Measurement Techniques, Vol. 52, No.

5, 2009

MECHANICAL MEASUREMENTS
DETERMINATION OF THE JOULETHOMSON
COEFFICIENT IN PROBLEMS OF MEASURING
THE FLOW RATE OF NATURAL GAS

E. P. Pistun, F. D. Matiko,
and O. Ya. Masnyak

UDC 681.121.84

Existing methods of calculating the JouleThomson coefficient for natural gas are analyzed. A procedure
for use in calculations of gas flow rate employing the variable pressure drop method is proposed which
gives increased measurement accuracy by eliminating the additional error in determining the temperature
of the natural gas.
Key words: JouleThomson coefficient, flowmeter, natural gas.

The variable pressure drop method is mainly used to measure the flow rate and quantity of natural gas in large-diameter pipelines. The new standards [16], which standardize the measurement of the flow rate and quantity of flowing media
using this method, provide the possibility of taking into account the systematic component of the error in measuring gas flow
rate due to the difference between the temperature of the gas at the point where it is measured after the constricting device
and the temperature before the constricting device. In accordance with the requirements in [5], the temperature after the constricting device must be corrected when there is a loss in pressure in the constricting device greater than 160 kPa, i.e., this
correction is obligatory. According to the standard [6], the correction for the change in temperature of the gas when it flows
through a constricting device must be introduced if high accuracy in measuring the flow rate is required, and also if there
are considerable pressure losses in the constricting device. Hence, in the standard [6] there is no limitation on the value of
the pressure loss as in [5], while the correction for the change in the gas temperature, measured after the constricting device,
is recommended only when a higher accuracy in measuring the flow rate is necessary. The same may apply when measuring
flow rate as described in [5].
The main obstacle to introducing such a correction has been the lack of a simple method of calculating it using a
calculation of the JouleThomson coefficient. Below we present a fairly simple method of calculating this coefficient, which
can be achieved using microprocessor calculators of the flow rate and quantity of a gas.
As is well known, the flow of a gas through the constricting device is accompanied by a change in the parameters of
the state of the gas (the pressure, temperature and density) along the measuring part of the pipeline. A change in the temperature of the gas in the throat of the constricting device and after it along the length of the pipeline means that the constricting
device of the flowmeter and its measuring temperature converter, placed a distance of 515 internal diameters of the measuring pipeline [5] after the constricting device, are situated in a zone of different temperatures of the gas flow. It is impossible to
measure the gas temperature immediately in front of the constricting device, as required by the variable pressure drop method,
without distorting the kinematic structure of the flow. Hence, the temperature difference of the gas immediately in front of the
constricting device and at the point where the converter is situated are found [5, 6] by calculation using the integral
Lviv National Politechnic University (Lvivska Politekhnika), Lviv, Ukraine; e-mail: epistun@polynet.lviv.ua. Translated
from Izmeritelnaya Tekhnika, No. 5, pp. 4649, May, 2009. Original article submitted March 11, 2009.
0543-1972/09/5205-0509 2009 Springer Science+Business Media, Inc.

509

Dh, K/MPa

p, MPa

Fig. 1. Dependence of the JouleThomson coefficient on the pressure and temperature.

JouleThomson effect equation. In order to use this equation, it is necessary to know the value of the JouleThomson coefficient, which is a function of the pressure, temperature and composition of the natural gas.
The equation of the change of temperature in the throttling of a gas has the form [7]
dT =

1
T ( / T ) p dp = Dh dp ,
cp

(1)

where p, T, and are the absolute pressure, temperature and specific volume of the gas, respectively; cp is the isobaric heat
capacity of the gas under the operating conditions; and Dh is the JouleThomson coefficient.
As follows from (1), to determine the coefficient Dh it is necessary to calculate the isobaric heat capacity of the
gas cp, the parameters of the equation of state of the gas (the specific volume ) and some derivatives of the equation of state.
The equation of state of a gas can be expressed differently depending on which independent parameters of the state it connects. To describe the state of a natural gas the existing standard documents (in particular [8]) propose the equations of state
in the form of the dependence of the compressibility Z on the density and the temperature T, i.e., Z = (, T). To calculate
the coefficient Dh using this equation of state, formula (1) must be converted to the following form [9]:

Z
Z
Dh = T

T
T

Z
c p Z + .

(2)

Using formula (2) and the equation of state from [8], we can obtain the following theoretical equation:
Dh = 103(A2 A1)/[cp(1 + A1)],

(3)

where Dh is the JouleThomson coefficient, K/MPa; cp is the specific isobaric heat capacity of the gas, kJ/(kgK); is the
density of the gas under the operating conditions, kg/m3; and A1 and A2 are dimensionless coefficients, which can be calculated as described in [8].
Equation (3) enables us to find the values of the JouleThomson coefficient in the region where the equation of state
given in [8] holds for the known total component composition of the gas.
510

TABLE 1. Coefficients of Eq. (4)

Values of the coefficients aij for different j

i
1

67.802

446.76

1102.4

1209.1

498.87

143.8

965.14

2433.4

2736.8

1162.1

113.04

775.41

2001.9

2307.9

1003.15

22.27

158.22

420

490.79

209.56

14.178

95.204

241.87

276.03

118.25

2.7783

20.996

62.704

90.622

56.442

In the international standard [10], a fundamental equation is presented which establishes the relation between the
free Helmholtz energy and the reduced density and temperature, and also the component composition of the gas. Equations
for determining the thermodynamic properties, including the JouleThomson coefficient, were obtained in [10] using partial
differential equations of the free Helmholtz energy. The method proposed in [10] is intended for calculating the properties of
mixtures in the gaseous state and has a wide area of application (p 30 MPa and 250 T 350 K).
The method based on formula (3) and the equation of state from [8] and the method based on [10] have a number of
drawbacks. They are both based on equations with interdependent parameters (Z, ), (p, ), which can only be calculated by
an iteration method, which increases the computational load of the processors. The methods require the introduction of data
on the total component composition of the gas, which involves a correction to the software of modern computers, the overwhelming majority of which use simplified data on the gas composition. Hence, it is urgent to develop a simplified procedure
for calculating the JouleThomson coefficient using three parameters of the composition of the natural gas: the densities under
standard conditions, containing nitrogen and carbon dioxide.
We have developed a method of calculating the JouleThomson coefficient in the absolute pressure range of 0.115 MPa
and the temperature range of 250350 K under these conditions. These ranges of variation of the pressure and temperature
cover problems of metering natural gas when it is distributed, transported, and extracted. The pressure dependences of the
JouleThomson coefficient for fixed values of the temperature for such important practical areas are shown in the figure. They
are constructed from values calculated using the equations from [10] for natural gas with a nitrogen content of 0.997% and
a carbon dioxide content of 0.612%, and a density under standard conditions of 0.701 kg/m3.
As follows from the figure, the coefficient Dh has maximum values at low pressures and temperatures. The sensitivity of Dh to a change in temperature increases as the pressure falls. On the other hand, at a high pressure (p > 10 MPa) both
the JouleThomson coefficient and its sensitivity to a change in temperature fall considerably. If we distinguish a region of
values of Dh for a gas pressure of up to 5 MPa, and this is the region in which the overwhelming majority of variable pressure-drop flowmeters are used, then the values of Dh will be close to the maximum in this region. Consequently, it is most
important to take the throttling process into account in this range of variation of the pressure.
The procedure is based on the use of the principle of the correspondence of states. The fundamental equation of the
method describes the dependence of the JouleThomson coefficient on the reduced temperature and density. The coefficients
of the fundamental equation are found by processing masses of values of Dh, obtained from [10], for the pressure range
0.115.0 MPa and the temperature range 250350 K. According to the procedure developed, the JouleThomson coefficient
is calculated from the formula
6

Dh =

aij 5 j 6 i ,

(4)

i =1 j =1

where and are the reduced temperature and density, and aij are coefficients given in Table 1.
511

TABLE 2. Example of the Calculation of the Error sys

Initial data
Absolute gas pressure before the constricting device

p1 = 0.5 MPa

Gas temperature, measured after the constricting device

T2 = 273.15 K
p = 100 kPa

Pressure drop on the diaphragm

Gas density under standard conditions

st = 0.7186 kg/m3

Nitrogen content

xn = 1.181 % mole

Carbon dioxide content

xc = 0.917 % mole
Results of the calculation
Dh = 5.59 K/MPa

JouleThomson coefficient
Relative diameter of the diaphragm

0.2

0.4

0.6

Pressure drop in the constricting device , kPa

95.3

82.5

62.1

Change in gas temperature due to throttling in the diaphragm T = Dh, K

0.53

0.46

0.35

Gas temperature before the diaphragm (according to Section 6.3.c [5]) T1 = T2 + T, K

273.68

273.61

273.5

Relative change in the gas flow rate when the calculated temperature changes by T,

0.100

0.087

0.065

sys = (QT2 QT 1)/QT1, %

Values of the reduced temperature and the reduced density were calculated from the formulas
= T/Tpc,

= /pc,

(5)

where T is the thermodynamic temperature of the natural gas, K; is the density of the natural gas under working conditions,
kg/m3,
= pTst st /(pstTK);
(6)
p is the absolute pressure of the natural gas, MPa; pst, Tst, and st are the pressure, temperature, and density of the natural gas
under standard conditions; and K is the compressibility coefficient, which can be calculated from simplified data on the composition of the gas and its pressure up to 12 MPa using methods NX19 mod., GERG91 mod. [8], and SD 7-2005 [11], and
at pressures higher than 12 MPa only using the method described in [11].
The pseudocritical temperature of the gas in kelvins is calculated from the formula [8]
Tpc = 88.25(0.9915 + 1.759st 1.681xn xc),

(7)

while the pseudocritical density (kg/m3) is calculated from the simplified formula, given in [12]:
pc = 163.5(st /0.6682)0.6 + 62.62xn + 163.359xc,

(8)

where xn and xc are the molar fractions of nitrogen and carbon dioxide.
We checked this method against the theoretical values of the JouleThomson coefficient, obtained using the method
described in [10] for natural gases, whose composition corresponds to the quality certificates of the laboratory in Ukrtransgaz
from different regions. The relative difference between the results obtained does not exceed 1.5% for these natural gases.
A detailed test of the method shows that for natural gases with a density under standard conditions of up to 0.75 kg/m3 and
a nitrogen and carbon dioxide content of up to 5% each, the relative deviation of the values obtained using the proposed
method from the values according to [10] does not exceed 3.0% in the pressure and temperature variation ranges indicated.
512

An example of the calculation on the systematic component of the error sys of the gas flow rate for a natural gas
metering point, operating at a limiting value of the pressure drop of p = 100 kPa and different values of the relative diaphragm
diameter , is shown in Table 2. The parameters of the meter, represented in the example, correspond completely to the requirements of the standards [15], which is confirmed by calculations of the values of the flow rate carried out using the Raskhod-RU
program [13], certified by the Testing Center of the All-Russia Research Institute of Metrological Service (VNIIMS).
As follows from Table 2, the systematic error in measuring the flow rate of a gas due to the reduction in the temperature when the gas flows through a constricting device can reach 0.1% for meters operating in complete correspondence
with the requirements [15] and at pressure losses not exceeding 160 kPa. The elimination of such a considerable systematic
error when measuring gas flow rate undoubtedly requires attention.
We can draw the following conclusions from the above result.
The JouleThomson coefficient can be calculated in problems of natural-gas metering, when the total component
composition of the gas is known, using the proposed formula (3) and the equation of state given in [8] or using the equations
from [10]. The proposed procedure for calculating the JouleThomson coefficient is more convenient for use in gas flow rate
calculators since it enables the calculations to be carried out using simplified data on the gas composition in ranges of variation of the pressure of 0.115.0 MPa and of the temperature of 250350 K.
The use of the procedure described in natural-gas flow-rate meters enables the systematic component of the error in
measuring the flow rate, due to the difference in the gas temperature at the point where the measurement is made after the
constricting device and the temperature before the constricting device, to be reduced.

REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.

State Standard GOST 8.586.1-2005 (ISO 8.586.1-2003), Measurement of the Flow Rate and Quantity of Liquids and
Gases Using Standard Constricting Devices. Pt. 1: Principle of the Method of Measurement and General Requirements.
State Standard GOST 8.5486.2-2005 (ISO 8.586.2-2003), Measurement of the Flow Rate and Quantity of Liquids
and Gases Using Standard Constricting Devices. Pt. 2: Diaphragms: Technical Requirements.
State Standard GOST 8.586.3-2005 (ISO 8.586.3-2003), Measurement of the Flow Rate and Quantity of Liquids and
Gases Using Standard Constricting Devices. Pt. 3: Nozzles: Technical Requirements.
State Standard GOST 8.586.4-2005 (ISO 8.586.4-2003), Measurement of the Flow Rate and Quantity of Liquids and
Gases Using Standard Constricting Devices. Pt. 4: Venturi Tubes: Technical Requirements.
State Standard GOST 8.586.5-2005, Measurement of the Flow Rate and Quantity of Liquids and Gases Using Standard
Constricting Devices. Pt. 5: Measurement Procedure.
ISO 5167-1:2003, Measurement of the Fluid Flow by Means of Pressure Differential Devices Inserted in Circular
Cross-Section Conduits Running Full. Pt. 1: General Principles and Requirements.
B. P. Porshakov et al., Thermodynamics and Heat Transfer (in Technological Processes of the Oil and Gas Industry)
[in Russian], Nedra, Moscow (1987).
State Standard GOST 30319.0:3-96, Natural Gas: Methods of Calculating Physical Properties.
E. E. Shpilrain and P. M. Kesselman, Principles of the Theory of the Thermal Properties of Materials [in Russian],
Energiya, Moscow (1977).
ISO 20765-1:2005, Natural Gas: Calculation of Thermodynamic Properties. Pt. 1: Gas Phase Properties for
Transmission and Distribution Applications.
SD 7-2005, Natural Gas: Procedure for Calculating Compressibility in the 1225 MPa Pressure Range.
DSSDD 8-2006, Natural Gas: Procedure for Calculating the Adiabatic Index in Pressure Intervals from 0.1 to 25 MPa
and at Temperatures from 250 to 320 K.
E. P. Pistun, L. V. Lesovoi, and O. Ya. Masnyak, Commercial Metering of Energy Carriers: Proc. of the 26th Int.
Conf., St. Petersburg (2007).

513