You are on page 1of 16

23rd World Gas Conference, Amsterdam 2006

CRACK ARRESTABILITY OF HIGH-PRESSURE GAS PIPELINES BY

X100 OR X120

Authors

Izumi TAKEUCHI
Hiroyuki MAKINO
Shuji OKAGUCHI
Nobuaki TAKAHASHI
Akio YAMAMOTO

Sumitomo Metal Industries Lt., Japan

To meet the growing demand of natural gas. new gas fields must be developed. Pipes produced by UOE process were proved to meet X100 or X120 grade high strength line pipes for high-pressure operation. The transportation cost of gas to market is an important factor for the development of new gas fields. Lower cooling finishing temperature has been applied to get higher strength. ABSTRACT The demand of natural gas has been increasing. . However the arrest energy depends on pipe arrangement in the burst test and the possibility of intrinsic arrest of X100 or X120 pipes is revealed by the HLP (High-strength Line Pipe committee in ISIJ) simulation method. API grade X100 and X120 line pipes for high-pressure gas pipeline have been successfully developed and the application technologies have been studied to construct safe and economical pipelines. High quality steel and TMCP (Thermo-Mechanical Control Process) technology in plate rolling made it possible to achieve high strength and high toughness of X100 or X120 grade large diameter line pipes. which are located in remote area. Optimum steel chemistry for high strength and high toughness has been selected to maintain good weldability. The discussion was started on the necessity of crack arrestors in high-pressure gas pipelines. A series of full scale burst tests were conducted to evaluate the arrest energy for high strength line pipes. It was reported that the prediction of arrest energy based on the previous experiments with conventional grade pipes was not suitable to apply in X100 or X120 line pipes and intrinsic arrest of propagating ductile fracture in high pressure pipelines was hard to achieve. One of the key issues for the application of these high strength line pipes is the arrestability of propagating ductile fracture in high-pressure gas pipelines. High-pressure gas pipelines by high strength steel pipes have been focused to find the solution of it.

SUMMARY . INTRODUCTION 2. DEVELOPMENT OF X100 AND X120 STEEL FOR LINE PIPES 3. SIMULATION OF PROPAGATING DUCTILE FRACTURE IN GAS PIPELINES 6. TABLE OF CONTENTS 1. INTRINSIC ARRESTABILITY OF X100 OR X120 IN REAL PIPELINES 7.2 ExxonMobil X120 Test 4.1 ECSC Program on X100 4. PROPERTIES OF X100 AND X120 LINE PIPES 4. REVIEW OF FULL-SCALE BURST TESTS USING X100 OR X120 LINE PIPES 4.5 Bp X100 Test 5.4 DemoPipe Project on X100 4.3 Advantica’s JIP on X100 4.

Pcm = 0. The steels for higher strength line pipes are state-of-the-art products of pipe manufactures. The properties of pipes produced by 4). Propagating ductile fracture is the unique fracture mode in high-pressure gas pipelines. The microstructure of X70 grade steel consists of ferrite and pearlite/bainite. as large amount of high quality steel is required in short delivery period. Lower transformation microstructure such as martensite and lower bainite is required to increase strength and fine low carbon martensite and lower bainite dominant microstructure is essential to achieve adequate fracture toughness at the strength level over 900 MPa. which are suitable for TMCP (Thermo-Mechanical Control Process). and . which are essential requirements to line pipes. The figure shows that addition of 10 ppm of B suppressed the ferrite formation entirely. DEVELOPMENT OF X100 AND X120 STEEL FOR LINE PIPES The target tensile properties of X100 and X120 were selected to be 690 MPa (SMYS) and 760 MPa (SMTS) for X100 and 827 MPa (SMYS) and 931 MPa (SMTS) for X120. respectively. Traditionally quench and tempered process was employed to get high strength of X100 or X120 grade.20 %). In order to maximize austenite hardenability while keeping the Pcm low for enhanced weldability.06%C-Mn-Cu-Ni-Mo-Nb-Ti steel (Pcm = 0. Lean chemistry is selected from the viewpoint of weldability and alloy cost. The development of X100 and X120 grade line pipes for high-pressure gas pipelines has been 1)~3) conducted as the common interest of steel manufactures and major oil companies .5) UOE process are carefully investigated for safe operation under high pressure . BODY OF PAPER 1. However the capacity of heat treatment process is not sufficient to adopt for line pipe steel production.05%C-Mn-Cu-Ni-Mo-Nb-V-Ti-B steel (B = 10 ppm. low carbon Mo-Nb bearing and boron free steel and low carbon Mo-Nb bearing steel with a minute amount of boron were designed for X100 and X120. 2. whose highest grade is X80. The possibility of intrinsic arrest of X100 or X120 in realistic pipelines will be presented. series of burst test conducted in conventional manner of test procedure by X100 or X120 are reviewed.20 %) and 0. INTRODUCTION One of the barriers to increase natural gas supply to market is the transportation cost. High-pressure gas pipelines are expected to give a solution to it. Figure 1 illustrates the continuous cooling transformation diagrams after hot deformation of 0. as high strength steel pipes result in less steel consumption and less construction cost comparing to ordinal API grade line pipes. Firstly high strength steels. as there is no specification for X100 or X120 in current API 5L. The accelerated cooling technology has been investigated to achieve strength and toughness. High strength steel pipes have an advantage to transport gas in high-pressure. To estimate the arrest energy against propagating ductile fracture in high-pressure gas pipelines. have to be developed. One of the critical issues in high-pressure gas pipelines is the arrest of propagating ductile fracture.

Nitrogen. Hv 280 for boron free steel and Hv 320 for boron steel were observed.05%C-Mn-boron steel can meet X120 grade. respectively. bainitic ferrite with martensite and lower bainite composed the microstructure of the boron free steel. The burst stress reasonably coincides with tensile stress of base metal. The low CFT below 300 °C is stable range for production in terms of strength and toughness. TMCP was applied in plate rolling with strict control of rolling and cooling conditions to achieve optimum microstructure for each grade. The pipes were produced by UOE process with double submerged-arc welding (SAW). Hydrostatic burst test was conducted to confirm the burst stress of X100 and X120 high strength line pipe.lowered the bainite transformation temperature. The steel for X100 is 0. However in case of X100 or X120 the yield strength measured by un-flattened round bar specimen is around 110 MPa higher than by flattened strip specimen. In order to achieve optimum properties. According to hardness test. restrict grain growth and adverse precipitation effects. After accelerated cooling (> 20 K/s). PROPERTIES OF X100 AND X120 LINE PIPES Prototype X100 and X120 line pipes were produced in Sumitomo Metal’s Kashima Steel Works. and martensite and lower bainite dominate microstructure of boron bearing steel. phosphorus and sulfur contents were reduced to gain cleanliness. The carbon and silicon contents were controlled in low level to minimize the amount of MA (Martensite-austenite) constituent. V and Ti) were carefully selected to help control of MA. so they are suitable for X100 and X120 tensile strength. Microalloys (Nb. Slabs for plate were made by the continuous caster. The yield strength by round bar specimen is more reasonable value to evaluate the pipe strength. .20 % for X120. and the hardenability was controlled to ensure fine lath base microstructure in HAZ during weld thermal cycle. Steel chemistry must be optimized for SAW welding in UOE process.05%C-1. The pipe sizes are 914 mm OD (Outer diameter) X 19 mm thick of X100 and 914 mm OD X 16 mm thick of X120.83%Mn-(Cu-Ni-Cr-Mo-Nb-Ti) and the steel for X120 is 0. Tensile strength and vTs are improved by lowering the CFT from 500 °C to below 300 °C . The converter of 250 tones was used to melt the steels and vacuum degassing was employed. Heat affected zone (HAZ) toughness is an important property of steel for welded large diameter pipes.06%C-Mn boron free steel can meet X100 and 0. It is reported that the Bauschinger effect is not ignored in case of 6) X100 or X120 .21 % for X100 and 0. The hoop tensile properties of line pipes were measured by flattened strip specimen and un-flattened round bar specimen. It is proved that the strength of pipes can meet X100 or X120 grade. 0. respectively. The flattened strip specimen has been employed for line pipes according to API 5L specification. Burst stress in hydrostatic burst test is also shown in the same table. reduction rate in austenite and cooling start temperature are required in addition to CFT. The Pcm for the steel are 0. Figure 2 shows the effect of the cooling finishing temperature (CFT) in TMCP on tensile strength and ductile to brittle transition temperature (vTs) of Charpy test of 16 mm thick plate.56%Mn-(Cu-Ni-Cr-Mo-Nb-V-Ti-B). Table 1 summarizes the tensile properties in transverse direction as average value of over 20 joints.07%C-1. The YS and TS measured by both specimens meet the target properties of X100 or X120.93. though the Y/T ratio exceeds 0. the precise control of rolling conditions such as re-heating temperature. 3. which is current requirement in API 5L for line pipes.

9) 4. Low C and low sulfur steel is essential for high strength line pipe.9456 where it arrested after about 5 meters progress.6129 out of the trench. length 2. went back in the base metal till they joined causing the severance of the test line and the ejection of the pipe No. respectively. The Charpy absorbed energy in HAZ and weld metal are enough high for low temperature service. Absorbed energy in base metal at –10 °C or –30 °C is over 250 J in both grades.0 mm wall thickness pipes pressurized to 20.9447. Each test results are shown in Figure 3(a) and 3(b). The premature severance invalidated the result from the west side. after running about 200 mm along the weld joint. Table 2 shows the Charpy impact test and DWTT result.3 Advantica’s JIP on X100 . thickness 19 mm. The pressurizing medium was lean natural gas containing approximately 98 % methane. Transition temperature is low enough to keep share area of DWTT at –10 °C more than 85 %. 4. SNAM and Europipe. The absorbed energy is the key factor to estimate arrestability of propagating ductile fracture in high-pressure gas pipelines. The two separate cracks. This research program included two full-scale fracture propagation tests on X100 pipes. nd In the 2 ECSC test. In this test.85 MPa corresponding to 72 % SMYS.1 ECSC Program on X100 A research program on behalf of ECSC (European Coal and Steel Community) had been carried out by a joint co-operation among CSM. 36” diameter 16. REVIEW OF FULL-SCALE BURST TESTS USING X100 OR X120 LINE PIPES 7). 36” diameter 16.0 m. the crack on the west side propagated through the initiation pipe but split in two different cracks at the girth weld with the adjacent pipe. The absorbed energy increases depending on sulfur content in X100 as the same manner of conventional grades. clearance 9. the crack on the west side propagated with a maximum speed of about 310 m/s and arrested in the last pipe No.6157 where it arrested at the end of the pipe. Test results are shown in Figure 3(c). 10) 4.5 mm) was placed in pipe 3E.2 ExxonMobil X120 Test In 2000.8) 4.0 mm wall thickness pipes pressurized to 18. The crack on the east side propagated with a maximum speed of about 300 m/s and entered in the following pipe No.1 MPa corresponding to 75 % SMYS. The two tests were undertaken in 1998 and in 2000 at the CSM Perdasdefogu test site in Italy.1 mm wall thickness pipes pressurized to 12. a loose sleeve crack arrestor (X65. On the east side.6058 where propagated and in pipe No.6 MPa corresponding to 68 % SMYS. The pressurizing medium was air. 56” diameter 19. st In the 1 ECSC test. the crack propagated through the initiation pipe and entered in pipe No. a full-scale fracture propagation test using X120 pipes was carried out by CSM on behalf of ExxonMobil. The initiated crack propagated to both ends of the test section indicating insufficient toughness in pipe for intrinsic arrestability at the test conditions and the ineffectiveness of the loose sleeve crack arrestor. The measured crack speed ranged between 350 and 280 m/sec.

36” diameter 16. rich natural gas and low temperature were acting simultaneously. Finally. and then decreased to 120-130 m/s on the last three pipes until finally stopping at the beginning of the crack arrestor. The two tests were undertaken in 2001 at Advantica’s Spadeadam test site in UK. On the east side. the crack on the west side propagated through the 1W pipe but bifurcated at the weld joint and arrested as a “ring-off”.1 test.64 MPa corresponding to 75 % SMYS. the crack propagated through the 1E and arrested in the 2E pipe. This project included two full-scale fracture propagation tests on X100 pipes. BP and TransCanada Pipelines. 36” diameter 13.3 MPa corresponding to 80 % SMYS. nd In the 2 DemoPipe test. the severance at weld joint between 4E and reservoir pipe was occurred and the section containing 4E and part of 3E pipe was left the trench. Details of this test remain confidential to Bp at this stage. This project included two full-scale fracture propagation tests on X100 pipes. 11) 4. it may be stated that the crack propagated through the test section and stopped at the crack arrestors .6 MPa corresponding to 69 % SMYS. after the initiation pipe.8831 and arrested in the middle of the second pipe No. Each test results are shown in Figure 4(a) and 4(b). the severance at weld joint between 2W and 3W pipe was occurred and the 2W pipe was left the trench. In this test. The ring-off lifted the section of pipes containing the initiation pipe. This test was executed under extreme severe conditions. However.0 mm wall thickness pipes pressurized 18. 36” diameter 15.0 mm wall thickness pipes pressurized to 22. 12) 4. On the west side. a new ECSC program with additional sponsorship from EPRG. The crack on the east side propagated quickly in the first pipe No. the severance in the 3E pipe was occurred and the section of pipes containing the initiation pipe was left the trench. fiberglass. Both tests were performed using lean natural gas containing approximately 98 % methane as the pressurizing medium. sponsored by Alliance Pipeline.5 Bp X100 Test In 2003. The premature ring-off invalidated the result from the west side. respectively.0 mm wall thickness pipes pressurized to 13. the crack speed decreased to 110 m/s in the first pipe No. the initiated crack propagated through two pipes and arrested in the third pipe on either side. respectively. BG. a full-scale fracture propagation test using X100 pipes was carried out by Advantica on behalf of Bp. In the Advantica’s JIP No. A Joint Industry Project had been undertaken by Advantica. Finally. 36” diameter 20.8834. Both tests used lean natural gas containing approximately 96 % methane as the pressurizing medium.0 m. thickness 40 mm). the initiated crack propagated to both ends of the test section indicating insufficient toughness in pipe for intrinsic arrestability at the test conditions. The two tests were undertaken in 2002 and in 2003 at the CSM Perdasdefogu test site. commenced in 2001 and the experimental activities concluded in 2003. composite reinforced arrestor and grouted sleeve arrestor were tested. Finally. since the high usage factor.0 MPa corresponding to 80 % SMYS.2 test. In the Advantica’s JIP No.0 mm wall thickness pipes pressurized to 19. a composite crack arrestor was positioned after the last pipe in the east side (length 2.4 DemoPipe Project on X100 The DemoPipe Project. st In the 1 DemoPipe test.8835 after the initiation pipe. Each test results are shown in Figure 4(c) and 4(d).

In order to predict the toughness requirement. the HLP Committee (High-strength Line Pipe Committee in ISIJ) developed a simulation model for the propagating ductile fracture in pipelines from a series of seven full-scale burst tests. instead of Charpy energy per unit area. and the interaction determines extent or distance of the propagating ductile fracture. In order to extend its applicability. and the corresponding toughness level is referred to as the arrest toughness by this Battelle Two Curve method. Therefore.indicating insufficient toughness in pipe for intrinsic arrestability at the test conditions and the effectiveness of the composite reinforced arrestor and the grouted sleeve arrestor (X80. 5. direct application of this model no longer correctly predicts the required toughness. the Battelle Two Curve method takes the comparison of curves expressing the variation of crack velocity and of gas decompression velocity with pressure. correction factors have been implemented to align the predicted toughness with results from full-scale experimental tests on higher grade pipe materials. 14) On the other hand. if an intersection exists between the crack and the gas decompression velocity curves. However. Another feature of this method is that the constant and exponent in the original Battelle’s crack velocity equation were recalibrated to accommodate pre-cracked DWTT energy per unit area directly to the index of material resistance. Pre-cracked DWTT energy can be converted to Charpy energy by the relationship developed by HLP Committee for the convenience of Charpy test. the pressure at the crack front will decreases and the crack will arrests. gas decompression velocity exceed crack velocity for all pressure levels. no further decrease of the pressure at the crack front is possible and crack will continue to propagate. the tangent condition between the two curves represents the boundary between arrest and propagation. pipelines are designed such that the toughness in pipe body is sufficient to intrinsically arrest a propagating crack. If no intersection exists between the crack and the gas decompression velocity curves. Traditionally. . Once a crack initiates. Thus. the Battelle 13) Two Curve method is the most popular predictive method.6 mm. so that a fast crack produces high pressure and a slow crack produces low pressure at the crack front. On the other hand. the pressure level where crack and gas decompression run together at the same velocity exists. as the strength of materials increases. The Battelle Two Curve method successfully accounted for a considerable quantity of full-scale experimental results on lower grade pipe materials (X70 and lower). Among these models. determination of the required toughness for the arrest of propagating ductile fracture has been the focus of much research over the past thirty years. SIMULATION OF PROPAGATING DUCTILE FRACTURE IN GAS PIPELINES The propagating ductile fracture is an interactive phenomenon between gas decompression and crack propagation. The crack chases after the decompression wave. This HLP method is a dynamic variant of the Battelle Two Curve method and be able to calculate 15) instantaneous crack velocity and propagating distance . clearance was filled with epoxy grouting). Many models and formulae have been developed for the determination of the required toughness. inner gas escapes from the ruptured portion of the pipeline and decompression wave proceeds in the pipeline. thickness 20.

increasing toughness arrangement is used in the full-scale fracture propagation test. for the full-scale crack propagation test on such high grade pipes. pipelines with even toughness arrangement are supposed and simulated fracture arrest distances (crack propagation distances) are plotted against Charpy energy of pipes used. as the gradient of the increasing toughness arrangement becomes steeper. Both the increase in pipe grade and the increase in design factor offer higher operating pressure but invite higher required toughness. since their arrest toughness may lie on the upper borderline. The unrealistic higher crack velocity produced from the unrealistic lower toughness initiation pipe is harmful to assess the crack arrestability of a pipeline. However. This tendency always exists in this kind of traditional style test. four cases of the simulated crack propagation for different pipe toughness arrangement are compared. the even toughness arrangement as same as a real pipeline instead of the increasing toughness arrangement and the judgment on the basis of the total propagation length is recommended. when we estimate the intrinsic arrestability of real pipelines on the basis of the full-scale burst test results. As the Charpy energy of pipes increases. in other words the crack propagation history. in case of X100 or X120 pipes. the required toughness may lie on the upper borderline and in this case the harmful aspect of the increasing toughness arrangement comes out. This effect mislead into underestimating the crack arrestability of a real pipeline with even toughness arrangement. As shown in this figure. Examples of simulated results of the propagating ductile fracture by HLP method are shown in Figure 5. In case of X100 or X120 pipes. This increasing toughness arrangement used to have an effective aspect that only one test can determine approximately the required toughness for the interested operating condition of a pipeline. Based on the . However. the fracture arrest distance decreases drastically showing that the fracture arrest distance is very sensitive to the toughness of pipes used. It is notable that transporting gas composition also changes the required toughness widely. This effective aspect works well so far as the required toughness lies in the middle between toughness ranges available. In particular. since the propagating ductile fracture is the interactive phenomenon between the gas decompression and the crack propagation. Traditionally. However. The gas compositions used for these simulations are shown in Table 3. it becomes a common view that intrinsic arrest of X100 or X120 in pipeline is hard to achieve and in this case the use of crack arrestors may be required. INTRINSIC ARRESTABILITY OF X100 OR X120 IN REAL PIPELINES Based on the full-scale burst tests. Figure 5(a) shows the effect of pipe grade and figure 5(b) shows the effect of design factor on crack propagation. Figure 6 shows the effect of the increasing toughness arrangement on crack propagation. crack will propagate longer and the arrest toughness in test tends to increase. the toughness of the initiation pipe controls the maximum crack velocity and has a strong effect on total crack propagation length. we have to take careful considerations on the difference between real pipelines and the full-scale fracture propagation tests. 6. And not so much attention seems to be paid to the effect of the toughness arrangement. In this figure. the fracture arrest toughness in test depends on the pipe toughness arrangement. this misleading may imply the misjudgment concerning the necessity of crack arrestors. Therefore. In this figure. as shown in figure 5(c).

low Pcm. Based on the simulated crack propagation length for different pipe toughness arrangement. The burst stress of these pipes reasonably coincides with tensile stress of base metal. Mn-Cu-Ni-Cr-Mo-Nb-V-B-Ti for X120. However. the fracture arrest toughness in test depends on the pipe toughness arrangement. The microstructures are fine bainitic ferrite with finer martensite/lower bainite for X100 and fine ausformed martensite and bainite for X120. and contains the following alloys: Mn-Cu-Ni-Cr-Mo-Nb-Ti for X100. since the propagating ductile fracture is the interactive phenomenon between the gas decompression and the crack propagation. as shown in figure 6. Traditionally. HLP method suggest the possibility of intrinsic arrest of X100 or X120 pipes in real pipelines. increasing toughness arrangement is used in the full-scale fracture propagation test. 7. . full-scale fracture propagation tests using X100 or X120 pipes are reviewed and analyzed.simulated crack propagation for even toughness arrangement. The steels are low carbon. The developed pipes have excellent combinations of strength and toughness in both base metal and heat affect zone of the longitudinal seam weld. HLP method reveals the possibility of intrinsic arrest of developed X100 or X120 pipes in real pipelines with even toughness arrangement. To estimate the arrest energy against propagating ductile fracture in X100 or X120 gas pipelines. SUMMARY API grade X100 and X120 grade UOE pipes for use in gas transmission pipelines have been successfully developed. The results of mechanical tests conducted on prototype pipes were satisfactory according to the target properties for X100 and X120 pipe.

.. Construction of Ultra High Strength Steel Pipelines. S. ISOPE. H. A. Developments in Fracture Control Technology for Gas Pipelines Utilising High Strength Steels. 11.. T. 4 International Symposium on Computational Technologies for Fluid/Thermal/Chemical Systems with Industrial Applications. Tensile properties and deformation behavior at the weld position of X100 line pipe steel. . J. Macia. and Maxey.. Belgium. G. Eiber. Calgary. Y. Petersen.. H. and Harris. Kawaguchi. Spinelli. G. Barsanti.. J. M. 14.. M. Koo. G. Final Report on Fracture Control Technology for Natural Gas Pipelines. and Bauer J. Demofonti. A. H.. J. Papka. Endo. P. Matsumoto. Demofonti.. G. 381-388. and Petersen. Bangaru. Takagi.. Yamamoto. L. 4th International Pipeline Conference. and Yoshimura. Papka. Ikeda. C. L. H.. M. P. 3 Int. J. M. Demofonti. 287-298. D. Tokyo. L. (2004). Offshore and Polar Engineering Conference. and Harris. Shiwaku.. th Development and Mechanical Properties of X120 Linepipe. 3. S. 69-88.. S. Pipe Dreamer’s Conference. Yokohama. P. (2002).. and Millwood.. 50-59. J. Belgium. (2002). W. Papka. C. Interactive Simulation of th Gas Decompression and Crack Propagation in Natural Gas Transmission Pipelines. Fonzo. H. Hamada.... ASME. Macia. L. A. Di... ISIJ International. (2003). H. 83-94.. G. Matsumoto Y.. Development of Large rd Diameter Pipe in X100. 6. W.. Ostend. T. Stevens.. 36-42.beyond X80. Columbus.. D. 12. Symposium on High Performance Materials in Offshore Industry. Takeuchi. Biagio. M. 5. and Luton. H. H. Crack arrestor design by finite element analysis for an X100 gas transportation pipeline. The prospect of high-grade steel pipe for gas pipelines.. 8. C.. Makino. M. Hillenbrand. N. Hammada. Barbas. H. Yamamoto. J. A. D.. Vancouver . Mannucci. Shitamoto. Burgge. T. Niederhoff K. (1993). Brugge. M. Bangaru. D. 4. Mannucci. Honolulu. L. C.. Prediction for Crack Propagation and Arrest of Shear Fracture in Ultra-high Pressure Natural Gas Pipelines. (2002). 509-520. Kubo. 3 Int. S. Fracture propagation resistance evaluation of X100 TMCP steel pipes for high-pressure gas transportation pipeline using full-scale burst tests. Volume II. D. Salvini. G. Andrews. Di.. 307-321. J. Makino. and Komizo. 9. 13 Int. High strength steel. G. Pipe Dreamer’s Conference. and Luton. S. 4th Internal Conference on Pipeline Technology. and Batte. Makino. G. and Edwards... 15. I. 467-482.. A.. Okaguchi. Biagio.. N. (2003). 3rd International Pipeline Technology Conference.. G. Ostend. P. REFERENCES rd 1. G. Y. Demofonti.. Mannucci. Mannucci. A. V. (2002).. S. Stevens. R. (2004). Lewssem. Hillenbrand. Full-Size Testing and Analysis of X120 Linepipe. W. S. Petersen. Large Diameter X100 Gas Linepipes : Fracture Propagation Evaluation by Full-Scale Burst Test. Yokohama. A... Honolulu.. M.. A. Kawaguchi. (2000).. T. and Okaguchi. T. Y. Fairchild. (2001). M. J. Okaguchi. IGU 22nd World Gas Conference... 10. ISOPE. Pipeline Technology Conference. (2000). Yokohama. Inoue. 41. 2. Fairchild. Pipeline Technology Conference. Hillenbrand. M. 7.. Belgium. J. D. M.. Fairchild. D. (2003). A.. Volume I. Pipe Dreamer’s Conference. BATTELLE. Hammond.. R.. D. D.. S. 469-482. (2002). Y. S. I. G. Takeuchi. Fujino. and Hillenbrand. Macia. Barsanti. (2000). Knauf. G. 185-202. H. Possible Use of New Materials for High Pressure Linepipe Construction: An Opening on X100 Grade Steel. G. Y. V. W. 289-305. Stevens. H. and Machida S. S..T. 483-500 13. N. 4th Internal Conference on Pipeline Technology. Bubenik.

6: Simulated results for the effect of pipe toughness arrangement on crack propagation Table 1: Tensile properties (Base metal: Transverse direction) and burst stress of the X100 and X120 linepipe Table 2: Toughness of the X100 and X120 linepipe (Transverse direction) Table 3: Gas compositions for analysis (mol%) . TITLE OF FIGURES/TABLES Fig.: 1100 °C.Cu-Ni-Cr-Mo-Nb-Ti boron free steel and boron bearing steel (Reheating temp.5: Simulated results of the propagating ductile fracture by HLP method (pipelines with even toughness arrangement are supposed) Fig.2: Effect of cooling finishing temperature on tensile strength and Charpy impact properties of low C-Cu-Ni-Cr-Mo-Nb-V-T-B steel plates (laboratory scale test: 16mm thickness) Fig.3: Results of full-scale burst tests using X100 or X120 pipes -1 Fig.4: Results of full-scale burst tests using X100 or X120 pipes -2 Fig.1: CCT diagram after hot deformation of low C-Mn. Deformatio n: 900 °C 50 % compression) Fig.

1: CCT diagram after hot deformation of low C-Mn. Deformatio n: 900 °C 50 % compression) -20 1100 Condition for stable production B steel -40 1000 TS (MPa) X120 900 -60 B free steel TS 800 X100 -80 vTs ( C) 700 vTs -100 600 500 -120 0 100 200 300 400 500 Cooling finishing temperature ( C) Fig.2: Effect of cooling finishing temperature on tensile strength and Charpy impact properties of low C-Cu-Ni-Cr-Mo-Nb-V-T-B steel plates (laboratory scale test: 16mm thickness) . FIGURES Cooling rate B added steel 25K/s B free steel 800 Temperature / C Ferrite 600 Bainite 400 Martensite 200 5 10 50 100 5001000 Cooling time /s Fig.Cu-Ni-Cr-Mo-Nb-Ti boron free steel and boron bearing steel (Reheating temp.: 1100 °C.

4: Results of full-scale burst tests using X100 or X120 pipes -2 .2 Test results Pipe 4W 3W 2W 1W Initiator 1E 2E 3E 4E CharpyV(J) 290 256 198 174 168 185 214 294 320 (ejected) (ejected) WEST EAST (Severance) (Severance) Propagation distance : 16m 25m (c) 1st DemoPipe Test results Pipe 8808 8795 8797 8786 8781 8783 8780 8799 8776 CharpyV(J) 291 249 237 215 193 228 223 258 355 WEST EAST Propagation distance : over45m over45m (d) 2nd DemoPipe Test results Pipe 8824 8826 8834 8831 8837 8835 8839 8836 8851 CharpyV(J) 267 240 252 247 211 206 223 249 257 (Composite Crack Arrestor) WEST EAST Propagation distance : abt.1 Test results Pipe 4W 3W 2W 1W Initiator 1E 2E 3E 4E CharpyV(J) 264 250 184 150 117 126 179 254 256 (ejected) WEST EAST (Severance) Propagation distance : 35m 34m (b) Advantica's JIP No.18.3: Results of full-scale burst tests using X100 or X120 pipes -1 (a) Advantica's JIP No. (a) 1st ECSC Test results Pipe 6020 6083 6129 6113 6058 6157 6061 CharpyV(J) 271 245 200 151 170 263 284 (ejected) WEST EAST Propagation distance : (Severance) 25m (b) 2nd ECSC Test results Pipe 9447 9458 9460 9461 9456 9457 9446 CharpyV(J) 297 252 202 165 259 253 274 WEST EAST Propagation distance : 26m 8m (c) ExxonMobil X120 Test results Pipe 5W 4W 3W 2W 1W Initiator 1E 2E 3E CharpyV(J) 270 278 276 225 223 151 216 215 226 (Loose Sleeve Crack Arrestor) WEST EAST Propagation distance : Over55m Over35m Fig.5m Over44m (Arrest at a crack arrestor) Fig.

30MPa) (b) Effect of (17.90MPa) crack propagation 10℃ "S gas".5: Simulated results of the propagating ductile fracture by HLP method (pipelines with even toughness arrangement are supposed) (a) Test result (Advantice' JIP No.04 YS(MPa) 836 821 844 836 834 876 839 846 782 TS(MPa) 850 832 856 849 853 919 897 887 856 (c) Pipe toughness arrangement (CharpyV(J)) Pipe 4W 3W 2W 1W Initiator 1E 2E 3E 4E Gradient Arrangement (1) 350 300 250 200 150 200 250 300 350 +50J/Pipe Arrangement (2) 320 285 250 215 180 215 250 285 320 +35J/Pipe Arrangement (3) 290 270 250 230 210 230 250 270 290 +20J/Pipe Arrangement (4) 250 250 250 250 250 250 250 250 250 0J/Pipe (d) Simulated results by HLP method Crack vel.6: Simulated results for the effect of pipe toughness arrangement on crack propagation . 50 X100.72SMYS. Cv(J) Fig.80SMYS 100 0. 16mmWT 0 100 200 300 "C gas" (c) Effect of 100 "S gas" gas composition on "L gas" crack propagation X100. 36inchOD.02 10.48MPa) 10℃ "S gas". (m/sec) Crack vel. 50 0. 36inchOD.72SMYS. X120 (a) Effect of 100 X100 (20.72SMYS (19. L(m) 50 0. Fracture arrest distance in one side. 16mmWT 0 100 200 300 Full-size Charpy energy of pipes used.37MPa) (13.2 Test) (ejected) (ejected) WEST EAST (Severance) (Severance) Propagation distance : 16m 25m (b) Pipe data Pipe 4W 3W 2W 1W Initiator 1E 2E 3E 4E Length(m) 10.08 11.01 11. (m/sec) 500 500 400 Pipe toughness arrangement : 400 300 300 200 (1) 200 (1) 100 100 (2) (3) (4) (4) (3) (2) 0 0 50 40 30 20 10 0 10 20 30 40 50 Propagating distance (m) Propagating distance (m) Fig.60SMYS crack propagation (14.85MPa) pipe grade on X80 (17. 36inchOD.02 10.00 10.00 10.03 11.37MPa) design factor on 0. 16mmWT 0 100 200 300 0.07 11.

00 4.02 0.70 3. ----. ----.50 “C gas” 89.97 API strip speci.50 0.50 0.56 0.00 2.47 0. ----- “S gas” 92.50 0.106 0.57 4.81 885MPa 1029MPa 0. TABLES Grade X100 X120 Tensile property YS TS Y/T ratio YS TS Y/T ratio Round bar speci.00 3. ----.00 0.15 0. 803MPa 863MPa 0.050 ----.075 0.86 Burst stress 856MPa 1056MPa Table 1: Tensile properties (Base metal: Transverse direction) and burst stress of the X100 and X120 linepipe X100 X120 Charpy impact property (at -10°C) DWTT Charpy impac t property (at -30°C) DWTT BaseMetal Weld HAZ SA-10°C BaseMetal Weld HAZ SA-20 °C 261J 132J 154J 97% 262J 88J 174J 88% Table 2: Toughness of the X100 and X120 linepipe (Transverse direction) Gas CH4 C2H6 C3H8 iC4H10 nC4H10 iC5H12 nC5H12 nC6H14 >=nC7H16 N2 CO2 “L gas” 96.30 0.100 0.72 Table 3: Gas compositions for analysis (mol%) .026 0.80 0.033 0.03 ----.24 0. ----. 686MPa 842MPa 0.93 993MPa 1020MPa 0.050 0. 0.