You are on page 1of 8

1 Copyright 2012 by ASME

Proceedings of the 2012 9th International Pipeline Conference


IPC2012
September 24-28, 2012, Calgary, Alberta, Canada
IPC2012-90345
INVESTIGATION OF THE EFFECTS OF PIPE WALL ROUGHNESS AND PIPE
DIAMETER ON THE DECOMPRESSION WAVE SPEED IN NATURAL GAS
PIPELINES


C. Lu, G. Michal, A. Elshahomi, A. Godbole
University of Wollongong
Wollongong, NSW, Australia
P. Venton
Venton & Associates Pty Ltd
Bundanoon, NSW, Australia


K. K. Botros
NOVA Research and
Technology Centre
Calgary, AB, Canada
L. Fletcher
Welding and Pipeline Integrity
Bright, Victoria, Australia
B. Rothwell
Brian Rothwell Consulting Inc.
Calgary, AB, Canada


ABSTRACT
The shock tube experimental results have shown clearly
that the decompression wave was slowed down in a pipe with a
rough inner surface relative to that in a smooth pipe under
comparable conditions. In the present paper a one-dimensional
dynamic simulation model, named EPDECOM, was developed
to investigate the effects of pipe wall roughness and pipe
diameter on the decompression wave speed. Comparison with
experimental results showed that the inclusion of frictional
effects led to a better prediction than that of the widely used
model implemented in GASDECOM. EPDECOM simulation
results showed that the effect of roughness on the
decompression wave speed is significant for pipe diameters less
than 250 mm. However the decompression wave speed is
nearly independent of the roughness for diameters above 250
mm as the frictional effect becomes negligible at such
diameters.

INTRODUCTION
Fracture propagation is a significant issue for pipelines
transporting gases and the need to arrest a running fracture in a
pipeline is paramount to the integrity and safety of the
pipelines operation. The Battelle Two-Curve Model (BTCM)
is the most commonly used approach for the prediction of the
minimum linepipe toughness (arrest toughness) required to
arrest a running fracture [1]. The BTCM involves the
superposition of two curves: the gas decompression wave speed
characteristic and the fracture propagation speed characteristic,
each being a function of the local gas pressure. The limit
between arrest and propagation of a running fracture is
identified by a condition of tangency between the
decompression curve and the fracture curve. The toughness
necessary to reach this condition is the arrest toughness.
A number of full-scale burst tests were conducted to
investigate pipeline gas decompression since the 1970s [2-8].
However, full scale burst tests are prohibitively costly, except
for major projects. Botros et al. have experimentally
investigated the decompression wave speed using a small
diameter NPS 2 shock tube [9, 10]. Phillips and Robinson have
reported decompression wave speed measurements using an
NPS 6 shock tube [11]. In both full scale burst tests and shock
tube tests, the decompression wave speed can be determined
from pressure-time traces measured by transducers mounted
along the pipe. For any pressure level below the initial
pressure, the time of arrival of the decompression wave at each
pressure transducer can be determined, and the corresponding
propagation speed W is calculated using a linear fit of the
traveling distance of the isobars against time. Such calculations
are repeated for progressively lower pressures. The results are
presented in terms of the decompression wave speed W as a
function of the pressure P.
Many models were developed to predict the
decompression wave speed [12]. GASDECOM is the most
widely used model. In GASDECOM, the decompression wave
speed (W) is calculated by
W=C u
(1)
2 Copyright 2012 by ASME
where C is the speed of sound behind the decompression wave
and u is the mean outflow speed behind the decompression
wave.
Increments in pressure and density are used to calculate C
and u. The outflow speed u at any given pressure is the sum of
incremental u determined from
u=

( u)
s
(2)
where

( u)
s
=C
( )
s

(3)
where is the mass density and the subscript s indicates a
transformation along an isentrope. In GASDECOM the
Benedict-Webb-Rubin-Starling (BWRS) Equation of State
(EOS) is used to calculate the thermodynamics properties
(speed of sound and density) of the gas.
Several models have followed the approach of
GASDECOM. A detailed review was given in Ref. [11]. The
Advantica model uses the cubic London Research Station
(LRS) EOS [13], which is similar to the Redlich-Kwong-Soave
(RKS) EOS. The results in the Advantica model are calibrated
through a scaling factor applied at the initial conditions. The
decompression model developed by Groves et al. adopts the
RKS EOS to determine the thermodynamic properties [14]. The
PipeDecom model allows the effects of non-equilibrium
thermodynamics to be represented in the calculation. Delayed
liquid droplet formation is included in the wave velocity
prediction through manual adjustments of the temperature of
nucleation [11]. The approach taken in the model developed by
Makino et al. [15] follows the approach of GASDECOM to
model single phase decompressions in pipes undergoing
fracture propagation. For two-phase conditions, the mass,
momentum and energy conservation equations are used to
calculate the velocity. All properties are calculated using the
BWRS EOS.
Recent experimental campaigns pointed out limitations in
current decompression wave speed models:
- The shock-tube tests conducted by Botros et al. [9]
showed that the decompression wave speed decreases as
the non-dimensional wall roughness (c/D) increases.
GASDECOM-type models do not include the effects of
wall roughness and pipe diameter on the decompression
wave speed.
- Botros [16] compared the densities in the dense phase
region as predicted by five EOSs (GERG, AGA-8,
BWRS, PR and RKS) with experimental data for
different hydrocarbon mixtures. It was found that the
GERG EOS outperforms all other equations in the
region up to P = 30 MPa and T > -8
o
C. However, GERG
has not been implemented in the commonly used
decompression wave speed models.
The present study aims at developing a new decompression
wave speed model that takes into account the effects of wall
roughness and pipe diameter using the GERG-2008 EOS.
SHOCK TUBE EXPERIMENT
Shock tube tests were carried out at the TransCanada
Pipeline Gas Dynamics Test Facility in Didsbury, Alberta,
Canada [9]. The tests were designed to understand and quantify
the effects of pipe diameter and wall friction on the
decompression wave speed.
The shock tube consisted of four spool pieces of NPS 2
stainless steel pipe making up a total length of 42 m. They
were made of NPS 2 x 11.1 mm WT, SCH XX, ASTM A312,
316 SS seamless tube (I.D.= 38.1 mm). All individual spools
were designed for 41.37 MPa pressure with a design factor of
0.8 and a location factor of 0.625. All spool pieces were
internally honed to a roughness RZ less than 0.635 m, except
for the rough tube tests, for which only Spool #1 was replaced
with one having internal surface roughness RZ = 3.81 m. A
total of 16 Endevco dynamic pressure transducers were
mounted along the length of the shock tube, 13 of which were
mounted on the front spool where the rupture disc was located.
Table 1 Shock tube test conditions and gas composition
Reference Test
(Lean gas)
Test 4
(Rich gas)
P
i
(MPa) 18.438 19.837
T
i
(
o
C) 17 11.37
C1 (mole%) 97.2754 82.3030
C2 (mole%) 1.43651 6.86748
C3 (mole%) 0.26693 7.13351
iC4 (mole%) 0.03208 0.75241
nC4 (mole%) 0.04322 0.98587
iC5 (mole%) 0.01015 0.00668
nC5 (mole%) 0.00749 0.00475
C6+ (mole%) 0.00919 0.0064
N
2
(mole%) 0.55554 0.47430
CO
2
(mole%) 0.363498 1.465631
The test program consisted of a total of 8 tests conducted
with various gas compositions representative of conventional
natural gas mixture (Reference Test) and three other medium
rich, rich and ultra rich mixtures. These 8 rupture tests were
conducted using the smooth shock tube, and then repeated
using the rough tube for the same nominal gas compositions
and initial pressures and temperatures. The detailed
experimental conditions and results have been given in Ref.
[9]. The shock tube experimental results have shown clearly
that the decompression wave was slowed down in a shock tube
with a rough inner surface relative to that in a smooth tube
under comparable conditions. Table 1 lists the condition and
the gas composition for two typical tests (Reference Test for
lean gas and Test 4 for rich gas) using the rough tube.
DECOMPRESSION WAVE SPEED MODEL
A new decompression wave speed model, named
EPDECOM, has been developed in the present study. This
model solves the following one-dimensional dynamic
differential equation (Eqn. (4)) using the finite difference
method. Eqn. (4) is derived assuming that the outflow speed
3 Copyright 2012 by ASME
and pressure are constant along the pipe radial direction. The
decompression is supposed to be isentropic.
P
t
+( C u)
P
x

Cf
2D
u
2
=0
(4)
where P is the pressure, C is the speed of sound, u is the
outflow speed, t is the time, x is the distance from the rupture,
is the density, f is the Darcy friction factor and D is the pipe
diameter. The friction factor is currently calculated by Eqns. (5)
and (6).
2
10
) ( log 8 . 1
1
|
|
.
|

\
|

=
A
f (5)
11 . 1
7 . 3
/ 9 . 6
|
.
|

\
|
+ =
D
Re
A
c
(6)
where Re is the Reynolds number and c is the pipe wall
roughness.
In addition to Eqn. (4), the change of outflow speed across
the decompression wave is obtained from Eqn. (7). The speed
of sound C, the density and the temperature T are obtained
from the GERG-2008 EOS using Eqns. (8), (9) and (10)
respectively. The thermodynamic quantities are expressed as
functions of the pressure P and the entropy s.

C = u
c
c (7)
s
|

P
= C
c
c
(8)
=( P,s)
(9)
T=T( P,s)
(10)
GERG-2008 EOS [17] covers the gas phase, the liquid
phase, the supercritical region, and vapour-liquid equilibrium
states for natural gases and other mixtures consisting of the 21
components methane, nitrogen, carbon dioxide, ethane,
propane, n-butane, isobutane, n-pentane, isopentane, n-hexane,
n-heptane, n-octane, hydrogen, oxygen, carbon monoxide,
water, helium, argon, n-nonane, n-decane, and hydrogen
sulphide. The normal range of validity of GERG-2008 covers
temperatures from 90 K to 450 K and pressures up to 35 MPa.
The uncertainty of GERG-2008 in estimating the gas phase
density and speed of sound is less than 0.1% for temperatures
ranging from 270 K to 450 K at pressures below 35 MPa. In
the liquid phase, the uncertainty on the density amounts for less
than 0.1 to 0.5% for many binary and multi-component
mixtures. The estimated uncertainty in liquid phase (isobaric)
enthalpy differences is less than 1%, down to 0.5% for some
mixtures. The vapour-liquid equilibrium is described with
reasonable accuracy. Accurate vapour pressure data for binary
and ternary mixtures consisting of the natural gas main
components are reproduced by GERG-2008 to within their
experimental uncertainty, which is approximately 1 to 3%.
Based on the fluid state (P, T, , C, u,) at (t, x), a finite
difference scheme of Eqn. 4 was used to evaluate the pressure
field forward in time. Following a flash point calculation of the
density for the updated pressures, the velocity of the flow was
obtained using Eqn. (7). The procedure was implemented in
Matlab with the integration of subroutines to call the GERG
EOS.

0 1 2 3 4 5 6
0
4
8
12
16
20


PT1
PT1A
PT1B
PT2
PT3
P
r
e
s
s
u
r
e
,

M
P
a
Time, ms

(a) Reference Test
0 1 2 3 4 5 6
0
4
8
12
16
20
PT1
PT1A
PT1B
PT2
PT3

P
r
e
s
s
u
r
e
,

M
P
a
Time, ms

(b) Test 4
Fig. 1 Simulated pressure-time traces
Fig. 1 shows the simulated time dependent pressure
profiles observed by five pressure transducers. The location of
the transducers from the full bore opening is: PT1 = 29.5 mm,
PT1A = 92.4 mm, PT1B = 102.8 mm, PT2 = 200 mm and PT3
= 350 mm. The pressure at each location drops rapidly once the
initial decompression wave reaches each transducer. For the
lean gas (Reference Test), the pressure gradually reaches a
steady value after the initial drop. However, for the rich gas
(Test 4), the pressure goes through a plateau after the initial
drop, before decreasing further to finally reach a steady
pressure.
To calculate the decompression wave speed (W), the time at
which a given pressure level reaches each transducers is first
determined. By presenting these locations against time, the
decompression wave speed is obtained by carrying out a linear
regression of each isobar curves. The slope of each regression
is the decompression wave speed for each isobar.
4 Copyright 2012 by ASME
0 100 200 300 400 500 600
0
4
8
12
16
20
P
r
e
s
s
u
r
e
,

M
P
a
Experiment
EPDECOM
GASDECOM
Decompression wave speed, m/s

(a) Reference Test
0 100 200 300 400 500 600
0
4
8
12
16
20
P
r
e
s
s
u
r
e
,

M
P
a
Experiment
EPDECOM
GASDECOM
Decompression wave speed, m/s

(b) Test 4
Fig. 2 Comparisons of simulation results and experimental
results for Reference test and Test 4
RESULTS
Fig. 2 compares the measured and simulated
decompression wave speeds for the Reference Test and Test 4.
The results from GASDECOM are also presented. It can be
seen that both EPDECOM and GASDECOM lead to accurate
results near the initial conditions. As the pressure decreases in
Fig. 2(a), the predicted decompression wave speed increasingly
differs between the two models. Compared to GASDECOM,
the prediction from EPDECOM matches more closely the
experimental observations. In Fig. 2(b), it should be noted that
the models converge towards the end of the pressure plateau
but then diverge again at the end of the decompression.
There are still gaps between EPDECOM simulation results
and the experimental results, especially for pressures around
the plateau. This may be partly due to an inappropriate Darcy
friction factor. The Darcy factor (f) calculated by Eqn. (5) was
increased fourfold to obtain the results presented in Fig. 3. The
enlarged friction factor improves the prediction. This may
indicate that friction is more important than expected and that
the frictional model needs to be improved in future research.
Other factors responsible for gaps between simulation results
and experimental results can include delayed nucleation and
inaccurate temperature calculation due to thermal exchanges of
the gas with the pipe wall [11].
0 100 200 300 400 500 600
0
5
10
15
20
P
r
e
s
s
u
r
e
,

M
P
a
Decompression wave speed, m/s

Experiment
EPDECOM
EPDECOM with an
enlarged friction factor

Fig. 3 Comparison of experiment, EPDECOM and EPDECOM
with an enlarged friction factor for Test 4
Fig. 4 shows the pressure vs. decompression wave speed
(P-W) curves for D = 38.1 mm and for several levels of
roughness from 0.635 m to 30 m. The gas composition is the
same as for the Reference Test. The insert at the left-top corner
of the figure gives details for lower pressures. Similar to the
experimental observation, the P-W curve moves upward when
the roughness increases. If this decompression speed
characteristic is used to calculate the arrest toughness in a 457
mm OD, X70 pipeline with a design factor of 0.72, the arrest
toughness would change from 47.4 J to 52.1 J for a change of
roughness from 0.635 m to 30 m (see the insert in Fig. 4).
Fig. 5 shows the effect of roughness on the decompression
wave speed for D = 250 mm. The calculation conditions in Fig.
5 are the same as Fig. 4 except for the pipe diameter. Two P-W
curves with = 3.81 m and = 30 m are compared. The
variation between the P-W curves from Fig. 5 is very small
compared to the results shown in Fig. 4. The arrest toughness
in the conditions of Fig. 5 was calculated for a 250 mm OD,
X70 pipeline with a design factor of 0.72. For roughness of
3.81 m and 30 m, the resulting arrest toughness equals to
45.2 J and 45.4 J respectively.
The results from Fig. 4 and 5 indicate that the effect of
roughness on the decompression wave speed depends on the
pipe diameter as observed in [11]. For pipe diameters less than
250 mm, the effect of roughness on the decompression wave
speed is significant. This effect becomes negligible for pipe
diameters above 250 mm.
5 Copyright 2012 by ASME
0 100 200 300 400 500 600
0
4
8
12
16
20
Arrest toughness:
c=0.635m: 47.4J
c=3.81m: 48.8J
c=10m: 49.4J
c=30m: 52.1J
0 50 100 150 200
5
6
7
8




c=0.635m
c=3.81m
c=10m
c=30m
c

in
c
r
e
a
s
e
s
P
r
e
s
s
u
r
e
,

M
P
a
Decompression wave speed, m/s

Fig. 4 Effect of roughness on decompression wave speed for
D=38.1 mm and the same gas composition as Reference Test
0 100 200 300 400 500 600
0
4
8
12
16
20
40 60 80 100
5.6
6.0
6.4
6.8


c in
c
re
a
s
e
s
Arrest toughness:
c=3.81m: 45.2J
c=30m: 45.4J


P
r
e
s
s
u
r
e
,

M
P
a
c=3.81m
c=30m
Average decompression wave speed, m/s

Fig. 5 Effect of roughness on decompression wave speed for D
= 250 mm and the same gas composition as Reference Test
Fig. 6 shows the pressure vs. decompression wave speed
(P-W) curves for several levels of roughness ranging from
0.635 m to 30 m. All the other conditions are equal to the
conditions in Test 4 (Rich gas). The pipe diameter used in the
simulation was equal to 38.1 mm. Fig. 7 shows the results for a
250 mm diameter pipe. The insert at the left-top corner of the
figure gives details for lower pressures.
Similar to the experimental observation, the P-W curve
moves upward when the roughness increases. The
decompression speed characteristics in Figs. 6 and 7 were used
to calculate the arrest toughness in a 457 mm OD, X70 pipeline
with a design factor of 0.72. The corresponding arrest
toughness is provided in the figures. For a 38.1 mm diameter
pipe, an increase of roughness from 3.81 m to 30 m leads to
an increase of the arrest toughness by 5.29% from 107.7 J to
113.4 J. For a 250 mm diameter pipe an equal change of
roughness leads to a variation of only 0.3% of the arrest
toughness.
0 100 200 300 400 500 600
0
4
8
12
16
20
c=0.635m
c=3.81m
c=10m
c=30m
0 40 80 120 160 200
7.8
8.1
8.4
8.7
9.0



Arrest toughness:
c=0.635m: 103.4J
c=3.81m: 107.7J
c=10m: 109.4J
c=30m: 113.4J
c

in
c
r
e
a
s
e
s
P
r
e
s
s
u
r
e
,

M
P
a
Decompression wave speed, m/s

Fig. 6 Effect of roughness on decompression wave speed for D
= 38.1 mm and the same gas composition as Test 4
0 200 400 600
0
4
8
12
16
20
0 20 40 60
5.6
6.0
6.4
6.8
7.2
7.6


c in
c
re
a
s
e
s


c=3.81m
c=30m
P
r
e
s
s
u
r
e
,

M
P
a
Arrest toughness:
c=3.81m: 101.5J
c=30m: 101.8J
Decompression wave speed, m/s

Fig. 7 Effect of roughness on decompression wave speed for D
= 250 mm and the same gas composition as Test 4
The results from the simulation of a rich gas confirm the
conclusions drawn from the simulation of a lean gas: the effect
of roughness on the decompression wave speed depends on
pipe diameter.
Fig. 8 shows the effect of pipe diameter (D) on the
decompression wave speed, all the other conditions being equal
to the conditions of the Reference Test (Lean gas). The insert
gives the details of the P-W curves for lower pressures. In the
low pressure region the characteristic changes significantly
with the increase of the pipe diameter from 38.1 mm to 250
mm. This strong change at low pressures is non linear and
asymptotic as shown by the decrease of the variation between
the P-W curves at diameters of 250 mm, 500 mm and 1000
mm. The variation between these last two diameters is
insignificant.

6 Copyright 2012 by ASME
0 100 200 300 400 500 600
0
4
8
12
16
20
0 30 60 90 120 150
4.5
5.0
5.5
6.0
6.5
7.0



D=38.1mm
D=250mm
D=500mm
D=1000mm
P
r
e
s
s
u
r
e
,

M
P
a
D

in
c
r
e
a
s
e
s
Decompression wave speed, m/s

Fig. 8 Effect of pipe diameter on decompression wave speed
for = 3.81 m and same gas composition as Reference Test
0 100 200 300 400 500 600
0
4
8
12
16
20
0 30 60 90 120 150
6
7
8
9




D=38.1mm
D=250mm
D=500mm
D=1000mm
P
r
e
s
s
u
r
e
,

M
P
a
Average decompression wave speed, m/s
D

i
n
c
r
e
a
s
e
s

Fig. 9 Effect of pipe diameter on decompression wave speed
for =3.81 m and the same gas composition as Test 4
Fig. 9 shows the effect of pipe diameter on the
decompression wave speed for the conditions of Test 4 (Rich
gas). Even though the gas composition and the initial pressure
in Fig. 9 are different to those in Fig. 8, equivalent observation
are made, namely the pipe diameter only plays a noticeable
role for pipes having a diameter less than 250 mm.
The analysis above indicates that the pipe diameter affects
significantly the decompression wave speed and arrest
toughness for small diameter pipes (roughly D < 250 mm),
whereas it can be neglected for pipes diameters above 250 mm.
This conclusion appears to hold for both lean and rich gases.
DISCUSSION
The local decompression wave speed can be calculated
from Eqn. (11), namely [9]
( )
2
/ 2
u
x P D
Cf
u C
x P
t P
W
c c
=
c c
c c
=

(11)
The third term in Eqn. (11) accounts for the effects of the
roughness and the pipe diameter on the decompression wave
speed due to the presence of the Darcy factor f and the pipe
diameter D. In addition, the change of roughness or pipe
diameter influences the pressure P, the density , the outflow
speed u and the speed of sound C. This influence on the flow
variables affects indirectly the decompression wave speed W.
Eqn. (11) can be rewritten as:
2
Ku u C W o = (12)
D
f
= o (13)
( ) x P
C
K
c c
=
/ 2

(14)
o represents the direct effect of the roughness and the pipe
diameter on the decompression wave speed while K represents
the indirect effect. Detailed calculations using EPDECOM
indicated that o predominantly affects the decompression wave
speed. Therefore, it is assumed in the following analysis that, at
the first order, K is independent of the roughness and the pipe
diameter.
Eqns. (5), (6) and (13) show that o is a non linear function
of the roughness and the pipe diameter. In Fig. 10, o is shown
as a function of D for four different roughness values using a
Reynolds number Re=10
8
. The magnitude of o is large at very
small pipe diameters but it decreases rapidly as the pipe
diameter increases to ~100 mm, indicating that the pipe
diameter affects significantly the decompression wave speed
for small diameters. As the diameter increases further, o
decreases to zero asymptotically. In the limit of an infinite
diameter the effect of friction on the decompression speed
wave becomes null and the standard model W = C - u holds
strictly. For a pipe diameter greater than 250 mm, o becomes
considerably small for the simulated four roughness cases as
shown in Fig. 10. This analysis explains why, in EPDECOM,
the effects of roughness and pipe diameter on the
decompression wave are small for pipe diameters above 250
mm.
0 200 400 600 800 1000
0
5
10
15
20
25


c=0.635m
c=3.81m
c=10m
c=30m
R
a
t
i
o

o
=
f
/
D
Pipe diameter, mm
D>250mm

Fig. 10 Direct influence factor as a function of pipe diameter
for four different roughness values.
7 Copyright 2012 by ASME

If the Reynolds number is assumed to be constant during
the decompression then o is independent of time and it can be
seen as a design variable. K, on the contrary, is a function
depending on the thermodynamics of the fluid due to the
product C as well as dependent on the pressure gradient inside
the pipe (cP/cx).
Since the pressure gradient is large at the beginning of the
decompression then (i) K is qualitatively small and (ii) the
property of the gas (lean or rich for example) will not play a
significant role. Inversely, in the limit of a uniform pressure in
the pipe, representative of the conditions towards the end of the
decompression, K can be very large due to a vanishingly small
pressure gradient. The sensitivity and the potential singularity
of the model at low pressure gradients is an indication that (i)
the pressure gradient needs to be accurately predicted and that
(ii) the model may lack the description of a mechanism that
physically balances the effect of K at low pressure gradients.
Further investigations are necessary to clarify the sensitivity of
Eqn. (11) and its use to safely predict the decompression
characteristic.

CONCLUSIONS
The conclusions of the present study include:
(a) A one-dimensional dynamic simulation model, named
EPDECOM, was developed to investigate the effects
of wall roughness and pipe diameter. Compared to the
shock tube experimental results, EPDECOM improves
the accuracy of the prediction compared to the
commonly used model GASDECOM for the simulated
mixtures.
(b) Parametric studies using EPDECOM show that an
increase of the pipe wall roughness decreases the
decompression wave speed and increases the arrest
toughness. The effect of roughness is significant for
small diameter pipes (roughly D < 250 mm), while
this effect is negligible for pipes with D > 250 mm.
This shows that the evaluation of the potential effect
on arrest toughness in DN450 diameter pipe presented
from the shock tube test results is conservative.
(c) It has been found that the decompression wave speed
is nearly independent on pipe diameter for pipes with
D > 250 mm for the tested mixtures.
(d) The influence of friction in the calculation of the
decompression speed is function of the pressure
distribution. Investigations on the applicability of the
model at low pressure gradient are necessary as
calculations can approach a singularity near a uniform
pressure distribution.
ACKNOWLEDGMENTS
This work was funded by the Energy Pipelines CRC,
supported through the Australian Governments Cooperative
Research Centre Program. The funding and in-kind support
from the APIA RSC is gratefully acknowledged. Authors also
appreciate PRCI, TransCanada Pipeline and Alliance Pipeline
for their supports to the shock tube experiments.
REFERENCES
[1] A. B. Rothwell, Fracture Propagation Control for Gas
Pipelines Past, Present and Future, Pipeline Technology, Vol.
1, 2000, 387-405
[2] W. A. Maxey, Fracture Initiation, Propagation and Arrest.
5th

Symposium on Line Pipe Research, p. J-1. PRCI Catalogue
No. L30174. 1974,
[3] R. J. Eiber, T. A. Bubenik and W. A. Maxey, Fracture
Control for Natural Gas Pipelines. PRCI Catalogue No.
L51691, 1993.
[4] S. Igi, S. Kawaguchi and N. Suzuki, Running Ductile
Fracture Analysis for X80 Pipeline in JGA Burst Tests, Pipeline
Technology Conference, Ostend, 12-14 October 2009, Paper
no: Ostend2009-068
[5] R. M. Andrews and N. A. Millwood, The Fracture Arrest
Behaviour Of 914 mm Diameter X100 Grade Steel Pipelines,
Proceedings of IPC 2004 International Pipeline Conference
October 4 - 8, 2004 Calgary, Alberta, Canada, IPC04-0596
[6] H. Makino, I. Takeuchi and R. Higuchi, Fracture
Propagation And Arrest In High-Pressure Gas Transmission
Pipeline By Ultra High Strength Line Pipes, Proceedings of
IPC2008, 7th International Pipeline Conference September 29-
October 3, 2008, Calgary, Alberta, Canada, IPC2008-64078
[7] S. Kawaguchi, K. Murai, Y. Hashimoto et al., Full-Scale
Burst Tests Of Ultra-High Pressured Rich-Gas Pipelines Under
Buried And Unburied Conditions, Proceedings of IPC2008, 7th
International Pipeline Conference September 29-October 3,
2008, Calgary, Alberta, Canada, IPC2008-64434
[8] H. Makino, I. Takeuchi, and R. Higuchi, Fracture
arrestability of high pressure gas transmission pipelines by high
strength linepipes, Pipeline Technology Conference, Ostend,
12-14 October 2009, Paper no: Ostend2009-008
[9] K. K. Botros, J. Geerligs, L. Fletcher, B. Rothwell, P.
Venton, L. Carlson, Effects of Pipe Internal Surface Roughness
on Decompression Wave Speed in Natural Gas Mixtures,
Proceedings of IPC 2010 8th International Pipeline Conference
September 27-October 1, 2010, Calgary, Alberta, Canada,
IPC2010-31667
[10] K. K. Botros, J. Geerligs and R. J. Eiber, Measurement of
Decompression Wave Speeds in Rich Gas Mixtures at High
pressures (370 bar) Using Specialized Rupture Tube, ASME
Journal of Pressure Vessel Technology, vol. 132 , p. 051303-1,
October, 2010.
[11] A. G. Phillips and C. G. Robinson, Gas Decompression
Behavior following the Rupture of High Pressure Pipelines
Phase 2: Modeling Report, PRCI Contract PR-273-0135,
November 2005.
[12] A. G. Phillips and C. G. Robinson, Gas Decompression
Behavior following the Rupture of High Pressure Pipelines
Phase 1, PRCI Contract PR-273-0135, November 2002.
[13] D. M. Johnson, P. S. Cumber, N. Horner, L. Carlson and
R. Eiber, Full-Scale Fracture Propagation Experiments: A
Recent Application and Future Use for the Pipeline Industry,
Proceedings of the ASME International Pipeline Conference,
Calgary, 1-5 October 2000.
[14] T. K. Groves, P. R. Bishnoi and J. M. E. Wallbridge,
Decompression Wave Velocities in Natural Gases in Pipe Lines,
8 Copyright 2012 by ASME
The Canadian Journal of Chemical Engineering,Vol. 56, pp 664
- 668, December 1978.
[15] H. Makino, T. Sugie, H. Watanabe, T. Kubo, T. Shiwaku,
S. Endo, T. Inoue, Y. Kawaguchi, Y. Matsumoto and S.
Machida, Natural Gas Decompression Behavior in High
Pressure Pipelines, ISIJ International, Vol. 41, pp 389 - 395,
2001
[16] K. K. Botros, Performance of Five Equations of State for
the Prediction of VLE and Densities of Natural Gas Mixtures in
the Dense Phase Region, Chem. Eng. Comm., 2002, Vol.
189(2), 151.
[17] W. Wagner, Description of the Software Package for the
Calculation of Thermodynamic Properties from the GERG-
2004 XT08 Wide-Range Equation of State for Natural Gases
and Other Mixtures, April 2009