Experiment: M19
2500-
2600 i 2700- 2800- 2900- 3000-
—B - 30 min
1
1
1
32 min
25002600
34 min
ciL'2700-
2500-2600-2700-2800-2900-3000-
—® 34.5 min
, m1 1J. |T I' 11
1 I
36 min
2500-2600-2700-
2800-2900-3000-
0.00 0.25 0.50 0.75 1.00 Gas fraction
§•2800-Q
29003000-
2500 ^2600 * 2700C^2800D
2900 i 3000
0.00 0.20 0.40 0.60 Gas fraction
0.0 0.2 0.4 0.6Gas fraction
0.00 0.25 0.500.751.00 Gas fraction
35 min
25002600
'2700-^2800^oa
2900j 3000
25002600
'2700-HI
HI
35.5
§-2800i D
2900 i3000-
0.00 0.25 0.50 0.75 1.00 Gas fraction
0.00 0.25 0.500.751.00 Gas fraction
0.00 0.25 0.50 0.75 1.00 Gas fraction
—M— 37 min — 38 min
2500 t2600-
'2700-
^2800-3<ua2900-J 3000
25002600
'2700-
^■2800-OQ
2900-3000-
0.00 0.100.200.30 0.40 Gas fraction
0.00 0.10 0.200.30 0.40 Gas fraction
Figure C.28 Gas fraction as a function of depth for experiment M19
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
136
Type of experiment: Circulation at 62 spm (>sl=1.24 ft/sec)Experiment: M20
Interval = 0 - 2 '
e 0.30 ‘ o
V)aO
0.10'
0.00'
D !0 40 60 80 100 120 14Time (mi
0 If nutes
Interv al = 2, 761 - 2. 861 ft
10 15 20 25 30 35 40 45Time (minutes)
e 0.30 qoZ 0.20cs* 0.10
$ 0.00
Interval = 2,361-2.951 ft
£
JL0 20 40 60 80 100 120 140 160
Time (minutes)
c 0.20-3 0.15
.£ 0.10-« 0 .0 5 1
° o.oo-
: Interv; il= 2,9 61-5 ,8 22 ftj:
j* 1 ‘ 1 1 I . i i 1 !
0 20 40 60 80 100 120 140 160Time(minutes)
Oft
On-line 2,761 ft
Top 2,861 ft
Middle 2,961 ft
Bottom 3,061 ft (failed)
5,822 ft
TD = 5,884 ft
Figure C.29 Gas fraction for different depths and times for experiment M20
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
Dcpt
ii (fi
) De
pth
(ft)
Dept
h (ft
)
137
Experiment: M20
2500-
2600-
2700-
2800-
2900-
3000-
I —® 45 min
j
II1
46 min
c. • u ■Q
2500
2600
2700
2800
2900
3000
—H- 48 min
TI1'
1
51 min
2500-q
2600
2800-
2900-
3000-
1
1 1 1J
1-111
2500
2600
'2700
■2800
2900-
3000-
47 min
1 B 11f v11iii
2500 n
2600 H
’ 2700
g-2800^ O
2900
3000'II
0.00 0.20 0.40 0.60 Gas fraction
0.0 0.2 0.4Gas fraction
0.6 0.000.25 0.500.751.00 Gas fraction
49 min
2500
2600
2700
g-2800-
2900-
3000-
Va
2500-2600-
•2700-
•2800-
2900'
3000
50 min
0.00 0.25 0.50 0.75 1.00 Gas fraction
0.000.25 0.500.75 1.00 Gas fraction
0.000.25 0.50 0.75 1.00 Gas fraction
52 min 55 min
oa
2500-
2600-
■2700-
•2800-
2900
3000ii—m
2500-
2600 •
2700-
oD
2900 i
30000.00 0.25 0.500.75 1.00
Gas fraction
0.000.25 0.50 0.75 1.00 Gas fraction
0.00 0.25 0.50 0.751.00 Gas fraction
Figure C.30 Gas fraction as a function of depth for experiment M20
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
138
Type of experiment: Migration with choke openExperiment: M21
_ 0.16:Interv il = 0 -2,76 I ft L;5 U.1Z”g
£ O.OSl a / m2 U.U4"Cq
/0.00“^ ♦ • * i * * *0 20 40 60 80 100 120 140 160 180 200
Time(minutes)
0.20'■x 0.15“ej «. i .«O
<so
0.10 ■: 0.051 0.00‘
0.30-
0.20-
0.10-
0.00-
■ Inter fal = j ,761- 2,861 Ift f ]/
i /= -r-rr-
1 i,
0 20 40 60 80 100 120 140 160 180 200Time(minutes)
Intei val = 2,861 -2,96 ft
f ]
i /
0 20 40 60 80 100 120 140 160 180 200Time(minutes)
*“ 0 15 :Inter val = 2,961 - 5,82: I ft
- U.15 ; c: a i n "> U*JLU -IM55 U»Uj -
° o .o o -
0 20 40 60 80 100 120 140 160 180 200Time (minutes)
Oft
On-line 2,761 ft
Top 2,861 ft
Middle 2,961 ft
Bottom 3,061 ft
(failed)
5,822 ft
TD = 5,884 ft
Figure C.31 Gas fraction for different depths and times for experiment M21
Reproduced w ith permission of the copyright owner. Further reproduction prohibited w ithout permission.
Dept
h (ft
) De
pth
(ft)
Dept
h (ft
)Experiment: M21
2500-2600-2700-2800-2900-3000-
j —® 118 min
jjIj |f
120 min 122 min
aa
2500-2600-
'2700--2800-2900-3000-it
2500-
2600-'2700-
§" 2800 iD
29003000
11-9i i - a
0.00 0.10 0.20 0.30 Gas fraction
0.00 0.10 0.20 0.30 Gas fraction
0.00 0.10 0.20 0.30 Gas fraction
2500-2600-2700-2800-2900-3000-
124 min 128 min
F ti i— ■
oQ
2600-2700-
•2800-29003000
L SI
1
<uQ
2500-2600-
'2700-•2800-2900-3000-
130 min
I
0.00 0.10 0.20 0.30 Gas fraction
0.00 0.10 0.20 0.30 Gas fraction
0.00 0.10 0.20 0.30 Gas fraction
132135 min 135 min
250026002700280029003000-
iTI■ i
i
aD
2500-2600-
'2700-•280029003000 3
2500-2600t
2700-oD
2900-3000-
3
0.00 0.10 0.20 0.30 Gas fraction
0.00 0.10 0.20 0.30 Gas fraction
0.00 0.10 0.20 0.30 Gas fraction
Figure C.32 Gas fraction as a function of depth for experiment M21
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
Gas
fr
actio
n ^
as fr
flCt'o
n G
as
frac
tion
Gas
fr
acti
on
140
Type of experiment: Circulation at 32 spm (■vsl=0.64 ft/sec)Experiment: M22
261-ftinterva0.16i
0 .0 8 i0.0410.00
0 20 40 60 80 100 120 140 160Time (minutes)
0 .3 0 10 .25=0.2010 .15=o.ion0 .0 5 10.001
In tp rv
I0 20 40 60 80 100 120 140 160
Time (minutes)
0 .3 0Interval = 2,861-2,96
0.20
0.0020 40 60 80 100 120 140 1600
Time(minutes)
0.20-0.15-0.10-0.05-0.00-
| Inter val = 2,961 - 5,82: : ft1111 ■ ■ • • ‘ ‘ ‘ i • *
0 20 40 60 80 100 120 140 160 180 200Time (minutes)
Oft
On-line 2,761 ft
Top 2,861 ft
Middle 12,961 ft
JL, Bottom 3,061 ft
(failed)
5,822 ft
TD = 5,884 ft
Figure C.33 Gas fraction for different depths and times for experiment M22
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
Depth
(ft)
De
pth
(ft)
Depth
(ft
)
141
Experiment: M22
2500260027002800-2900-q 3000
—B — 38 min
i i
IE
i i
40 min
2500-2600-i
45 min
^Q O S'2800-
25002600270028002900 i
3000
50 min
60 min
25002600270028002900-}3000 *
2900-3000-
25002600
'2700
oQ
2900 i 3000
a
0.00 0.10 0.20 0.30 Gas fraction
0.00 0.10 0.20 0.30 Gas fraction
0.00 0.10 0.20 0.30 Gas fraction
52 min
2500-2600-
'2700-^ 2800 -oQ
2900-3000-
ua
2500-2600-
'2700-•2800-2900-
3000-
55 min
0.00 0.10 0.20 0.30 Gas fraction
0.00 0.10 0.20 0.30 Gas fraction
0.00 0.10 0.20 0.30 Gas fraction
70 min 70 min
2500-32600-
'2700-<DG
2900 i 3000
2500-q 2600
'2700-
<DQ
2900 3000
0.00 0.10 0.20 0.30 Gas fraction
0.00 0.10 0.20 0.30 Gas fraction
0.00 0.10 0.20 0.30 Gas fraction
Figure C.34 Gas fraction as a function of depth for experiment M22
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
142
Type of experiment: Circulation at 62 spm (vsl=1.24 ft/sec)Experiment: M23
s 0.03-
Z 0.02- «u o.oru 0.00-
In ter ■al = 0 - 2,761 ft
1
[ ^ k i ,0 20 40 60 80 100 120 140 160 180
Time (minutes)
C 0.3(T . 2
Interv al = 2 , '61 - 2 . 861 f t
« 0.2CF
O.lflFC5
r •*w O.OCr ‘ * 1 1 . . .
0 20 40 60 80 100 120 140 160 180Time(minu
c 0.30
Z 0.20cz0.10 -=
u 0.00
j In ten al = 2i A
,861 -: :,96l i t
jLa
' ‘ 'it/j • A
0 20 40 60 80 100 120 140 160 180Time (minutes)
B 0 .2 0 -q-2 n
In te r a! = 2 ,961 - i1,822 1
?• ft 1 f t- \ f " *--------.U v*lU
n n c — \ l / ]C5 U.VO - ° 0 .0 0 J
0 20 40 60 80 100 120 140 160 180Time (minutes)
0 ft
O n-line
2,761 f t
Top
2,861 ft
M iddle
J 2,961 ft
Bottom
3,061 ft
(failed)
5,822 ft
TD = 5,884 ft
Figure C.35 Gas fraction for different depths and times for experiment M23
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
Depth
(ft
) De
pth
(ft)
Depth
(ft
)
143
Experiment: M23
2600-2700-
2900-3000-
— 20 min
1
1
1
22 min
2500 n 2600
25 min
'2700-
28 mm
35 min
2500-2600-2700-2800^29003000
2500-2600-
'2700-oQ
2900-3000-
0.00 0.10 0.20 0.30 Gas fraction
0.00 0.10 0.20 0.30 Gas fraction
0.00 0.10 0.20 0.30 Gas fraction
30 min
25002600
'2700-
Q2900 i 3000
250012600 ~
'2700-
32 min
8“ 2800 D
2900 i3000
0.00 0.10 0.20 0.30 Gas fraction
0.00 0.10 0.20 0.30 Gas fraction
0.00 0.10 0.20 0.30 Gas fraction
40 min 50 min
25002600
'2700-^ 2800a>D
2900-3000-
iiII !■
i M
25002600
'2700§-2800iQ
29003000 II
0.00 0.10 0.20 0.30 Gas fraction
0.00 0.10 0.20 0.30 Gas fraction
0.00 0.10 0.20 0.30 Gas fraction
Figure C36 Gas fraction as a function of depth for experiment M23
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
Gas
fr
actio
n G
as
frac
tion
Gas
fr
actio
n G
as
frac
tion
144
Type of experiment: Circulation at 32 spm (vsl=0.64 ft/sec)Experiment: M24
0.20i
0.05-
1 J n te rv s 1 = 0 - 100 f t5 3) 4
5 J "
) T-t t t - T T —t - j
0 20 40 60 80 100 120 140 160 180Time (minutes)
0.301
0.201
0.101
0.00
o.oi-
o.oi-
o.oo-
Inteival = 100 - 200 ft
0 20 40 60 80 100 120 140 160 180Time (minutes)
Interval = 200 -303 ft
J0 20 40 60 80 100 120 140 160 180
Time(minutes)
0.20- 0.15-j 0.10-! 0.05 0.00 J
In ten al = 3 00 - 5,1■22 ftj[l
* ‘ ‘ * * • ' • ‘ 1 . * .0 20 40 60 80 100 120 140 160 180
Time (minutes)
Oft
On-line 100 ft
Top 200 ft
Middle 300 ft
Bottom 400 ft
(failed)
5,822 ft
TD = 5,884 ft
Figure C.37 Gas fraction for different depths and times for experiment M24
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
Depth
(ft
) De
pth
(ft)
Depth
(ft
)
145
Experiment: M24
200-
— 75 min
B-------
if
■
B_____
78 min
100
200
—M— 85 min
K-------
M■
f
i ■_____
100
0
100
200
300
S ioo.c
£200
0.00 0.10 0.20 0.30 Gas fraction
0.00 0.10 0.20 0.30 Gas fraction
300-
80 min
s ioo. s
£ 200
300-
*
0.00 0.10 0.20 0.30 Gas fraction
0.00 0.10 0.20 0.30 Gas fraction
s ioo
£ 200
300
90 min
0
gioo
£ 2 0 0
300 -p-
95 min
0.00 0.10 0.20 0.30 Gas fraction
0.00 0.10 0.20 0.30 Gas fraction
0.00 0.10 0.20 0.30 Gas fraction
—fit- 115 min —® 125 min
0-
« ioo
£ 2 0 0
300
« -
T
0
g 100Q 200-*-
300410.00 0.10 0.20 0.30
Gas fraction0.00 0.10 0.20 0.30
Gas fraction
Figure C.38 Gas fraction as a function of depth for experiment M24
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
146
Type of experiment: Circulation at 62 spin (vsl=1.24 ft/sec)Experiment: M27
__ 0.2(h•I o.isig 0.10-
0.051O 0.00^
In ten al = 0 2.761 f
Ifctrl20 40 60 80 100 120 140 160
s 0.40 -3 ••§ 0.30wi 0.20 1S 0.10 -j ° 0.00
In ten a! = 2,/ 61 - 2.8 SI ft ,1 1j j i L i , 11nHi
M i M B
20 40 60 80 100 120 140 160Time(minutes)
s 0.401.2 A *5 A " Interv; J = 2,? 61-2 ,9 SI ft
".jU •
« 5 A ‘‘J A 1> U.ZU -
r \cs U-lU : u 0.00 J J . .
20 40 60 80 100 120 140 160Time (minutes)
c 0.40-g0.30H
c:<2= 0.20-2V5 0.10 H
0.00 J
Interva
Os*fSII ;i - 3,o< l ftw>
\\
- J . \ - 4
20 40 60 80 100 120 140 160
s 0 .2 0 -g• 2 n i c ■
Interval = i i,061 - 5,822 ftz °*15 =J ? a i n *> U . I U -
CZ V « u J ;
° 0 .0 0 J50 100 150 200 250
Time(minutes)
0 ft
On-line 2,761 ft
Top Lj-12,861 ft
Middle J 2,961 ft
Bottom 3,061 ft
5,822 ft
TD = 5,884 ft
Figure C.39 Gas fraction for different depths and times for experiment M27
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
147
Experiment: M 27
2850-'2950-
§■305013150-3250-
—® 65 min
B-----
1
1
1
70 min
D
0.0 0.1 0.2 0.3 0.4 Gas fraction
75 min
2750 -BW28501I
'2950"§•3050-* Q
3150
32500.0 0.1 0.2 0.3 0.4
Gas fraction
79 min
27502850
2950§-3050-fe= a
3150-3250
0.0 0.1 0.2 0.3 0.4 Gas fraction
2750-2850-
'2950-•3050-3150-3250-
72 min
3 1
2750 ■»
2850 H■2950-
8" 3050-jfa
31503250-
I
0.0 0.1 0.2 0.3 0.4 Gas fraction
0.0 0.1 0.2 0.3 0.4 Gas fraction
77 min
2750-2850-
'2950-
oO
3150-
3250 4-
275028502950
8-3050^1 D
3150i
3250-
78 min
m = m
0.0 0.1 0.2 0.3 0.4 Gas fraction
0.0 0.1 0.2 0.3 0.4 Gas fraction
H 82 min —H — 85 min
27502850
2950^-3050<x>O
31503250
0.0 0.1 0.2 0.3 0.4Gas fraction
0.0 0.1 0.2 0.3 0.4 Gas fraction
Figure C.40 Gas fraction as a function of depth for experiment M27
Reproduced w ith permission o f the copyright owner. Further reproduction prohibited w ithout permission.
Type o f experiment: Migration with choke closedExperiment: M28
s ° - 1 0 1 S
g 0 . 0 5 H
3 o.oo-
Interval = <) - 2,761 ft
5 0 100 1 5 0 200 2 5 0
0 . 8 0 -
0 . 6 0 -
0 . 4 0 -
3 0. 2 0 -
Time (minutes3 Interval = *.761 - 2.861 ft
i 1............. , A tflpBBLk—
5 0 1 0 0 1 5 0 2 0 0 2 5 0
Time(minutes)s
° * 4 0 1
0 . 3 0 H
c s* 0 . 2 0 - j
C/3C3 0 . 1 0 - ^
u0 . 0 0 -
Interval = 2 ,861-2 ,96 | ft
1
5 0 1 0 0 1 5 0 2 0 0 2 5 0
T:me(minutes)- 0 . 4 0 ^
0 . 3 0 i
c:
,h 0 . 2 0 H
C/3c5 0 . 1 0 H
u o.oo-
Interval = 2,961 - 3,06 ft
r —
j5 0 100 1 5 0 200 2 5 0
s 0 . 2 0 - q
I 0 . 1 5 - i
. £ 0 . 1 0 - 3
3 0 . 0 5 -
u 0.00-
Time [minutes)Interval = .5,061 - 5,822 ft
::
=
5 0 1 0 0 1 5 0 2 0 0 2 5 0
Time(minutes)
Oft
O n - l i n e
2,761 ft
Top 2,861 ft
Middle 2,961 ft
□ Bottom 3,061 ft
5,822 ft
TD = 5,884 ft
Figure C.41 Gas fraction for different depths and times for experiment M28
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
Depth
(ft
) De
pth
(ft)
Depth
(ft
)
149
Experiment: M28
2750-
2850-2950-3050-3150-3250-
— I5l min
f
j f
f
f
160 min
2750-
2850-2950-
I
^3050-OQ
275012850129503050-
165 min
180 min
2750-
2850- 2950- 3050- 3150- 3250
at0.0 O.l 0.2 0.3 0.4
Gas fraction
3150- 3250-
2750-■
163 min
0.0 O.l 0.2 0.3 0.4 Gas fraction
0.0 O.l 0.2 0.3 0.4 Gas fraction
0.0 O.l 0.2 0.3 0.4 Gas fraction
170 min
275 0H 2850
CD
D
'2950-•3050-3150-
3250 3
27502850
2950
QJQ
3150
3250
175 min
I
0.0 O.l 0.2 0.3 0.4 Gas fraction
0.0 O.l 0.0 O.lGas fraction Gas fraction
—® 190 min — 200 min
oQ
2750-2850
'2950•305031503250
' L -
- H ----- II
I F = = 1-
2750 2850-fc
'2950-g-3050-a
3150 - 3250-
m=W
0.0 O.l 0.2 0.3 0.4 Gas fraction
0.0 O.l 0.2 0.3 0.4 Gas fraction
Figure C.42 Gas fraction as a function of depth for experiment M28
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
Gas
fr
actio
n G
»s fr
actio
n G
as
frac
tion
Gas
fr
actio
n G
as
frac
tion
Type of experiment: Migration with choke openExperiment: M29
0.60“
0.40
0. 20"
0.00
Interval = ) - 2,7 51 ft
0.4010.30i
0 20 40 60 80 100 120 140 160 180 200Time (minutes i
Interval = !.761
0.40-0.30-0.20 -
0.10-0.00-
0 20 40 60 80 100 120 140 160 180 200Time(minutes)
3 Intei val = 2,861 -2,96 ft1 \1 r J
N i
0 20 40 60 80 100 120 140 160 180 200Time(minutes)
0.40-0.30-0.200.10
E Intei val = 2,961 -3,06E / Vj r V ij
i ,X
w* ‘ ‘ • * ‘
0.400.300.200.100.00
0 20 40 60 80 100 120 140 160 180 200Time (minutes)
d Inter val = ; 1,061 ■5,822 ft
3 / r1 /3 / r1 ■ ‘ • * 1 . * 1 »0 20 40 60 80 100 120 140 160 180 200
Time(minutes)
Oft
On-line 2,761 ft
Top LJ2,861 ft
Middle 2,961 ft
Bottom 3,061 ft
5,822 ft L
TD = 5,884 ft
Figure C.43 Gas fraction for different depths and times for experiment M29
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
Depth
(ft)
De
pth
(ft)
Depth
(ft
)
151
Experiment: M29
2750-II2850-iff2950iff
3250-
—B - 60 min
1-----
B
I
1
65 min
275012850
70 min
<oQ
2750Tjff 2850 2950 3050
3150 3250
80 min
3
110 min
2750t
2850-1---- i « -2950------ * -3 0 5 0 il= = « -3150i3250
'2950•305031503250
E0.0 0.1 0.2 0.3 0.4
Gas fraction0.0 0.1 0.2 0.3 0.4
Gas fraction
90 mm
0.0 0.1 0.2 0.3 0.4 Gas fraction
120 min
27502850
'2950§•3050
31503250
* = *
0.0 0.1 0.2 0.3 0.4 Gas fraction
JCn. -
27502850
'2950
31503250
—S 100 min
1-----
1* !8 —i i
IS—
0.0 0.1 0.2 0.3 0.4 Gas fraction
0.0 0.1 0.2 0.3 0.4 Gas fraction
130
27501B = =*ff28501 H 12950"
*5I1
o ivjU "1Q
n B3150 . 3250 J
0.0 0.1 0.2 0.3 0.4 Gas fraction
0.0 0.1 0.2 0.3 0.4 Gas fraction
0.0 0.1 0.2 0.3 0.4 Gas fraction
Figure C.44 Gas fraction as a function of depth for experiment M29
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
VITA
Shiniti Ohara, bom in Sao Paulo, Brazil, received a Bachelor of Science degree in
Civil Engineering from Universidade Estadual de Campinas, Sao Paulo, Brazil, in 1979.
Following graduation, he joined Petroleo Brasileiro S.A. (Petrobras) where he took two
years of graduate courses in petroleum engineering at Petrobras Training Center in
Salvador. Upon finishing the course, he moved to Petrobras headquarters in Rio de
Janeiro. There he worked for five years as a drilling engineer in offshore rigs and well
design activities. In August 1987, he entered Universidade Estadual de Campinas where
he earned his Master of Science degree in petroleum engineering in March 1989. After
his graduation he was relocated to the Special Techniques Sector of the Drilling
Department, where he worked with new drilling bits, air drilling in the Amazon jungle,
and turbo-drilling. In 1990, he was the head of the Directional Drilling Sector, where he
worked on the first Latin American offshore horizontal well. Following this, in 1991, he
was the head of exploration drilling for South and Southeast Regions o f Brazil. He
worked there until Spring 1992 when he entered the doctoral program in the petroleum
engineering at Louisiana State University. He is married to Neuza F. Suguihara Ohara
and they have two children: Sara Hitomy (12 years old) and Fabio Shinji (10 years old).
152
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
DOCTORAL EXAMINATION AND DISSERTATION REPORT
Candidate: Shiniti Ohara
Major Field: Petroleum Engineering
Title of Dissertation: Improved Method for Selecting Kick ToleranceDuring Deepwater Drilling Operations
Approved:
fk h r tn ~T. /WMajor P rofessor send Ghai
'LDean of the—Graduate School
EXAMINING COMMITTEE:
7/ 0. £&
£
Date of examination:
1 2 /0 8 /9 5 ___________
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.