Proceedings of the 2012 9th International Pipeline Conference IPC2012 September 24-28, 2012, Calgary, Alberta, Canada

IPC2012-90660

MULTI-TIER TENSILE STRAIN MODELS FOR STRAIN-BASED DESIGN PART 3 MODEL EVALUATION AGAINST EXPERIMENTAL DATA
Ming Liu and Yong-Yi Wang Center for Reliable Energy Systems Dublin, OH USA David Horsley BP Calgary, Alberta, Canada Steve Nanney PHMSA, US DOT Houston, TX, USA

ABSTRACT This is the third paper in a three-paper series related to the development of tensile strain models. The fundamental basis [1] and formulation [2] of the models are presented in two companion papers. This paper covers the evaluation of the models against large-scale experimental data which include a total of 24 full-scale pipe tests with and without internal pressure [3,4] and 30 curved wide plate (CWP) tests [5,6]. The 24 full-scale pipe specimens are nominally X65 grade (12.75 OD and 12.7-mm wall thickness) and made by two manufacturers. The actual yield strength of the two pipes differs by approximately 14 ksi. The girth welds are made with three welding procedures, creating three weld strength levels. The full-scale test program are designed to evaluate the effects of internal pressure, weld strength mismatch, pipe strength, pipe Y/T ratio, flaw location, flaw size, and toughness The 30 CWP specimens are from 36" OD and 19.1 mm wall thickness X100 pipes. The girth welds are made with two welding procedures, creating two slightly different weld strength mismatch levels. The CWP test specimens expand the range of material grade and wall thickness for the model evaluation. The model evaluation demonstrates that the overall correlations between the experimental test data and model predations are similar when the model predictions are made with Level 2 and 3 procedures and various toughness options. The Level 2 procedure with Charpy energy option and Level 3b provide the best overall one-to-one correlation between the test data and model prediction. The Level 3b shows greater scatter than Level 2 with the Charpy energy option. The most significant contributor to the TSC variations and the difference between the measured and predicted TSCs is the strength variation in the pipes. A small variation in the strength can lead

to a large variation of the measured remote strain even when the flaw behavior is essentially the same. For the 24 full-scale pipe tests, a strength variation of 1 ksi in the pipes would explain the large variations of the measured TSC in comparison to the model predictions. The TSC models produce consistent results that capture the overall trend of the test data. KEYWORDS Strain-based design, tensile strain capacity, full-scale test, curved wide plate test, TSC model evaluation INTRODUCTION The fundamental basis and the multi-tier tensile strain models are described in two companion papers [1,2]. This paper first covers the background of experimental tests and model development. The experimental test matrices are then introduced. A few key features of the test data are shown to facilitate the understanding of the nature of the test data. A major part of the paper is then devoted to the comparison of the model prediction and experimental test data. The determination of the various toughness representations is described. The major conclusions from the model evaluation are given at the end of the paper. Tensile strain capacity (TSC) can be affected by many factors, including internal pressure [7,8,9], girth weld high-low misalignment [10,11,12], HAZ softening [13], and weld strength mismatch. The influence of these factors was systematically investigated in a recently completed project funded by the US DOT and PRCI [3,4]. The tensile strain models incorporating the effects of these factors have been developed and evaluated against full-scale test data [4]. These models are a significant extension of the approach developed in prior PRCI projects [14,15,16,17,18,19] which form the basis of the tensile strain design procedure in the Annex C of CSA Z662 2007 Edition [20].

1 Copyright 2012 by ASME This work is in part a work of the U.S. Government. ASME disclaims all interest in the U.S. Governments contributions.

TEST MATRIX Full-Scale Pipe Test Matrix Twenty four (24) full-scale pipe tests were completed in a DOT and PRCI sponsored project (i.e., PRCI ABD-1) [3,4]. The test matrix and results are shown in Table 1. The pipe and weld properties are summarized in Table 2. Two different pipes, namely X65 high and low Y/T pipes, were tested. It should be noted the designation of low Y/T was simply for convenience, as it too was relatively high Y/T as shown in Table 2. The pipes were of 12.75 OD and 12.7-mm wall thickness. The high Y/T pipes had two types of girth welds which were referred to as the evenmatched and overmatched welds, respectively. The low Y/T pipes had one type of weld which overmatched the pipe strength. Both pressurized and non-pressurized tests were conducted. Each test specimen had either one or two girth welds and each girth weld had two flaws of identical size and notch location. Three flaw locations were tested, i.e., base metal, weld centerline, or HAZ. Table 1 Test matrix and results of X65 full-scale pipe tests
Test ID Pipe Material Temp. (C) 1.1 1.2 1.7 1.8 1.22 1.11 1.24 1.12 1.13 1.14 1.18 1.5 1.6 1.21 1.9 1.19 1.10 1.17 1.23 1.3 1.4 1.15 1.16 1.20 X65 High Y/T X65 High Y/T X65 High Y/T X65 High Y/T X65 High Y/T X65 High Y/T X65 High Y/T X65 High Y/T X65 High Y/T X65 High Y/T X65 High Y/T X65 High Y/T X65 High Y/T X65 High Y/T X65 High Y/T X65 High Y/T X65 High Y/T X65 High Y/T X65 High Y/T X65 Low Y/T X65 Low Y/T X65 Low Y/T X65 Low Y/T X65 Low Y/T 20 20 20 20 20 20 20 20 -20 -20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 N/A N/A Even Even Even Over Over Over Over Over Over Even Even Even Over Over Over Over Over N/A N/A Over Over Over 0.8 0 0.8 0 0.8 0.8 0.8 0 0.8 0 0.8 0.8 0 0.8 0.8 0.8 0 0.8 0.8 0.8 0 0.8 0 0.8 BM BM HAZ HAZ HAZ HAZ HAZ HAZ HAZ HAZ HAZ WM WM WM WM WM WM WM WM BM BM HAZ HAZ WM Weld Pressure Flaw Mismatch Factor Location a 2c TSC at Max Load (%) 0.83 1.88 4.74 8.07 2.01 1.24 2.28 2.69 1.59 3.12 7.73 1.58 4.64 0.72 0.64 1.39 3.10 2.13 0.69 1.51 2.77 4.20 6.81 3.97 Nominal Stress at Max Load (Mpa) 590 555 606 565 609 599 602 568 614 579 606 592 560 596 585 592 558 605 588 487 460 515 474 515

Curved Wide Plate Test Matrix Thirty (30) CWP specimens of an X100 pipe (36 OD and 19.1-mm wall thickness) were tested in a DOT/PRCI sponsored project (i.e., PRCI Project MATH-1) [5,6]. The test matrix and results are shown in Table 3 and the pipe/weld properties are given in Table 4. Two rounds of welds were made, i.e., Round 1 and Round 2. The Round 1 welds were made with a singletorch mechanized GMAW process. The Round 2 welds were made with a dual-torch GMAW process. The two rounds of welds had similar weld strength mismatch levels (i.e., slightly overmatched pipe strength). Three flaw locations were tested, i.e., base metal, weld centerline, or HAZ. Table 3 Test matrix and results of X100 CWP tests
Test ID No ID No ID CWP-16 CWP-03 CWP-11 CWP-19 CWP-01 CWP-15 CWP-10 CWP-12 CWP-02 CWP-05 CWP-06 CWP-07 CWP-08 CWP-09 CWP-13 CWP-14 CWP-17 CWP-18 CWP-20 CWP-21 CWP-22 CWP-23 CWP-24 CWP-25 CWP-26 CWP-27 CWP-28 CWP-29 Pipe Materials X100 X100 X100 X100 X100 X100 X100 X100 X100 X100 X100 X100 X100 X100 X100 X100 X100 X100 X100 X100 X100 X100 X100 X100 X100 X100 X100 X100 X100 X100 Temp. Weld Flaw Round Location (C) 25 -20 25 25 -20 -20 25 25 -20 -20 -20 -20 25 -20 25 -20 -40 -40 -40 -40 25 25 -20 -20 25 25 -20 -20 -40 -40 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 BM BM Weld HAZ Weld Weld Weld HAZ Weld HAZ HAZ HAZ Weld Weld HAZ HAZ Weld Weld HAZ HAZ Weld HAZ Weld HAZ Weld HAZ Weld HAZ HAZ Weld a (mm) 3.00 3.11 3.00 3.00 3.14 3.27 6.00 6.00 6.06 6.14 2.98 2.91 2.12 2.23 2.17 1.97 2.95 3.21 3.09 2.81 2.88 3.09 2.96 2.94 6.36 6.29 6.15 6.19 3.08 3.35 2c (mm) 50.0 50.0 50.0 50.0 50.0 50.0 30.0 30.0 30.0 30.0 50.0 50.0 75.0 75.0 75.0 75.0 50.0 50.0 50.0 50.0 50.0 50.0 50.0 50.0 30.0 30.0 30.0 30.0 50.0 50.0 TSC-Max Load (%) 4.21 3.13 2.33 1.69 2.52 2.29 1.83 2.64 1.76 3.48 3.45 2.31 2.33 3.32 4.17 3.48 2.28 2.44 2.80 3.55 1.66 1.85 1.85 2.23 1.54 1.55 1.61 2.13 2.08 2.54

(mm) (mm) 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 2 3 3 3 3 3 50 50 35 35 50 50 50 50 50 50 35 35 35 50 50 50 50 35 70 50 50 50 50 50

Table 4 Tensile properties of X100 CWP pipe (longitudinal) and all-weld metal
Materials Tensile Properties Flaw Location Pipe Weld HAZ Weld HAZ Yield Strength Y/T (ksi) 113 121 113 120 113 (MPa) 781 835 781 827 781 0.91 0.91 0.91 0.93 0.91

Table 2 X65 pipe (longitudinal) and weld tensile properties

X65 High Y/T Pipe X65 Low Y/T Pipe

Pipe Material

Weld Attribute No Single Torch (Round 1) Dual Torch (Round 2)

Pipe Property YS (ksi) Minimum Maximum Median 74.6 82.2 79.0 UTS (ksi) 83.8 87.9 85.9
Y/T

Weld Mismatch at UTS

Pipe Property UTS (ksi) 67.6 71.7 69.4

Y/T

Evenmatch Overmatch YS Weld Weld (ksi) -2% 9% 5% 12% 18% 15% 55.9 64.4 61.8

Weld Mismatch at UTS 14% 25% 16%

X100

0.87 0.94 0.93

0.83 0.93 0.88

Copyright 2012 by ASME

KEY OBSERVATIONS OF THE PIPE TESTS Some of the key observations from the 24 full-scale pipe tests are summarized here. TSC Variations from Duplicate Tests Two sets of duplicate tests of specimens with nominally identical conditions were performed, i.e., Tests 1.9/1.19 and Tests 1.11/1.24. The nominal stress vs. remote strain relations of the Tests 1.9/1.19 are shown in Figure 1. The TSCs were the strains at the maximum stress. The TSCs of Tests 1.9 and 1.19 are 0.61% and 1.26%, respectively. The maximum stress of Test 1.19 is only about 7 MPa (1 ksi) higher than that of Test 1.9. Similarly, as shown in Table 1, the two duplicate tests, 1.11 and 1.24, had TSCs of 1.15% and 2.21%, respectively. The maximum stresses of Tests 1.11 and 1.24 differ by only about 3 MPa (0.4 ksi). These data demonstrate that the TSCs of duplicate tests can vary by a factor of two, while the maximum stresses at the point of the TSCs are very close. The maximum stress variation is well within normally expected pipe strength variation of the same joint of pipe. These duplication specimens essentially behave identically; yet the reported/measured TSCs can differ by a factor of two due to the very flat stress-strain curves beyond approximately 0.7% strain.
800 X65 High Y/T - Overmatched Weld OD = 12.75", WT = 12.7 mm Flaw: 3 mm x 50 mm (Weld Center) With Pressure

same pipe and weld. Therefore, nominally identical flaws in the same specimen can sample different local materials and different levels of weld mismatch. The local property variation can be inferred from the CMOD curves measured from the same test specimen. For example, the CMOD vs. remote strain relations from Tests 1.19 and 1.14 are shown in Figure 2 (a) and (b), respectively. Each test specimen had four nominally identical flaws and the CMOD of all flaws were recorded. It shows that the CMOD of the nominally identical flaws behaved differently. In general, one flaw eventually became dominant and led to the final failure. In Test 1.19, the CMOD of one flaw became dominant at very low strain values and led to the final failure. On the other hand, in Test 1.14, the CMOD of all four flaws were very close until one of four flaws started to open rapidly against the remote strain. The opening of this flaw effectively determined the tensile strain capacity of the test specimen. The difference in the CMOD of the nominally identical flaws is due, in part, to the variation of the material properties in the local area near the flaw.
6 5 4 (a) Pipe 1.19 X65 High Y/T - Overmatched Weld OD = 12.75", WT = 12.7 mm Flaw: 3 mm x 50 mm (Weld Center) With Pressure
CTF 1-45

CMOD (mm)

3
2 1 0 0.000

CTF 2-225 CBF 3-135 CBF 4-315

600

Stress (MPa)

400

200

0.002

0.004

0.006

0.008

0.010

0.012

0.014

0.016

Pipe 1.9 Pipe 1.19

Remote Strain (mm/mm)

6 5 4 (b) Pipe 1.14 (-20C) X65 High Y/T - Overmatched Weld OD = 12.75", WT = 12.7 mm Flaw: 3 mm x 50 mm (HAZ) No Pressure
CTF 1-45

0 0.000

0.002

0.004

0.006

0.008

0.010

0.012

0.014

0.016

Remote Strain (mm/mm)

Variation of Tensile Properties The tensile properties of the two pipes (X65 high Y/T and X65 low Y/T) and their welds are given in Table 2. The two pipes show similar strength variations, i.e., about 8 ksi for YS and 4 ksi for UTS. The Y/T ratio of the high and low Y/T pipes vary from 0.87 to 0.94 and from 0.83 to 0.93, respectively. The weld strength mismatch (measured by UTS) of the high Y/T pipe varies from -2% to 9% for the nominally evenmatched weld; and 12% to 18% for the overmatched weld. The weld strength mismatch (measured by UTS) of the low Y/T pipe varies from 14% to 25%. Variation of Flaw Response from Full-Scale Tests Pipe and weld properties are usually not uniform along the

CMOD (mm)

Figure 1 Nominal stress vs. remote strain of duplicate tests

3
2 1 0 0.000

CTF 2-225 CBF 3-135 CBF 4-315

0.010

0.020

0.030

0.040

Remote Strain (mm/mm)

Figure 2 Variation of measured CMOD from identical flaws Flaw Behavior and Remote Strain Measurement To highlight the relationship between the flaws in a girth weld and the remote strain, it is useful to look at the pipe crosssection that contains the flaws (termed flawed plane) and the

Copyright 2012 by ASME

regions where the remote strain is measured (i.e., uniform strain zones). As shown in Figure 3, the connection between the flawed plane and the uniform strain zones is the equivalence of longitudinal load. If the cross-sectional areas are the same at those locations, this equivalence of load leads to the equivalence of nominal stress. The action of the remote region is transmitted to the flawed plane through the equivalence of the nominal stress.
Flawed Plane Uniform strain zone Load/stress Load/stress Uniform strain zone

behavior can translate to relatively large differences in strain values when the stress-strain curves are flat. For example, Figure 4 shows three slightly different stressstrain curves. The baseline stress-strain curve was taken from the full-scale Test 1.18. Two additional stress-strain curves were created by adding or subtracting 0.5 ksi (3.5 MPa) to the strength of the baseline curve at 0.5% strain and beyond. The range of strain values due to the slight strength difference is shown in Figure 5. The baseline curve establishes a one-toone correlation. The other two curves provide the upper and lower bound strain values corresponding to the same stress produced by the baseline curve at strains greater than 1.0%. The result shows that when a flaw is subjected to the stress level corresponding to the stress at 3.0% strain of the baseline stress-strain curve, the reported strain in the uniform strain zone can vary from 2.0% to 5.4% if the stress-strain curves vary by 0.5 ksi from the baseline stress-strain curve.
2.5 (a) X65 Low Y/T - Base Metal and HAZ 2.0

Figure 3 Schematic illustration of the remote regions where the strains are measured and the flawed plane where the flaw failure events are initiated
630

possible strain range

Nominal Stress (MPa)

610

590

CTOD R (mm)

1.5

570

Baseline s-s curve f rom Test 1.18 Upper s-s curve (baseline + 0.5 ksi) Lower s-s curve (baseline - 0.5 ksi)

1.0
Pipe Material (Test 1) Pipe Material (Test 2)

0.5

Pipe Material (Test 3) HAZ

550 0.00

0.02

0.04

0.06

0.08

0.10
0.0 0.0 2.5 0.5 1.0

Remote Strain (mm/mm)

Figure 4 Stress-strain curves of slightly different strength

a (mm)

1.5

2.0

2.5

(b) X65 Low Y/T - Weld Metal

2.0

CTOD R (mm)

1.5

1.0
Weld Metal (Test 1)

0.5

Weld Metal (Test 2) Weld Metal (Test 3)

0.0 0.0 0.5 1.0

a (mm)

1.5

2.0

2.5

Figure 5 Strain values corresponding to the three stressstrain curves at the same stress level as a function of the strain of the baseline stress-strain curve If the materials in the uniform strain zones have slightly different stress-strain behaviors from those near the flawed plane, the required remote strain to provide the same nominal stress may be different. A small difference in the stress-strain

Figure 6SENT resistance curves (CTODR) of X65 low Y/T pipe for (a) pipe and HAZ flaw; (b) weld metal flaw DETERMINATION OF FRACTURE TOUGHNESS CTOD Resistance Curves (CTODR) - X65 CTOD resistance curves were obtained from SENT tests following CANMET procedures [21,22]. The resistance curves of the X65 low Y/T and high Y/T pipes are shown in Figure 6

Copyright 2012 by ASME

and Figure 7, respectively. The resistance curves were fitted to power functions in the form of CTODR = A aB. The fitted coefficients A and B are given in Table 5. Multiple tests were conducted for X65 low Y/T pipe and weld materials. The test repeatability is very reliable, as shown in Figure 6. In general, the CTOD resistance curves of the pipe and HAZ materials are higher than those of the weld metals. The resistance curves of the pipe and HAZ materials are similar. Apparent Toughness (CTODA) - X65 The CTODA was calculated from Charpy energy, high constraint SENB single value CTOD, and SENT resistance curve. The procedures to determine the apparent toughness (CTODA) can be found in [4]. The CTODA calculated from different methods are shown in Table 6. The upper shelf Charpy energy and upper shelf CTOD value used for CTOD A calculations are also given for reference. Material Toughness (CTODR and CTODA) - X100 Similar procedures as above were followed to obtain the material toughness properties for the X100 materials. The CTOD resistance curve (CTODR) and apparent toughness (CTODA) are summarized in Table 7.
2.0 (a) X65 High Y/T Pipe - Evenmatched Weld 1.5
Pipe Material

Table 5 Coefficients of fitted R-curves - X65

Materials Pipe Material Weld Attribute None Even X65 High matched Y/T Over matched None X65 Low Y/T Over matched Flaw Location Pipe Weld HAZ Weld HAZ Pipe Weld HAZ Resistance Curve (CTODR = A a B)* A 1.38 0.84 1.38 1.05 1.69 1.75 1.34 1.75 B 0.41 0.50 0.41 0.47 0.72 0.40 0.52 0.40

*The units of CTODR and a are mm.

Table 6 Summary of apparent toughness (CTODA) - X65

Material Property Charpy and SENB Upper Shelf Upper Shelf CTOD ( SENB Charpy - at 99.9% Energy Max Load) (J) 0.92 0.85 0.92 0.87 0.92 0.88 0.80 0.88 390.0 170.0 390.0 235.0 245.0 300.0 185.0 300.0 (mm) 0.72 0.58 0.72 0.54 0.61 0.77 0.44 0.77 Apparent Toughness (CTODA) From CTOD From SENT (Conversion R-curve (a Factor = 1.5) = 0.5 mm) (mm) 1.07 0.87 1.07 0.81 0.91 1.15 0.65 1.15 (mm) 1.04 0.59 1.04 0.76 1.03 1.33 0.93 1.33

Weld Flaw Attribute Location

Yield Strength

Y/T

From Charpy (mm) 1.41 0.65 1.41 0.83 0.87 1.09 0.66 1.09

(ksi) None X65 High Y/T Even matched Over matched None X65 Low Y/T Over matched Pipe Weld HAZ Weld HAZ Pipe Weld HAZ 79.3 76.3 79.3 85.9 79.3 61.0 66.2 61.0

(MPa) 547.0 526.2 547.0 592.3 547.0 420.5 456.4 420.5

CTOD R (mm)

1.0
HAZ 0.5 Pipe Material Weld Metal
No Pipe Material

Table 7 Summary of material toughness - X100

Materials Charpy and SENB Upper Shelf Charpy Energy (J) Pipe 287 (ft-ldf) 212 Upper Shelf CTOD (SENB) (mm) 0.38* 6 3 1.264 0.424 0.424 1.027 0.800 0.459 0.459 0.891 0.734 0.574 0.774 0.774 0.747 0.647 0.805 0.805 0.836 0.726 1.21 0.69 0.69 0.99 0.99 0.61 0.61 0.97 0.97 0.76 0.40 0.40 0.76 0.76 0.44 0.44 0.76 0.76 1.07 0.34 0.34 0.83 0.66 0.36 0.36 0.70 0.60 Flaw Depth (mm) 3 1.364 0.833 CTODR = A*aB From Charpy (mm) 1.21 CTODA From From CTOD SENT R(Conversi curve (a on(mm) Factor = 0.75 (mm) 0.76 1.07

Weld Attribute

Flaw Location

0.0 0.0 0.5 1.0

a (mm)

1.5

2.0

2.5
X100

3.0
(b) X65 High Y/T Pipe - Overmatched Weld 2.5 2.0 1.5

Single Torch (Round 1)

Weld

162

119

0.20 6 3

HAZ

234

173

0.38 6 3

Dual Torch (Round 2)

Weld

141

104

0.22 6 3

HAZ

231

170

0.38* 6

CTODR (mm)

* Not available, used single torch data

Comparison of Model Prediction and Test Data X65 Full-Scale Pipe Tests All Test Data
HAZ
Weld Metal

1.0
0.5 0.0 0.0 0.5 1.0

a (mm)

1.5

2.0

2.5

Figure 7 SENT resistance curves (CTODR) of X65 high Y/T pipe for (a) evenmatched weld; (b) overmatched weld

The TSCs were calculated with Level 2 and 3 procedures for all 24 full-scale tests. For the Level 2 approach, the TSCs were calculated based on the initiation-control limit state where the CTODA was obtained from Charpy and high-constraint CTOD toughness tests (Table 6). The Level 3 predictions were made using the toughness obtained from low-constraint SENT tests and with both the initiation-control and ductile-instability based limit states. The R-curves (CTODR) were obtained from

Copyright 2012 by ASME

SENT tests (Table 5) and the CTODA was calculated from the R-curves at 0.5-mm flaw growth (Table 6). For both Level 2 and 3 procedures, predictions were made using median pipe Y/T and weld mismatch levels obtained from the small-scale tests (Table 2). The predictions are shown in Table 8. The comparison of the predicted and measured TSCs of all 24 tests is shown in Figure 8, where the predictions were made using the initiation-control based limit state and the CTOD A obtained from SENB tests (Level 2). In 2 out of the 24 tests, the measured strains are less than 0.7% (in two small circles in Figure 8) and the measured values are lower than the predicted ones. In 7 of the 24 tests, the measured strains are greater than 3.5% (in a big circle in Figure 8) and the measured values are much higher than the predicted values. For the remaining 15 tests where the measured strains are between 0.7% and 3.5%, the measured strains match the predicted ones well. Table 8 Input parameters and predicted TSCs - X65 pipe
Predicted TSC (Initiation Control) Test Information Weld Mismatch 1.00 1.00 1.05 1.05 1.05 1.15 1.15 1.15 1.15 1.15 1.15 1.05 1.05 1.05 1.15 1.15 1.15 1.15 1.15 1.00 1.00 1.16 1.16 1.16 TSC Max Load (%) 0.83 1.88 4.74 8.07 2.01 1.24 2.28 2.69 1.59 3.12 7.73 1.58 4.64 0.72 0.64 1.39 3.10 2.13 0.69 1.51 2.77 4.20 6.81 3.97 Toughness Charpy CTOD A (mm) 1.41 1.41 1.41 1.41 1.41 0.87 0.87 0.87 0.87 0.87 0.87 0.65 0.65 0.65 0.83 0.83 0.83 0.83 0.83 1.09 1.09 1.09 1.09 0.66 TSC (%) 1.28 2.27 2.18 3.88 1.69 1.64 1.64 2.92 1.71 3.03 2.12 1.05 1.86 0.81 1.57 1.57 2.79 2.03 2.17 1.25 2.23 2.42 4.30 1.52 Toughness SENB CTOD A (mm) 1.07 1.07 1.07 1.07 1.07 0.91 0.91 0.91 0.91 0.91 0.91 0.87 0.87 0.87 0.81 0.81 0.81 0.81 0.81 1.15 1.15 1.15 1.15 0.65 TSC (%) 0.98 1.74 1.70 3.03 1.31 1.71 1.71 3.04 1.78 3.16 2.21 1.40 2.48 1.08 1.53 1.53 2.73 1.98 2.12 1.32 2.34 2.54 4.51 1.49 Toughness SENT CTOD A (mm) 1.04 1.04 1.04 1.04 1.04 1.03 1.03 1.03 1.03 1.03 1.03 0.59 0.59 0.59 0.76 0.76 0.76 0.76 0.76 1.33 1.33 1.33 1.33 0.93 TSC (%) 0.95 1.43 1.66 2.49 1.28 1.92 1.92 2.88 1.99 2.99 2.47 0.95 1.42 0.74 1.44 1.44 2.17 1.87 2.00 1.51 2.26 2.86 4.30 2.10 Predicted TSC (Ductile Instability) Toughness SENT CTOD R=A*a A 1.38 1.38 1.38 1.38 1.38 1.69 1.69 1.69 1.69 1.69 1.69 0.84 0.84 0.84 1.05 1.05 1.05 1.05 1.05 1.75 1.75 1.75 1.75 1.34 B 0.41 0.41 0.41 0.41 0.41 0.72 0.72 0.72 0.72 0.72 0.72 0.50 0.50 0.50 0.47 0.47 0.47 0.47 0.47 0.40 0.40 0.40 0.40 0.52
B

The predictions using the initiation-control and ductileinstability based limit states are very similar. The correlation of TSCs between the initiation-control and ductile-instabilitycontrol limit states is shown in Figure 9 for all 24 tests. The CTOD resistance curves are obtained from SENT tests and the CTODA values are determined from the resistance curves at a = 0.5 mm. The results demonstrate that the initiation-control and ductile-instability-control limit states produce very similar TSCs.
6.0

Predicted TSC (Instability Based) (%)

5.0

Instability based prediction: SENT R-curve Initiation based prediction: CTOD A obtained from SENT R-curve at a = 0.5 mm

4.0

3.0

2.0

1.0

Test ID 1.1 1.2 1.7 1.8 1.22 1.11 1.24 1.12 1.13 1.14 1.18 1.5 1.6 1.21 1.9 1.19 1.10 1.17 1.23 1.3 1.4 1.15 1.16 1.20

2c

Y/T

TSC (%) 0.84 1.49 1.60 2.84 1.13 2.30 2.30 4.10 2.43 4.33 3.22 1.07 1.90 0.74 1.38 1.38 2.45 1.95 1.48 1.35 2.41 2.57 4.57 2.12

(mm) (mm) (mm) 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 2 3 3 3 3 3 50 50 35 35 50 50 50 50 50 50 35 35 35 50 50 50 50 35 70 50 50 50 50 50 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.92 0.92 0.92 0.92 0.92 0.92 0.92 0.92 0.91 0.91 0.92 0.92 0.92 0.92 0.92 0.92 0.92 0.92 0.92 0.88 0.88 0.88 0.88 0.88

0.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0

PredictedTSC (Initiation Based) (%)

Figure 9 Comparison of TSCs between Level 3a (initiationcontrol) and Level 3b (ductile-instability) of all 24 full-scale pipe tests Test Data with TSC between 0.7% and 3.5% The full-scale tests with measured TSCs of 0.7-3.5% were further examined with Levels 2 and 3 procedures. The comparison of the measured and predicted TSCs from Level 2 procedures is shown in Figure 10. Two initiation-control based toughness options, Charpy upper shelf energy and highconstraint SENB CTOD tests, were exercised. The comparison of the measured and predicted TSCs from Level 3 procedures is shown in Figure 11. Overall all Levels 2 and 3 toughness options provide similar predictions. The scatter in Level 3 predictions is slightly higher than that in Level 2 predictions. Test Data of Low Measured TSCs To examine the two tests with low strain values (< 0.7%), the measured nominal stress vs. remote strain relations for Tests 1.9, 1.19, and 1.23 are shown in Figure 12. Test 1.18 is also shown to indicate the typical stress-strain curve shape. Tests 1.9 and 1.19 were duplicate tests and the predicted TSC was about 1.6%. While this prediction matches well with the result of Test 1.19 (1.3%), it is considerably higher than the result of Test 1.9 which failed at 0.61%. The specimens were inspected for high-low misalignment after welding and notch geometry was checked pre and post-

5.0
1.16

4.0
TSC > 3.5%

Predicted TSC (%)

1.8

3.0

1.23

1.15 1.17

1.6

1.18

2.0
1.20

1.7

1.0 1.9 1.19 (Duplicate tests) 0.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0
CTODA (SENB)

Measured TSC (%)

Figure 8 Predicted (Level 2) and measured TSCs of 24 tests

Copyright 2012 by ASME

test. No irregularities were found. In addition, the fracture surfaces were examined and no hidden weld defects were discovered. The strength and chemical compositions of those welds were thoroughly analyzed in a post-test examination. It was confirmed that the weld has proper strength as expected. However, unexpected high Oxygen contents were discovered in those welds of Tests 1.9 and 1.23 (low TSCs). The oxygen contents in the weld of Test 1.19 were within the normal range. It is suspected that the high oxygen contents may deteriorate the weld toughness. However, it should be noted that some tests failed at very high TSCs also showed high oxygen contents. The details of the strength and chemical composition analyses can be found in [4]. The CTODA measured from the residual profiles of those non-breaching-wall flaws in Tests 1.9 and 1.23 also indicated the possibility of low weld toughness. The CTOD A of some crack profiles in those two tests was about 0.5 mm which is on the lower bound of the toughness spread of the welds from the same welding procedure.
4.0 3.5

4.0 3.5

(a)

CTODA (SENT)

3.0

Predicted TSC (%)

2.5 2.0 1.5 1.0 0.5 0.0 0.0

4.0 3.5

0.5

1.0

1.5

2.0

2.5

3.0

3.5

Measured TSC (%)

(b)
R-Curve (SENT)

3.0

Predicted TSC (%)

2.5 2.0 1.5 1.0 0.5

(a)

CTODA (Charpy)

3.0

Predicted TSC (%)

2.5

0.0
2.0 1.5 1.0 0.5 0.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

Measured TSC (%)

Figure 11 Comparison of predicted (Levels 3a and 3b) and measured TSC of 15 tests with strain values of 0.7-3.5% (a) CTODA from SENT; (b) R-curve from SENT
650 Predicted TSC: 1.6% - 2.2% 600

4.0 3.5 3.0

Measured TSC (%)

(b)

CTODA (SENB)

Nominal Stress (MPa)

550
500 450 400 350 0.000
1.9 (a = 3 mm, 2c = 50 mm) 1.19 (a = 3 mm, 2c = 50 mm) 1.23 (a = 2 mm, 2c = 70 mm) 1.18

Predicted TSC (%)

2.5 2.0 1.5 1.0

X65 High Y/T Weld Mismatch = 15% Weld Flaws With Pressure

0.020

0.040

0.060

0.080

0.100

0.5
0.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5

Remote Strain (mm/mm)

Measured TSC (%)

Figure 12 Measured nominal stress vs. remote strain relations for selected tests (Tests 1.9, 1.19, and 1.23) In summary, high oxygen level and low toughness may have contributed to the measured low TSCs of Tests 1.9 and 1.23. Since the nominal failure stresses of those tests are fairly close to other tests which produced high failure strains, the

Figure 10 Comparison of predicted (Level 2) and measured TSC of 15 tests with strain values of 0.7-3.5% (a) CTODA from Charpy; (b) CTODA from SENB

Copyright 2012 by ASME

other contributing factor is the flat stress-strain curve of the tested materials. Test Data of High Measured TSCs The tests with high measured TSCs were examined by grouping them using pipe Y/T ratio and internal pressure. The measured nominal stress vs. remote strain relations for Tests 1.7, 1.17, and 1.18 (see those marked in Figure 8) are shown in Figure 13. The conditions of the three tests were similar (i.e., X65 high Y/T pipe with pressure) and the predicted TSCs of all three tests were approximately 2.1%. The measured TSC of Test 1.17 was 1.9% which matches the prediction well. However, the measured TSCs of Tests 1.7 and 1.18 were 4.2% and 5.9%, respectively, which are much higher than the predicted TSCs. On the other hand, of the three tests (1.7, 1.17, and 1.18) shown in Figure 13, the largest difference between the nominal stresses at the predicted and measured TSCs is only about 5.6 MPa (as marked in Figure 13). Similar results can be found in Figure 14 for the X65 high Y/T pipe tests without pressure (Tests 1.6 and 1.8). The largest difference between the nominal stresses at the predicted and measured TSCs of the two tests is about 8.2 MPa (from Test 1.8). The measured strains are greater than the predicted strains by a factor of more than 2. The measured nominal stress vs. remote strain curves of the X65 low Y/T pipes tested under pressurized and nonpressurized conditions showed similar results. Although the measured and predicted TSCs can vary by a factor of two, the maximum difference between the nominal stresses at predicted and measured TSCs is less than 14.5 MPa. Impact of Small Strength Variations The results of Figure 13 and Figure 14 clearly demonstrate that, while the measured strains at the failure events can have large differences, the differences in the corresponding stresses at the same failure events are very small. As illustrated in Figure 3, the flaws can only sense the stress transmitted from the remote regions. The flaws should be expected to behave the same if the applied stress on the flawed plane is the same 1. The remote strains that are needed to produce this same level of stress may be different, depending on the stress-strain relations. The impact of pipe strength variation was further analyzed by the linepipe groups, i.e., by the Y/T ratio. In Figure 15, a representative stress-strain relation was obtained from Test 1.8 for high Y/T pipes, and is shown as the baseline curve. Two more curves with a strength difference 1 ksi (6.9 MPa) were created by parallel-shifting the baseline curve. Using a process similar to that generated Figure 5, the strain range at
1 The context of this discussion is the stress being applied the pipe cross section where the flaw is located. The micro-scale mechanism of flaw failures is a separate subject which is not discussed here.

the same stress level was estimated from the upper and lower curves. It should be noted that the strain range due to the strength level variation is dominated by the shape of the stress-strain curve and the amount of strength variation. The absolute strength level of the baseline curve has little influence on the strain range so the single baseline curve can be used to represent the influence of strength variation with respect to the baseline curve.
650 Predicted TSC: ~ 2.1% 600

5.6 MPa

Nominal Stress (MPa)

550

X65 High Y/T Flaw: a =3 mm, 2c = 35 mm With Pressure

1.17 (Weld Flaw; Mismatch = 15%) 1.18 (HAZ Flaw; Mismatch = 15%) 1.7 (HAZ Flaw; Mismatch = 5%)

500
450 400 350 0.00

0.02

0.04

0.06

0.08

0.10

Remote Strain (mm/mm)

Figure 13 Measure nominal stress vs. remote strain relations for selected tests (X65 high Y/T pipe with pressure Tests 1.7, 1.17, and 1.18)
600 8.2 MPa 550

Nominal Stress (MPa)

500 450

Predicted TSC of Test 1.6

Predicted TSC of Test 1.8

400
350 300 0.00

X65 High Y/T Flaw: a = 3 mm, 2c = 35mm Mismatch = 5% No Pressure

1.6 (WM Flaw) 1.8 (HAZ Flaw)

0.02

0.04

0.06

0.08

0.10

Remote Strain (mm/mm)

Figure 14 Measure nominal stress vs. remote strain relations for selected tests (X65 high Y/T pipe no pressure Tests 1.6 and 1.8) The experimentally measured and the predicted TSCs of the high Y/T pipes are compared in Figure 16 for all 19 high Y/T pipe tests. The predicted TSCs using the median values of the pipe and weld tensile properties are given as the x values. The corresponding measured TSCs are given as the y values. Three lines are also given in the plot. The middle line represents the one-to-one, i.e., perfect agreement between the measured and predicted TSCs. The upper line represents the

Copyright 2012 by ASME

remote strain value needed to produce the same stress level as the baseline curve when the stress-strain relation is 1 ksi (6.9 MPa) weaker than the baseline curve. By the same process, the lower line represents the remote strain when the stress-strain relation is 1 ksi (6.9 MPa) stronger than the baseline curve. The figure shows that the middle line of the prediction goes through the center of the cluster of the test data. The upper and lower bound curves due to a 1 ksi strength variation follow the upper and lower bounds of the experimental data.

those of Figure 16. Most of the test data treads the upper bound curve.

Figure 17 Baseline stress-strain relation of low Y/T pipe and two variations from the baseline relation
Measured TSC and Expected Range (%)
9.0
Measured TSC

8.0 7.0 6.0 5.0

One-to-one line Expected TSC f rom upper strength curve Expected TSC f rom lower strength curve

Figure 15 Baseline stress-strain relation of high Y/T pipe and two variations from the baseline relation
Measured TSC and Expected Range (%)
9.0
Measured TSC

4.0
3.0 2.0 1.0 0.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0

8.0
One-to-one line

7.0 6.0 5.0

Expected TSC f rom upper strength curve Expected TSC f rom lower strength curve

4.0
3.0 2.0 1.0 0.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0

TSC Predicted from Median Tensile Property (%)

Figure 18 Comparison of measured and predicted TSCs of the low Y/T pipes. The predicted TSCs are from Level 2 using SENB toughness option. The possible ranges of the TSCs from the strength variation of 1 ksi are also shown.
TSC Predicted from Median Tensile Property (%)

X100 CWP Tests The TSCs of the 30 CWP tests are calculated with Level 2 and 3 models. The TSC predictions use the zero-pressure option described in the companion paper [2]. The predictions from different approaches are shown in Table 9. The comparison between the predicted and measured TSCs are further shown in Figure 19 and Figure 20 for Level 2 and Level 3 models, respectively. All comparisons show a certain level of scatter. In general, the Level 2 predictions with CTOD A from Charpy are the closest to the measured TSCs. The Level 2 predictions with CTODA from SENB are generally lower than the measured TSCs. The ductile-instability based Level 3 predictions with CTODR from SENT are similar to the Level 2 predictions with CTODA from Charpy, but with larger scatter.

Figure 16 Comparison of measured and predicted TSCs of the high Y/T pipes. The predicted TSCs are from Level 2 using SENB toughness option. The possible ranges of the TSCs from the strength variation of 1 ksi are also shown. A similar process similar to the one described above was repeated for the low Y/T pipes. In Figure 17, a representative stress-strain relation was obtained from Test 1.16 for the low Y/T pipes. Two more curves with a strength difference 1 ksi (6.9 MPa) were created by parallel-shifting the baseline curve. The experimentally measured and the predicted TSCs of the low Y/T pipes are compared in Figure 18 for all 5 tests. The upper and lower lines were generated in the same way as

Copyright 2012 by ASME

Table 9 Input parameters and predicted TSCs - X100 CWP

Predicted TSC (Initiation Control) Test Information TSCMax Load (%) 4.21 3.13 2.33 1.69 2.52 2.29 1.83 2.64 1.76 3.48 3.45 2.31 2.33 3.32 4.17 3.48 2.28 2.44 2.80 3.55 1.66 1.85 1.85 2.23 1.54 1.55 1.61 2.13 2.08 2.54 Toughness Charpy Weld Mismatch 1.00 1.00 1.06 1.08 1.06 1.07 1.08 1.06 1.06 1.06 1.08 1.07 1.07 1.07 1.07 1.06 1.06 1.06 1.06 1.06 1.03 1.03 1.03 1.03 1.04 1.04 1.04 1.03 1.03 1.03 CTODA (mm) 1.21 1.21 0.69 0.99 0.69 0.69 0.69 0.99 0.69 0.99 0.99 0.99 0.69 0.69 0.99 0.99 0.69 0.69 0.99 0.99 0.61 0.97 0.61 0.97 0.61 0.97 0.61 0.97 0.97 0.61 TSC (%) 3.36 3.17 2.24 4.20 2.59 2.57 1.72 2.57 1.81 2.49 3.71 2.56 3.15 2.80 3.95 4.74 2.45 1.85 3.71 3.22 1.67 2.90 2.08 2.71 1.38 2.17 1.43 2.10 3.16 1.79 Toughness SENB Toughness SENT CTODA (mm) 0.76 0.76 0.40 0.76 0.40 0.40 0.40 0.76 0.40 0.76 0.76 0.76 0.40 0.40 0.76 0.76 0.40 0.40 0.76 0.76 0.44 0.76 0.44 0.76 0.44 0.76 0.44 0.76 0.76 0.44 TSC (%) 2.18 2.05 1.24 3.31 1.44 1.44 1.00 2.01 1.06 1.95 2.91 1.99 1.75 1.54 3.15 3.85 1.36 1.03 2.91 2.51 1.16 2.29 1.46 2.14 1.00 1.72 1.04 1.67 2.51 1.26 CTODA (mm) 1.08 1.08 0.34 0.83 0.34 0.34 0.34 0.67 0.34 0.67 0.83 0.83 0.34 0.34 0.83 0.83 0.34 0.34 0.83 0.83 0.36 0.70 0.36 0.70 0.36 0.60 0.36 0.60 0.70 0.36 TSC (%) 3.04 2.86 1.03 3.59 1.20 1.20 0.85 1.77 0.90 1.71 3.17 2.17 1.44 1.27 3.41 4.14 1.13 0.86 3.16 2.74 0.94 2.11 1.18 1.97 0.83 1.36 0.86 1.32 2.32 1.02 Predicted TSC (Ductile Instability) Toughness SENT CTODR=A*aB A 1.36 1.36 0.42 1.03 0.42 0.42 0.42 0.80 0.42 0.80 1.03 1.03 0.42 0.42 1.03 1.03 0.42 0.42 1.03 1.03 0.46 0.89 0.46 0.89 0.46 0.73 0.46 0.73 0.89 0.46 B 0.83 0.83 0.77 0.75 0.77 0.77 0.77 0.65 0.77 0.65 0.75 0.75 0.77 0.77 0.75 0.75 0.77 0.77 0.75 0.75 0.81 0.84 0.81 0.84 0.81 0.73 0.81 0.73 0.84 0.81 TSC (%) 3.21 3.10 1.34 3.58 1.53 1.56 1.78 2.39 1.80 2.35 3.23 2.35 1.23 1.16 2.54 2.79 1.42 1.19 3.23 2.77 1.24 2.52 1.50 2.36 1.80 2.24 1.83 2.16 2.72 1.39

Test ID No ID No ID CWP-16 CWP-03 CWP-11 CWP-19 CWP-01 CWP-15 CWP-10 CWP-12 CWP-02 CWP-05 CWP-06 CWP-07 CWP-08 CWP-09 CWP-13 CWP-14 CWP-17 CWP-18 CWP-20 CWP-21 CWP-22 CWP-23 CWP-24 CWP-25 CWP-26 CWP-27 CWP-28 CWP-29

a (mm) 3.00 3.11 3.00 3.00 3.14 3.27 6.00 6.00 6.06 6.14 2.98 2.91 2.12 2.23 2.17 1.97 2.95 3.21 3.09 2.81 2.88 3.09 2.96 2.94 6.36 6.29 6.15 6.19 3.08 3.35

2c (mm) 50.0 50.0 50.0 50.0 50.0 50.0 30.0 30.0 30.0 30.0 50.0 50.0 75.0 75.0 75.0 75.0 50.0 50.0 50.0 50.0 50.0 50.0 50.0 50.0 30.0 30.0 30.0 30.0 50.0 50.0

Y/T 0.91 0.91 0.91 0.91 0.91 0.91 0.91 0.91 0.91 0.91 0.91 0.91 0.91 0.91 0.91 0.91 0.91 0.91 0.91 0.91 0.91 0.91 0.91 0.91 0.91 0.91 0.91 0.91 0.91 0.91

the variation of pipe strength, the measured strains on two sides of the specimen were usually different. The averaged value of the measured strains from two sides is shown in Figure 19 and Figure 20. In Figure 21, three values are shown for each test. Two end values connected by a bar are from each side of the girth weld. The averaged value of the two sides is shown in the middle of the bar. Figure 21 clearly demonstrates the variation of measured strains in the same CWP specimen. The measured strains from two sides of the specimen can differ by a factor of two. The absolute difference between two sides increases with the increase of the overall strain value. Figure 21 further highlights the possibility of large scatters in measured TSCs due to small variations in pipe strength.
5.0

CTOD R from SENT R-Curve

4.0

Predicted TSC (%)

WMC HAZ BM

3.0

5.0

(a) CTODA from Charpy

4.0

2.0

Predicted TSC (%)

WMC HAZ BM

1.0

3.0

0.0 0.0 1.0 2.0 3.0 4.0 5.0

2.0

Measured TSC (%)

1.0

Figure 20 Comparison of predicted (Level 3b) and measured TSCs of 30 CWP tests (ductileinstability, CTODR from SENT)
0.0 1.0 2.0 3.0 4.0 5.0
6.0

0.0

5.0

Measured TSC (%)

CTODA from Charpy

5.0

Predicted TSC (%)

(b) CTOD A from SENB

4.0

WMC HAZ BM

WMC HAZ BM

4.0

Predicted TSC (%)

3.0

3.0

2.0

2.0
1.0

1.0

0.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0

0.0 0.0 1.0 2.0 3.0 4.0 5.0

Measured TSC (%)

Measured TSC (%)

Figure 19 Comparison of predicted (Level 2) and measured TSCs of 30 CWP tests (a) CTODA from Charpy; (b) CTODA from SENB In the CWP tests, separate strain measures were taken in the pipe body material on both sides of the girth weld. Due to

Figure 21 Comparison of predicted (Level 2) and measured TSCs of 30 CWP tests with CTODA from Charpy

10

Copyright 2012 by ASME

CONCLUSIONS The general observations from the model evaluation are summarized as follows. (1) TSC predictions from Levels 2 and 3 procedures are compared with experimental test data. Different toughness options are applied in Level 2 procedures. The overall correlations between the experimental test data and model predictions are similar when the model predictions are made with Level 2 and 3 procedures and various toughness options. (2) The Level 2 procedure with Charpy energy option and Level 3b provide the best overall one-to-one correlation between the test data and model prediction. The Level 3b shows greater scatter than Level 2 with the Charpy energy option. This observation is consistent for the 24 full-scale pipe tests and the 30 CWP tests. (3) The predicted TSCs trend well and go through the center of the test data cluster. Given the consistency in predicted TSC by two levels of the models with various toughness options, the variations in the correlation between the test data and predictions are primarily due to the variation of the test data. (4) The most significant contributor to the TSC variation and the difference between the measured and predicted TSCs is the strength variation in the pipes. A small variation in the strength can lead to a large variation of the measured remote strain even when the flaw behavior is essentially the same. (5) For the 24 full-scale pipe tests, a strength variation of 1 ksi in the pipes would explain the large variations of the measured TSC in comparison to the model predictions. (6) For the X65 linepipe materials tested in this project, the plastic part of the stress-strain curves is very flat. In the high strain regime, the flaw cannot sense the difference in the strains in remote regions as these strains produce almost identical stress. From a practical viewpoint, it is unlikely that the large strain values produced in some of the tests can be repeated consistently given the required precision and repeatability of the stress-strain curves. Therefore, large TSC variations are expected even under nominally identical conditions for this type of materials with flat stress-strain curves. (7) The Levels 2 and 3 models produce consistent results that capture the overall trend of the test data. Since the Level 1 models is effectively a tabular version of Level 2 models, it is concluded that the TSC models are capable of providing TSC predictions that are consistent with experimental data with the understanding that the experimental data can have large variations from linepipe materials with flat stress-strain curves.

ACKNOWLEDGMENTS The financial support from DOT/PHMSA and PRCI is gratefully acknowledged. The full-scale pipe tests were done at C-FER Technologies under the supervision of Mr. Mark Stephens and Mr. Randy Petersen. The CWP tests were done at NIST Boulder under the supervision of Dr. Timothy Weeks, Dr. Mark Richards, and Mr. J. David McColskey. The contributions of those individuals and their test programs constitute an integral part of the tensile strain model development and evaluation. Their dedication and efforts are greatly appreciated. REFERENCES 1 Wang, Y.-Y., Liu, M., Zhang, F., Horsley, D., and Nanney, S., Multi-tier Tensile Strain Models for Strain-Based Design Part I - Fundamental Basis, Proceedings of the 9th International Pipeline Conference, Paper No. IPC201290690, Calgary, Alberta, Canada, September 24-28, 2012. 2 Liu, M., Wang, Y.-Y., Song, Y., Horsley, D., and Nanney, S., Multi-tier Tensile Strain Models for Strain-Based Design Part II - Development and Formulation of Tensile Strain Capacity Models, Proceedings of the 9th International Pipeline Conference, Paper No. IPC201290650, Calgary, Alberta, Canada, September 24-28, 2012. 3 Wang, Y.-Y., Liu, M., Long, X., Stephens, M., Petersen, R., and Gordon, R., Validation & Documentation of Tensile Strain Limit Design Models for Pipelines, PRCI Project ABD-1, US DOT Agreement DTPH56-06-T000014, Final report, August 2, 2011, http://primis.phmsa.dot.gov/matrix/PrjHome.rdm?prj=200. 4 Wang, Y.-Y., Liu, M., and Song, Y., 2011, Second Generation Models for Strain-Based Design, PRCI Project ABD-1, US DOT Agreement DTPH56-06-T000014, Final report, August 30, 2011, http://primis.phmsa.dot.gov/matrix/PrjHome.rdm?prj=201 . 5 Weeks, T., McColskey, J. D., and Richards, M., Weld Design, Testing, and Assessment procedures for HighStrength Pipelines Curved Wide Plate Tests, DOT Agreement DTPH56-07-T-000005, final report 277-T-09, September, 2011. 6 Wang, Y.-Y. and Zhou, H., Curved Wide Plate Test Results and Transferability of Test Specimens, DOT Agreement DTPH56-07-T-000005, topical report 277-T-11, January 21, 2012. 7 Mohr. W. C., Strain-Based Design of Pipelines, final report to MMS, October 2003.

11

Copyright 2012 by ASME

8 Mohr, W., Weld Area Mismatch and Pressure Effects in Strain-Based Design, Proceedings of the 4th International Conference on Pipeline Technology, Edited by Rudi Denys, Ostend, Belgium, May 9-13, 2004, pp. 279-290. 9 Liu, M. and Wang, Y.-Y., Significance of Biaxial Stress on the Strain Concentration and Crack Driving Force in Pipeline Girth Welds with Softened HAZ, Proceedings of the 26th International Conference on Offshore Mechanics and Arctic Engineering (OMAE 2007), San Diego, California, USA, June 10-15, 2007. 10 Wang, Y.-Y. and Liu, M., The Role of Anisotropy, Toughness Transferability, and Weld Misalignment in the Strain Based Design of Pipelines, Proceedings of the 17 th International Offshore and Polar Engineering Conference (ISOPE 2007), Lisbon, Portugal, July 1-6, 2007. 11 Wang, Y.-Y. and Liu, M., Weld Integrity from the Perspective of Strain-Based Design of Large Diameter and High Strength Pipelines, Special Issue of the Proceedings of Pipeline Technology Now and Then at the 8 th International Welding Symposium, 8WS, November 16-18, 2008, Kyoto, Japan. 12 Kibey, S. A., Minnaar, K., Issa, J. A., and Gioielli, P. C., Effect of Misalignment on the Tensile Strain Capacity of Welded Pipelines, Proceedings of the 18th International Offshore and Polar Engineering Conference (ISOPE 2008), Vancouver, Canada, July 6-11, 2008. 13 Liu, M., Wang, Y.-Y., and Horsley, D., Significance of HAZ Softening on Strain Concentration and Crack Driving Force in Pipeline Girth Welds, Proceeding of the 24th International Conference on Offshore Mechanics and Arctic Engineering , 2005, June 13-17, Kalkidiki, Greece. 14 Wang, Y.-Y., Rudland, D., Denys, R., and Horsley, D. J., A Preliminary Strain-Based Design Criterion for Pipeline Girth Welds, Proceedings of the International Pipeline Conference 2002, Calgary, Alberta, Canada, September 29October 3, 2002. 15 Wang, Y.-Y., Cheng, W., McLamb, M., Horsley, D., Zhou, J., and Glover, A., Tensile Strain Limits of Girth Welds with Surface-Breaking Defects Part I an Analytical Framework, in Pipeline Technology, Proceedings of the 4th International Conference on Pipeline Technology, Edited by Rudi Denys, Ostend, Belgium, May 9-13, 2004. 16 Wang, Y.-Y., Horsley, D., Cheng, W., Glover, A., McLamb, M., and Zhou, J., Tensile Strain Limits of Girth Welds with Surface-Breaking Defects Part II Experimental Correlation and Validation, in Pipeline Technology, Proceedings of the 4th International Conference on Pipeline Technology, Edited by Rudi Denys, Ostend, Belgium, May 9-13, 2004.

17 Horsley, D. and Wang, Y.-Y., Weld Mismatch Effects on the Strain Limits of X100 Girth Welds, in Pipeline Technology, Proceedings of the 4th International Conference on Pipeline Technology, Edited by Rudi Denys, Oostende, Belgium, May 9-12, 2004. 18 Wang, Y.-Y., Cheng, W., and Horsley, D., Tensile Strain Limits of Buried Defects in Pipeline Girth Welds, Proceedings of the International Pipeline Conference 2004, Calgary, Alberta, Canada, October 4-8, 2004. 19 Wang, Y.-Y., Liu, M., Horsley, D., and Zhou, J., A Quantitative Approach to Tensile Strain Capacity of Pipelines, 6th International Pipeline Conference, Paper No. IPC2006-10474, September 25-29, 2006, Calgary, Alberta, Canada. 20 CSA Z662, Oil and Gas Pipeline Systems, Canadian Standards Association, 2007. 21 Shen, G. and Tyson, W.R., Evaluation of CTOD from JIntegral for SE(T) Specimens, Pipeline Technology 2009. 22 Shen, G., Gianetto, J.A. and Tyson, W.R, Measurement of J-R Curves Using Single-Specimen Technique on Clamped SE(T) Specimens, Proceedings of the Nineteenth International Offshore and Polar Engineering Conference, June 21-26, 2009.

12

Copyright 2012 by ASME