July 2012, Vol. 239 No. 7

Features

Natural Gas Metering Technology - Past, Present, and Future

As everyone knows, shale gas finds have substantially increased natural gas production in the U.S. over the last few years. This, among other reasons, has attracted many new people to the natural gas industry.

For the benefit of those who are responsible for natural gas measurement and who are relatively new to the industry, and also those who are industry veterans, we’ll review some of the key developments in natural gas measurement technology over the past 20 years or so. This will provide perspective on how the technology has evolved over that time.

We’ll also speculate on the advancements to expect in the years ahead. This discussion focuses primarily on metering technology for high-volume pipeline applications and covers the principal flow elements (i.e., orifice, ultrasonic, turbine, and Coriolis flow meters) as well as natural gas sampling methods and gas composition determination.

Orifice Flow Meters
Orifice flow meters have been widely used in the natural gas industry for decades. Orifice meters are based on a set of mechanical specifications and, hence, do not require a flow calibration before being installed in the field. However, it is critically important that the meter be designed, built, installed, operated, and maintained in accordance with the applicable specifications in order for it to measure flow within the inherent uncertainty of the standard discharge coefficient (Cd). The applicable specifications are found in Chapter 14.3 of the API “Manual of Petroleum Measurement Standards” (MPMS).

A proper orifice meter installation is one key to accurate flow measurement. The meter installation configuration specifications contained in the current version of API MPMS Chapter 14.3, Part 2 are based primarily on experimental work performed in the 1990s. This research specifically addressed the length of straight pipe required upstream of the orifice plate, with and without flow conditioning, such that accurate flow rate measurement would be achieved. The results of these experiments are documented in an API White Paper published in 2001 by what was then the Gas Research Institute (GRI) and now the Gas Technology Institute (GTI). This research concluded that, depending on the upstream piping configuration, as much as 145 diameters of straight pipe may be required upstream of the orifice plate. Prior to that finding (up until the early 1990s), it had been assumed that 44 diameters of straight pipe upstream was sufficient to produce accurate flow measurement. Upstream straight lengths shorter than 145 pipe diameters may be acceptable, depending on the configuration of the upstream piping and whether or not a flow conditioner is used, and if so, what type is used.

The research documented in the API White Paper also found differences in the performance of the various commercially-available flow conditioners tested. For instance, it was found that a 19-tube-bundle flow straightener was effective at eliminating flow swirl in a pipe, but was not effective at eliminating velocity profile asymmetry. In contrast, several perforated-plate-type flow conditioners, such as the CPA 50E, the GFC™, and the Daniel Profiler, proved effective at reducing or eliminating velocity profile asymmetry, as well as swirl. An appropriate flow conditioner placed upstream of an orifice plate can reduce the minimum length of straight pipe required upstream, yet still produce accurate flow measurement.

Certain operating conditions can also bias the orifice flow measurement. Research on this topic has been ongoing for many years and continues today. Examples of field conditions that can cause a bias error in the flow measurement include liquid in the flow stream (e.g., compressor oil, hydrocarbon condensate, water, etc.), valve grease deposits in the meter tube or on the orifice plate, pulsations in the flow, an orifice plate installed backwards, a bent plate, nicks or notches in the orifice bore leading edge, etc. In 2001, a compilation of the information in the public domain that pertained to this subject was published in a GRI report. This report cited over 40 research references that dealt specifically with operational effects and serves as an excellent information resource for common operational effects encountered in the field.

Table 1 shows a condensed version of key results from the GRI operational effects report. Table 1 is color coded to illustrate that most operational errors in orifice meter measurement are biases that can be either positive or negative, depending on the specific operating condition that creates the error. Operational problems or effects that always produce a positive bias in the flow measurement (i.e., an over-registration of the flow relative to the actual amount) are highlighted with a green background, problems that always result in a negative bias are highlighted with a red background, and problems that could result in either a positive or negative bias error are highlighted with a yellow background. The interested reader is encouraged to review the GRI report for more information on how the experiments were conducted and the applicability of the results to specific field applications.

Table 1. Flow Rate Errors for Various Operational Effects for Orifice Meters.
In the mid-2000s, there was also a research effort launched to improve the expansion factor correlation used in the orifice meter flow equation.

Even with the introduction of newer and more accurate metering options, orifice meters are still used extensively in the natural gas industry. For the foreseeable future, orifice meters will continue to play a significant role in natural gas measurement. Future enhancements to orifice meters could include greater use of diagnostic methods to identify common operational situations that may introduce measurement errors.

Ultrasonic Flow Meters
Multi-path ultrasonic flow meters began to gain prominence for high-pressure, high-capacity natural gas flow applications in the early to mid-1990s. However, their use for custody transfer applications was limited due to the lack of an industry standard that could be referenced in natural gas sales contracts and transmission pipeline tariffs. In 1994, the Transmission Measurement Committee (TMC) of the American Gas Association (AGA) set out to remedy that situation. The AGA TMC first published an Engineering Technical Note in May 1996 and a recommended practice in June 1998. The recommended practice is AGA Report No. 9.

Report No. 9 is a performance-based standard that includes performance limits on meter accuracy and repeatability. Once Report No. 9 was published, use of multipath ultrasonic flow meters for custody transfer applications began to surge. Development of multipath ultrasonic flow meters in the 1990s advanced relatively quickly. Although research on this metering technology was progressing, much was still to be learned about how the meter installation configuration and the operating environment affected meter performance. Consequently, the original AGA Report No. 9 was limited in its guidance to end users on how to effectively handle these effects.

In the late 1990s and early 2000s, a number of research initiatives were launched to better understand the performance envelope of this emerging measurement technology. Much of the research focused on installation and operational effects testing. One of the early findings of the research was that ultrasonic meters were very sensitive to the shape of the velocity profile of the flow passing through the meter.

Soon, a flow conditioner was routinely being placed upstream of each meter to provide a consistent flow profile at the meter and to minimize any adverse effects of flow distortions created by the upstream piping configuration. Furthermore, it was observed that each flow meter/flow conditioner combination produced unique meter performance characteristics, as illustrated in Figure 1. In this example, two different brands of 12-inch diameter multipath meters were flow tested, first, without a flow conditioner upstream of each meter and then with various flow conditioners placed upstream of each.

It became commonplace for ultrasonic meters to be flow calibrated with the flow element, flow conditioner, and adjacent upstream and downstream piping joined in a single assembly before the meter was installed in the field. In fact, the latest edition of AGA Report No. 9 recommends this.

Brand “A”

Brand “B”
Figure 1: Effect of a Flow Conditioner Upstream of an Ultrasonic Flow Meter in which “Bare Meter” configuration was 100 diameters of straight pipe upstream of the meter.

The evolution and research on ultrasonic meters continued through the 2000s. Recent research has evaluated the onboard diagnostic capabilities of ultrasonic flow meters to help identify installation and operational effects that adversely affect measurement accuracy.

A second edition of AGA Report No. 9 was issued in April 2007. This edition provided more detailed guidance on meter design, installation, operation, and maintenance than did the first edition. Since this technology continues to evolve, expect another update to Report No. 9 in the not too distant future.

“Other” Flow Meter Types (i.e., Turbine and Coriolis Flow Meters)
Space constraints prevent a detailed discussion, but major initiative over the past 15 years have culminated in an update to the industry recommended practice, i.e., American Gas Association Report No. 7 (Measurement of Natural Gas by Turbine Meter), in 2006.

Coriolis flow metering technology began to gain popularity in the natural gas industry, primarily for high-pressure, low-volume applications, in the late 1990s. An effort was launched in 1999 by the AGA TMC, with assistance from the API Committee on Gas Fluids Measurement (COGFM), to develop an industry standard for this technology. The AGA first published an Engineering Technical Note on the subject (i.e., Coriolis Flow Measurement for Natural Gas Applications) in 2001, followed by a recommended practice, AGA Report No. 11 (Measurement of Natural Gas by Coriolis Meter) in 2002. An update to Report No. 11 is scheduled to be published in 2012. Research in support of the original development of AGA Report No. 11 was funded by GRI, with additional assistance from several Coriolis meter manufacturers and end users. Coriolis metering technology continues to evolve, with meter size and flow range continuing to expand. Expect the pending revision to AGA Report No. 11 to address the evolutionary changes in the technology achieved over the past decade.

Natural Gas Sampling Methods And Gas Composition Determination
In 1993, the API COGFM began a 12-year research effort to evaluate all of the generally-accepted natural gas sampling techniques used to acquire a representative gas sample from a pipeline. This effort reviewed and revised the standard practices in the United States for collecting and handling spot, composite, and on-line natural gas samples. The sampling methods reviewed during this study were described in the fourth edition of Chapter 14.1 (i.e., Collecting and Handling of Natural Gas Samples for Custody Transfer) of the API MPMS. Some of the sampling methods referred to in Chapter 14.1 were described in detail in the Gas Processors Association (GPA) Standard 2166 (i.e., Obtaining Natural Gas Samples for Analysis by Gas Chromatography).

Five GRI research reports and three Minerals Management Service research reports were issued summarizing the findings of this large body of work. The GRI reports covered an array of research tasks, including identification of root causes of natural gas sample distortion, an assessment of common spot sampling methods when the source gas was at or near its hydrocarbon dew point, an assessment of the effect of ambient conditions on composite gas sampler performance, an assessment of the effectiveness of common cleaning methods for floating-piston sample cylinders, a comparison of measured and calculated gas dew points for a range of gas mixtures, an assessment of the repeatability of dew point measurements using a chilled mirror dew point detector, an assessment of analytical methods for predicting hydrocarbon dew points, and an assessment of the performance of commonly used water vapor sampling methods, among other tasks.

The MMS research reports presented information on the best practices for preparation of natural gas blends used as calibration standards, a review of the state-of-the-art of sampling methods for wet gas flows, and an evaluation of a proposed gas sampling method performance verification test protocol that would eventually be included in Chapter 14.1 of the MPMS. The wealth of information produced from this work is too lengthy to discuss in detail here. However, some of the key findings included the following.
• Cleanliness of the gas sampling system and gas sample bottles is critically important for extracting and transporting gas samples that are representative of the flowing gas stream. Saturated steam was shown to be the most effective cleaning agent and gas sample bottles should be cleaned between each use.
• Control of the hydrocarbon dew point in the gas sampling process is also critically important. The gas sample must stay above its hydrocarbon dew point from the time the gas sample is extracted from the pipeline to the time it is run through a gas analyzer. Otherwise, condensation of the sample will occur and the compositional analysis will not be representative of the flowing gas stream. It may be necessary to heat the sample system and/or sample bottle to achieve this goal.
• Also, predicting hydrocarbon dew points from equation-of-state calculations can be problematic if the gas composition analysis of interest contains detailed hydrocarbon component information only though hexane (C6). Without detailed characterization of those hydrocarbon components heavier than hexane, the predicted hydrocarbon dew point may be substantially different than the actual value – which may suggest that a phase change (condensation) in the sampling process is not imminent, when it actually is. As an example, Figure 2 shows several phase diagrams (calculated using the Peng-Robinson equation of state) for a single gas composition (1,050 BTU/scf), all computed by assuming different characterizations for the hexanes and heavier hydrocarbons. (The dashed lines on the figure denote 95% confidence intervals on the dew point curve of the same color.)

Figure 2: Effect of Using Various C6+ Characterizations on Predicted Hydrocarbon Dew Point Curves.

As a result of this research initiative, a fifth edition of API MPMS, Chapter 14.1 was published in 2001 and a sixth edition was published in 2006 (and reaffirmed in 2011). A new edition of GPA Standard 2166 was also published in 2005. Because of this research work, some of the less accurate sampling methods were culled from Chapter 14.1 and additional guidance was provided for some of the preferred methods. Suffice to say, the current sampling methods referenced in Chapter 14.1 have proven to be effective for dry natural gas sampling – as long as these methodologies are followed precisely.
The API focus on sampling methods has now moved on to wet gas applications, where an ideal solution has yet to be found. Expect advancements in this area in the years ahead.

Closing
We have given a very brief summary of a broad subject. The interested reader is referred to the reference documents for more detailed information. In the years ahead, expect natural gas flow metering technology to continue to evolve, with greater use of microprocessor technology. Expect more onboard diagnostic capabilities to be added to the various flow metering types to provide better tracking of meter performance problems in a timelier manner. Expect gas sampling technology to become more robust (i.e., able to better handle less than ideal or contaminated flow streams) and environmentally-friendly in the years ahead. Sampling methods that vent natural gas to the atmosphere will likely be phased out because of their adverse effect on the environment. Also, new sensing technology will likely allow for integrated metering packages that measure energy rate rather than gas flow rate and gas composition (or heating value) independently. The applicable metering standards will continue to be updated to reflect changes in technology, although those documents inherently lag behind the latest technological developments.

Authors
Edgar B. Bowles, Jr. is director of the Fluids and Machinery Engineering Department at Southwest Research Institute, San Antonio, TX. (Ph: (210) 522-2086, e-mail: ed.bowles@swri.org)
Adam G. Hawley is a research engineer in the Fluid Dynamics and Multiphase Flow Program at Southwest Research Institute. (Ph: (210) 522-3427, e-mail: adam.hawley@swri.org)

Literature List
“Concentric, Square-edged Orifice Meters,” Manual of Petroleum Measurement Standards, Chapter 14.3, Parts 1-4, API, Washington, D.C., August 2011.
Studzinski, W., U. Karnik, P. LaNasa, T. Morrow, D. Goodson, Z. Husain, and J. Gallagher, “Topical Report: Orifice Meter Installation Configurations With and Without Flow Conditioners White Paper Prepared for API 14.3 Part 2,” Gas Research Institute, GRI Report No. GRI-99/0262, Chicago, IL, March 2001.
Morrow, T. B., D. L. George and M. G. Nored, “Topical Report: Operational Factors that Affect Orifice Meter Accuracy,” Southwest Research Institute, SwRI Project No. 18-8890, Gas Research Institute, GRI Report No. GRI-00/0141, Chicago, IL, April 2002.
Morrow, Thomas B., “Topical Report: Additional Studies of Orifice Meter Installation Effects and Expansion Factor,” Southwest Research Institute, SwRI Project No. 10315, Gas Research Institute, GRI Report No. GRI-04/0246, Chicago, IL, March 2005.
Hodges, Casey and Thomas Morrow, “White Paper Summarizing API Investigative Testing for Existing Expansion Factor Equation,” Report to the Chapter 14.3.1 Working Group, API Committee on Petroleum Measurement Standards Meeting, Phoenix, AZ, March 2005.
Beeson, J., “Ultrasonic Metering – A Field Perspective,” AGA Operations Conference, 1995.
“Engineering Technical Note: Ultrasonic Flow Measurement for Natural Gas Applications,” Transmission Measurement Committee, American Gas Association, Arlington, VA, May 1996.
“Measurement of Gas by Multipath Ultrasonic Meter,” Report No. 9, Catalog No. XQ0701, Transmission Measurement Committee, AGA, Washington, D.C., June 1998.

van Bloemendaal, K. and P. M. A. van der Kam, “Installation Effects on Multipath Ultrasonic Flow Meters: The ‘ULTRAFLOW’ Project,” Third International Symposium on Fluid Flow Measurement, San Antonio, TX, April 1995.
Groupe Européen de Recherches Gazières (GERG), “Present Status and Future Research on Multi-path Ultrasonic Gas Flow Meters,” GERG Technical Monograph 8, 1995.
Grimley, T. A., “Topical Report: Performance Tests of 12-inch Multipath Ultrasonic Flow Meters,” Southwest Research Institute, SwRI Project No. 8890, Gas Research Institute, GRI Report No. GRI-96/0291, Chicago, IL, August 1996.
Grimley, T. A., “Topical Report: Performance Testing of 12-inch Ultrasonic Flow Meters and Flow Conditioners in Short Run Meter Installations,” SwRI Project No. 18-8890, GRI Report No. GRI-01/0129, Chicago, IL, January 2002.
Grimley, T. A., “Topical Report: Performance Testing of 8-inch Ultrasonic Flow Meters for Natural Gas Measurement,” Southwest Research Institute, SwRI Project No. 18-8890, Gas Research Institute, GRI Report No. GRI-00/0124, Chicago, IL, November 2000.
Grimley, Terrence A., “Topical Report: Effects of Diameter Mismatch and Line Pressure Variations on Ultrasonic Gas Flow Meter Performance,” Southwest Research Institute, SwRI Project No. 18-4435, Gas Research Institute, GRI Report No. GRI-02/0031, Chicago, IL, April 2002.
Morrow, Thomas B., “Topical Report: Line Pressure and Low Flow Effects on Ultrasonic Gas Flow Meter Performance,” Southwest Research Institute, SwRI Project No. 18-10315, Gas Research Institute, GRI Report No. GRI-05/0133, Chicago, IL, March 2005.
Hawley, Adam G., “Investigation of Ultrasonic Meter Accuracy Shifts Resulting from Grime on the Meter Body and Acoustic Transducers,” Southwest Research Institute, SwRI Project No. 18-16427, Pipeline Research Council International, PRCI Contract No. PR-015-11700, Church Falls, VA, March 29, 2012.
George D. L., “Topical Report: Turbine Meter Research in Support of the Planned AGA Report No. 7 Revision,” Southwest Research Institute, SwRI Project No. 18-4435, Gas Research Institute, GRI Report No. GRI-00/0226, Chicago, IL, January 2003.
George D. L., “Topical Report: Effects of Line Pressure and Gas Density on Turbine Meter Measurement Accuracy at Conditions from Atmospheric Air to 700 psig Natural Gas,” Southwest Research Institute, SwRI Project No. 18-6584, Gas Research Institute, GRI Report No. GRI-03/0172, Chicago, IL, August 2004.
Siebenaler, Shane P. and Darin L. George, “Topical Report: Effects of Turbine Meter Cartridge Change-out on Measurement Uncertainty,” Southwest Research Institute, SwRI Project No. 18-10315, Gas Research Institute, GRI Report No. GRI-04/0207, Chicago, IL, March 2005.
“Measurement of Natural Gas by Turbine Meter,” Report No. 7, Catalog No. XQ0601, Transmission Measurement Committee, American Gas Association, Washington, D.C., February 2006.
“Engineering Technical Note: Coriolis Flow Measurement for Natural Gas Applications,” Catalog No. XQ0112, Transmission Measurement Committee, American Gas Association, Washington, D.C., December 2001.
“Measurement of Natural Gas by Coriolis Meter,” Report No. 11, Catalog No. XQ0311, Transmission Measurement Committee, American Gas Association, Washington, D.C., January 2003.
Grimley, T. A., “Topical Report: Coriolis Mass Flow Meter Performance with Natural Gas,” Southwest Research Institute, SwRI Project No. 18-4435, Gas Research Institute, GRI Report No. GRI-01/0222, Chicago, IL, January 2002.
“Collecting and Handling of Natural Gas Samples for Custody Transfer,” Manual of Petroleum Measurement Standards, Chapter 14.1, 6th Edition, American Petroleum Institute, Washington, D.C., August 1993.
“Obtaining Natural Gas Samples for Analysis by Gas Chromatography,” GPA Standard 2166-86, Gas Processors Association, Tulsa, Oklahoma, 1986.
Behring, Kendricks, A., III and Eric Kelner, “Topical Report: Natural Gas Sample Collection and Handling – Phase I,” Southwest Research Institute, SwRI Project No. 18-8890, Gas Research Institute, GRI Report No. GRI-99/0194, Chicago, IL, August 1999.
Kelner, Eric, Charles L. Britton (CEESI), Kendricks A. Behring, III, and Cecil R. Sparks, “Topical Report: Natural Gas Sample Collection and Handling – Phase II: Experimental Testing Under Simulated Field Conditions,” Southwest Research Institute, SwRI Project No. 18-8890, Gas Research Institute, GRI Report No. GRI-01/0069, Chicago, IL, January 2003.
Kelner, Eric, Cecil R. Sparks, and Kendricks A. Behring, III, “Topical Report: Natural Gas Sample Collection and Handling – Phase III: Experimental Investigation of Gas Sampling Techniques and Equipment,” Southwest Research Institute, SwRI Project No. 18-8890, Gas Research Institute, GRI Report No. GRI-01/0070, Chicago, IL, August 2002.
George, D. L., A. M. Barajas, E. Kelner, and M. Nored, “Topical Report: Natural Gas Sample Collection and Handling – Phase IV,” Southwest Research Institute, SwRI Project No. 18-6584, Gas Research Institute, GRI Report No. GRI-03/0049, Chicago, IL, November 2004.
George, D. L., A. M. Barajas, R. C. Burkey, and T. B. Morrow, “Topical Report: Natural Gas Sample Collection and Handling – Phase V,” Southwest Research Institute, SwRI Project No. 18-10315, Gas Research Institute, GRI Report No. GRI-05/0134, Chicago, IL, March 2005.
George, D. L., “Preparation of Natural Gas Blends Used as Calibration Standards: Sources of Uncertainty and Best Preparation Practices,” Southwest Research Institute, U.S. Minerals Management Service, Herndon, VA, April 2003.
Nored, M. and D. L. George, “Final Report: A Review of the Current State and Direction of Methods for Sampling Wet Gas Flows,” Southwest Research Institute, U.S. Minerals Management Service, Herndon, VA, May 2003.
Barajas, A. M. and D. L. George, “Assessment of Sampling Systems for Monitoring Water Vapor in Natural Gas Streams,” Southwest Research Institute, U.S. Minerals Management Service, Herndon, VA, March 2006.
George, D. L., A. M. Barajas, E. Kelner, and M. Nored, “Topical Report: Natural Gas Sample Collection and Handling – Phase IV,” Southwest Research Institute, SwRI Project No. 18-6584, Gas Research Institute, GRI Report No. GRI-03/0049, Chicago, IL, November 2004.