Verification and Testing of a Differential Flow Meter Using the API MPMS Chapter 22.2 Protocol

1
Verification and Testing of a Differential Flow
Meter Using the API MPMS Chapter 22.2 Protocol

• Eric Kelner, P.E., Southwest Research Institute,
  San Antonio, TX
• Russell Burkey, Southwest Research Institute, San
  Antonio, TX
• Eric Reid, Rosemount-Dieterich Standard, Boulder,
  CO

2
Presentation Outline

• Metering Research Facility
• Background of API MPMS, Chapter 22.2
• General requirements of the standard
• Testing procedure and test results
• Conclusions

3
Metering Research Facility
Built in 1991, the MRF is a high-accuracy natural
gas flow test and calibration facility capable of
simulating field operating conditions.
4
API MPMS Chapter 22.2 Background

• In 2003, a standard defining a testing and
  reporting protocol for flow meters that produce a
  change in velocity by creating a differential
  pressure was developed by the American Petroleum
  Institute (API) Manual of Petroleum Measurement
  Standards (MPMS), Chapter 5.7 Working Group,
  entitled Differential Pressure Flow Measurement
  Devices. It was subsequently moved to MPMS,
  Chapter 22, Section 2.
• Objectives of the standard
• Establish performance characteristics of the
  meter.
• Facilitate understanding and introduction of new
  metering technologies.
• Provide a method for validating manufacturers
  performance specifications.
• Provide a means to compare different types of
  differential pressure flow meters.
• Quantify the measurement uncertainty of the
  device.

5
API MPMS, Chapter 22.2 General Requirements

• Test Facility Verification
• Traceable to NIST.
• Must measure Cd within the 95 confidence
  interval of the R-G equation over the Reynolds
  number range of testing.
• Baseline Tests
• Intended to establish meter performance under
  good flowing conditions.
• Meter located at least 30D downstream of a
  perforated plate flow conditioner.
• Disturbance Tests
• Intended to establish meter performance under
  worst-case velocity profile distortions.
• Two close-coupled elbows oriented out-of-plane.
• Half-moon plate or half-open gate valve (upstream
  and downstream).
• Swirl Generator producing minimum of 24 degrees
  of swirl.
• Combined upstream and downstream disturbances,
  based on previous test results.
• Special installation tests for specific
  installations.
• Meter location, relative to disturbances, is
  specified by the manufacturer.

6
API MPMS, Chapter 22.2 General Requirements

• Uncertainty
• Facility and test result uncertainty and
  uncertainty calculation methods, must be
  reported.
• Acceptance Criteria
• Defines whether additional testing must be done
  to eliminate (statistically) an installation
  effect.
• Alternatively, can provide a statement of bias or
  a new Cd curve for cases with significant
  installation effects.
• The standard allows flexibility in determining
  the acceptance criteria.
• Reporting
• Standardized format to facilitate comparison of
  different meters.

7
Test of Rosemount Dieterich-Standard Conditioning
Orifice Plate Flow Meters

• In 2005, several Rosemount Dietrich-Standard
  Model 405 C and Model 1595 conditioning orifice
  plate flow meters were tested in accordance with
  API MPMS, Chapter 22.2 under BLM oversight.
• The testing was performed in the Low Pressure
  Loop (LPL) at the Metering Research Facility
  (MRF) at Southwest Research Institute (SwRI).
  The gas flow tests performed at the MRF included
  baseline and disturbance tests.
• The testing included five 2-inch diameter meters,
  with various beta ratios (i.e., ß 0.20, 0.40,
  0.60), two 3-inch diameter meters with a 0.65
  beta ratio, and five 4-inch diameter meters with
  various beta ratios (i.e., ß 0.20, 0.40, 0.65).
• Results from the tests of the Rosemount Model 405
  C conditioning orifice plate flow meter are
  presented.

8
Facility Verification - MRF Low Pressure Loop

• MRF measured Cd is within the 95 confidence
  interval of the Reader-Harris-Gallagher equation.

9
Reference Flow Meters

• The MRF uses a bank of critical flow Venturi
  nozzles as the working flow rate standard.

10
Test Meter

• Rosemount Model 405 C
• Four small orifices.
• Less sensitive to velocity profile effects than a
  standard orifice meter.

11
Baseline Tests

• Over 100 diameters of straight pipe upstream of
  test meter.

12
Disturbance Tests Out-of-Plane Elbows
13
Disturbance Tests

• Half-moon plate and swirl generator.
• The swirl generator was provided by Rosemount
  Dieterich-Standard.

14
Confirmation of Swirl Angle
15
Swirl Measurement System
16
Contributions to Uncertainty

• The following variables were accounted for in the
  uncertainty calculation
• Meter bore diameter
• Meter tube diameter
• Gas temperature
• Gas pressure
• Meter differential pressure
• Reference mass flow rate
• Gas composition
• Gas density (computed from equations of state)
• Gas expansion factor (computed from ISO-5167)
• Facility reproducibility
• Point-to-point scatter in Cd (from six
  consecutive 90-second runs at each flow rate)
• Primary element installation effects

17
Acceptance Criteria

• Discussions between BLM and Rosemount
  Dietrich-Standard produced the following
  acceptance criteria.
• Acceptance Criterion 1
• Compare point by point. If the 95 confidence
  intervals for Cd for the baseline and disturbance
  tests overlap, no installation effect exists for
  that configuration.
• If no overlap, use Acceptance Criterion 2.
• Acceptance Criterion 2
• Compare average Cd values over the tested range
  of Reynolds numbers.
• If no overlap, retest using longer upstream pipe
  lengths.

18
Acceptance Criterion 1
19
Acceptance Criterion 2
20
Reporting Requirements

• Test facility information.
• Name, location, test fluid.
• Pressure and temperature instrumentation
  calibration certificates.
• Surface roughness of piping used for tests.
• Meter information.
• Name, type, S/N, model, size, schedule.
• Geometry and critical dimensions.
• Flow conditioner positioning.
• Manufacturers Cd and flow equations.
• Description of full test matrix and presentation
  of results.
• Test type, meter orientation, upstream/downstream
  piping.
• Sketch or diagram.
• Table and plot of measured Cd, including
  uncertainty.

21
Reporting - Test Record Sheets
22
Reporting - Test Record Sheets
23
Test Results

• For this beta ratio, the meter showed no
  additional error due to the disturbances.
• In general, the installation effect tended to
  increase as beta ratio increased.

24
Manufacturers Recommendations Based on Test
Results

• The table summarizes the straight run
  requirements for all models, as determined from
  the test results.

25
Conclusions

• The testing protocol described in API MPMS,
  Chapter 22.2 can provide a thorough and rigorous
  method for testing the performance of any
  differential pressure flow meter.
• The tests are severe and quantify meter
  performance under extreme, if not worst-case,
  flowing conditions.
• The protocol is sufficiently standardized to
  provide apples-to-apples comparisons of various
  differential pressure flow meters, regardless of
  meter geometry.
• The standard is subject to interpretation with
  respect to the definition of an installation
  effect and the statistical analysis.
• These issues should be resolved during
  discussions between the manufacturer, user, and
  test laboratory operator during the development
  of the test plan.