X100 - Girth Welding, Joint Properties and Defect Tolerance
X100 - Girth Welding, Joint Properties and Defect Tolerance
X100 - Girth Welding, Joint Properties and Defect Tolerance
2
= (1)
If there is a defect of fixed size a subject to a stress , by substituting
a Y K
I
=
the toughness can
be expressed as:
E
a Y
Y
=
(2)
If, as is usual, the design factor is kept constant, the required toughness will increase linearly with
the grade of the material. This analysis is simplified, as it ignores factors such as welding residual
stresses, but it does serve to illustrate the challenge when attempting to ensure defect tolerance in a
high strength pipeline. Toughness levels which may have been adequate in a pipeline constructed
from a conventional grade such as X60 or X65 may no longer be adequate in a very high strength
material.
Crack Initiation Resistance
In practice it is usually found that the defect tolerance of pipelines produced from conventional
grade materials is not dependent on the toughness, but is controlled by the plastic collapse
behaviour. As will be noted below, it is not yet established whether this is the case for the very high
strength materials. The most commonly used crack initiation model for pipelines is that originally
due to Kiefner et al.[10]. For a throughwall defect this model takes the general form:
flow T
M = (3)
where the quantities are defined as:
hoop stress, N/mm
2
flow
flow stress, defined below
M
T
is the Folias factor, which accounts for the stress concentrating effect of bulging at the ends of
the crack, and is a function of the diameter, wall thickness and crack length. The models, and
subsequent variations, differ in their definition of the flow stress, and this can have an effect on the
tolerable crack length as the pipe grade increases. Shannon [11] defined the flow stress as 1.15 *
SMYS, and using this definition it can be seen from (10) that the tolerable crack size is independent
of material grade when operating at a constant design factor. A problem with this definition for high
strength steels with relatively limited work hardening capability is that it can result in a flow stress
that exceeds the tensile strength. For example, with X100 (690 N/mm
2
) material the flow stress
would become 793 N/mm
2
, which is above the typical minimum tensile strength of 770N/mm
2
usually assumed for this material. This can be avoided by adopting the definition used in BS 7910
[13], which defines the flow stress as the average of the yield and ultimate strengths.
HOME
Figure 4: Critical axial through wall crack length for a 914 mm diameter 12.7 mm wall thickness
pipeline operating at 80% SMYS
Figure 4 shows the critical through wall crack length using (10) for a typical high strength steel
pipeline application as a function of pipe grade. The crack length is not strongly dependent on
grade. The upper shelf Charpy toughness required for a collapse controlled failure, calculated using
the toughness dependent model in [10], is also shown; the levels exceed 100J for the very high
strength grades. These levels are in fact below those typically achieved by the pipe manufacturers.
There is only a small amount of published data to show whether these models will apply to very
high strength material. Two sets of data with conflicting results are shown in Figure 5; the ring data
from [8] show that the collapse model is satisfactory, whilst data from pipe burst tests from [14]
show that the toughness dependent variant of the model is unconservative. Further work is required
to resolve these issues.
Volumetric Corrosion Defect Resistance
Internal stress corrosion cracking (SCC) is not expected to be a problem for very high strength
linepipe as the expected application is sweet dry gas. Hence the resistance of the materials to
volumetric corrosion is expected to be the most serious integrity issue related to corrosion. Again,
only limited data are available in the public domain for this type of defect. Some of these data are
also shown for metal loss defects in Figure 5; they indicate that existing models based on plastic
collapse [16] are likely to be satisfactory. Further work is required to confirm this conclusion.
Figure 5: Parent metal axial crack tolerance results for X100 materials. Left, data from [8] for
surface breaking cracks; right data from [14] for a through-wall crack
HOME
Girth Weld Defect Tolerance
The curved wide plate test has been widely used for assessing the defect tolerance of pipeline girth
welds. The test uses a 300mm wide axial strip with the girth weld located centrally. This is loaded
in tension to failure, with the parent metal strain inferred from the overall extension and the crack
mouth opening. Figure 6 shows the results obtained in the Demopipe project [5], plotted on axes of
parent metal strain and non-dimensional defect area. All tests were carried out at -20C; all the
welds had achieved Charpy impact energies in excess of 40 J average at temperatures of -40C or
lower. The diamond symbols are the Demopipe results. Two out of the ten points are inside the
EPRG Tier 2 [12] limits. One of these test welds was significantly undermatched, so can be
disregarded, as the EPRG guidelines require an overmatched weld. The other point inside the EPRG
limit failed as a brittle fracture. This was a 25 mm thick SMAW procedure which was
approximately matching and so should be included in the analysis. The failure initiated in the
cellulosic root, which undermatches locally due to the tensile strength of an E6010 consumable.
The stronger cap and fill overmatch the parent metal and should have shielded the root. Further
work is required to address these issues.
0
0.5
1
1.5
2
2.5
0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00
Defect area %
P
i
p
e
s
t
r
a
i
n
a
t
m
a
x
l
o
a
d
%
X100
X80
Tier 2
Undermatched
Brittle fracture
0
0.5
1
1.5
2
2.5
0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00
Defect area %
P
i
p
e
s
t
r
a
i
n
a
t
m
a
x
l
o
a
d
%
X100
X80
Tier 2
Undermatched Undermatched
Brittle fracture Brittle fracture
Figure 6: Curved wide plate test results for X100 [5] and X80 [15] girth welds
For comparison, Figure 6 also shows results on large diameter X80 girth welds from [15]. These
welds were made using automatic GMAW procedures and show comparable performance levels to
the X100 data. Overall, it appears that it will be possible to obtain adequate levels of defect
tolerance in X100 pipeline girth welds, although in the short term project specific testing may be
required to optimise the acceptance criteria for a specific application and design conditions.
As an alternative to the semi-empirical curved wide plate approach, these test results have also been
analysed using the failure assessment diagram approach of BS 7910 [13]. Figure 7, from the final
Demopipe report [5], shows the results from the wide plate tests and also four full scale bending
tests. The analysis used the Level 2A (generic) assessment diagram using specification minimum
tensile properties and the CTOD toughness appropriate to the notch location and weld process. For
the wide plate tests the stresses were estimated from the test records as force over area at point of
failure, while for the full scale bending tests the local stress was estimated from the strain gauge
data and converted into stress using the stress-strain curve. As is commonly found for pipeline
HOME
applications, the fracture mechanics analysis is conservative. All four of the bending tests achieved
100% SMYS so the EPRG Tier 3 criteria [12] were satisfied. Note that a constraint corrected
analysis [9] would expand the collapse region so these might be considered as collapse controlled
rather than mixed collapse-fracture failures.
Figure 7: X100 girth weld tests from the Demopipe project [5] plotted using the BS 7910 failure
assessment diagram.
5 CONCLUSIONS
The numerous welding procedure trials undertaken within the Demopipe project have
enabled the selection of a range of candidate welding consumables and processes for the
construction of X100 pipelines. The final weld metal overmatch required will be subject to
the given pipeline design, but the current paper has shown that electrodes exhibiting
toughness levels in excess of the EPRG guidelines but with weld metal yield strengths not
necessarily overmatching the equivalent pipe longitudinal strength have resulted in burst
tests devoid of problems associated with the girth welds.
As expected, a decrease in toughness (both Charpy V and CTOD) was found with
decreasing temperature, but the change was considered to be relatively small in the range of
-20C to +20C.
CTOD values exhibited a decrease in value with increasing weld metal yield strength; this
was not so prevalent with the impact toughness values.
An increase of pipe wall thickness indicated a general increase in weld metal yield strength
and hardness values when the same SMAW and GMAW consumable/WPS was used,
highlighting the important effect of wall thickness within high strength weld metal
procedures.
The results of the curved wide plate testing undertaken within the Demopipe programme
have in general indicated that it will be possible to obtain adequate levels of defect tolerance
in X100 pipeline girth welds, although in the short term project specific testing may be
required to optimise the acceptance criteria for a specific application and design conditions.
An analysis using the Level 2A assessment diagram of BS7910 incorporating the wide plate
tests and full scale bending tests in conjunction with the specification minimum tensile
properties and CTOD toughness appropriate to the notch location and weld process resulted
HOME
in a conservative result as is commonly found for pipeline applications. All four of the
bending tests achieved 100% SMYS such that the EPRG Tier 3 criteria were satisfied.
REFERENCES
1 A. Glover, Application of Grade 555 (X80) and 690 (X100) in Arctic climates, Proc. of Conf. On Application &
Evaluation of High Grade Linepipe in Hostile Environmnents, Yokohama, (2002) p33
2 L. Barsanti et al., From X80 to X100: Know-how reached by ENI Group on High Strength Steel, Proc. of Conf. On
Application & Evaluation of High Grade Linepipe in Hostile Environmnents, Yokohama, (2002) pp231-244
3 N. Millwood, N Sanderson and J. Hammond, Design and Construction in Ultra-High Strength Linepipe, Pipes and
Pipelines International, March-April 2001, pp17-22
4 D. Fairchild et al., High Strength Steels Beyond X80, Proc. of Conf. On Application & Evaluation of High Grade
Linepipe in Hostile Environments, Yokohama, (2002) pp 307-321
5 G. Demofonti et al., Ultra high strength pipeline prototyping for natural gas transmission DEMOPIPE project
final report, Centro Sviluppo Materiali Report No. 11880R, (2004)
6 M.G.Hudson, Welding of X100 Linepipe, PhD Thesis, Cranfield University, Cranfield, Beds, (2004)
7 JIS Z 3158, Method of Y-Groove Weld Cracking Test, (1993)
8 B. Fu, N.A. Millwood, D.Q. Vu, S. Matthews, M. Jacobs, Failure behaviour of ultra high strength line pipe, In:
Pipeline Technology Proceedings of the third international pipeline technology conference Volume I, Brugge,
Belgium, May 21-24 2000. Denys RM. Amsterdam: Elsevier Scientific, (2001), pp 497-507
9 R.M. Andrews, G.C. Morgan, W.J. Beattie, The significance of low toughness areas in the seam weld of linepipe,
Paper IPC04-0422. In: Proceedings of 2004 International Pipeline Conference. October 4-8, 2004. Calgary,
Alberta, Canada. New York: American Society of Mechanical Engineers; (2004).
10 W.A. Maxey, J.F. Kiefner, R.J. Eiber, A.R. Duffy, Ductile fracture initiation, propagation and arrest in cylindrical
vessels. In: Proceedings of the 1971 National Symposium on Fracture Mechanics, Part II, ASTM STP 514.
Philadelphia: American Society for Testing and Materials; (1972), pp 70-81
11 R.W.E. Shannon, The failure behaviour of linepipe defects. International Journal of Pressure Vessels and Piping
(1974), 2:243-55
12 G. Knauf, P. Hopkins, The EPRG guidelines on the assessment of defects in transmission pipeline girth welds. 3R
International 1996;35 (10/11):pp 620-624.
13 BS 7910:1999. Guide to methods for assessing the acceptability of flaws in metallic structures Incorporating
Amendment 1, April 2000. London: British Standards Institution; (2000)
14 S. Kawaguchi, N. Hagiwara, M. Ohata, M. Toyoda, Modified equation to predict leak/rupture criteria for axially
throughwall notched x80 and x100 linepipes having higher charpy energy. Paper IPC04-0332. In: Proceedings of
2004 International Pipeline Conference October 4-8, 2004 Calgary, Alberta, Canada. New York: American Society
of Mechanical Engineers; (2004)
15 R.M. Andrews and L.L. Morgan, Integration of automated ultrasonic testing and engineering critical assessment
for pipeline girth weld defect acceptance. In Fourth International Conference on Pipeline Technology, Ostende,
May 9-13 2004. Denys R. Beaconsfield: Scientific Surveys,. (2004) 2, pp 655-667
16 B. Fu and A.D. Batte, An overview of advanced methods for the assessment of corrosion in linepipe, HSE
Offshore Technology Report OTO 1999 051. London: HSE Books; (1999)
HOME