Fluke 721, 721EX Application Note

Calibrating Gas Custody

Transfer Flow Computers

Gas custody transfer flow computers that calculate flow by measuring the

differential pressure across a flow restriction, such as an orifice plate, require

special calibration to perform at optimum accuracy. In custody transfer

applications where the buying and selling of commodities like natural gas is

involved, calibration checks are performed frequently as a matter of fiduciary

responsibility. For the purpose of this application note, the use of gas custody

transfer flow computers in the natural gas transmission industry is referenced.

Flow computers need multiple measurements to calibrate each device. In the

normal application, three measurements are made: volumetric flow, static

(line) pressure and temperature. A calculation is performed using this data to

determine the actual mass of the gas flowing through the pipeline.

The Fluke 721 Precision Pressure Calibrator has
special features that support the complete calibration
of natural gas multi-variable electronic flowmeters
and other types of flow computers. With two internal
pressure ranges, external Fluke 750P pressure mod­ules and an optional precision RTD probe, all of the
calibrations required for the flow computer can be
performed with just one instrument.

The Fluke 721 is available with two built-in
pressure ranges from from 16 psi/1 bar up to
5000 psi/345 bar. For this application, the
configuration with the low pressure sensor (P1)
16 psi/1 bar and high pressure sensor (P2) of
1500 psi/100 bar is frequently the best fit. Since the
Fluke 721 has an accuracy specification as a percent
of full scale, it is important to closely match the full
scale of the calibrator to the scale of the application
in order to get the best performance. (See sidebar
on system accuracy calculation and importance of
maintaining an adequate accuracy ratio).

In addition to the calibrator itself, a high and
low pressure calibration pressure source will be
needed. An accessory RTD probe is also required
for measuring temperature. An appropriate low
pressure source with 0.01 inH2O of resolution is
needed for the low pressure test, and a high pres­sure source such as a regulated nitrogen bottle or
hydraulic hand pump are required. One pump style
typically does not work well for both tests or exten­sive cleaning is required to switch from hydraulic
(oil or water) to pneumatic testing. High pressure
pumps typically do not have the desired resolution
for the low pressure test.

Application Note

PRESSURE

721

F1 F2 F3

V

mA

COM

CALIBRATOR

ZERO

30V
24mA
MAX

stream

7.21psi

Fluid

stream

Figure 1. Energ y is ex pressed in real, reactive, and apparent power.

Gas custody transfer
flow computer operational theory

Custody transfer flow computers are called by a
variety of names including electronic flowmeters
(EFMs) and multi-variable flow computers, but they
all feature some common principles of operation.

1. Volumetric flow measurement uses some type of
flow restriction such as an orifice plate to gener­ate a pressure drop. The differential pressure

Fluid

From the Fluke Digital Library @ www.fluke.com/library

created across this pressure drop is measured by

n

2

1

2

2

2

the flow computer as the primary measurement.
It is based on the principle that flow velocity is
proportional to the square root of the pressure
drop. The volumetric rate is then calculated from
the velocity by knowing the diameter of the pipe
in which the gas is flowing.

The measured pressure drop (differential) is
typically around 200 inches of water column
(“WC) or 8-10 psi.

2. To convert volumetric flow to mass flow you
also need to know the density of the mass per
volume of the flowing media. The flow com­puter makes this calculation using 2 additional
measurements, plus a range of factors/constants
based on the flowing media. The two additional
measurements are the static pressure of the gas
in the pipeline and the temperature of the gas in
the pipeline.

The static pressure in these applications
ranges widely from a low of about 300 psi /
20 bar to a high of about 2000 psi / 138 bar.

The temperature of the gas is usually at
ambient, so it is within the range of normal
environmental conditions.

3. A final consideration about flow computers is
how they are typically installed and used.

Industrial applications use either the analog

output of the flowmeter (4 to 20 mA) or a digital
output like the HART signal to get data from the
flowmeter to a control system or data acquisition
system.

This analog output is generally not used in
gas pipeline applications. Instead, the flowmeter
is a specialized device that operates standalone
to measure and record the total mass flow
through the pipeline. The total is periodically
“downloaded” from the flowmeter to be used in
an accounting of gas flow and custody transfer.
This information is also often sent through wire­lessly to a central control point in operations.

The flowmeter may be packaged with other
electronic devices to be able to perform this
function or it may be purpose manufactured,
which is the most common type.

How to calibrate the flow computer

Each flow computer manufacturer has created
a proprietary method of calibration, but they all
use the same general technique, which will be
described here.

In these proprietary calibrations, the manufac-

turer has provided a software application, which
runs on a notebook computer (PC). The PC is con­nected to the serial port or USB port of the flow

System Accuracy Determination

In order to effectively calibrate an instrument, the
calibrator used must be more accurate than the
instrument by some factor. The factor will vary
according to the application, but it should be as
large as is practical. The minimum factor is gener­ally considered to be 3 to 4 times. The common
term for expressing this factor is Test Uncertainty
Ratio or TUR. If the calibrator is 4 times more accu­rate than the device being tested it is referred to as
having a TUR of 4:1.

The rationale behind this comes from a tech­nique for the statistical analysis of the error in a
system. This technique is called Root Square Sum
or RSS. To determine the error in a system you take
the square root of the sum of the errors squared
for all elements in the system. Note that this is not
the maximum possible error in a system, but is the
largest error which is statistically likely.

This formula describes the calculation, where Et
is the total error and E1, etc. are the errors of the
individual components of the system.

2

=+++EEEE

t

By using a TUR of 4:1, the effect of the error
in the calibrator is reduced to a small percentage
of the error of the instrument under test and can

...

therefore generally be disregarded. As an alterna­tive to having a calibrator with the appropriate
ratio, users may elect to de-rate the performance
of the instrument to a value four times that of the
calibrator.

For example, using a calibrator with ±0.05%
accuracy would have a TUR of 4:1 testing an
instrument with an accuracy of ±0.2%. Due to
the continual advances in instrument technology,
calibration technology may, from time to time,
fail to provide the necessary TUR to calibrate to
the instrument manufacturer’s rated specification.
Alternately you can tighten the test tolerance to
80% of the desired specification to gain the same
confidence using a technique called guardbanding.
The fundamental concept of guardbanding is to
restrict the Pass/Fail limits applied to a calibration
test based on a defined criterion. The purpose of
guardbanding is to control the risk of accepting an
out-of-tolerance unit, or rejecting an in-tolerance
unit. Without guardbanding the result of a test will
be Pass or Fail. With guardbanding the result of a
test will be Pass, Fail, or Indeterminate. A Pass or
Fail test result without guardbanding may change
to a result of Indeterminate with guardbanding.
For more information on guardbanding refer to the
application note “Guardbanding with Confidence”
at www.fluke.com/guardbanding.

2 Fluke Corporation Calibrating Gas Custody Transfer Flow Computers