Fine Tuning Coriolis Flow Meter Calibrations Utilizing Piece-

Wise Linearization

Tonya Wyatt Global Chemical Industry Manager

Emerson Process Management – Micro Motion, Inc.

AGA Operations Conference and Biennial Exhibition 2015

New Developments to Improve Natural Gas Custody Transfer Applications with Coriolis Meters Including Application of Piecewise Linear Interpolation (PWL)

Marc Buttler

Midstream O&G Marketing Manager, Micro Motion Emerson

Ron Gibson

Senior Engineer, ONEOK, Inc.

Gary McCargar

Senior Engineer, ONEOK, Inc.

Karl Stappert

Americas Flow Solutions Advisor, Emerson Process Management

Tonya Wyatt

Process Gas and Chemical Marketing Manager, Micro Motion Emerson

Agenda Calibration Adjustment Methods for Gas Meters

Coriolis Meter Calibration Options

– Principle of Operation Review

– Calibration Fluid Flexibility

– Pressure Effect Compensation

– Multi-Point Piecewise Linear Interpolation (PWL)

Implementation of PWL

– PWL and Pressure Effect Compensation Together

– Results

In-situ Secondary Verification of Calibration

Impact of Zero

Calibration Adjustment Methods from AGA 11

Flow Weighted Mean Error (FWME)

Polynomial Algorithm

Multi-Point Piecewise Linear Interpolation (PWL)

Piecewise Linearization (Step-wise)

-0.50%

-0.30%

-0.10%

0.10%

0.30%

0.50%

0 2 4 6 8 10Err

or

Flow Rate, lbm per second

As Found Error

Step-wise Correction

Corrected Data

-0.50%

-0.30%

-0.10%

0.10%

0.30%

0.50%

0 2 4 6 8 10Erro

r

Flow Rate, lbm per second

As Found Error

FWME Correction

Corrected Data

-0.10%

-0.05%

0.00%

0.05%

0.10%

0 2 4 6 8 10Erro

r

Flow Rate, lbm per second

As Found Error

Polynomial Algorithm Correction

Corrected Data

-0.10%

-0.05%

0.00%

0.05%

0.10%

0 2 4 6 8 10Erro

r

Flow Rate, lbm per second

As Found Error

Multi-Point Piecewise Linear Interpolation Correction

(PWL)

Corrected Data

Coriolis Meter Principle of Operation

The principle of operation dictates a Coriolis meter’s performance characteristics

• Sensitive to bulk inertial forces of the fluid

• Insensitive to change in fluid properties and velocity profile

As a particle inside a rotating body moves toward or away

from the center of rotation, the particle generates inertial

forces that act on the body.

Theory of Operation

Process fluid enters the sensor and flow is split with half the flow through each tube

Drive coil vibrates tubes at natural frequency

Pick-off coils on inlet and outlet sides generate raw measurement signals

Isometric View

At No Flow

No Flow

Flow

Causes

Twist

Direct Mass Flow Measurement

Coriolis Meter Raw Sensitivity Varies with Design

Raw Sensitivity Depends on Design of Tube Geometry

Signal to Noise Ratio Depends on Raw Sensitivity and Stability

Calibration Flexibility, Immunity to Secondary Effects, and Diagnostic Capabilities Depend on Signal to Noise Ratio

0

1

2

3

4

5

6

Flow Sensitivity

Examples

Large TubeGeometry

Medium TubeGeometry

Short TubeGeometry

nse

c of

signal

per

g/s

ec o

f fl

ow

AGA Report No. 11 / API MPMS Ch. 14.9 Measurement of Natural Gas by Coriolis Meter

First Edition Published by the American Gas Association (AGA) December 2003

Adopted by American Petroleum Institute (API) upon publication

API Manual on Petroleum Measurement Standards (MPMS) Chapter 14.9

Similar to other AGA technology performance based specifications

AGA 3: Orifice

AGA 7: Turbine

AGA 9: Multipath Ultrasonic

AGA Report No. 11 / API MPMS Ch. 14.9 Measurement of Natural Gas by Coriolis Meter

2nd Edition Published February 2013

Covers all single phase natural gases as pure or a mixture of hydrocarbons and diluents

API Standard API MPMS Chapter 14.9

Recommended Practice Specification, calibration,

installation, operation, maintenance, and verification

(Air)(Gas) x Gr

x x

(Gas) x

MassSCF

TbRZb

MrPb

MassSCF

b

MassSCF

Conversion of Mass to Volume at Standard Conditions

AGA8 Detail

AGA8 Gross 1 or 2

Non-ideal gas law:

Pb, Tb, R are constants

lbs/day ÷ lbs/ft3 = ft3/day

Molar weight, Base Compressibility, and Specific Gravity

Are All Determined by Gas Composition

Note: Zb does not vary more

than 0.02% at base conditions.

AGA11 Eqn. D.2

No pressure or Temperature Measurement Required

to Convert from Mass to Standard Volume

Calibration Fluid Flexibility Purpose and Benefits

“Calibration fluid flexibility” is a capability that allows a traceable liquid calibration to be used for traceable gas

measurements

Liquid medium meter calibration system benefits

– Easier to control liquid (e.g., water) system uncertainties

– Lower cost

– Greater safety

Recognized in AGA Report No. 11 / API MPMS Ch. 14.9

– Must demonstrate acceptable provenance for each Coriolis meter design

NMi Declaration of Water-to-Gas Calibration Transferability for Micro Motion Coriolis Meters

Testing

Included ½ to 12 inch Coriolis meters

Conducted on following mediums

Natural gas

Nitrogen

Ethylene

Results

CMF Model Coriolis meters calibrated on water at manufacturer may be applied to the following applications without requiring a field calibration or gas calibration

Every gaseous medium with density greater than 4 kg/m3

Super-critical ethylene with density up to 450 kg/m3

NMi Euroloop Test for CMFHC Meters

Effect of Pressure on Coriolis Meters

• Internal pressure changes the shape of the flow tube • Tube ovality becomes round • Tube bends straighten

• Changes in flow tube shape increases stiffness of flow tube • Changes in tube stiffness directly affects sensor calibration • Magnitude of effect varies by meter size and design

stiffness tube

delay time*

FCF

FCFm

P

P

P

FMass

SCF

F

TbRZb

MrPb

MassSCF

F

b

MassSCF

(Air)(Gas) x Gr

x x

(Gas) x

Application of Compensation for the Effect of Pressure on the Meter

FP compensation is

only required for

some of the larger

meter sizes

•FP factor can be employed to compensate for the

effect of pressure on the meter.

• FP factor is not a correction for behavior

of the gas.

Flow Pressure Effect Correction and Potential Error

Potential Error w/o Pressure Correction (Natural Gas)

Example Application of Pressure Effect Compensation - CMFHC2 Gas Test Results

• All data collected on natural gas using meter factory calibration on water

• Data shown with and without standard FP pressure compensation

• Max deviation of all compensated data < 0.25%

-0.10%

-0.05%

0.00%

0.05%

0.10%

0 2 4 6 8 10Err

or

Flow Rate, lbm per second

As Found Error

Multi-Point Piecewise Linear Interpolation Correction

(PWL)

Corrected Data

Multi-Point Piecewise Linear Interpolation (PWL)

Correction applied at selected linearization points is equal and opposite to the average of the as-found values at the same flow rate

Correction values applied between neighboring points are determined by linear interpolation between the two points

Correction above the highest flow rate point are held constant

Correction below the lowest point is based on linear interpolation to zero error at zero flow to allow meter zero adjustment to control accuracy below Qt

PWL Procedure

Optional PWL software must be ordered separately when the order is placed for a Micro Motion ELITE CMF meter

PWL calibration and configuration is conducted by qualified third-party calibration laboratories selected and hired by end users

– Emerson has trained, equipped, and qualified three independent gas labs in NA to execute the procedure (to date) Colorado Engineering Experiment Station, Inc., Southwest Research Institute, and TransCanada Calibrations

Basic Procedure at the third-party lab

1. Install and zero meter as necessary

2. Collect as-found data (pressure compensation active)

3. Select and program meter with up to 10 linearization points from the as-found data

4. Collect as-left data to verify accuracy of linearization

PWL and Pressure Comp. Together PWL As-Found Data Collected with Pressure Compensation Active

– PCal remains the original factory water calibration pressure

– This method keeps pressure compensation and linearization independent from each other

Alternative Method: PWL As-Found Data Collected with Pressure Compensation Inactive

– PCal becomes the gas lab As-Found pressure for future pressure compensation

– This method resets the pressure compensation baseline pressure to the gas test pressure

-0.60%

-0.40%

-0.20%

0.00%

0.20%

0.40%

0.60%

0 2 4 6 8 10Erro

r

Flow Rate, lbm per second

As Found Error(PRESS. COMP ENABLED)

Multi-Point Piecewise Linear Interpolation Correction

(PWL)

Corrected Data

-0.60%

-0.40%

-0.20%

0.00%

0.20%

0.40%

0.60%

0 2 4 6 8 10Erro

r

Flow Rate, lbm per second

As Found Error(NO PRESS. COMP -

CMFHC2 at 230 psig)

Multi-Point Piecewise Linear Interpolation Correction

(PWL)

Corrected Data

Results with PWL – 1-inch Meter CMF100

-0.5

-0.4

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

0.4

0.5

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5% E

rror

lbm per second

As Found Data

Verification Data

Observations

Test turndown ≈ 58 : 1

All verification averages better than ± 0.08%

Verification averages above 0.13 lbm/sec better than ± 0.027%

All verification data better than ± 0.11%

-0.5

-0.4

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

0.4

0.5

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5% E

rro

r

lbm per second

As Found Averages by Flow Rate

Verification Averages by Flow Rate

Averages at Each Flow Rate

All Data

-0.5

-0.4

-0.3

-0.2

-0.1

0.0

0.1

0.2

0.3

0.4

0.5

0 2 4 6 8 10 12% E

rror

lbm per second

As Found Data

Verification Data

Results with PWL – 2-inch Meter CMF200

Observations

Test turndown ≈ 45 : 1

All verification averages better than ± 0.09%

All verification data better than ± 0.29%

Averages at Each Flow Rate

All Data -0.5

-0.4

-0.3

-0.2

-0.1

0.0

0.1

0.2

0.3

0.4

0.5

0 2 4 6 8 10 12% E

rro

r

lbm per second

As Found Averages by Flow Rate

Verification Averages by Flow Rate

-0.5

-0.4

-0.3

-0.2

-0.1

0.0

0.1

0.2

0.3

0.4

0.5

0 5 10 15 20 25% E

rro

r

lbm per second

As Found Averages by Flow Rate

Verification Averages by Flow Rate

Results with PWL – 3-inch Meter CMF300

Observations

Test turndown ≈ 10 : 1

All verification averages better than ±0.08%

All verification data better than ± 0.22%

Averages at Each Flow Rate

All Data

-0.5

-0.4

-0.3

-0.2

-0.1

0.0

0.1

0.2

0.3

0.4

0.5

0 5 10 15 20 25% E

rror

lbm per second

As Found Data

Verification Data

Secondary Verification Purpose and Benefits

“Secondary Verification” is a capability that allows a measuring device to use an alternative method to confirm

flow measurement accuracy without the need for a calibration to a traceable flow reference standard

Reduce lost and unaccounted for (LAUF) product and reconciliations with the capability to perform more frequent checking without adding cost

Work practices may use statistical data and secondary verification results to extend primary calibration intervals

Recognized in AGA Report No. 11 / API MPMS Ch. 14.9

Secondary Verification Methods used for Coriolis Meters

Structural observations

– Ignore all fluid effects to measure changes to the meter structure

– Example: Micro Motion Smart Meter Verification (SMV) method to measure stiffness of flow tubes

Fluid dependent observations

– Uses knowledge of current fluid properties to test meter response to fluid

– Fluid properties must be accurately measured and controlled

– Example: Micro Motion Known Density Verification method (KDV) to observe Compact Density Meter (CDM) behavior when fluid density is known

2

1XFRF

F M jC K

Coriolis Meters are like Simple Springs – Stiffness Influences Measurement

Density proportional to

• Tube Stiffness

• Tube Mass

• Fluid Mass

Mass Flow proportional to • Tube Stiffness

• Mass Inertial Forces

Spring Natural Frequency proportional to

• Spring Stiffness

• Spring Mass

• Weight

Using Resonant Modal Analysis to determine Coriolis Flow Tube Stiffness

• Quick, reliable indication of complete sensor health – from tube structure to electronics

• Proactive maintenance to eliminate calibrations and improve process uptime

Easily verify meter performance - in line and on demand

Test tones

Response

Okay

Not Okay

Span vs. Zero

bmxy

Meter

Zero

FCF

Zeroing Best Practices

Most applications – Use factory zero

Insure no flow condition

Insure meter is full

Insure process conditions are stable

Example: Micro Motion Zero Verification monitors 8 parameters to check stability of process and check current zero value

Conclusions

Some Coriolis meters have demonstrated the capability to be calibrated on liquid in order to measure gas

Some Coriolis meters can be adjusted with PWL in gas calibration labs to improve gas flow measurement

PWL must be implemented together correctly with pressure compensation

Secondary verification methods exist to confirm Coriolis meter accuracy after primary flow calibrations including span and zero