BASICS OF HIGH-PRESSURE MEASURING AND REGULATING STATION DESIGN

Class No. 1010

James N. Witte
USGO Gas Quality Lead
TransCanada Pipelines USA
700 Louisiana Street
Houston, Texas USA

Introduction
This paper presents a review of the criteria necessary for designing high-pressure natural gas measuring and
regulating stations. For the purpose of this discussion, high pressure is assumed to be gauge pressure values
above typical distribution mainline pressures (greater than 60 psig). Therefore, it is most applicable to station
designs with gathering, midstream, and transmission operations.
The design engineer has many factors to consider when designing a metering facility. These factors must include
considerations for their impact on measurement accuracy, facility capital cost, environmental stewardship, public
stewardship, and long-term maintenance cost. The successful designer will seek to achieve high accuracy at an
appropriate capital cost, while minimizing the risks of environmental hazards, public nuisance, such as noise, and
maintenance requirements.
Regardless of the haste with which a meter station is constructed, it is most commonly in operation for many years.
It is not unusual for a facility to be in place for more than 30 years. Considering this, each facility must meet the
purpose for which it is intended, regardless of whether gas flow rates might increase or decrease over time. It is,
therefore, prudent for a designer to consider the potential for facility expansion or flow rate reduction and how that
might affect the facility design. Later modifications are likely to be costlier than the initial construction.
Standards
Several codes and standards are applicable to measuring facilities. The following is a list of applicable regulatory
agencies and standard writing entities:
1) American Gas Association (AGA)
2) American Petroleum Institute (API)
3) American Society of Mechanical Engineers (ASME)
4) American National Standards Institute (ANSI)
5) Gas Processors Association (GPA)
6) U.S. Department of Transportation (DOT) Office of Pipeline Safety (OPS)
7) U.S. Department of the Interior (BLM, BOEMRE, and BSEE)
8) National Fire Protection Association (NFPA)
9) National Electric Code (NEC)
10) U.S. Environmental Protection Agency (EPA)
11) U.S. Occupational Safety and Health Administration (OSHA)
12) Applicable local, state, and federal requirements not governed by the previous agencies
Initial Data
In order to design a measuring facility properly, some initial data are needed for equipment sizing and selection.
One of the classic challenges of the designer is trying to interpret the commercial representative’s statement of
range, i.e., consider a station with a range stated as 0 – 250 million standard cubic feet per day (MMscfd). The first
consideration is that this will be a rather large meter station. Secondly, there is no indication of what is being fueled
so that the range requirements can be verified. Also, it is appropriate for the design engineer to advise that zero is
not a flow rate and determine what the actual minimum station flow rate will be.
To aid with these types of discussions, let’s consider what data are necessary to do a proper design. Table 1
provides a list of items to consider as one is beginning the design process. Note that it is important for all of the
parties involved to pay attention to the engineering units for each parameter. It is normal for transmission pipeline
companies to describe flow rates in terms of daily values and local distribution companies to use hourly values.
By considering each of the items listed in Table 1, one can best arrive at the necessary design requirements. It is
best to talk to the other party’s engineers and learn exactly what is being connected and how it will likely be operated.
Note that one needs to know the maximum, minimum, and expected normal flow rates for proper sizing. For
instance, a power plant design engineer will have precisely worked through the unit fuel requirements and the
staging of units required by the operating plan. This will be critical information for determining station “turndown”
(i.e., the ratio of maximum flow rate to minimum flow rate) and range.
Table 1. Design Considerations
Design Considerations Units
Expected flow rates (maximum, normal, and minimum) MMscfd
Peak hourly flow rate Mscfh
Projected growth MMscfd
Pressure (maximum, normal, and minimum) psig
Maximum Allowable Operating Pressure (MAOP) of both systems psig
Maximum Operating Pressure (MOP) of both systems psig
Overpressure protection (type, method, and responsibility)
Expected flowing temperatures (maximum, normal, and minimum) ºF
Base pressure psia
Base temperature ºF
Atmospheric pressure psia
Gas relative density unit less
Gas heating value Btu/scf
Hydrocarbon dew point temperature ºF
Water content lbs. / MMscf
Maximum carbon dioxide, nitrogen, and oxygen (diluents) mole %
Maximum hydrogen sulfide grains/100 scf
Gas composition and gas quality determination method
Maximum delivery pressure (downstream of the meter) psig
Maximum allowable noise dBA
EGM software and hardware specifics per operating company
Control method requirements (e.g., flow control with pressure override)
Remote monitoring and control requirements (SCADA)
Frequency of monitoring required (real-time/daily/weekly/monthly)
Location (onshore/offshore/wetland/residential area, etc.)
Availability of utilities (electricity, telephone, etc.)
Local building or other permit requirements
Liquid removal, measurement, and re-injection requirements
Condensate removal/storage/handling requirements
Criticality of service (is shutdown of service for inspection allowable or not)
Filter separator requirement
Heating requirement Btu/hr
Odorization requirement
Requirements for meter proof or reproof

Another sizing consideration is that city gate stations may have peak hourly rates that are about six times as great
as the daily total demand. In this case, all of the flow is needed at peak demand times, once in the morning and
once in the evening. Facility sizing based on daily total demand would likely cause the facility to have inadequate
capacity.
Major Piping Design
All piping assemblies in a high-pressure facility must meet the codes for pressure piping design,
i.e., API Specification 5L[i], API Specification 6D[ ii], ASME/ANSI B16.5[iii], and ASME/ANSI B16.9[iv]. In accordance
with DOT 49 CFR Part 192[v], pipe used for measurement facilities must be designed with a Class 3 design factor
of 0.50 (50% of specified minimum yield strength (SMYS)) or a factor of 0.40 in a Class 4 location.
Tap valves and interconnect piping are sized on the basis of a velocity limit. This velocity limit is usually assigned
by company convention; however, it is relatively common practice to use a velocity limit of 70 feet per second. It is
important to consider whether liquids might be present in the flow stream to avoid reaching the erosion velocity
threshold, i.e., the flow velocity at which the onset of pipe wall erosion may occur. The erosion velocity threshold
may be very high for clean dry natural gas, but may drop below the 70 feet per second threshold when free liquids
or particulates are introduced into the flow stream.
Headers are necessary for multi-tube meter stations to ensure acceptable velocity balance across the parallel runs.
If properly sized, the headers will allow for velocity reduction prior to meter tube entry. This velocity reduction is
necessary to reduce turbulence intensity in the meter tube entry.
Upstream piping design is an important part of ensuring accurate measurement. The metering technologies
currently used in the natural gas industry were designed assuming homogeneous, steady-state, non-swirling,
axi-symmetric, and bulk turbulent flow. Another way of stating this is that the inlet velocity profile is assumed to be
a function of the “law of the wall” that assumes that the velocity profile shaping occurs as a function of pipe wall
roughness and radial position.
Perhaps the best way to think about this approach is to envision very gradual flow direction changes that will result
in essentially straight flow streamlines. Compact meter station designs with close-coupled pipe fittings make it a
challenge to produce smooth, straight streamlined flow. Too much turbulence caused by convoluted piping
geometry upstream of an ultrasonic flow meter can create problems in transmitting acoustic signals across the
flowing gas stream. Figure 1 is a photograph of a meter station that produces a high turbulence intensity and
skewed flow profile at the meter tube inlets due to several tee junctions upstream of the meter tubes.

Figure 1. Meter Station Piping Showing Several Short-Coupled Turns Prior to Measurement
Another consideration for station piping is pressure drop. Under-sizing pipe and piping elements may create
excessive pressure drop and a waste of horsepower in compression.
Check valves are a necessary item for unidirectional flow stations and should always be located at the station tap
rather than in line with the meter tube. This practice helps avoid the potential for distortion of the velocity profile in
the meter tube, which can produce flow measurement bias errors.
Metering
The selection of meter type will be a function of the application and flow range required. Prudent design practice
limits the applied range to flow rates for which the transduced flow transmitter inputs are limited to an operating
range of 10% – 90% of calibrated span or frequency range. A typical differential pressure span value is 300 inches
water column, and a typical input frequency span value is 5,000 Hz.
Typical maximum velocity values for different inferential meter types are as follows:
1) Orifice meter = 30 feet per second
2) Turbine meter = 45 feet per second
3) Ultrasonic meter = 70 feet per second
Each company will have its own preference for meter type as a function of application. Some will almost exclusively
use orifice meters for production receipt measurement. Others might use turbine meters as the meter of choice for
city gate applications. Ultrasonic meters may be preferred for mainline to mainline high-flow, interconnect facilities.
Rarely is a single meter tube going to be adequate for an application. Having multiple meter runs allows for better
applicability of the facility within the optimum performance region of the flow meters. Overall metering uncertainty
may be reduced as flow is distributed across multiple meter runs. With multiple meter runs, the turndown of a meter
station is a function of the number of parallel runs and the control set points for switching flow among the individual
meters.
Different approaches to run switching are possible, but the most popular method is cascading flow. For cascading
flow, meter runs are placed in service in sequential order. Usually the primary meter run remains in service at all
times as flow increases with the opening of the parallel runs. Run switching control is handled in a gas flow
computer with user-entered values for run switching set points. These values are selected to keep flow on each
meter run within the optimum performance for that meter.
Calculation of run switching set points may be accomplished by assuming balanced flow across each run for meter
runs of equivalent diameters. If the meter station design will have meter runs of different diameters, the individual
meter run capacities will not be equally balanced across the meter runs, and the smallest diameter meter run will
have a lower tube velocity than the other parallel runs of larger diameter. Failure to consider this flow balancing
phenomenon may limit station capacity if meter capacities were simply selected from a meter capacity chart.
To determine the maximum velocity on the smaller diameter runs at station capacity, the Darcy Equation may be
used to calculate the pressure drop across the largest diameter meter run at full capacity. This pressure drop value
may be applied to the other parallel meter run capacity calculations, and a rearrangement of the Darcy Equation
may be solved for the individual run velocities.
It is important for the meter station designer to consider run switch points to assure that the flow rates are always
within an acceptable range when switching occurs.
Gas Quality
Gas quality criteria are established by the operating companies’ tariffs. Limitations on non-hydrocarbon
components, hydrocarbon dew point temperature, and other contaminants are defined. A transportation pipeline
may choose to reject a gas source if they find that the gas has materials which are commercially objectionable.
These materials could include sand, dust, gums, crude oil, condensed liquids, or impurities.
Two-phase flow may produce metering bias errors (positive or negative) in excess of 2%. For most operations, this
magnitude of error is unacceptable. The solution is to include liquid separation as part of the meter station design
to eliminate free liquids from the gas stream. Separators are necessary on the downstream side of glycol
dehydration processes and where operating conditions may allow for condensation of hydrocarbon liquids in the
pipeline upstream of the meter station. The design engineer needs to know that not all separators are the same,
and the selection of an appropriate unit will depend on the flow regime. In some cases, a slug catcher could be
required in addition to a separator.
Compression
The use of gas compressors in close proximity to the meter station may lead to unacceptably high measurement
bias errors. As stated earlier, one of the primary assumptions of currently applied metering technologies is the
presence of steady-state flow. If flow pulsation is expected or determined to exist, an engineering study must be
made to design pulsation dampening equipment. The pulsation dampeners may be installed either on the
compressor piping or at the meter station. Either way, the dampener must be positioned between the meter station
and the compressor (or any other pulsation source) to be effective. Pulsation dampening can be expensive to
implement, and eliminating the pulsation at the source is the preferred method.
Station Control, Control Valves, and Pressure Regulation
Station design would be incomplete without the installation of either flow control valves or pressure regulators.
Many transmission pipelines will decline the maintenance requirements associated with flow/pressure regulation.
However, one of the two parties must take on the responsibility.
Several different control valve configurations are available. Some examples of control valves are globe valves,
noise-attenuated ball valves, axial-flow plug or boot-type regulators, and diaphragm and grid-type regulators.
Usually, the selection is a matter of operating company choice.
Most meter stations are installed as a part of a SCADA system that allows a gas control center to remotely control
the facility. The majority of the facilities are simple flow control stations in which the gas controller can write a
remote set point to the unit, and it will throttle a control valve to deliver a prescribed (nominated) quantity of gas for
a desired length of time.
Certain control philosophies are important to understand. If the application is the only supply to a power plant, it
must be fed as a pressure control supply. Power plants cannot tolerate limitations on the unit speed or unit starting
sequence. A steady pressure during the sequence of load changes is desirable. The amount of pressure drop that
is permissible will be a function of the sizing of the unit fuel gas regulators. Usually, control within 10 psi is desired.
If an existing power plant already has other gas suppliers, it is necessary to meet with the plant operators to
determine the control requirements for the new feed. For example, suppose there are three suppliers to the plant,
one (taking the swing) will be for pressure control, while the other two will be for flow control. Having all three for
flow control would produce unsatisfactory results, as pressure on the plant header may drift if the scheduled flow
rates do not precisely match the demand of the power generation.
There are several control options available, such as flow control, pressure control, flow control with pressure
override, pressure control with flow override, and flow control with pressure under-ride. These are normally
produced as a product of software controller interactions through a selector function. Figure 2 presents an example
diagram for software controller interactions with override functions.
 Figure 2. Control Configuration Diagram
The best design policy concerning controls is to keep it simple. Nesting of controls, unique devices, and complicated
interacting scenarios can result in a technical support nightmare. Designs that require frequent ongoing
outside-consultant support are costly to maintain and should be avoided.
Failure Modes
Consideration of equipment failure is important to ensure safe operation. Design standards such as DOT 49 CFR
Part 192 require the design engineer to consider one pressure control failure and plan for an overpressure protection
device, such as a relief valve or pressure monitor valve. The author suggests that the design engineer consider
how each critical element will respond during a failure event. Most operating companies have determined how
control valves and regulators should fail as a function of application. Table 2 provides an example of a failure mode
analysis. Note that some applications are more critical to the operation and have provisions for maintaining service.
Other less critical applications may be allowed to fail off-line.
The case for the failure mode analysis shown in Table 2 assumes an independent pressure monitor that is a
pneumatic system. Using an independent pneumatic pressure monitor valve eliminates considerations of electrical
power failure, which could affect electronic controls.
For those applications having electric or electronic valve controls, considerations for power failure must be made.
Usually, a backup uninterruptable power supply or DC battery backup power supply is installed as a part of the
station controls design.
 Table 2. Example Failure Mode Analysis Where the Monitor Valve is a Standalone Pneumatic Unit

Facility Use Failure Regulator Position Monitor Position

Loss of power gas Fail in last position Fail closed
Loss of position signal Fail open N/A
Power plant delivery Loss of pressure sensing Fail open Fail open
Positioner power fail Fail in last position N/A
Loss of positioner feedback signal Fail open N/A

Loss of power gas Fail in last position Fail in last position

Loss of position signal Fail open N/A
Mainline pressure limiting station Loss of pressure sensing Fail open Fail open
Positioner power fail Fail in last position N/A
Loss of positioner feedback signal Fail open N/A

Loss of power gas Fail in last position Fail closed

Loss of position signal Fail open N/A
City gate (pressure reducing) station Loss of pressure sensing Fail open Fail open
Positioner power fail Fail in last position N/A
Loss of positioner feedback signal Fail open N/A

Loss of power gas Fail in last position Fail closed

Loss of position signal Fail open N/A
Pipeline to pipeline interconnect Loss of pressure sensing Fail open Fail open
Positioner power fail Fail in last position N/A
Loss of positioner feedback signal Fail open N/A

Loss of power gas Fail in last position Fail closed

Loss of position signal Fail open N/A
Receipt from gas storage Loss of pressure sensing Fail open Fail open
Positioner power fail Fail in last position N/A
Loss of positioner feedback signal Fail open N/A

Energy Measurement
Energy measurement requires proper gas sampling at the point of measurement. Some well-defined guidelines
exist in API Manual of Petroleum Measurement, Chapter 14.1[vi] and GPA 2166[vii].
The goal should be to sample gas from the least-turbulent point in the meter run. A location immediately
downstream of an orifice plate is not an advisable choice since any free liquids may be atomized as they pass
through the orifice plate and may bias the sample.
Good sampling practices avoid the opportunity for hydrocarbon dew point temperature to be reached in the sampling
system. Attention to the installation of heated assemblies and heat-traced sample tubing is necessary to maintain
the gas temperature above the hydrocarbon dew point, thus eliminating condensation of heavier gas components.
Sample probes need to be installed in a straight run of horizontal pipe. The sample probe should extend into the
middle third of the pipe diameter.
Online gas chromatographs must be installed with attention to the same sampling criteria, with the additional
concern of avoiding hydrocarbon dew point temperature in the calibration gas cylinder. The slightest condensation
of hydrocarbon liquids inside a calibration gas cylinder will cause measurement bias if the cylinder is used for
calibration without preheating.
Conclusions
Proper meter station design requires knowledge of several standards and regulations and requires experience to
develop good practice. It is hoped that this paper has provided some insight into the basics of high-pressure
measuring and regulating station design, but certainly does not eliminate the need to study and understand the
reference material. For guaranteed success, the designer must maintain a current working knowledge of equipment
and recent research that drives changes in industry standards and meter station design.
It has been the experience of the author that industry members are most willing to share their knowledge and
experience on this subject. The new station designer is encouraged to attend industry schools and question
experienced attendees for the best learning experience.
References
1. “Specification for Line Pipe – Upstream Segment,” Specification 5L, 45th Edition, American Petroleum
Institute, 1220 L Street, N.W., Washington, D.C. 20005, December 1, 2012 (effective date: July 1, 2013).
2. “Specification for Pipeline and Piping Valves,” Specification 6D, 24th Edition, American Petroleum Institute,
1220 L Street, N.W., Washington, D.C. 20005, August 2014 (effective date: August 1, 2015).
3. “Pipe Flanges and Flanged Fittings: NPS 1/2 through NPS 24 Metric/Inch Standard,” ASME/ANSI B16.5-
2013, American Society of Mechanical Engineers, Two Park Avenue, New York, NY 10016, April 29, 2013.
4. “Factory-Made Wrought Buttwelding Fittings,” ASME/ANSI B16.9-2012, American Society of Mechanical
Engineers, Two Park Avenue, New York, NY 10016, February 28, 2013.
5. 49 CFR Part 192 – Transportation of Natural and Other Gases by Pipeline: Minimum Federal Safety
Standards, U.S. Code of Federal Regulations (CFR), U. S. Department of Transportation, October 1, 2011.
6. API Manual of Petroleum Measurement Standards, Chapter 14 – Natural Gas Fluids Measurements,
Section 1 – Collecting and Handling of Natural Gas Samples for Custody Transfer (R2011), 6th Edition,
American Petroleum Institute, 1220 L Street, N.W., Washington, D.C. 20005, January 1, 2006.
7. “Obtaining Natural Gas Samples for Analysis by Gas Chromatography,” Standard 2166-05, Gas Processors
Association, 6526 East 60th Street, Tulsa, Oklahoma 74145, January 1, 2005.