0

ASME Conference Presenter Attendance Policy and Archival Proceedings

2014;():V003T00A001. doi:10.1115/IPC2014-NS3.
FREE TO VIEW

This online compilation of papers from the 2014 10th International Pipeline Conference (IPC2014) represents the archival version of the Conference Proceedings. According to ASME’s conference presenter attendance policy, if a paper is not presented at the Conference, the paper will not be published in the official archival Proceedings, which are registered with the Library of Congress and are submitted for abstracting and indexing. The paper also will not be published in The ASME Digital Collection and may not be cited as a published paper.

Commentary by Dr. Valentin Fuster

Materials and Joining: Crack Propagation and Arrest

2014;():V003T07A001. doi:10.1115/IPC2014-33042.

This work addresses a two-parameter description of crack-tip fields in bend and tensile fracture specimens incorporating the evolution of near-tip stresses following stable crack growth with increased values of the crack driving force as characterized by the J-integral. The primary objective of this study is to assess the coupled effects of geometry and ductile tearing on crack-tip constraint, as characterized by the JQ theory, to correlate fracture behavior in circumferentially cracked reeled pipes and common fracture specimens. 3-D finite element computations including stationary and growth analyses were conducted for 3P SE(B) and clamped SE(T) specimens having different notch depth (a) to specimen width (W) ratio. Additional 3-D finite element analyses were also performed for circumferentially cracked pipes with a surface flaw having different crack depth (a) over pipe wall thickness (t) ratios. A cell methodology to model Mode I crack extension in ductile materials was utilized to describe the evolution of J with the evolving near-tip opening stresses. Laboratory testing of an API 5L X70 steel using deeply cracked C(T) specimens was used to measure the crack growth resistance curve for the material and to calibrate the cell parameter defined by the initial void fraction, f0. The present results provide further understanding of crack growth resistance measurements in pipeline steels using SE(T) and SE(B) specimens while eliminating some restrictions against the use of shallow cracked bend specimens in defect assessment procedures.

Commentary by Dr. Valentin Fuster
2014;():V003T07A002. doi:10.1115/IPC2014-33047.

The influence of crack speed on dynamic fracture toughness of pipeline steel has been observed in some recent tests, although it is still a challenge to obtain a specific relationship between dynamic fracture toughness and crack speed due to the expensive costs of experiments. Meanwhile, the understanding of the dependence of fracture toughness on crack speed is critical for material selection and crack-arrest design in high-strength steel pipelines. The present work develops a strain rate-dependent cohesive zone model and related finite element model to analyze speed-dependent dynamic fracture of pipeline steels observed in recent drop-weight tear tests. Different than most of existing cohesive zone models, the traction-separation law of the present model considers the role of rate of separation, and a strain rate-dependent elastic-viscoplastic constitutive model is employed for the bulk material. The speed-dependences of crack-tip-opening angle (CTOA) and energy dissipation observed in experiments are reproduced in our simulations for crack speed up to 150 m/s. A remarkable feature of the present work is that the present rate-dependent model can predict speed-dependent fracture as a consequence of the strain rate effect even when all fixed material parameters are speed-independent. These results suggest that the strain rate effect in the bulk material could be largely responsible for the speed-dependent dynamic fracture of pipeline steels, and the present rate-dependent model could be used to simulate dynamic fracture of pipeline steels especially when experiments are difficult or too expensive.

Commentary by Dr. Valentin Fuster
2014;():V003T07A003. doi:10.1115/IPC2014-33080.

Two full-scale fracture propagation tests have been conducted using dense phase carbon dioxide (CO2)-rich mixtures at the Spadeadam Test Site, United Kingdom (UK). The tests were conducted on behalf of National Grid Carbon, UK, as part of the COOLTRANS research programme.

The semi-empirical Two Curve Model, developed by the Battelle Memorial Institute in the 1970s, is widely used to set the (pipe body) toughness requirements for pipelines transporting lean and rich natural gas. However, it has not been validated for applications involving dense phase CO2 or CO2-rich mixtures. One significant difference between the decompression behaviour of dense phase CO2 and a lean or rich gas is the very long plateau in the decompression curve.

The objective of the two tests was to determine the level of ‘impurities’ that could be transported by National Grid Carbon in a 914.0 mm outside diameter, 25.4 mm wall thickness, Grade L450 pipeline, with arrest at an upper shelf Charpy V-notch impact energy (toughness) of 250 J. The level of impurities that can be transported is dependent on the saturation pressure of the mixture. Therefore, the first test was conducted at a predicted saturation pressure of 80.5 barg and the second test was conducted at a predicted saturation pressure of 73.4 barg.

A running ductile fracture was successfully initiated in the initiation pipe and arrested in the test section in both of the full-scale tests.

The main experimental data, including the layout of the test sections, and the decompression and timing wire data, are summarised and discussed.

The results of the two full-scale fracture propagation tests demonstrate that the Two Curve Model is not (currently) applicable to liquid or dense phase CO2 or CO2-rich mixtures.

Commentary by Dr. Valentin Fuster
2014;():V003T07A004. doi:10.1115/IPC2014-33121.

A fracture control plan is often required for a gas transmission pipeline in the structural design and safe operation. Fracture control involves technologies to control brittle and ductile fracture initiation, as well as brittle and ductile fracture propagation for gas pipelines, as reviewed in this paper. The approaches developed forty years ago for the fracture initiation controls remain in use today, with limited improvements. In contrast, the approaches developed for the ductile fracture propagation control has not worked for today’s pipeline steels. Extensive efforts have been made to this topic, but new technology still needs to be developed for modern high-strength pipeline steels. Thus, this is the central to be reviewed.

In order to control ductile fracture propagation, Battelle in the 1970s developed a two-curve model (BTCM) to determine arrest toughness for gas pipeline steels in terms of Charpy vee-notched (CVN) impact energy. Practice showed that the BTCM is viable for pipeline grades X65 and below, but issues emerged for higher grades. Thus, different corrections to improve the BTCM and alternative methods have been proposed over the years. This includes the CVN energy-based corrections, the drop-weight tear test (DWTT) energy-based correlations, the crack-tip opening angle (CTOA) criteria, and finite element methods. These approaches are reviewed and discussed in this paper, as well as the newest technology developed to determine fracture arrest toughness for high-strength pipeline steels.

Commentary by Dr. Valentin Fuster
2014;():V003T07A005. doi:10.1115/IPC2014-33133.

One of the major research topics in the area of gas pipeline fracture control is the suitability of using Charpy energy for ductile fracture control for modern and/or high strength line pipes. A common understanding is that, for pipe body crack self-arresting, the deviation of the actual required Charpy energy from those predicted using the traditional procedure of Battelle Two-Curve Method (TCM) is getting larger with higher strength pipes. DWTT is being paid more attention to because of its larger and full thickness specimen that can better capture the fracture process than a Charpy specimen does. Previous work at TransCanada indicated that various fracture speeds can be achieved in DWTT specimens and it is the steady-state fracture speed that is representative to the actual fracture propagation in a gas pipeline. It has also been found that the steady-state fracture toughness, in terms of either fracture energy or CTOA, is fracture speed dependent with lower fracture toughness for higher fracture speeds. Previous analysis also indicated by considering the speed dependent toughness, better predictions can be obtained for both self-arresting fracture toughness requirement and the fracture propagation speed. Previous DWTT fracture toughness data published by the authors exhibited a strong speed dependence and it was demonstrated that if the actual speed dependence is plugged into the modified TCM, both the fracture toughness and fracture speed would be over predicted. The assumption was that the original TCM was calibrated using pipe fracture data that also had speed dependent fracture toughness but the speed dependence was less strong than those for the modern pipes.

This paper presents an updated DWTT fracture dataset that expands the previously published data by adding high speed DWTT test results of modern line pipe steels with a range of grades X70-X100 and three old vintage pipe materials that is representative to the pipes that were used for the original TCM testing and calibration. The toughness data for the low grade pipes also shows speed dependence which purports the previous assumption.

Commentary by Dr. Valentin Fuster
2014;():V003T07A006. doi:10.1115/IPC2014-33141.

Transition welds joining pipe segments of unequal wall thickness are typically designed using back-bevel butt welds in accordance with industry recommended practices. An alternative approach, for joining transition pieces, would be the use of a counterbore-taper design, which has been successively utilized by TransCanada PipeLines.

In comparison with the back-bevel joint design, the counterbore-taper design provides a simple geometry that facilitates the welding process for joints of unequal wall thickness, improves the NDT quality and reliability, and increases the process efficiency for welding and NDT tasks. The counterbore-taper design reduces the effect of stress concentrations at the weldment and enhances fatigue life.

A parameter study, using continuum based finite element methods, was conducted to comparatively examine the mechanical performance of a pipe joint, using back-bevel and counterbore-taper designs, with unequal wall thickness and different material grade. The parameters examined include pipe diameter, D/t ratio, axial force and moment. The numerical study assessed the mechanical stress response, including stress path, initial yield and onset of plastic collapse, for back-bevel and counterbore-taper joint designs.

Based on these preliminary investigations, the performance of each transition joint design was evaluated and guidance on the selection of the joints design method was provided.

Commentary by Dr. Valentin Fuster
2014;():V003T07A007. doi:10.1115/IPC2014-33143.

The possibility of fracture propagation is a significant concern for pipelines transporting gases. The need to arrest a running fracture in a pipeline is paramount to the integrity and safety of the pipeline’s operation. The most commonly used method to predict pipe resistance to a ductile fracture is the Battelle Two Curve Method (BTCM). Recent full-scale fracture propagation tests have shown that the BTCM cannot give accurate predictions for modern high-toughness steels.

A Computational Fluid Dynamics (CFD) model has been developed to identify the limitations of the BTCM. The predictions of the developed model are in good agreement with the available experimental data. It has been found from the simulation that the pressure drops very fast in the initial stages of fracture propagation, leading to an intersection between the fracture curve and the decompression curve. However, the fracture speed does not remain constant at the speed corresponding to the intersection point, as assumed in the BTCM. It generally decreases, following the decompression curve as the fracture propagates. This observation indicates that the J-type fracture propagation curve adopted in the BTCM does not accurately represent the fracture propagation behaviour.

Commentary by Dr. Valentin Fuster
2014;():V003T07A008. doi:10.1115/IPC2014-33164.

The safety assessment of flawed pressurized pipes requires the knowledge of toughness properties which are usually provided in terms of impact energy from standard full-sized CVN notch specimens. For pipes with wall thickness less than 10mm different Charpy standards allow for the application of sub-sized specimens. However, it is still not fully clear how the impact energy from sub-sized specimens can be used to evaluate the fracture resistance of the pipes and how this energy correlates to the one from the full-sized specimen. Although different empirical correlations between sub-sized and full-sized specimens exist in the literature their validity is questionable since they are based on the results for older generation of steels. In the recent years the application of damage mechanics models has been promoted to assess the fracture behavior and deformation capacity of pipelines. The main advantage of these models can be found in their capability to link the damage evolution and the underlying stress/strain condition.

In this paper damage mechanics approach is applied to describe fracture behavior of X65 pipeline material. Within the damage mechanics approach, Gurson-Tvergaard-Needleman (GTN) model is considered to be adequate for the simulation of ductile fracture. For brittle fracture, GTN model is extended by a propagation criterion which examines if the cleavage fracture stress is reached by the maximal principal stresses. The model parameters are calibrated and verified by means of load-displacement curves obtained from instrumented impact tests on different sized CVN specimens. This damage model is subsequently employed to simulate ductile-brittle transition behavior.

Topics: Steel , Pipelines
Commentary by Dr. Valentin Fuster
2014;():V003T07A009. doi:10.1115/IPC2014-33169.

Fracture propagation control in gas transmission gas pipelines belongs to the major design requirements for safe operation at high internal pressures. However, the current tests such as Drop-Weight-Tear Test (DWTT) and full-scale West-Jefferson (WJ) test reach the limits of their applicability with respect to transition temperature evaluation for seamless quenched and tempered small diameter pipes reflecting nowadays alloying concepts related to mechanical properties. Hereby, different geometry and material effects are evident which might lead to misinterpretation and unreliability of testing results. This paper aims to discuss open issues addressed in the literature and in own experimental findings with respect to reliability and transferability of testing methods, fracture parameters and their representativeness of seamless quenched and tempered pipeline behavior. By applying damage mechanics approach, it is possible to quantify the prevailing stress state and thus to understand the mechanisms controlling specific fracture appearance (ductile or brittle). Furthermore, studies were performed with objective to quantify the effect of different parameters (geometry, material and loading) on the fracture performance of the pipeline. The results from these investigations will serve as a basis for a safe pipeline design against propagating fracture.

Commentary by Dr. Valentin Fuster
2014;():V003T07A010. doi:10.1115/IPC2014-33251.

Fracture mechanics methods for engineering assessment of acceptable flaw sizes in pipeline girth welds have been widely and successfully embraced by the pipeline industry. Advancements driven by strain-based design have identified elevated conservatism in assessment of material toughness by standardized high constraint fracture toughness test methods. Methods of reducing conservatism include the use of constraint adjustment factors or constraint-matched test specimens. Variants of the single edge-notched tensile (SENT) specimen have been widely reported as appropriate constraint-matched laboratory-scale specimens. This paper presents the results of SENT and SENB toughness testing of pipeline girth welds in both ductile and brittle/transitional temperature regimes.

Testing of 19.2mm weldments was conducted at room temperature (RT) and −5°C, with the intent of assessing the practicality of the single-specimen SENT methodology for low constraint fracture toughness assessment of typical high toughness production welds. Typical SENT specimens exhibited up to 50% higher upper shelf toughness results compared to SENB specimens. The majority of specimens failed E1820 crack straightness validity criteria, while the majority of specimens met E2818 (ISO 15653) criteria.

Testing of 10.4mm weldments was conducted on pipe known to exhibit low HAZ toughness (brittle pop-ins) at −5°C in the SENB configuration. SENT testing was conducted over temperatures spanning typical operating, design, and winter construction lowering-in temperatures (i.e. RT to −35°C), with the intent of investigating material sensitivity to brittle response under constraint-matched conditions. Brittle responses were observed in SENT specimens at both −20°C and −35°C, and ductile (upper shelf) behavior at −5°C and warmer; SENB specimens exhibited consistently brittle behavior at RT and −5°C, suggesting a HAZ transition temperature shift of at least −30°C for the constraint-matched test geometry.

Commentary by Dr. Valentin Fuster
2014;():V003T07A011. doi:10.1115/IPC2014-33326.

A full gas burst test at low temperature below −40°C was performed using a high frequency welded (HFW) linepipe with high-quality weld seam, “MightySeam®,” [1–4] in order to verify the applicability of the Drop Weight Tear Test (DWTT). Residual stress exists in the pipe body of HFW linepipe because the manufacturing method includes a sizing process. Therefore, it is necessary to clarify the difference between the arrestability in the DWTT without residual stress in the specimen and that in the full gas burst test with residual stress in the pipe body.

The full gas burst test is performed using a test pipe specimen in which a notch is introduced into the base material by an explosive cutter. In addition, a test pipe specimen with a notch introduced into the weld seam was used in this study because the developed HFW linepipe, “MightySeam®,” has excellent low-temperature toughness as a result of control of the morphology and distribution of oxides generated in the welding process by temperature and deformation distribution control. The Charpy transition temperature of “Mighty Seam®” was much lower than −45 °C.

Ductile cracks were initiated from the initial explosive notch, and these cracks were arrested after ductile crack propagation of about 1 m in base material on both sides. The fracture behavior was similar in appearance in the DWTT without residual stress and the full gas burst test with residual stress.

Topics: Low temperature
Commentary by Dr. Valentin Fuster
2014;():V003T07A012. doi:10.1115/IPC2014-33334.

The crack-tip opening angle (CTOA) has been investigated as a fracture propagation resistance parameter for prevention of fast ductile fracture in gas pipelines. A CANMET simplified single specimen CTOA method has been proposed as a mill test and is being applied to characterize critical CTOA (CTOAc) of typical pipe steels to develop a toughness database and improve the test method.

In this work, tests using standard machined V-notch and API pressed notch (PN) drop-weight tear test (DWTT) specimens at quasi-static and impact rates were performed on modern X65 and X70 pipe steels typical of those used for CO2 pipelines. The rotation factor of the X65 steel deduced from the deformed ligament geometry is equivalent to those of X70 to X100 steels. Pressed-notch DWTT specimens were successfully fractured in impact and yielded CTOAc values equivalent to those of V-notched specimens following the CANMET recommended practice for determination of CTOAc. The effect of loading rate on CTOAc between the quasi-static and impact rates (covering five orders of magnitude) is small or negligible, being within experimental scatter. This is in agreement with data in the CANMET database, except for a low-toughness X52 steel that showed an increase in CTOAc between quasi-static and impact loading rates. The effect of flattening on CTOAc was also investigated and is small or negligible for the large-diameter (at least 508 mm) pipes tested in this work. The results may be used to support and improve a proposed test method for determination of CTOAc being considered by an ASTM Task Group and currently being evaluated in a round-robin testing program.

Topics: Steel , Pipes
Commentary by Dr. Valentin Fuster
2014;():V003T07A013. doi:10.1115/IPC2014-33373.

The critical conditions of hydrogen content and residual stress in the high strength steel weld with the tensile strength level of over 980MPa were investigated. The critical hydrogen concentration for the cold cracking in the Y-grooved constraint weld joint was evaluated with intentionally introducing hydrogen gas. Residual stress conditions at the “root” portion in the weld joint were evaluated by the neutron diffraction technique. It was found that the root portion showed highest tensile stress of over 1110MPa in the transverse direction, and cracking occurred when the average hydrogen content was over 2ppm. In order to clarify the critical conditions of the principal tensile stress and local accumulated hydrogen concentration of the weld metal, the delayed fracture testing by using the notched round bar specimen with electrochemically hydrogen charged was conducted. It was seen that the cold cracking behavior in the Y-grooved weld joint was explained by the critical conditions of the maximum principal stress and the local accumulated hydrogen content obtained from the delayed fracture with the small specimen.

Commentary by Dr. Valentin Fuster
2014;():V003T07A014. doi:10.1115/IPC2014-33440.

The bulging factor for an external constant-depth axial surface crack in a pipe was calculated by 3D FE simulations. This was done in a manner consistent with Folias’s original work for the axial through-wall-cracked pipe bulging factor (MT), but was evaluated in the elastic to full-plastic conditions. The results demonstrated that the actual surface-cracked pipe bulging factor is considerably lower than the bulging factor empirically derived by Maxey/Kiefner (Mp) back in the 1970s. Based on the results of the present study, it is suggested that Mp function in the Ln-Secant equation is not truly a bulging factor for axial surface crack. Rather it is an empirically developed equation with many correction factors embedded in it to apply the Dugdale model for prediction of maximum pressure of axial surface-cracked pipes. However, due to this empiricism, this method becomes invalid (or overly conservative) when it is applied in predicting the crack-driving force using the J-based Ln-Secant equation.

Topics: Pipes , Surface cracks
Commentary by Dr. Valentin Fuster
2014;():V003T07A015. doi:10.1115/IPC2014-33449.

Single-Edge-Notch-Tension, SEN(T), specimens have been found to provide good similitude for surface cracks in pipes, where a surface-cracked structure has lower constraint condition than bend-bars and C(T). The lower constraint condition gives higher upper-shelf toughness values, and also a lower brittle-to-ductile transition temperature. Also, the SENT specimen eliminates concern of material anisotropy since the crack growth direction in the SENT is the same as in a surface-cracked pipe. While the existing recommended and industrial practices for SEN(T) have been developed based on assumption of monomaterial across the crack, their applicability for the evaluation of fracture toughness of heat-affected-zone (HAZ) is evaluated in this paper. When conducting tests on SEN(T) specimens with prescribed notch/crack in the heat-affected-zone (HAZ), the asymmetric deformation around the crack causes the occurrence of a combination of Mode-I (crack opening) and Mode-II (crack in-plane shearing) behavior. The extent of this mode mixity is dependent on the relative difference between the material properties of the adjacent girth weld and pipe base metals, as well as the amount of crack growth in the test. This mode mixity affects the measurement of the crack-tip-opening-displacement (CTOD) and evaluation of fracture mechanics parameter, J. The CTOD-R curve depicts the change in toughness with crack growth, in a manner similar to the J-R curve methodology. Observations also show a mismatch in the length of the crack growth that is measured on the fracture surface, attributable to the material deformation differences across the two adjacent materials (weld and base metals). This paper discusses the experimental observations of Mode-I and Mode-II behavior seen in tests of SEN(T) specimens with notch/crack in the HAZ and as the crack propagates through the weld/HAZ thickness. The paper addresses the issues related to and the changes needed to account for such behavior in the development of recommended practices or standards for SEN(T) testing of weld/HAZ. The effects of mode mixity in HAZ testing is critical to the development of crack growth resistance, CTOD-R and J-R curves employed in Engineering Critical Assessment (ECA) of pipelines.

Commentary by Dr. Valentin Fuster
2014;():V003T07A016. doi:10.1115/IPC2014-33457.

Since the late 1960s’ the Battelle Two-curve (BTC) model is the standard method applied in setting up design requirements with regard to the prevention of long-running ductile fracture in pipelines. It is a straightforward tool employing Charpy-V notch (CVN) toughness as key-measure for material resistance against crack propagation. On basis of pipe dimensions, material strength, and under consideration of decompression behavior of the transferred media, it enables to set up requirements for a minimum CVN toughness level to achieve crack arrest. Overall applicability of the BTC model is based on calibration of the underlying equations to a sound data-base, including both full-scale burst test results and small-scale laboratory testing data involving typical line-pipe grades at that period, i.e. up to grade X70 steels with below 100 J upper-shelf CVN toughness. Now over the last decades, mechanical behavior of line-pipe steels was improved significantly. Responding to market demands, higher grades were designed and also toughness levels were raised as outcome of R&D efforts within the steel industry. Unfortunately, stepping outside the original material data-base from BTC model calibration, this method did forfeit its reliability. At the beginning, mispredictions were mainly related to higher grade steels and elevated operating pressures. But more recent full-scale tests did reveal discrepancies in application of the BTC model also for so-called new vintage steels, i.e. grades actually being inside the original data base for model calibration but from current production routes.

With regard to applicability/reliability of BTC model based predictions for crack arrest, the origin of uncertainty has particularly been traced back to the involved material toughness measure. Nowadays, it is common sense that the CVN upper-shelf toughness value inadequately describes the resistance against running ductile fracture. More recent thoughts coherently argue towards closer involving stress-strain response and plastic deformation capacities of the material. On basis of results for grades X65, X80 and X100, the general relation between ductility and toughness is discussed. Finally, an elastic-plastic fracture mechanics related analytical approach is introduced which enables to quantify the resistance against ductile fracture propagation. The objective is to provide a reliable procedure for crack arrest prediction in line-pipe steels.

Commentary by Dr. Valentin Fuster
2014;():V003T07A017. doi:10.1115/IPC2014-33476.

The transition temperature behavior of pipes subjected to propagating fracture is assessed by means of Battelle Drop Weight Tear (BDWT) tests. These consist of notched specimens which are subjected to impact loading under a three point bend test configuration. The fracture surface is assessed to derive shear or brittle portions that are macroscopically visible. These have been shown to correspond well to the fracture surface of a pipe subjected to a propagating fracture. Historically, it is required for the test to either exhibit entirely ductile fracture or a combination of brittle and ductile provided that the test initiates in a brittle fashion with propagation in a ductile manner. Conversely, tests exhibiting ductile initiation with brittle or mixed brittle and ductile propagation are not acceptable to some standards/specifications. In recent times, this requirement has been softened in certain standards/specifications but it remains to be explained what these two diverging opinions are based upon and which one of the two is the correct one. This nominally unacceptable fracture mode has been termed inverse or abnormal fracture and is frequently observed with modern pipeline steels. Other than inverse fracture character, these specimens show every sign of highly ductile material being tested. The unanswered question is: does the brittle area reflect material properties or is it caused by the specific test conditions meaning that it is simply a testing issue? Furthermore, the reason to retain the requirement is not obvious and a procedure is missing on how to assess this type of specimen. The only possibility outlined in testing standards to avoid inverse fracture is the use of an alternative notch, the so-called Chevron notch that is supposed to facilitate brittle crack initiation. Mostly, the problem remains unsolved and further guidance is not given. Current research aims at avoiding invalid results by choosing different notches that could potentially be successful but also on checking the correspondence of results of testing inverse material to pipe behavior in West Jefferson tests. Alternatives discussed to avoid inverse fracture are, apart from the above named one, static pre-cracked notches and fatigue notches. Within this paper, test results of specimens notched with all of these notches are presented. None of these were successful in eliminating inverse fracture. Literature studies were conducted to understand the background of the requirement. Investigations of material behavior after having been subject to heavy pre-strain were carried out. The question is, on basis of the available evidence, which conclusions can be drawn and if these are sound enough to derive general guidance.

Commentary by Dr. Valentin Fuster
2014;():V003T07A018. doi:10.1115/IPC2014-33514.

Current line-pipe steels have significantly higher Charpy upper-shelf energy than older steels. Many newer line-pipe steels have Charpy upper-shelf energy in the 300 to 500J range, while older line-pipe steels (pre-1970) had values between 30 and 60J. With this increased Charpy energy comes two different and important aspects of how to predict the brittle fracture arrestability for these new line-pipe steels.

The first aspect of concern is that the very high Charpy energy in modern line-pipe steels frequently produces invalid results in the standard pressed-notch DWTT specimen. Various modified DWTT specimens have been used in an attempt to address the deficiencies seen in the PN-DWTT procedure. In examining fracture surfaces of various modified DWTT samples, it has been found that using the steady-state fracture regions with similitude to pipe burst test (regions with constant shear lips) rather than the entire API fracture area, results collapse to one shear area versus temperature curve for all the various DWTT specimens tested. Results for several different materials will be shown. The difficulty with this fracture surface evaluation is that frequently the standard pressed-notch DWTT only gives valid transitional fracture data up to about 20-percent shear area, and then suddenly goes to 100-percent shear area.

The second aspect is that with the much higher Charpy energy, the pipe does not need as much shear area to arrest a brittle fracture. Some analyses of past pipe burst tests have been recently shown and some additional cases will be presented. This new brittle fracture arrest criterion means that one does not necessarily have to specify 85-percent shear area in the DWTT all the time, but the shear area needed for brittle fracture arrest depends on the pipeline design conditions (diameter, hoop stress) and the Charpy upper-shelf energy of the steel. Sensitivity studies and examples will be shown.

Commentary by Dr. Valentin Fuster
2014;():V003T07A019. doi:10.1115/IPC2014-33553.

Addition of alloying elements can alter the properties of high-strength linepipe steel. Particularly the addition of Chromium and Molybdenum acts to suppress ferrite nucleation and promote the formation of acicular bainite microstructures and thereby increase the tensile properties of modern linepipe steel. While chemistry is a factor, welding parameters can also be influential and affect the HAZ toughness. The present work compares the effect of C, Cr, and Mo on the girth weld HAZ fracture toughness of X70 in identical welds.

Three pipes of size 48″ OD × 0.528″ WT with different combinations of C, Cr, and Mo were produced. Identical welding procedures were employed to produce two girth welds so that a low-C, Cr pipe (CE = 0.238) was joined to a high-C, Cr pipe (CE = 0.268) which in turn was joined to a low-C, Mo pipe (CE = 0.224). By evaluating the HAZ properties on either side of a weld, it was possible to accurately assess chemistry affects on HAZ properties.

These girth welds were subjected to different testing for the evaluation of girth weld HAZ impact and fracture toughness. These included all-weld metal and pipe body tensile testing, micro hardness testing of HAZ, microstructure analysis, Charpy V-notch testing of weld metal and HAZ, and CTOD testing of weld metal and HAZ at −5 °C and −20 °C. In addition, to investigate the transformation behaviour, Gleeble simulations of coarse-grain heat affected zone (CG-HAZ) were conducted using skelp samples, which were taken from the same coils as the pipe samples.

The results demonstrated that among the low and high carbon equivalent (CE) alloys, materials with low CE values showed better toughness properties. While among the low CE materials, the material with high Mo performed better in terms of toughness. No clear effect of weld position around the pipe circumference on the CTOD values was observed.

Commentary by Dr. Valentin Fuster
2014;():V003T07A020. doi:10.1115/IPC2014-33684.

The tensile properties of line pipe are usually determined using a flattened strap tensile sample which is obtained by cutting a long transverse sample from the pipe and then flattening it prior to machining the final tensile coupon. Although, several documents have been published to standardize this test, variability in the reported yield stress for the same material tested by different labs continues to be an issue particularly in high strength line pipe (X70 and X80).

Pipe properties are influenced by the pipe forming operations which introduce plastic strain into the steel. As well, the flattening of the tensile blank reverses the deformation and leads to Bauchinger effects which further alter the tensile properties of the material. There is no standard available for the flattening process and pipe manufacturers and operators continue to seek a best practice for the process. In addition, other factors such as placement of extensometer on a flattenend tensile specimen during the tensile testing have been considered a source of variation in the results.

Several projects were conducted to identify the source of variability and to standardize the flattening and testing process among the Evraz QA Labs. These initiatives included: a round robin tensile testing program in which tensile tests were performed on flat plates and subsequently on flattened strap specimens produced from a sister plate; examination of a 2-step flattening procedure against a 1-step method, and investigation of the extensometer placement (ID, OD or side mounted) on the recorded stress-strain behaviour.

The flattening process was found to be the main source of variability of yield stress. No real trend was observed resulting from extensometer position. Other testing practices such as specimen gripping, zeroing the load and positioning at the start of the test, and the dimensional variability within the reduced section of a specimen were also found to contribute to yield stress variability. Best practices for determination of yield stress using flattened strap tensile samples are discussed.

Topics: Yield stress
Commentary by Dr. Valentin Fuster
2014;():V003T07A021. doi:10.1115/IPC2014-33709.

Prevention of unstable ductile crack propagation is one of crucial issues in pipeline industry. Some numerical methods for predicting unstable ductile crack propagation/arrest have been developed, of which the Battelle Two-Curve Method (TCM) and the HLP method have been most widely used. Although these two methods are simple and useful, they contain empirical parameters based on the data of full-scale pipe burst tests. Therefore, their applicable range might be limited; their prediction accuracy reportedly drops when applied to the burst tests using new grade pipes and CO2 pipeline burst tests. To overcome their shortcomings, the authors developed a new model, hereafter called “UT model”, without empirical parameters. The UT model describes pipe deformation and gas decompression by one-dimensional partial differential equations, and judges crack propagation/arrest using dynamic energy balance. In addition, the UT model is fully-coupled model which means that the interaction among pipe deformation, gas decompression and crack propagation is considered. Also, soil backfill effect, which constrains pipe deformation, is incorporated into the UT model; kinetic energy of soil surrounding a pipe is dealt as added density of a pipe. In the present paper, the UT model is described in detail, and applied to natural gas pipe burst tests both with and without soil backfill.

Commentary by Dr. Valentin Fuster
2014;():V003T07A022. doi:10.1115/IPC2014-33712.

Current defect assessment procedures of large engineering structures, including pipeline systems and their welded components such as field girth welds, employ crack growth resistance curves in terms of J-resistance or CTOD-resistance curves. Standardized techniques for crack growth resistance testing of structural steels are based upon laboratory measurements of load-displacement records and adopt two related estimation formulas for fracture toughness values: 1) estimating J from plastic work based on crack mouth opening displacement (CMOD), and 2) determining the CTOD value from first evaluating the plastic component of J using the plastic work defined by the area under the load vs. CMOD curve and then converting it into the corresponding value of plastic CTOD. This work addresses an investigation on the relationship between J and CTOD for three-point SE(B) and clamped SE(T) fracture specimens based upon extensive numerical analyses conducted for crack configurations with varying crack sizes. These analyses include stationary and crack growth plane-strain results to determine J and CTOD for the cracked configurations based on load-displacement records. The numerical computations show strong similarities between the J-CTOD relationship for stationary and growth analysis with important implications for experimental measurements of CTOD-resistance curves. The study provides a body of results which enables establishing accurate relationships between J and CTOD for use in testing protocols for toughness measurements.

Commentary by Dr. Valentin Fuster
2014;():V003T07A023. doi:10.1115/IPC2014-33718.

Fracture toughness is an important material property in describing material resistance against fracture with a point value or in the format of a resistance curve. For ductile materials, the commonly used fracture parameters are the J-integral and the crack-tip opening displacement (CTOD, or δ). ASTM E1820 provides standard procedures for determining the JIc, δIc, J-R curve and δ-R curve using bending specimens with deep cracks. This usually leads to high crack-tip constraint conditions and conservative fracture resistance curves.

Actual cracks found in pipelines and welds are often shallow and dominated by tensile forces, resulting in low constraint conditions and elevated resistance curves. Thus the standard resistance curves can be overly conservative for a shallow crack. To obtain realistic fracture toughness values to meet the practical needs for pipelines, different test methods have been developed using a single edge-notched tension (SENT) specimen. This includes the multiple specimen method, the single specimen method, the J-R curve test procedure, and the δ-R curve test procedure. This paper presents a critical technical review of existing fracture toughness test methods and procedures using SENT specimens, with discussions on the toughness estimation equation, key parameter calibration, rotation correction, and test procedure limitation. Historical efforts related to the SENT testing and applications of ASTM fracture test standards to the SENT specimens are also reviewed briefly.

Commentary by Dr. Valentin Fuster

Materials and Joining: Pipe Manufacturing and Welding

2014;():V003T07A024. doi:10.1115/IPC2014-33069.

The challenging exploration conditions appearing in ultra deep offshore projects promoted the development of high strength linepipe steel grades with yield strength of 80 ksi and higher in recent years. With increasing strength more attention has to be paid to welding procedures to realise the required mechanical properties of the weld seam. The combination of demanding toughness requirements at low temperatures and adequate corrosion resistance of welded joints is a key for complex deep offshore riser and linepipe applications. The welding process was optimised by Vallourec with respect to heat input and preheating temperature for joining seamless quenched and tempered pipes in grade X80. A root welding strategy has been developed particularly with regards to sour service applications. Extensive mechanical test results including Charpy impact testing, hardness, CTOD and SSC testing will be presented. In addition Gleeble trials were carried out using different thermal cycles to simulate multilayer welding. The aim was to improve the understanding of the base material behaviour in the heat affected zone (HAZ) during welding. The microstructure was characterized by LOM, SEM and furthermore hardness and Charpy impact tests were executed. Based on the gathered knowledge and test results welding recommendations and welding strategies for high strength steel X80 seamless line pipes are deduced.

Topics: Steel , Welding
Commentary by Dr. Valentin Fuster
2014;():V003T07A025. doi:10.1115/IPC2014-33079.

To meet the increasing worldwide demand for natural gas, there is a need to safely and economically develop remotely located resources. Pipeline construction is a major activity required to connect these remote resources to markets. Such pipeline routes may cross areas containing geohazards such as discontinuous permafrost, active seismicity and offshore ice gouging. These pipelines may be subjected to longitudinal strains above 0.5%. To safely design pipelines for such conditions, a strain-based design (SBD) approach can be used in addition to conventional allowable stress designs (ASD).

Significant pipeline construction cost savings can be achieved with the use of higher strength steels (X70+) due to reduced pipe wall thicknesses (less steel) and faster girth welding. However, a robust welding technology for higher strength SBD pipelines is often a technology gap depending on the target level of longitudinal strain that needs to be accommodated, since such applications often demand excellent weld toughness at low temperatures (−15°C) and high tensile strength (>120ksi). This paper discusses the development of an enabling welding technology that offers a superior combination of strength and toughness compared to commercially available technologies.

Acicular ferrite interspersed in martensite (AFIM) has been previously identified as a useful high strength weld metal microstructure that can be applied in field pipeline construction. This paper describes how this microstructure has been used to create welds with excellent strength overmatch and good ductile tearing resistance for X80 SBD pipelines. This approach has been implemented for mainline, double-joining and repair welding applications. This paper describes the welding procedures, mechanical properties achieved, estimated strain capacities, and the results of a full-scale pipe strain capacity test.

Topics: Welding , Design , Pipelines
Commentary by Dr. Valentin Fuster
2014;():V003T07A026. doi:10.1115/IPC2014-33082.

In this study, defect formation mechanisms in the ERW (electric resistance welding) process for API pipe production were investigated. The results showed that defects observed in the weld joint of ERW pipe are a main factor in the deterioration of the mechanical properties of welded joints. From systematical research, it was clear that the crucial main defects of ERW pipe are caused by large inclusions with complex compositions after the steel making process and penetrators formed during ERW welding with excessive heat input condition. In order to guarantee the toughness of the weld joints, after theoretical and experimental considerations ERW pipe weld defect reducing methods can be recommended. According to previous test results, it was clear that the hook cracks caused by large inclusions were reduced by selecting the slag composition ratio of CaO and Al2O3 at the eutectoid point during steel making process. On the other hand, in case of the penetrator formation type defects, it was defined that by controlling of the heat input in the optimization range was the best solution for decreasing the penetrator formation.

Commentary by Dr. Valentin Fuster
2014;():V003T07A027. doi:10.1115/IPC2014-33099.

The general trend in oil and gas industry gives a clear direction towards the need for high strength grades up to X100. The exploration in extreme regions and under severe conditions, e.g. in ultra deep water regions also considering High Temperature/High Pressure Fields or arctic areas, becomes more and more important with respect to the still growing demand of the world for natural resources. Further, the application of high strength materials enables the possibility of structure weight reduction which benefits to materials and cost reduction and increase of efficiency in the pipe line installation process.

To address these topics, the development of such high strength steel grades with optimum combination of high tensile properties, excellent toughness properties and sour service resistivity for seamless quenched and tempered pipes are in the focus of the materials development and improvement of Vallourec.

This paper will present the efforts put into the materials development for line pipe applications up to grade X100 for seamless pipes manufactured by Pilger Mill. The steel concept developed by Vallourec over the last years [1,2] was modified and adapted according to the technical requirements of the Pilger rolling process. Pipes with OD≥20″ and wall thickness up to 30 mm were rolled and subsequent quenched and tempered. The supportive application of thermodynamic and kinetic simulation techniques as additional tool for the material development was used. Results of mechanical characterization by tensile and toughness testing, as well as microstructure examination by light-optical microscopy will be shown. Advanced investigation techniques as scanning electron microcopy and electron backscatter diffraction are applied to characterize the pipe material up to the crystallographic level. The presented results will demonstrate not only the effect of a well-balanced alloying concept appointing micro-alloying, but also the high sophisticated and precise thermal treatment of these pipe products.

The presented alloying concept enables the production grade X90 to X100 with wall thickness up to 30 mm and is further extending the product portfolio of Vallourec for riser systems for deepwater and ultra-deep water application [1, 3, 4].

Topics: Steel
Commentary by Dr. Valentin Fuster
2014;():V003T07A028. doi:10.1115/IPC2014-33101.

Various accessories such as buckle arrestors and J-lay collars are needed in some cases to successfully lay and secure an offshore pipeline on the sea bed. For such applications the using of high strength seamless pipes in Grade X70 and X80 with heavy wall are necessary. However, there is only small information regarding the welding procedure for such grades in heavy wall dimensions.

In comparison to steels used for lower strength level, the chemistry of high strength steel pipes includes higher amounts of micro-alloying elements as well as requires a more complex heat treatment. Due to the higher carbon equivalent these steel grades react more sensitive on heat input during welding. Consequently, the range of welding parameters which ensure suitable mechanical properties has to be adapted.

This article presents the results of weldability trials carried out on seamless API X80 heavy wall (> 50mm) line pipe. The welding trials were performed using different preheating temperatures and heat inputs followed by microstructure investigations and mechanical tests of the multilayer welds. The sour gas resistance has to be demonstrated by SSC-tests because it stays challenging to guarantee values below 250 HV10.

Commentary by Dr. Valentin Fuster
2014;():V003T07A029. doi:10.1115/IPC2014-33109.

Critical performance of modern high strength linepipe is related to the ability of the steel to maintain mechanical properties in the weld heat affected zone (HAZ). The region most susceptible to mechanical property degradation is the coarse grained HAZ, however in multipass welds, the intercritically reheated CGHAZ (ICCGHAZ) also presents challenges to maintain toughness.

Currently Ti is employed to minimise austenite grain coarsening through the grain boundary pinning action of TiN precipitates. This is effective because of the high thermal stability of TiN but control of the precipitate size distribution is very much dependent on alloy design and processing conditions to ensure final weld HAZ properties, particularly toughness. This can be difficult to maintain and alternative methods are required to further improve performance of the weldments.

It is now evident that increased additions of Nb in modern high temperature processed (HTP) steels have demonstrated increased control of HAZ microstructures with improved fracture toughness [1, 2]. The present paper details the microstructure - property relationship of two pipe steel grades with different alloy designs. Evaluation of the critical CGHAZ was achieved by simulation techniques, calibrated using real weld thermal cycles, to determine the influence of alloy design and specifically level of Nb on weld zone properties.

The results reveal that the fracture toughness of the simulated CGHAZ in the HTP steel is superior to that of a conventional microalloyed pipeline steel grade. Toughness was related to the distribution of martensite-austenite (M-A) constituent and the effective grain size which appeared to correspond to prior austenite grain size as evidenced by examination of cleavage facet size (CFS) on fractured Charpy specimens.

Topics: Steel , Pipelines
Commentary by Dr. Valentin Fuster
2014;():V003T07A030. doi:10.1115/IPC2014-33122.

Structural integrity of submarine risers and flow lines transporting corrosive and aggressive hydrocarbons represents a key factor in operational safety of subsea pipelines. Advances in existing technologies favor the use of C-Mn steel pipelines (for example, API X65 grade steel) either clad or mechanically lined with a corrosion resistant alloy (CRA), such as Alloy 625, for the transport of corrosive hydrocarbons. However, while cost effective, specification of critical flaw sizes for their girth welds become more complex due to the dissimilar nature of these materials. In particular, effective fracture assessments of undermatched girth welds remain essential to determine more accurate acceptable flaw sizes for the piping system based upon engineering critical assessment (ECA) procedures. This work focuses on development of an evaluation procedure for the elastic-plastic crack driving force (as characterized by the J-integral) in pipeline girth welds with circumferential surface cracks subjected to bending load for a wide range of crack geometries and weld mismatch levels based upon the GE-EPRI framework. The study also addresses the effects of an undermatching girth weld on critical flaw sizes for a typical clad pipe employed in subsea flowlines having an Alloy 625 girth weld. The extensive 3-D numerical analyses provide a large set of solutions for J in cracked pipes and cylinders with mismatched girth welds while, at the same time, gaining additional understanding of the applicability of ECA procedures in welded cracked structural components.

Commentary by Dr. Valentin Fuster
2014;():V003T07A031. doi:10.1115/IPC2014-33150.

Demand for CRAs (Corrosion Resistant Alloys) clad steel is getting increased for pipeline application of oil and gas industry because of economic advantage over solid CRAs. CRAs clad steel consists of a CRAs layer for corrosion resistance and a carbon steel for mechanical properties. Nickel based Alloy625 is known to be suitable for harsh environmental condition such as high temperature and high pressure H2S (hydrogen sulfide) condition.

In this paper, the corrosion resistance of Alloy625/X65 clad steel plate for pipe produced by TMCP (Thermo-Mechanical Control Process) was investigated. TTP (Time - Temperature - Precipitation) and TTS (Time - Temperature - Sensitization) diagram of Alloy625 indicated precipitation nose, e.g. M6C and M23C6 which would cause deterioration of corrosion resistance. TMCP enable Alloy625 to avoid long time exposure to the precipitation nose. In Huey test, the corrosion rate in TMCP was almost the same as that of solution treated Alloy625 and smaller than that in Q-T (Quench and Temper). In ferric chloride pitting test, no pitting was observed in Alloy625 layer of TMCP type clad steel. In addition, the corrosion test simulating service environment using autoclave apparatus was conducted under the condition of 0.39MPa H2S - 0.53MPa CO2 - Cl solution at 200°C. Alloy625 clad steel produced by TMCP showed neither SSC (Sulfide stress corrosion cracking) nor crevice corrosion. All the mechanical properties of base carbon steel satisfied API 5L grade X65 specification by optimizing TMCP conditions. It is notable that 85% SATT of DWTT was below −10 °C. Thus, Alloy625/X65 clad steel plate for pipe produced by TMCP with both superior corrosion resistance and low temperature toughness has been developed.

Topics: Steel , Pipes
Commentary by Dr. Valentin Fuster
2014;():V003T07A032. doi:10.1115/IPC2014-33206.

Based on the appreciable progress being made in quality control and assurance technology for the electric resistance welding process, the number of applications for high-frequency electric resistance welded (HFW) linepipe in highly demanding, severe environments, such as offshore and sour environments, has gradually increased. Resistance to hydrogen-induced cracking (HIC) is the most important property for a linepipe to possess for use in sour environments. However, resistance to HIC, especially along the longitudinal weld seam, has not yet been fully related to metallurgical factors.

In this study, to clarify the effects of inclusions on the sour resistance properties of X60- to X70-grade steels, their resistances to HIC were numerically simulated. For the simulation, the steels were assumed to have a yield strength of 562 MPa and a tensile strength of 644 MPa. To estimate the effect of nonmetallic inclusions, a virtual inclusion was situated at the center of a 10-mm-thick HIC test specimen. Tests were performed using NACE test solution A.

The crack propagation rate was calculated as a function of the content of diffusible hydrogen, the diameter of the inclusion, and the fracture toughness of the matrix after hydrogen absorption. In the propagation calculation, the resistance to chemical reactions at the interface of the inclusion matrix was also considered to be a delaying factor. By assuming a resistance to chemical reactions at the interface, the crack propagation rate could be fitted to the actual HIC propagation rate.

Based on the numerical simulation results, HFW linepipe with a high-quality weld seam was developed. Controlling the morphologies and distributions of oxides generated during the welding process is the key factor for improving the resistance to HIC. Using a combination of optimized chemical composition, microstructure and oxide content, the weld seam of the developed X70-grade HFW steel pipe showed excellent resistance to HIC.

Commentary by Dr. Valentin Fuster
2014;():V003T07A033. doi:10.1115/IPC2014-33219.

The crack tip opening displacement (CTOD)-based fracture toughness has been widely used for structural integrity assessment and strain-based design of oil and gas pipelines. The double-clip on gauge method has been used to experimentally determine CTOD. In this study, three-dimensional (3D) finite element analyses of clamped single-edge tension (SE(T)) specimens are carried out to investigate the accuracy of the CTOD evaluation equation associated with the double-clip on gauge method. The analysis considers SE(T) specimens with ranges of crack lengths (0.3 ≤ a/W ≤ 0.7) and specimen thickness (B/W = 0.5, 1 and 2). Based on the analysis results, a modified CTOD evaluation equation based on the double-clip on gauge method is developed to improve the accuracy of the CTOD evaluation. This study will facilitate the application of the fracture toughness determined from the SE(T) specimen in the strain-based design of pipelines.

Topics: Gages
Commentary by Dr. Valentin Fuster
2014;():V003T07A034. doi:10.1115/IPC2014-33223.

Whilst there is extensive industry experience of under pressure welding onto live natural gas and liquid pipelines, there is limited experience for Carbon Dioxide (CO2) pipelines, either in the gaseous or dense phases. National Grid has performed a detailed research programme to investigate if existing natural gas industry under pressure welding procedures are applicable to CO2 pipelines, or if new specific guidance is required. This paper reports the results from one part of a comprehensive trial programme, with the aim of determining the preheat decay times, defined by the cooling time from 250 °C to 150 °C (T250–150), in CO2 pipelines and comparing them to the decay times in natural gas pipelines. Although new build CO2 pipelines are likely to operate in the dense phase, if an existing natural gas pipeline is converted to transport CO2 it may operate in the gaseous phase and so both cases were considered. The aims of the work presented were to:

• Determine the correlations between the operating parameters of the pipeline, i.e. flow velocity, pressure etc. and the cooling rate after removal of the preheat, characterised by the (T250–150) cooling time.

• Compare the experimentally determined T250–150 cooling times with the values determined using a simple one dimensional heat transfer model.

• Define the implications of heat decay for practical under pressure welding on CO2 pipelines.

Small-scale trials were performed on a 150 mm (6″) diameter pressurised flow loop at Spadeadam in the UK. The trial matrix was determined using a one dimensional heat transfer model. Welding was performed on a carbon manganese (C-Mn) pipe that was machined to give three sections of 9.9 mm, 19.0 mm and 26.9 mm wall thickness. Trials were performed using natural gas, gaseous phase CO2 and dense phase CO2; across a range of flow velocities from 0.3 m/s to 1.4 m/s.

There was relatively good agreement between the T250–150 cooling times predicted by the thermal model and the measured T250–150 times.

For the same pipe wall thickness, flow velocity and pressure level, the preheat decay cooling times are longest for gaseous phase CO2, with the fastest cooling rate recorded for dense phase CO2.

Due to the fast cooling rate observed on dense phase CO2, the T250–150 times drop below the 40 second minimum requirement in the National Grid specification for under pressure welding, even at relatively low flow velocities. The practical limitation for under pressure welding of pipelines containing dense phase CO2 will be maintaining sufficient preheating during welding.

The results from this stage of the technical programme were used to develop the welding trials and qualification of a full encirclement split sleeve assembly discussed in an accompanying paper (1).

Commentary by Dr. Valentin Fuster
2014;():V003T07A035. doi:10.1115/IPC2014-33254.

The objective of this research was to gain a better understanding of the influence of essential welding variables on microstructure and properties of the grain-coarsened heat-affected zone (GCHAZ) regions formed in pipeline girth welds. In this study, thermal simulation techniques were used to provide a detailed evaluation of the GCHAZ microstructure evolution and intrinsic toughness for two different pipe steels subjected to known welding thermal cycles.

The continuous cooling transformation (CCT) diagrams for the GCHAZ were determined by means of dilatometric techniques with a peak temperature (Tp) = 1350°C and a range of cooling times (Δt800–500 = ∼1 to 100 s). The transformation start and finish temperatures were used to create GCHAZ CCT diagrams for two X80 pipe steels. To further assist with the interpretation of CCT results both light optical microscopy (LOM) and microhardness surveys were used. The results revealed that transformation to predominantly low carbon lath martensite or fine bainite occurred for short cooling times, while bainite formed at intermediate cooling times and upper or granular bainite was obtained for longer cooling times. Some of the detailed features of these simulated GCHAZ microstructures were characterized by scanning electron and transmission electron microscopy (SEM and TEM) in order to better quantify the phases in selected samples. This analysis clearly indicates that despite similar carbon equivalents (CEs), the response of each steel to given GCHAZ thermal was quite different.

The GCHAZ Charpy-V-notch (CVN) impact energy transition curves for the series of single thermal cycles with cooling times, Δt800–500 = 6, 15 and 30 s and were compared against those obtained for the respective pipe steels. The results showed that there were upward shifts in transition temperature for the simulated GCHAZs relative to the respective pipe steels. This overall reduction of notch toughness was attributed to variations in microstructural features for the respective GCHAZs.

Commentary by Dr. Valentin Fuster
2014;():V003T07A036. doi:10.1115/IPC2014-33261.

This paper deals with the development of the overlay weld repair method and the weld sleeve repair method applicable to L555 (X80) pipeline. Mechanical properties of the repaired pipes were investigated and finally proper welding materials and conditions were determined which could ensure hardness limit, required strength and toughness for in-service weld repair methods. Applicability of the in-service repair methods was validated and it was consequently confirmed that the weld repair methods will be able to apply to the repair of operating or pressurized pipeline ensuring the integrity of pipeline.

Commentary by Dr. Valentin Fuster
2014;():V003T07A037. doi:10.1115/IPC2014-33265.

Along with the increasing demand of oil and natural gas by various world economies, the operating pressure of the pipeline is also increasing. Large diameter heavy wall X80 pipeline steel is widely used in the long distance high pressure oil and gas transportation in China today. In addition, development of X90/X100 has begun in earnest to support the growing energy needs of China. With the wide use of X80 steels, the production technology of this grade has become technically mature in the industry. Shougang Group Qinhuangdao Shouqin Metal Materials Co., Ltd. (SQS) since 2008 has been steadily developing heavier thicknesses and wider plate widths over the years. This development has resulted in stable mass production of X80 pipeline steel plate in heavy wall thicknesses for larger pipe OD applications.

The technical specifications of X80 heavy wall thickness and X90/X100 14.8–19.6 mm wall thicknesses, large OD (48″) requiring wide steel plates for the 3rd West-to-East Natural Gas Transmission Pipeline Project and the third line of Kazakhstan-China Main Gas Pipeline (The Middle Asia C Line) and the demonstration X90/X100 line (part of the 3rd West-East Project) in China required changes to the SQS plate mill process design. Considering the technology capability of steelmaking and the plate mill in SQS, a TMCP+OCP (Optimized Cooling Process) was developed to achieve stable X80 and X90/X100 mechanical properties in the steel plates while reducing alloy content.

This paper will describe the chemistry, rolling process, microstructure and mechanical properties of X80 pipeline steel plates produced by SQS for 52,000 mT of for the 3rd West-to-East Natural Gas Transmission Pipeline Project and 5,000 mT for the Middle Asia C Line Project along with 1000 tons of 16.3 mm X90/X100 for the 3rd West-East demonstration pipeline. The importance of the slab reheating process and rolling schedule will be discussed in the paper. In addition, the per pass reductions logic used during recrystallized rough rolling, and special emphasis on the reduction of the final roughing pass prior to the intermediate holding (transfer bar) resulting in a fine uniform prior austenite microstructure will be discussed. The optimized cooling (two phase cooling) application after finish rolling guarantees the steady control of the final bainitic microstructure with optimum MA phase for both grades. The plates produced by this process achieved good surface quality, had excellent flatness and mechanical properties. The pipes were produced via the JCOE pipe production process and had favorable forming properties and good weldability. Plate mechanical properties successfully transferred into the required final pipe mechanical properties. The paper will show that the TMCP+OCP produced X80 heavy wall and 16.3 mm X90 wide plates completely meet the technical requirements of the three pipeline projects.

Topics: Cooling , Steel , Gages , Pipelines
Commentary by Dr. Valentin Fuster
2014;():V003T07A038. doi:10.1115/IPC2014-33268.

Oil and gas exploration around the world continues at a rapid pace. This rapid pace of oil and gas exploration in North America has been fueled primarily thorough the development of horizontal drilling and the “fracking” process of underground shale formations. The demand for various grades and dimensions of API casing and pipe has and will continue to increase in the foreseeable future as these shale formations are exploited.

To support this demand in North America, Gallatin Steel has embarked on a program to develop API casing and pipe coil skelp via their Compact Strip Plant (CSP). A key characteristic of API grade pipeline steels is excellent fracture toughness. This is one area where historically CSP facilities have struggled, especially in gauges greater than 8.8 mm (0.350″) due to overall lack of reduction from the thin slab design of a CSP facility. In addition, utilizing the typical Thermomechanical Control Processing (TMCP) separating recrystallized and non-recrystallized rolling used in API coil skelp production for strength and toughness of a traditional HSM or plate mill is difficult to achieve in the continuous CSP facility. Gallatin Steel has successfully developed, through a controlled combination of slab quality, alloy design, process modifications and process control, excellent toughness in both charpy and DWTT performance from a 65 mm (2.56″) slab in final coil thicknesses up to 12.7 mm (0.500″).

This paper will describe the results achieved to date on various thicknesses from 7.6 mm to 12.7 mm API skelp development at Gallatin Steel. Mechanical property performance along with microstructures/grain size will be presented. In addition, future work that Gallatin Steel will undertake to further improve the capability to produce quality API coil skelp will be discussed.

Commentary by Dr. Valentin Fuster
2014;():V003T07A039. doi:10.1115/IPC2014-33321.

Thick-walled steel pipes manufactured through the UOE process are used in deep-water pipeline applications for the safe and cost-effective transmission of hydrocarbon energy resources. Such pipes are subjected to bending loads in the presence of high external pressure during their installation stage. The combination of bending and external pressure often triggers the development of structural instability due to excessive ovalization of the pipe with catastrophic effects. In the present study, the effect of UOE line pipe manufacturing process on the bending response of externally-pressurized thick-walled pipes is examined, using finite element simulation tools.

Commentary by Dr. Valentin Fuster
2014;():V003T07A040. doi:10.1115/IPC2014-33341.

It has been well documented that slab internal quality is one of the key factors for reduced susceptibility of hydrogen induced cracking (HIC) in line pipe steels designed for sour gas service. In addition, the creation of a homogeneous microstructure which is heavily influenced by the slab internal quality is also a critical key parameter to reduce the HIC susceptibility in higher strength line pipe steel grade X60 and above. For the application of deep sea linepipe exposed to higher external pressure environments, heavy gauge in combination with higher strength steel is essential. Homogeneity of the steel microstructure is a key to success for thicker plates used in sour service HIC applications in combination with a deep sea environment.

In this paper, various microstructures were compared along with an evaluation of the effects of the various microstructures on HIC susceptibility in grades X52, X65 and X70 designed for sour service. The various microstructures compared consisted of polygonal ferrite and pearlite in the X52 and polygonal ferrite, pearlite, acicular ferrite and bainite in the X65 and X70. The effect of microstructural inhomogeneity on HIC susceptibility was comparatively lower for the X52 than that of the X65 and X70. The microstructure of grade X65 and X70 were different due to the different conditions of rolling and cooling that were applied. Grades X65/X70 had a microstructure of polygonal ferrite/pearlite with bainite islands that resulted in a high crack length ratio (CLR) value caused by different hardness regions across the microstructural matrix. A homogeneous fine acicular ferrite microstructure produced by optimizing temperature control during rolling and cooling showed no hydrogen induced cracking. In addition, this alloy/process/microstructure design resulted in improved toughness results in low temperature drop weight tear test (DWTT). This paper will describe the successful production results of plate and pipe for high strength heavier gauge line pipe steels with highly homogeneous microstructures designed for sour service by controlling chemical design and process conditions in rolling and cooling.

In addition, HIC evaluation methods utilizing both a traditional NACE TM0284 method versus that of a Scan-UT method were conducted and compared. A proposal to make the NACE TM0284 testing method more reliable by using Scan-UT method will be presented.

Commentary by Dr. Valentin Fuster
2014;():V003T07A041. doi:10.1115/IPC2014-33372.

This paper describes metallurgical design concept and mass production result of heavy plates for Bovanenkovo-Ukhta (Yamal-Europe) Gas Pipeline project.

The metallurgical design concept was taken as (1) a dual phase microstructure consisting of fine dispersed ferrite and bainite for high strength and high toughness, (2) control of accelerated cooling stop temperature for achieving high yield strength (YS).

In the mass production, the narrow range of strength and elongation and excellent low temperature toughness was achieved (CVN energy at −40°C, and DWTT shear area at −20°C).

Commentary by Dr. Valentin Fuster
2014;():V003T07A042. doi:10.1115/IPC2014-33374.

Back-beveled transition welds for joining unequal wall thickness are often used in gas and oil transmission pipelines, as recommended in the pipeline codes and standards, such as CSA Z662, ASME B31.8, and ASME 31.4. However, one North American pipeline operator has successively utilized a counterbore-tapered design for transition of unequal wall thicknesses for over 30 years. The design philosophy of the counterbore-tapered joint is to reduce the stress concentrations in the heat affected zone, facilitate the welding of unequal wall thickness and NDT while achieving better quality, reliability, and productivity.

By conducting a comparative finite element analysis of the two joint designs, the present study evaluated the pressure containment capacities, the stress concentration factors, the stress intensity factors, and the limit loads of plastic collapses for both the counterbore-tapered and the back-beveled designs. The effect of a key design parameter, the counterbore length, on the integrity of the counterbore design was also examined.

The results of the comparative analysis showed that compared to the back-beveled joint, when the pipe materials of unequal wall thickness have the same strength, the counterbore-tapered joint has the same pressure containment capacity as the back-beveled joint. The back-bevel design offers lower stress concentration factor, lower stress intensity factor, and higher limit load of plastic collapse than the back-beveled design.

Commentary by Dr. Valentin Fuster
2014;():V003T07A043. doi:10.1115/IPC2014-33379.

The mechanical properties of X100 pipeline girth welds are quite sensitive to welding parameters and the design range for a viable welding procedure is narrower compared to pipeline steels of lower grades. The use of a high-productivity welding process, such as dual-torch gas metal arc welding (GMAW), further compounds the dependency of weld properties on welding parameters. Consequently, for X100 pipe welding procedure development, the path to achieve the required weld performance can be a time-consuming and costly process.

Developed in a recently completed project, the essential welding variable methodology provides an effective approach to optimize the development process for X100 pipe welding, with the benefits of reducing development time and saving cost. The present paper presents a practical case study of the methodology for girth welds.

The present paper focuses on the information needed and the analyses performed in the application of the methodology to the process of welding procedure development for a dual-torch pulsed GMAW (GMAW-P) procedure.

Using an analysis tool that can predict the thermal cycles from welding parameters and the available knowledge of microstructure and mechanical responses of both pipe materials and weld metals to welding thermal cycles (cooling rate), several candidates of dual-torch pulsed GMAW procedures were evaluated first for cooling times to help the determination of the final welding procedures. The finalized welding procedures used for the production of the qualification welds were evaluated to estimate the mechanical properties of the girth welds. The estimated weld properties will be compared to those from the test results when they become available.

Commentary by Dr. Valentin Fuster
2014;():V003T07A044. doi:10.1115/IPC2014-33380.

A weld quality control approach developed for the welding of high-strength pipeline steels has demonstrated its effectiveness in achieving reliability and consistency in the mechanical performance of girth welds. Using a predictive tool that can relate cooling times of welding thermal cycles with welding parameters and with the knowledge of microstructure responses of both pipe materials and weld metals to welding thermal cycles, the approach can evaluate the effects of welding parameters on weld properties and identify the essential welding variables. As a result, the essential welding variable approach can be used to optimize and help shorten the process of welding procedure development.

The current paper presents the application of the essential welding variable approach to the girth welding of X80 pipeline steels.

The application started with the selection of pipe materials, welding consumables, and candidate welding procedures. The selection of actual weld procedures and a welding matrix were made after the candidate welding procedures were analyzed in terms of cooling times. Girth welds for two X80 pipes of different chemical compositions, outside diameters, and wall thicknesses were made with single and dual torch GMAW-P processes and a range of welding consumables. The welding parameters were monitored and recorded for all welds; and the thermal cycles of selected welds were measured by thermocouples.

Small-scale testing, including all-weld-metal tensile test, Charpy impact toughness and CTOD fracture toughness tests, were evaluated and correlated with microstructures formed in the HAZ of the girth welds. The material responses of heat-affected zone (HAZ) to thermal cycles of typical GMAW-P single and dual torch processes were experimentally simulated (Gleeble®).

Detailed welding thermal cycle analyses were conducted based on the measured welding parameters. Cooling times of welding thermal cycles for the girth welds were calculated and correlated with the material responses, of X80 pipe steels to welding thermal cycles. The correlation demonstrated very good consistency between the cooling times, the results of the Gleeble simulation, and the mechanical properties of the girth welds.

The dependency of the weld properties on welding parameters was analyzed in terms of cooling times, and the optimization strategy for development of welding procedures that offer more balanced welding properties between strength and toughness was evaluated by adjusting the essential welding variables.

In summary, the process of applying the essential welding variable approach and the results from the tests and the analyses showed that the approach is capable of evaluating the effects of welding parameters on weld properties, identifying the essential welding variables, and ultimately optimizing welding procedures.

Topics: Steel , Welding , Pipes
Commentary by Dr. Valentin Fuster
2014;():V003T07A045. doi:10.1115/IPC2014-33388.

Over the past decades, the complexity of requirements regarding the properties of large-diameter linepipes has increased steadily. This is driven by factors such as increasing operating pressures or more hostile environmental conditions. Steel producers all over the world have responded to these demands by continuous development along the entire processing route from steelmaking to thermomechanical rolling and pipe production. Understanding the influence of the microstructure on pipe properties is a key element to extend the use of linepipe steels to more challenging conditions. For this reason, the techniques that are used for microstructure characterization are constantly refined.

The microstructure of modern microalloyed linepipe steels that are produced by thermomechanical rolling in combination with accelerated cooling depends strongly on the processing parameters during production. The grain size of the base metal is typically below 10 μm and may contain fractions of ferrite, bainite and M/A-constituents. Because of their size, these microstructure constituents are often not readily accessible to a quantitative analysis by classical light-optical microscopy. This was also found to be true within the heat-affected zone (HAZ) of large-diameter pipes. High-resolution scanning electron microscopy in combination with electron backscatter diffraction was found to offer a wide range of possibilities to characterize the microstructure quantitatively with regard to the effective grain size, the volume fraction of constituents and their variation over the wall thickness. The effects of variations in processing parameters in laboratory-scale trials on the microstructure and properties are illustrated. Based on these investigations, it was possible to refine the alloy design and processing parameters in order to improve the low-temperature toughness of the base metal of high strength plate material and the HAZ of longitudinal weld seams.

Commentary by Dr. Valentin Fuster
2014;():V003T07A046. doi:10.1115/IPC2014-33430.

The effect of fracture speed on the ductile fracture resistance of line-pipe steels can have an important effect in the basic understanding of the toughness requirements for crack arrest. Recently, the authors have extended the drop-weight tear test (DWTT) work and developed a modified back-slot (MBS) DWTT specimen to obtain higher fracture speeds. The initial experimental observations demonstrated that this type of specimen can be used to obtain higher fracture speeds. Furthermore, the experimental results clearly showed the effect of fracture speed on the ductile fracture resistance.

In this paper, an in-depth study was carried out to further investigate why higher fracture speeds are obtained in the MBS-DWTT specimens. For this purpose, quasi-static and dynamic/impact DWTT experiments were conducted for both standard DWTT and MBS-DWTT specimens. In addition, finite element analyses using cohesive zone model were carried out to investigate the fracture behavior in these tests. In summary, the higher fracture speeds in the MBS-DWTT come from two major factors. First, as demonstrated by the quasi-static test results, the natural unloading characteristics of the MBS-DWTT specimen (even under pure displacement-controlled loading) leads to higher fracture speeds. Second, the steep unloading curve of the MBS-DWTT specimen produces a higher possibility of an unstable ductile fracture even during the impact event, which will result in higher fracture speeds.

Commentary by Dr. Valentin Fuster
2014;():V003T07A047. doi:10.1115/IPC2014-33464.

High levels of high-low misalignment in pipeline girth welds have been identified as one of the possible contributing factors to some of the recent pre-service hydrostatic test failures or subsequent service failures. However, pipeline service experience indicates that nominally defect-free girth welds with high levels of misalignment and proper weld profiles can provide satisfactory long-term service. In this paper, recent analytical and experimental work aimed at understanding the impact of high-low misalignment in girth welds is described. In nominally defect-free welds, the performance of the welds is found to be predominantly determined by the misalignment ratio, weld strength mismatch ratio, and the weld profile. Iso-load-capacity relations are developed through finite element analysis (FEA) to capture the interdependence of those key parameters. The analysis procedure is validated by cross-weld tensile testing of girth welds with various levels of misalignment and weld strength mismatch. The effects of the circumferential extent of misalignment, alternatively termed local misalignment, are also analyzed. The effects of misalignment in girth weld with planar flaws are examined in the context of the tensile strain capacity.

The analytical and experimental evidence indicate that the absolute level of misalignment is not a sole indicator of girth weld performance. Weld transition profile, pipe wall thickness, and weld strength mismatch all play an important role. With proper weld profiles, minimal or small reduction of load capacity is observed even at very high levels of misalignment. Work is continuing to further examine the effects of high-low misalignment with a goal of making practical recommendations to be included in codes and standards.

Commentary by Dr. Valentin Fuster
2014;():V003T07A048. doi:10.1115/IPC2014-33492.

In the energy market, there is an increasing demand for oil & gas transmission pipelines with larger wall thicknesses and from higher strength linepipe steels. Addition of niobium (Nb) to the steel chemistry in combination with thermo-mechanical controlled processing allow increasing the thickness of the linepipe steels on coil while maintaining good strength and toughness. However, pipeline construction companies often indicate their concerns about the weldability of high Nb containing linepipe steels and in addition, Nb levels are restricted in some of the steel specifications for linepipe applications.

In this study, field weldability of industrially produced helical pipes made from 23.7 mm thick, high Nb containing X70 linepipe steel was evaluated. Welding procedure development was realized for narrow-groove mechanized gas metal arc welding (GMAW). Characterization of the girth welds produced revealed the suitability of the material for typical field welding procedures for onshore pipe laying. The details and the results of the investigations are presented and discussed in the paper.

Topics: Steel , Pipes
Commentary by Dr. Valentin Fuster
2014;():V003T07A049. doi:10.1115/IPC2014-33502.

The unique combination of high strength and low temperature toughness on heavy wall thickness coils allows higher operating pressures in large diameter spiral welded pipes and could represent a 10% reduction in life cycle cost on long distance gas pipe lines. One of the current processing routes for these high thickness grades is the thermo-mechanical controlled processing (TMCP) route, which critically depends on the austenite conditioning during hot forming at specific temperature in relation to the aimed metallurgical mechanisms (recrystallization, strain accumulation, phase transformation). Detailed mechanical and microstructural characterization on selected coils and pipes corresponding to the X80M grade in 24 mm thickness reveals that effective grain size and distribution together with the through thickness gradient are key parameters to control in order to ensure the adequate toughness of the material. Studies on the softening behavior revealed that the grain coarsening in the mid-thickness is related to a decrease of strain accumulation during hot rolling. It was also observed a toughness detrimental effect with the increment of the volume fraction of M/A (martensite/retained austenite) in the middle thickness of the coils, related to the cooling practice. Finally, submerged arc weldability for spiral welded pipe manufacturing was evaluated on coil skelp in 24 mm thickness. The investigations revealed the suitability of the material for spiral welded pipe production, preserving the tensile properties and maintaining acceptable toughness values in the heat-affected zone.

The present study revealed that the adequate chemical alloying selection and processing control provide enhanced low temperature toughness on pipes with excellent weldability formed from hot rolled coils X80 grade in 24 mm thickness produced at ArcelorMittal Bremen.

Commentary by Dr. Valentin Fuster
2014;():V003T07A050. doi:10.1115/IPC2014-33520.

Line pipe manufacturers always have to verify the mechanical properties on pipe to make sure that the pipe meets the requirements specified by the standard and/or customer. This involves measurement of mechanical properties along the hoop direction. The most accurate way to do so is by performing a ring expansion test, which, however, requires dedicated tools. The two other methodologies consist of standard tensile tests on either non-flattened round bar samples or so called ‘flattened tensile samples’. Round bar samples have the disadvantage that only part of the pipe’s wall thickness is considered. Furthermore they can only be used in case of larger OD/t ratios. Tests on flattened samples, on the other hand, require a flattening operation, which induces additional plastic deformation. However, this flattening operation is not standardized. Moreover, it was observed that the mechanical properties — especially the yield strength — resulting from tensile tests on flattened samples largely depend on test parameters such as residual deflection, extensometer position, flattening procedure, etc. Most manufacturers prefer to test flattened samples, because sample preparation is straightforward and cheap. Moreover it only requires a standard tensile bench.

An extensive FEA (Finite Element Analysis) study was launched to investigate the influence of those parameters on the measured yield strength. The applied FEA methodology consists of three steps. First the complete pipe forming process is modeled (in a simplified way). Next a pipe sample is flattened. Finally a tensile sample is cut from the flattened pipe sample and loaded in tension. The mechanical material behaviour is described by a combined kinematic-isotropic hardening model, which allows taking into account the Bauschinger effect. The results are also compared to simulations of ring expansion tests and tests on round bar samples.

Next a dedicated experimental test campaign was performed to verify the results of FEA. Results of ring expansion tests are compared to results obtained on round bar samples and flattened tensile samples.

The results of this study have shown that the applied methodology significantly affects the measured yield strength. Moreover tests on insufficiently flattened samples could considerably underestimate the actual yield strength on pipe. Finally some guidelines are provided to improve the reproducibility of the measured yield strength when using flattened samples.

Commentary by Dr. Valentin Fuster
2014;():V003T07A051. doi:10.1115/IPC2014-33551.

The benefits of mechanized welding for pipeline construction are well known, as reflected by the high industrial acceptance and usage of its variations. However, the engineering and qualification costs associated with the preparation of alternative acceptance criteria for typical pulsed and short-circuit MIG (GMAW-P and GMAW-S) girth welds can make the implementation of mechanization too costly and/or time consuming for small projects.

A multi-wire welding technology, employing a high-deposition consumable that possesses excellent positional capability, along with paired digitally controlled asynchronous inverter power sources, is presented.

Trials were performed on CSA Z245.1 914 mm (NPS 36) OD × 20.4 mm WT Grade 483 heavy wall high strength line pipe. One variant used an 8-head internal welding machine for the root pass, and a conventional single torch short-circuit GMAW hot pass in a compound narrow-groove configuration. A second variant utilized an externally applied controlled short-circuit GMAW-S process for the root pass in a factory-style pipe bevel configuration. Both variants employed fill and cap passes using tandem pulsed gas-shielded flux-cored arc welding (T-FCAW-G/P), using rutile consumables, with the “bug and band” MOW II mechanized welding system.

Basic mechanical testing was performed on the first weld variant, along with single-edge notched bend (SENB) crack tip opening displacement (CTOD) tests, and results are presented.

A productivity comparison is then shown, using weld data from the second weld variant against alternative processes, showing considerably lower fill and cap pass arc time using the T-FCAW-G/P process.

Given the process’s low tendency for the formation of planar discontinuities, the process is appealing for the use of “workmanship” acceptance criteria.

With further procedure development and fine-tuning of the process, tandem flux-cored arc welding may prove viable, particularly for “short” pipelines, where the costs of comprehensive engineering critical assessment/fitness-for-purpose weld procedure qualification and associated engineering work aren’t justified; as a higher productivity alternative to single wire flux-cored arc welding for mechanized tie-in welding; as a much higher productivity alternative to SMAW for tie-ins; or with a narrow groove design, mainline applications for longer-distance projects.

Topics: Arc welding , Pipes
Commentary by Dr. Valentin Fuster
2014;():V003T07A052. doi:10.1115/IPC2014-33630.

The modification of microstructure and mechanical properties of steels after a welding process has received considerable attention in the literature. In the case of welding HSLA steels for pipeline applications, the filler metal employed usually is overmatched (i.e. higher strength) compared to base metal to avoid fracture in this zone of the weld. For this reason, considerable work has emphasized microstructure evolution in the heat affected zone and the associated modification of mechanical properties in this region. In this study, the combined effect of microstructure/property distribution and the geometry of the weld are examined to understand where localization, necking and fracture occurs under tensile loading of a laboratory weld.

To achieve that, a series of tests were conducted on two different types of X80 submerged arc welds: Single and tandem wire welds. The tensile samples were machined transverse to the weld on plates 16 mm thick. Samples were tested where the geometry of the weld was preserved (i.e. the weld cap is left intact) and where the cap was removed in order to remove its effect. The local plastic strain during testing was determined using the Digital Image Correlation technique (DIC).

For the single wire weld, the influence of the cap geometry seems to be of second order, as the fracture location is the same with or without caps. But for the tandem wire welding, the fracture location is very different depending on the geometry: In the case where caps are kept, the fracture occurs outside the HAZ, in the base metal suggesting the important interplay between local mechanical properties and the weld geometry.

A Finite Element Model (FEM) was developed to gain insight into the geometrical effects on the local strain field distribution. The experimental strain distribution is compared to the FEM results to rationalize the effect of geometry. The model results are then used to discuss the position of fracture for the different samples, which correspond well to the experimental results.

Commentary by Dr. Valentin Fuster
2014;():V003T07A053. doi:10.1115/IPC2014-33645.

A previous IPC conference paper (1) described the technical challenges associated with the installation of a new hot tap connection, supplementary to an existing hot tap connection, on the Bord Gáis Éireann Brighouse Bay high pressure gas export terminal in the UK. Work carried out to verify that the hot tap connection would be fit for purpose included a pipe stress analysis, Finite Element Analysis (FEA) and Engineering Critical Assessment (ECA). These assessments were performed because the split tee shell thickness and consequently also the circumferential fillet weld leg lengths did not achieve the 2 × carrier pipe thickness criterion required by UK specifications for applications where design stress levels exceed 30% specified minimum yield strength.

Subsequently, it was identified that the existing hot tap connection installed in 2001 also did not meet the 2 × carrier pipe thickness criterion. Furthermore the material grade was lower than that for newer hot tap, i.e. P355 compared with P460 and the tee had been chamfered down from 50 mm to 40 mm at the ends, leading to reduced section circumferential fillet welds. This resulted in a leaner design than that for the newer hot tap and an ASME B31.3 area replacement calculation revealed that the area replacement ratio barely achieved the 1.0 requirement of the code suggesting a limited tolerance to system loading. Consequently similar stress analysis, FEA and ECA assessments to those previously undertaken were also subsequently performed for the existing hot tap connection. This paper provides details of the analyses and results obtained to determine the integrity of the existing hot tap split tee assembly which required a bespoke approach and a need to challenge conventional thinking.

Commentary by Dr. Valentin Fuster
2014;():V003T07A054. doi:10.1115/IPC2014-33655.

Pipeline infrastructure is the backbone of the energy industry and remains the safest and most cost effective method for transporting oil and gas. For decades corrosion has presented a significant challenge to pipeline operators. According to Alberta Energy Regulator data, internal corrosion is considered to be the root cause for more than 54% of all documented pipeline failures in Alberta [1]. Spoolable composite pipeline technologies have become a mainstream corrosion solution over the last 10 years, however these products are limited to smaller pipelines, typically less than 6 inches in diameter. Traditional slip-lining (field installed plastic lined steel pipe) is used for internal corrosion protection of larger pipelines, however it is costly, requiring labour intensive field construction, often completed in inhospitable environmental conditions. As a result project delays and cost over-runs are commonplace.

Recognizing the need for a cost effective pipeline corrosion solution for larger gathering pipelines, an innovative technology was developed that combines a unique mechanical pipe joining system with an integrated electro-fusion coupler. The new joining system enables insertion of an HDPE liner in a factory environment where costs and quality can be tightly controlled. The new joining system eliminates conventional welding of the pipeline in the field and instead uses a custom field press to quickly energize the mechanical pipe joint. Field scope is significantly reduced, construction completed in less time, and associated costs greatly reduced. This paper discusses the testing completed to qualify the new joining system for use in oilfield gathering pipelines.

The qualification test plan includes all requirements identified in applicable regulatory standards (primarily CSA Z662-11), and prudent engineering requirements based on anticipated field handling and anticipated operating conditions. The test regime was ultimately designed to ensure the suitability of the pipeline system for intended service. Testing included hydrostatic burst, static gas pressure, bend, cyclic pressure and thermal, vacuum, tensile, and compressive tests on the joint.

The test results show that in all cases the jointing system successfully met the established design performance criteria and in most cases exceeded the actual mechanical properties of the parent pipe, thus proving the joining system ready for field installations.

Topics: Joining , Pipes , Testing
Commentary by Dr. Valentin Fuster
2014;():V003T07A055. doi:10.1115/IPC2014-33668.

State-of-the-art linepipe steels are microalloyed low-carbon steels that combine high strength and fracture toughness with good weldability. During welding of pipe sections the heat affected zone (HAZ) experiences rapid thermal cycles resulting in a graded microstructure that can be significantly different from that of the base metal. In particular a variety of bainitic microstructures can form in the HAZ. Depending on the type of bainite mechanical properties may be improved or may lead to poor fracture resistance and be detrimental to the overall HAZ performance. Optical microscopy is not sufficient to differentiate bainitic morphologies which vary with the transformation temperature. The investigated X80 linepipe steel also contains retained austenite at room temperature. Based on the retained austenite it is possible to characterize the orientation relationship (OR) between austenite and the transformation products. It is found that bainite shows an orientation relationship near Kurdjumov-Sachs with the prior austenite. Variant selection is related to the driving force for the bainite reaction and hence depends on the transformation temperature. In the current study Electron BackScatter Diffraction (EBSD) mapping is used to characterize transformation products based on their orientation relationship. This approach offers a quantitative way to determine volume fractions of different types of bainite in complex HAZ microstructures which is necessary to establish structure-property relationships of the HAZ.

Topics: Steel
Commentary by Dr. Valentin Fuster
2014;():V003T07A056. doi:10.1115/IPC2014-33696.

Pipe elbow is a common feature in pipelines and piping systems as a means to changing directions of otherwise straight pipelines. Irrespective of the processes involved in manufacturing pipe elbows, it is of interest to investigate whether they have any geometric imperfections. Researchers at the University of Alberta have devised a technique to measure initial imperfection of straight pipes prior to testing, using high resolution 3D surface profiler in conjunction with 3D reverse engineering software. The objective of the current study is to extend the imperfections measurement technique from measuring straight pipe segments to pipe elbows. Six (6) ninety (90) degree elbows are measured in this research with outside diameters ranging from NPS 12 inch to NPS 42 inch. A 3D laser scanner is used to acquire surface data and create 3D models corresponding to each elbow. A method for the geometric analysis of the elbows is developed using 3D inspection and reverse engineering software Geomagic®. The geometric idealization of a pipe elbow is a torus, which can be defined by a circle revolving around an axis, coplanar with the circle. The idealized geometry for each elbow is obtained through the developed method of geometric analysis, which includes the diameter of the circle defining the torus, and its distance from the axis of revolution. The difference between the ideal torus and the scanned geometry is considered as imperfection of each pipe elbow. The wall thickness values at the ends or edges of select pipe elbows are also measured from the scanned data and are reported as percentage deviation from the specified wall thickness around the perimeter at different cross sections. The 3D reverse engineering of the elbows indicated that they resemble the ideal geometry very closely. The ovalization imperfections are seen to be well within the value specified by CSA Z662-11. The wall thickness deviations are seen to vary between −10% to +25% of the specified value, with increased thickness being more prominent in the elbows. Finite element analysis of an elbow with thickness imperfection shows that higher hoop stress appears on the intrados than initially intended.

Commentary by Dr. Valentin Fuster
2014;():V003T07A057. doi:10.1115/IPC2014-33714.

The microstructure and fracture behavior of the base metal of different X80 steel line pipe lots from several pipeline projects were analyzed. The resistance of the pipes to ductile fracture propagation was determined by the full-scale burst tests. The high intensity of fracture surface separation (secondary brittle cracks parallel to the rolling plane of the plate) appeared to be the main factor reducing the specific fracture energy of ductile crack propagation. A method for quantitative analysis of microstructure allowing estimation of the steel’s tendency to form separations is proposed. The procedure is based on the EBSD data processing and results in Cleavage Morphology Clustering (CMC) parameter evaluation which correlates with full-scale and laboratory mechanical test results. Two special laboratory mechanical test types utilizing SENT and Charpy test concepts for prediction of ductile fracture arrest/propagation in a pipe were developed and included into Gazprom specifications.

Commentary by Dr. Valentin Fuster
2014;():V003T07A058. doi:10.1115/IPC2014-33731.

This paper presents a method for estimating the brittle fracture limit of a weld with a notch in the heat-affected zone (HAZ) and residual stress based on the Weibull stress criterion. A constraint loss correction procedure using the Weibull stress criterion is specified in ISO 27306. However, this standard is applicable only to structural steel components with defects, not to welded joints. Therefore, we conducted fracture tests and finite element analyses to propose a new evaluation method for welded structural components.

In this study, three-point bending (3PB) tests and wide-plate (WP) tension tests of HAZ-notched welds made of 780-MPa-class high-strength steel were conducted at −40°C. Brittle fractures occurred in the HAZ regions of all the specimens, and the critical crack tip opening displacement (CTOD) values obtained in the 3PB and WP tests were approximately 0.02–0.07 mm and 0.08–0.11 mm, respectively. The minimum critical CTOD of the WP specimen fracturing at the coarse-grained region of the HAZ (CGHAZ) was approximately four times that of the 3PB specimen. These results confirmed that the difference of specimen geometry affects the brittle fracture resistance of a HAZ-notched weld with residual stress. Hence, the assessment of the brittle fracture limit of a welded structural component with a defect obtained by the fracture toughness of a 3PB specimen would be excessively conservative.

The effects of specimen geometry, residual stress, crack-front shape and HAZ microstructure classification on the Weibull stress were investigated to clarify the difference of experimental critical CTOD for 3PB and WP by using a finite element analysis. The results of this analysis showed that the Weibull stress of WP specimen was larger than one of 3PB specimen in all CTOD region due to difference of geometry. The welding residual stress increased the Weibull stress only for WP. Compressive residual stress and crack front shape for 3PB specimen did not affect the Weibull stress. The difference of HAZ microstructure distribution for same welded joint affects the Weibull stress for 3PB and WP specimens.

Finally, it was confirmed that the brittle fracture limit of a HAZ-notched weld with residual stress could be predicted from the Weibull stress criterion because critical CTOD of WP specimens predicted by critical CTOD of 3PB specimens fracturing at the CGHAZ included critical CTOD of WP specimens obtained by experiments.

Commentary by Dr. Valentin Fuster

Risk and Reliability

2014;():V003T12A001. doi:10.1115/IPC2014-33036.

The consequences of the accidental release of petroleum based liquids or natural gas from onshore pipelines are studied. Damage to property, environment and society are considered. Property damage and environmental reparation costs are evaluated directly from publicly available data. Straight forward regression models are proposed to quantify these types of consequence, considering the released fluid and the characteristics of the environment. Societal impact, taken as the number of casualties, is evaluated by combining approximated fire models, heat vs. mortality correlations, population density and the statistical value of life. For gas, a jet fire model is employed; the heat flux is parameterized by the pressure and the failure size. For liquid releases, either pool or jet fire model might be employed, according to the size of the hole. The heat flux of the pool fire model depends on the size of the pool, which is determined by a correlation between released volume and affected area. On the jet fire model the heat flux is parameterized by the release rate and the heat of combustion. This study may serve as basis for the estimation of the consequences of failure in the evaluation of the risk of operating hazard liquids and natural gas pipelines.

Topics: Pipelines , Failure
Commentary by Dr. Valentin Fuster
2014;():V003T12A002. doi:10.1115/IPC2014-33052.

In order to support its goals of zero incidents through risk-based asset management, Access Midstream (Access) requires a solution for providing decision support by modeling integrity threats that indicate potential safety and environmental incidents on all of its pipeline assets. The objective is to tailor existing pipeline risk methods to properly account for the unique conditions inherent to midstream gathering assets while still being flexible enough to account for traditional gas transmission and liquids transportation lines.

Access has created a unique risk-based spatial decision support system (RB-SDSS), customized to the needs of a gas gathering company that also owns and operates incidental gas transmission and hazardous liquids pipelines. Access’ business needs call for a comprehensive solution that includes a risk model that can handle a diverse asset base. Most of the company’s pipelines are unregulated, but risk-based management and decision-making are still desired for regulated and non-regulated pipelines alike. Access has designed a decision support system that automatically retrieves and integrates data, estimates risk with directly and indirectly related inputs in its algorithm, and disseminates the data to business users across the enterprise. The processing model is calibrated for risk assessment and is capable of distinguishing between high and low risk lines. It is able to generate results that can be understood and upon which appropriate action can be taken. These results are refreshed monthly as new assets are constructed or purchased and existing asset attributes are updated. Access’ decision support system is repeatable and scalable owing to maximum use of automated processes.

The focus of this paper is to outline Access’ approach to building a risk assessment solution for its pipeline assets. Each component of the RB-SDSS is reviewed: (1) Data Integration, (2) Risk Modeling, and (3) Organizational Utilization. This includes strategies for assessing traditional pipeline threats, geospatial techniques used to infer data, how to assess data quality and completeness, challenges, lesson’s learned, and future improvements.

Topics: Risk
Commentary by Dr. Valentin Fuster
2014;():V003T12A003. doi:10.1115/IPC2014-33108.

Pipeline risk assessment has been a major concern for Mexico’s Oil Sate Company (PEMEX) for the last couple of decades. During this period of time, isolated efforts from the different subsidiary bodies (upstream, midstream, and downstream) have been made regarding this matter, including the implementation of powerful software tools. However, due to several factors, none of these has provided an optimum solution for the company until now as they do not serve specific needs from PEMEX according to the country’s social and geographical environments. Besides that, a different software tool has been used by each subsidiary body, generating variations in practice and dissimilar criteria when assessing risk.

A project sponsored by the Ministry of Energy in Mexico and the National Council of Science and Technology seeks to address these issues by developing a unique and customized software solution for pipeline risk assessment in PEMEX. This paper presents the progress made so far regarding the development of this tool.

Commentary by Dr. Valentin Fuster
2014;():V003T12A004. doi:10.1115/IPC2014-33117.

Sinkholes are common features in parts of Florida, and the Florida Geological Survey maintains an online database of sinkhole incident reports (SIRs) that was started in 1965. The incident reports are accepted “as-is” without verification; sinkhole location, length, width, and depth are included in SIRs. A desktop assessment of sinkhole activity in northern Florida was developed on the basis of SIRs that were available in GIS (shape file) format from the Florida Geological Survey website and an understanding that sinkhole activity needed to be normalized to length for use in pipeline risk assessments.

The rate of sinkhole development in northern Florida was quantified by extracting sinkhole locations within 10 miles of a 230-mile-long hypothetical alignment of a pipeline and lateral. Over 500 sinkholes were located within the approximately 4,700-square mile polygon. Sinkhole trends aligned to highways indicate more complete reporting; therefore, 33 road segments comprising a combined length of about 944 miles within the polygon were used for statistical analysis. The SIR database was accepted as an accurate portrayal of sinkholes during its 47-year existence. Sinkhole activity was portrayed as annual frequency for sinkhole width or length ranging from 1 to 500 feet and normalized to 1 square mile and 1 lineal mile. A sinkhole 2 feet or larger in width occurs on average 8 times per year somewhere within 10 miles of the hypothetical alignment; whereas, a similar sinkhole occurs on average 4 times per year along a length of 944 miles. A 2-foot or larger sinkhole occurs on average about every 600 years within any 1 square mile of the 4,700-square-mile area and about every 200 years along any 1 mile of alignment length. On a per-lineal-mile basis, the expected 1,000-year sinkhole would be at least 9 feet wide; whereas the 1,000-year sinkhole would be at least 22 feet wide. Performance of specific pipelines under loading conditions associated with sinkholes of various widths can be assessed and used in a pipeline risk assessment.

Commentary by Dr. Valentin Fuster
2014;():V003T12A005. doi:10.1115/IPC2014-33129.

To introduce and apply Reliability-Based Design and Assessment (RBDA) method to China’s onshore natural gas pipelines, China Petroleum Pipeline Engineering Corporation (CPPE) undertook a research project based on achievements from a series of researches sponsored by PRCI. RBDA method aims to maintain a consistent risk level throughout the lifecycle of pipelines by rational designs, professional operations and scientific maintenance. The basis of RBDA method is a set of risk-based reliability targets for pipelines, especially the target value of Ultimate Limit State (ULS). CPPE has developed a database for 37,000km natural gas pipelines in China and defined 148 operating conditions corresponding to various pressures, pipeline diameters, steel grades and pipeline lengths in different location classes. Failure calculation models are modified according to the corrosion and equipment impact under each specified operating condition. While calculating the failure consequences, 20,000 kilometers pipes from different locations classes were selected and statistics of average population density was made. Statistics of failure consequences were made again. Finally, the overall risk level of built natural gas pipelines was calculated. This paper introduces 148 operating conditions, failure probabilities, calculation method regarding failure consequences and average population density of all locations of different classes. Based on target reliability of pipelines set on country level, design and construction plan for newly-built pipelines are optimized by using RBDA method for rationally guiding subsequent operation and maintenance to reach the most optimal and cost-efficient plan in whole lifecycle of pipelines.

Commentary by Dr. Valentin Fuster
2014;():V003T12A006. doi:10.1115/IPC2014-33146.

Corrosion is a common degradation process for most oil and gas pipelines in operation that can lead to leak and rupture failures. To avoid failures due to corrosion, integrity management plans for pipelines require fitness-for-service (FFS) assessments and remaining life analysis of the corrosion features that are detected by in-line inspections (ILIs). The objective of the present paper is to support the deterministic integrity and remaining life assessment of pipelines by introducing a pragmatic approach for the determination of corrosion rates from two inspections. The proposed approach is primarily tailored towards upstream and subsea pipelines that are subject to very high density internal corrosion rather than transmission pipelines with low to moderate densities of external features.

ILI data may be subject to significant measurement errors and feature matching for two ILIs can become highly unreliable if high-density corrosion is present. To address these uncertainties, the backbone of the proposed approach is to focus on corrosion clusters rather than individual corrosion pits and a filtering process is utilized to identify true corrosion growth. The introduced approach is supported by theoretical knowledge and practical experience. The approach can be easily executed in spreadsheet software tools without the application of advanced statistical and probabilistic methods for the deterministic remaining life assessment in practice.

Topics: Corrosion , Pipelines
Commentary by Dr. Valentin Fuster
2014;():V003T12A007. doi:10.1115/IPC2014-33147.

In order to apply the Reliability Based Design and Assessment (RBDA) methodology to evaluate the equipment impact on the onshore natural gas transmission pipelines in China, a research project was undertaken by China Petroleum Pipeline Engineering Corporation (CPPE) based on the framework developed by C-FER Technologies (C-FER) in “Guidelines for Reliability Based Design and Assessment of Onshore Natural Gas Pipelines” (sponsored by PRCI). The objective of the project was to collect native data and calibrate the probability models[1] in order to make it suitable for the situations in China where there is dense population and many newly-built high pressure and large diameter pipelines.

The equipment impact model consists of two components: a) the impact probability model which calculates the frequency of mechanical interference by excavation equipment; and b) the failure model which calculates the probability of failure in a given impact. A detailed survey was undertaken in 2012 in order to collect the data required to calculate the impact frequency and the load applied by an excavator to a pipeline. The survey data for impact frequency calculation was gathered based on 19,300km of transmission pipelines from 4 operating companies in China. They reflect current prevention practices and their effectiveness. The frequencies of basic events summarized in this survey used to calculate the probabilities of the fault tree are generally agreement with the data summarized in PRCI’s report. The impact frequencies calculated by the fault tree under typical prevention measures are 400%, 200%, 20% and 0% higher than that in PR-244-9910 report for class 1, class 2, class 3 and class 4 areas respectively, which is due to dense population and more construction activities. Bucket digging forces of 321 types of excavators from 20 manufacturers were gathered. The survey data of the forces are slightly higher than that in the PR-244-9729 report as a whole due to the increase in mechanical efficiency of excavators in recent years. The excavator maximum quasi-static load model was calibrated correspondingly. Equipment impact probability calculation and model sensitivity analysis results are described to present several characteristics of onshore transmission natural gas pipelines in China.

Commentary by Dr. Valentin Fuster
2014;():V003T12A008. doi:10.1115/IPC2014-33151.

Recently a long time discussion among specialists about the meaning of the probabilities of failure (POF) produced by different reliability analysis methods surfaced in pipeline journals. This paper, which was a long time in the making, is a follow-up on the discussion and analyses the actual reliability level which was empirically embedded in codes for pipeline design [B31G, B31Gmod, Shell92 and Battelle (PCORRC)] and Building Standard (BR) Main Pipelines #2.05.06-85, using a real pipeline as an example.

Assessment of the actual reliability level empirically embedded in BR is based on assessing the order of the quintiles of strength parameters (design values of tensile strength and yield strength of the pipe material) and load (internal pressure) on the pipeline.

This approach allows direct connection of the deterministic safety coefficients used in the BR with the level of reliability of the pipelines associated with these coefficients.

The actual reliability level, empirically embedded in international codes, is calculated as the probability that the limit state function (LSF) of ideal pipeline (without defects) is positive. LSF = PfPop, where Pf is the failure pressure of an ideal pipe, which is estimated by any design code; Pop is the operating pressure. The failure and operating pressure are considered as random variables. The expression for this probability was obtained analytically and in closed form.

Recommendations are also presented for choosing probability distributions and statistical parameters for random variables RVs. Extensive calculations permitted revealing the reliability levels which are actually present in the analyzed international pipeline design codes. In a nutshell, the paper proves that the international codes under consideration are very reliable, as they produce very safe designs of pipelines with very low POF, and, hence, provide large safety coefficients, and that the algorithm developed in the paper permits connecting the current level of pipeline degradation (in terms of POF), with its current safety coefficient, which, in this case, is a function of time.

All calculating in the paper where performed using MathCAD. Illustrations of these calculations are also presented in the paper.

Commentary by Dr. Valentin Fuster
2014;():V003T12A009. doi:10.1115/IPC2014-33171.

This paper compares a site specific quantitative risk analysis for a gas transmission pipeline using traditional “average” risk ranking methods to a more complex Monte Carlo analysis using a range of possibilities and consequences for the various areas of the site—each with their own probability. The comparison is based on the California “Guidance Protocol for School Site Pipeline Risk Analysis”, a quantitative risk analysis protocol which uses average probability and consequence values, and extends it to explain how a more complex Monte Carlo analysis of those same risk factors can give a more comprehensive understanding of anticipated risks and consequences.

Commentary by Dr. Valentin Fuster
2014;():V003T12A010. doi:10.1115/IPC2014-33213.

This paper presents a methodology to evaluate the reliability of corroding pipelines by simultaneously considering the growth and generation of corrosion defects. The non-homogeneous Poisson process is employed to model the generation of corrosion defects, whereas the non-homogeneous gamma process is used to characterize the growth of corrosion defects once generated. The parameters included in the non-homogeneous Poisson process and non-homogeneous gamma process are evaluated from the inline inspection data using a hierarchical Bayesian model. The measurement errors associated with the inline inspection tools are taken into account in the Bayesian updating. The time-dependent failure probability of the corroding pipeline is evaluated using the Monte Carlo simulation technique. The methodology is illustrated using a natural gas pipeline that has been subjected to multiple inline inspections over a period of time. The results illustrate the necessity to incorporate the generation of new corrosion defects in the reliability analysis of corroding pipelines.

Commentary by Dr. Valentin Fuster
2014;():V003T12A011. doi:10.1115/IPC2014-33218.

The United Kingdom Onshore Pipeline Operators Association (UKOPA) was formed by UK pipeline operators to provide a common forum for representing pipeline operators interests in the safe management of pipelines. This includes ensuring that UK pipeline codes include best practice, and that there is a common view in terms of compliance with these codes.

Quantitative risk assessment (QRA) is used by operators in the UK to determine if individual and societal risk levels at new developments adjacent to existing pipelines are as low as reasonably practicable (ALARP). In 2008 the UKOPA Risk Assessment Working Group developed codified advice on the use of QRA applied to land use planning assessments, which was published by the Institution of Gas Engineers & Managers (IGEM) and the British Standards Institute (BSI). This advice was designed to ensure a standard and consistent approach, and reduce the potential for disagreement between stakeholders on the acceptability of proposed developments.

Since publication of IGEM/TD/2 and PD8010-3 in 2008, feedback from users of the guidance together with new research work and additional discussions with the UK safety regulator, the Health & Safety Executive (HSE), have been undertaken and the codified advice has been revised and reissued in June 2013.

This paper describes the revisions to the guidance given in these codes in relation to:

• Clarification on application

• Update of physical risk mitigation measures (slabbing and depth of cover)

• Update of HSE approach to Land Use Planning

• Update of failure frequency data:

○ Weibull damage distributions for external interference

○ Generic failure frequency curve for external interference

○ Prediction of failure frequency due to landsliding

The revised codes, and their content, are considered to represent the current UK best practice in pipeline QRA.

Commentary by Dr. Valentin Fuster
2014;():V003T12A012. doi:10.1115/IPC2014-33260.

Reduction of human error can have a significant impact on the potential for spills and leaks and translate into better safety performance and financial gains for an organization. As important as the technical components of a design, construction, operation, and maintenance program is the human component of the activities being performed.

In the Pipeline Industry, human factors can create the potential for a human error at many points along the life cycle of a pipeline. Using a life cycle approach to manage human factors can provide an organization the capability to integrate human factors into programs, standards, procedures and processes using a disciplined approach. This paper reviews the life cycle of a pipeline and identifies areas where the potential for human error can have catastrophic results. Guidance is provided on the development of a human factors life cycle for the organization and illustrates available industry resources as well as opportunities for further research and development.

Commentary by Dr. Valentin Fuster
2014;():V003T12A013. doi:10.1115/IPC2014-33324.

Some pipeline operators evaluate risk on an individual pipeline basis even if the right of way (ROW) is shared with other pipelines. Determining a ROW strip risk condition may be complex or quite simple, according to the model adopted by the analyst. If the pipelines allocated in a shared ROW belong to different operators it is very likely that they apply different methods to evaluate a risk condition. The relative risk contribution cannot be added to estimate the risk of a ROW strip. In Mexico insurance companies request studies of collective risk in pipelines to decide whether to increase a prime or reduce coverage. This request does not have technical support or engineering guidelines to perform the analysis. In Pemex there are few documented events where a pipeline failure affects parallel pipelines, known as collateral damage. There are some methods to estimate a potential collateral damage as a function of soil damping and separation between pipelines (Ref.2). This scheme applies for gas pipelines and has to be complemented with an ignition scenario probabilistic analysis. In the case of hazardous liquids scenarios of leak and rupture have to be considered, including potential shed routes, product concentration sites and operator response capability. Since risk is assessed with particular and specific attributes of a pipeline the probability of failure cannot be directly added to adjacent pipelines. There are some failure mechanisms common for pipelines sharing the ROW, such as external corrosion and stress corrosion cracking (SCC), with different intensity when considering coating and corrosion protection (CP) efficiency. Internal corrosion depends on other factors such as product features so it does not necessarily repeat with the same magnitude in all pipelines. Pipeline threats can be expected to be the same in this case — with different intensity. For instance, third party activity and weather can threaten all pipelines allocated in the same ROW. These pipelines may present similar symptoms with different magnitude. Cover depth, additional protection and wall thickness play an important role in reducing third party (TP) and weather and outside forces (WOF) threats. The paper provides risk results of a ROW strip based on probability of failure values. Pipelines with biggest risk contribution were identified and integrity management alignment diagrams were obtained to correlate with risk values. A simple algorithm was developed to process risk results in terms on shared ROW buffer dimensions. The study is complemented with the results of a consequence simulation analysis for a gas pipeline

Topics: Pipelines , Risk
Commentary by Dr. Valentin Fuster
2014;():V003T12A014. doi:10.1115/IPC2014-33366.

Natural hazards can be initiating events for accidents in oil and gas pipelines. Severe past incidents bear testimony to the risk associated with pipeline accidents triggered by natural hazards (natechs). Post-incident analysis is a valuable tool for better understanding the causes, dynamics and impacts of such accidents. To identify the main triggers of onshore transmission pipeline natechs in the USA, natural gas and hazardous liquid incident reports collected by the Pipeline and Hazardous Materials Safety Administration were analyzed. Potential natech incidents were identified by automated data-mining followed by expert review. The analysis covered ∼21,000 incidents, about 6% of which were identified as natechs. Geological hazards triggered 50% of the identified natechs, followed by meteorological (25%), climatic (11%), and hydrological (11%) hazards. Landslides are the main geological hazard with 43% of the incidents within the category. Among meteorological hazards, lightning is the major hazard with 36%. 84% of the hydrological hazard related natechs were found to be due to floods. Cold-related hazards make up 93% of the natechs caused by adverse climatic conditions. Some preliminary qualitative results on consequences are provided as well.

Commentary by Dr. Valentin Fuster
2014;():V003T12A015. doi:10.1115/IPC2014-33368.

Correct assessment of the remaining life of distributed systems such as pipeline systems (PS) with defects plays a crucial role in solving the problem of their integrity.

Authors propose a methodology which allows estimating the random residual time (remaining life) of transition of a PS from its current state to a critical or limit state, based on available information on the sizes of the set of growing defects found during an in line inspection (ILI), followed by verification or direct assessment.

PS with many actively growing defects is a physical distributed system, which transits from one physical state to another. This transition finally leads to failure of its components, each component being a defect. Such process can be described by a Markov process.

The degradation of the PS (measured as monotonous deterioration of its failure pressure Pf (t)) is considered as a non-homogeneous pure death Markov process (NPDMP) of the continuous time and discrete states type. Failure pressure is calculated using one of the internationally recognized pipeline design codes: B13G, B31Gmod, DNV, Battelle and Shell-92.

The NPDMP is described by a system of non-homogeneous differential equations, which allows calculating the probability of defects failure pressure being in each of its possible states. On the basis of these probabilities the gamma-percent residual life of defects is calculated. In other words, the moment of time tγ is calculated, which is a random variable, when the failure pressure of pipeline defect Pf (tγ) > Pop, with probability γ, where Pop is the operating pressure. The developed methodology was successfully applied to a real life case, which is presented and discussed.

Commentary by Dr. Valentin Fuster
2014;():V003T12A016. doi:10.1115/IPC2014-33384.

A large Research and Development programme has been executed by National Grid to determine the feasibility of transporting carbon dioxide (CO2) by pipeline. Such pipelines would be required to form a transportation system to take the CO2 from its place of capture at an emitter’s site to a place of safe storage within a Carbon Capture and Storage (CCS) scheme. This programme received financial support from the European Union. As part of this programme, National Grid commissioned a series of experimental studies to investigate the behaviour of releases of CO2 mixtures in the gaseous and the liquid (or dense) phase. This has included simulating accidental releases in the form of punctures or ruptures of a buried pipeline and deliberate releases through different venting arrangements. This work is required, as CO2 has the potential to cause some harm to people if they are exposed to it for long enough at high concentrations. This paper gives an overview of the findings from this work and shows how the data has been used to help develop a number of the more pragmatic, predictive models for outflow and dispersion. This work complements the more theoretical studies carried out using state of the art advanced computational fluid dynamic models, employed by other UK based participants (University College London, University of Leeds, Kingston University and the University of Warwick) in the research programme.

Commentary by Dr. Valentin Fuster
2014;():V003T12A017. doi:10.1115/IPC2014-33394.

Unlike the circumstance associated with transmission pipelines, where variables that are attributes of risk are typically widely available in GIS systems or in other databases that are geo-referenced to linear assets, risk data for distribution systems are not typically linearly referenced to what is essentially a network system. Therefore the manner in which risk is calculated and displayed for distribution systems must differ significantly from the way these functions are performed on transmission pipelines. In distribution systems, failure (defined as the loss of containment) and the contributors to the likelihood of failure, is often highly correlated to system-specific circumstances, such as type of material used, installation era, and operating environment. These correlations between cause-and-effect as they relate to failure likelihood in distribution systems are not widely recognized on a universal basis, such as they might be in transmission pipeline environments, but are typically unique to each operating system.

Because system data for distribution networks is not typically available in a manner that can be linearly geo-referenced to pipeline coordinates the way it is for transmission systems, the convention of mapping risk to pipeline dynamic segments as a function of risk attributes that exist within those dynamic segments is not achievable for distribution systems the way that it is for transmission systems. Therefore, the most effective strategy for performing risk assessments in distribution systems is to create a database in which existing incident data can be correlated to system attributes, and then to use those correlations to create cause-and-effect relationships between system attributes and failure likelihood. Consequences are characterized in terms of the operating environment (e.g., wall-to-wall, residential, etc.), leak magnitude, type of facility (mains vs. service lines), and special mitigating or exacerbating factors, such as availability of excess flow valves, or the presence of inside meters.

A risk assessment methodology has been developed that accommodates the above constraints and that meets the stated objectives, and which is well-suited to the distribution system data infrastructure that is typical of most operators. Because the risk assessment approach leverages existing databases and incident reporting structures, it lends itself to automation, and re-evaluation on a regular basis. Reporting is facilitated by a ‘heat map’, which provides immediate insight as to the drivers of risk for each system sub-group having similar design, materials, and operating characteristics.

Commentary by Dr. Valentin Fuster
2014;():V003T12A018. doi:10.1115/IPC2014-33471.

The risk of pipeline failure is a measure of the state of knowledge of the pipeline; improved knowledge of the pipeline reduces the uncertainty and therefore can reduce the associated risk. Specifically for corrosion defects, the knowledge of the number and size of defects is often obtained using in-line inspection tools which have uncertainty associated with their measurement capabilities. Quantitative Risk Assessment (QRA) is a methodology that objectively assesses a range of pipeline integrity threats including the threat of corrosion failure. QRA can incorporate the impact of significant sources of analysis uncertainty, such as feature sizing in risk estimates. This paper discusses an application of QRA used to evaluate the operating risk of high pressure transmission pipeline segments in the TransGas system. Specific examples are described in which the inspection tool sizing uncertainty was shown to exert a significant influence on the calculated risk levels.

In carrying out the analysis, the failure probability models selected were dependent on the nature of the integrity threat and the type of information available for each pipeline. For the assessment of corrosion integrity, the results of in-line inspections were used directly in determining failure likelihood. For the other threats including equipment impact, geotechnical hazards, manufacturing cracks and stress corrosion cracking, the probability of failure was estimated from historical failure rates with adjustments to reflect line-specific conditions. Failure consequences were estimated using models that quantify the safety implications of loss of containment events. Using these models, safety risk measures were calculated along the length of each pipeline. The results of the analysis show the benefit of the use of inspection technologies with improved sizing accuracy, in terms of reduction in expected operating risk.

Commentary by Dr. Valentin Fuster
2014;():V003T12A019. doi:10.1115/IPC2014-33474.

System Wide Risk Assessment (SWRA) is an integral part of an Integrity Management Program (IMP), and it is the first step in most IMPs. Risk is the expected value of loss (often expressed as damage per year, i.e. expected number of annual injuries or fatalities). Risk is calculated as the product of the Probability/Likelihood of Failure (LoF) and the consequence of failure, where failure is defined as a loss of containment event. Hence, it is necessary to calculate both the Likelihood and the consequences of failure in order to accurately model risk.

For natural gas pipelines, consequence is primarily human safety-based. The primary threat to the population is the effect of the thermal radiation due to ignited pipeline ruptures. Currently, most pipeline industry system wide risk assessment models are qualitative risk models, where consequence is generally characterized by class, relative population measures, or some other relative measure. While this may be adequate for some relative risk ranking purposes, it is generally not accurate in representing the true consequences and the arbitrary nature leads to poor representation of actual consequences. Qualitative risk models are also highly subjective, and can have a high degree of bias. Thus, in this study, quantitative LoF assessment and a rigorous quantitative consequence model were used to make the risk assessment process more accurate, more objective, and transparent. The likelihood algorithm developed in this study is described in a companion paper. It should be noted that a quantitative estimate is never completely objective as subjective assumptions and idealizations are still involved, however it provides a framework to make it as objective as possible.

The consequence model implemented in this study is highly quantitative, and it depends on the pipeline properties (i.e. diameter, MAOP etc.) in addition to the structure properties (i.e. precise location and type of structures). The lethality zone utilized in the consequence model is a curve which has 100% lethality at the point of rupture but recedes in lethality as the point of concern moves away from the rupture location. The lethality curve is calculated using the PIPESAFE software [6] that is developed by rigorous analytical, experimental, and verification work. This ensures that the lethality curves are pipeline specific. Furthermore, the position of the structures inside the lethality zones is taken into consideration, which means the structures located closer to the pipeline see a higher degree of lethality than the structures further away from the pipeline.

Risk is represented by two specific, well defined measures: Individual Risk (IR), and Societal Risk (SR). These two measures are well accepted concepts of risk that go beyond the pipeline industry, and are particularly used in the pipeline industry in countries where quantitative risk is required by regulation (e.g. UK and Nederlands). IR takes into account the inherent risk of the pipeline to the single individual who may happen to be in the vicinity of the pipeline. SR, on the other hand, takes into account known population centers, settlements, and structures to define the risk to communities. When risk is calculated quantitatively, it is possible to use well defined and widely accepted criteria to determine the acceptability of risk in terms of IR and SR criteria for all pipelines. The advantages of using IR and SR are discussed and shown through implemented examples.

Commentary by Dr. Valentin Fuster
2014;():V003T12A020. doi:10.1115/IPC2014-33477.

A lethality zone due to an ignited natural gas release is often used to characterize the consequences of a pipeline rupture. A 1% lethality zone defines a zone where the lethality to a human is greater than or equal to 1%. The boundary of the zone is defined by the distance (from the point of rupture) at which the probability of lethality is 1%. Currently in the gas pipeline industry, the most detailed and validated method for calculating this zone is embodied in the PIPESAFE software. PIPESAFE is a software tool developed by a joint industry group for undertaking quantitative risk assessments of natural gas pipelines. PIPESAFE consequence models have been verified in laboratory experiments, full scale tests, and actual failures, and have been extensively used over the past 10–15 years for quantitative risk calculations. The primary advantage of using PIPESAFE is it allows for accurate estimation of the likelihood of lethality inside the impacted zone (i.e. receptors such as structures closer to the failure are subject to appropriately higher lethality percentages).

Potential Impact Radius (PIR) is defined as the zone in which the extent of property damage and serious or fatal injury would be expected to be significant. It corresponds to the 1% lethality zone for a natural gas pipeline of a certain diameter and pressure when thermal radiation and exposure are taken into account. PIR is one of the two methods used to identify HCAs in US (49 CFR 192.903).

Since PIR is a widely used parameter and given that it can be interpreted to delineate a 1% lethality zone, it is important to understand how PIR compares to the more accurate estimation of the lethality zones for different diameters and operating pressures. In previous internal studies, it was found that PIR, when compared to the more detailed measures of the 1% lethality zone, could be highly conservative. This conservatism could be beneficial from a safety perspective, however it is adding additional costs and reducing the efficiency of the integrity management process. Therefore, the goal of this study is to determine when PIR is overly conservative and to determine a way to address this conservatism.

In order to assess its accuracy, PIR was compared to a more accurate measure of the 1% lethality zone, calculated by PIPESAFE, for a range of different operating pressures and line diameters. Upon comparison of the distances calculated through the application of PIR and PIPESAFE, it was observed that for large diameters pipelines the distances calculated by PIR are slightly conservative, and that this conservativeness increases exponentially for smaller diameter lines. The explanation for the conservatism of the PIR for small diameter pipelines is the higher wall friction forces per volume transported in smaller diameter lines. When these higher friction forces are not accounted for it leads to overestimation of the effective outflow rate (a product of the initial flow rate and the decay factor) which subsequently leads to the overestimation of the impact radius. Since the effective outflow rate is a function of both line pressure and diameter, a simple relationship is proposed to make the decay factor a function of these two variables to correct the excess conservatism for small diameter pipelines.

Commentary by Dr. Valentin Fuster
2014;():V003T12A021. doi:10.1115/IPC2014-33511.

In order to understand the probability of failure (PoF) of any system, we must first quantify the uncertainties in the system we are modeling. Common industry practice is to assess potential pipeline defects deterministically and account for uncertainty by taking conservative estimates or adding margins of safety to each parameter. This creates very conservative values for use in integrity management. Instead, probabilistic modeling enables sources of uncertainty, such as measurement accuracy, and their influence on PoF to be quantified and independently accounted for.

Assessing threats probabilistically allows for ease of integration of likelihood with consequence for risk modeling and enables a quantitative comparison of crack, corrosion, deformation and other potential threats.

Enbridge is developing probabilistic models to obtain PoF results and an overall line condition for crack threats on its crude oil transmission pipelines. These results will enable Enbridge to quantify the effects of its integrity programs, and allow for a comparison with other potential mitigations such as hydro-testing and pressure reductions.

Enbridge’s probabilistic modeling methodology includes the selection and justification of model parameters, analysis methods, inputs, and uncertainties. This includes but is not limited to a selection of a failure model, failure criteria, pipe properties, feature sizing, and operating conditions. The selected input distributions are then sampled using a Monte Carlo method in order to calculate mean burst pressure and the resulting PoF.

Commentary by Dr. Valentin Fuster
2014;():V003T12A022. doi:10.1115/IPC2014-33546.

This paper describes how past pipeline incidents can be used to test whether the consequences of major hazard events as predicted by computer model are realistic.

Most materials transported by cross-country pipeline are flammable in nature, e.g. crude oil, petroleum products and natural gas. The hazardous consequences of a loss of containment may be modelled by a variety of computer models. One of the key inputs is how the failure may be modelled, in terms of the initial source terms and how the released material behaves subsequently. Modelling appropriate behaviour of the release is incumbent on the modeller / engineer, as well as accurately interpreting the output from the computer model.

There have been a number of high profile pipeline incidents in recent years that have had a devastating effect on the local community. Although one recognises the distressing effects of such incidents, these also provide an opportunity to test the consequences predicted by computer models. One of the key questions is whether it is likely that the adverse effects of an incident are overpredicted by the modelling inputs / technique and therefore whether the outputs from the model present a conservative thermal radiation dose.

This paper presents such a benchmarking exercise, which has been carried out to assess the degree of realism provided by computer modelling and the way in which the modelling is carried out. This exercise was conducted following a number of pipeline risk assessments where it was predicted that in some cases, hundreds of fatalities may occur following rupture of a pipeline transporting natural gas. It was felt that there may be a level of over-conservatism in the modelling, particularly as many incidents that have occurred have not resulted in the predicted level of fatal injuries.

Topics: Pipelines
Commentary by Dr. Valentin Fuster
2014;():V003T12A023. doi:10.1115/IPC2014-33552.

When installing subsea pipelines on an uneven seabed, the free spans can be vulnerable to fatigue damage caused by vortex induced vibrations (VIV). Indeed, even moderate currents can induce vortex shedding, alternately at the top and the bottom of the pipeline, at a rate determined by the flow velocity. Each time a vortex sheds, a force is generated in both the in-line and cross-flow direction, causing an oscillatory multi-mode vibration. This vortex induced vibration can give rise to fatigue damage of submarine pipeline spans, especially in the vicinity of the girth welds. Traditional design for VIV is recommended in DNV-RP-F105, which limits the allowable free span length and implies whether (and when) seabed interventions are required.

The traditional DNV-RP-F105 design method is based on a semi-empirical approach, where the allowable span length depends on the pipe properties (diameter, wall thickness, coating, steel SN_curves, …), the sea state (current velocity, wave induced velocity and period) and the soil conditions (submerged unit weight, undrained shear strength, bearing capacity,…). All these input parameters, however, exhibit a certain extent of scatter and uncertainty.

This paper presents a risk based evaluation of free spans, by applying the principles of structural reliability theory to the problem of long free spanning pipelines subjected to VIV. First, a fully deterministic on-bottom roughness analysis is performed to introduce numerical tools for free span analysis. Then, a sensitivity analysis on soil parameters is presented to show significant influence of soil properties on free span predictions.

To study the implications of uncertainty in soil properties, a First Order Reliability Method (FORM) analysis is presented at the end of this paper, where the soil properties are introduced as stochastic variables.

Commentary by Dr. Valentin Fuster
2014;():V003T12A024. doi:10.1115/IPC2014-33624.

As pipeline operators strive for safe and robust operations, the desire for improved risk awareness and operational safety, or process safety, culture continues to grow. The need for improvements related to operational safety has been felt for years throughout the oil and gas industry, but in recent years, it has also come to the forefront of the minds of pipeline operators. While most operators do not “process” anything, the principles of effective operational safety management are being stressed in pipeline incident investigations and communications from regulators.

While many organizations have found ways to improve occupational safety concerns, operational safety has remained overwhelming. It is often easier for an employee to envision the consequence that could result if he or she is splashed with a chemical; however, even an experienced operator may have a hard time imagining how what seems like a minor integrity event could escalate to a major incident.

Two critical building blocks in developing awareness of risk and operational safety are 1) ownership of risks, and 2) the ability to speak one common risk language. By giving field personnel the opportunity to maintain registers of the risks that are important to them, not necessarily the largest risks, both of these building blocks can be developed concurrently.

This paper outlines how the use of field-owned risk registers can help companies of all sizes, heritages, and cultures to improve methods for hazard identification, risk analysis, and risk control. As field personnel learn the language of risk, become familiar with ways to analyze potential consequences, and begin to understand how likely it is that an operational upset or incorrect operation could result in a major incident, personnel who otherwise might not participate in these types of activities begin to take interest. The paper provides insight into how, if implemented correctly, these risk registers can introduce risk management at all levels of the organization and provide a sense of ownership in the field regarding risk and operational safety, while still improving integrity, personal safety, and environmental protection.

Topics: Safety , Risk
Commentary by Dr. Valentin Fuster
2014;():V003T12A025. doi:10.1115/IPC2014-33639.

The System Wide Risk Assessment (SWRA) is an essential first step in the pipeline integrity management program. It is required by both Canadian and US regulators and is expected to estimate risk due to all threats, interaction of threats, and consequences. The main objective of the SWRA is to identify high risk segments so that segments with excessive risk can be mitigated.

The SWRA models developed in this study employs quantified likelihood models and consequence models. A companion paper explains the consequence models. This paper presents the framework and rationale used to produce quantifiable measures of likelihood for each threat. The quantification enables sensible comparisons between threat likelihood values and also enables realistic combining of likelihood values to produce total likelihood of failure due to all threats. It also facilitates identification of key parameters that contribute to each threat.

It is important to have a consistent risk framework that systematically applies to all the threats and accommodates all the different aspects and mitigative actions in each threat management process. For effective continuous improvement it is essential that the models are transparent and updatable. A consistent framework that is systematic, rigorous, transparent and updatable is utilized with explicit consideration to threat interactions.

The main advantages of the likelihood models developed in this study are:

• It is based on all evidence that is available for each threat (failure histories, observations from assessments, i.e., digs, HTs, and ILIs, and mechanistic understanding)

• It considers all nine threat categories and relevant subcategories where causal factors are different (such as SCC and Circumferential SCC within the crack threat category)

• It clearly considers all three types of threat interactions (Interacting coincident defects, Interacting-activating threats, and Interacting common-mode conditions) among all threat categories.

• It is based on subsystem specific historical failure rates for each threat, where subsystem is defined as a subset of pipelines that have different performance characteristics with respect to at least one threat. This basis enables the failure frequencies predicted to be more in line with reality and consequently improves accuracy of predictions and appropriate quantification.

• The subsystem specific historical failure rates are then calibrated to correlate to different mechanistic characteristics so that within-pipeline-subsystem variation due to changes in parameters is represented.

• Finally assessments or observations are used to appropriately update threat likelihood with latest knowledge from measured local observations.

All of the improvements mentioned above have helped the SWRA 2013 to produce more representative results. The comprehensive set of validation exercises verify that the results are realistic.

Commentary by Dr. Valentin Fuster
2014;():V003T12A026. doi:10.1115/IPC2014-33641.

Risk management of pipelines is a complex challenge due to the dynamic environment of the real world coupled with a wide range of system types installed over many decades. Various methods of risk assessment are currently being used in industry, many of which utilize relative scoring. These assessments are often not designed for the new integrity management program (IMP) requirements and are under direct challenge by regulators.

SemGroup had historically used relative risk assessment methodologies to help support risk management decision-making. While the formality offered by these early methods provided benefits, it was recognized that, in order to more effectively manage risk and better meet the United States IMP objectives, a more effective risk assessment would be needed.

A rapid and inexpensive migration into a better risk assessment platform was sought. The platform needed to be applicable not only to pipeline miles, but also to station facilities and all related components. The risk results had to be readily understandable and scalable, capturing risks from ‘trap to trap’ in addition to risks accompanying each segment.

The solution appeared in the form a quantitative risk assessment that was ‘physics based’ rather than the classical statistics based QRA. This paper will outline the steps involved in this transition process and show how quantitative risk assessment may be efficiently implemented to better guide integrity decision-making, illustrated with a case study from SemGroup.

Topics: Risk assessment
Commentary by Dr. Valentin Fuster
2014;():V003T12A027. doi:10.1115/IPC2014-33659.

During the regulatory phase of the Enbridge Northern Gateway Project (Northern Gateway), the Joint Review Panel (JRP) requested information on “how the risk factors resulting from the geotechnical and geographic aspects of the pipeline will be taken into account” and to demonstrate “the integration of risk factors with the environmental and socio-economic consequences from potential hydrocarbon releases”. Furthermore, the JRP required Northern Gateway to identify where a risk-based approach to design would be used to address geotechnical and seismic hazards, valve locations for spill consequence reduction and risk reduction in consequence areas”. [1]

To meet this requirement a semi-quantitative risk assessment (SQRA) was undertaken. Risk was defined as a function of probability and consequence, where the probability (expressed as a frequency) of loss of pipe integrity was quantitatively determined and the consequence of failure was qualitatively determined. The frequency of failure was a probabilistic combination of the calculated probability of failure from reliability methods, historical frequencies and assessed geo-hazard failure frequency rates. Consequence scoring was based on intersection of theoretical spills with “consequence areas” for environmental or socio-economic effects Frequency and consequence were then combined to provide risk scoring and ranking.

Failure frequencies were developed using reliability methods where appropriate. The use of reliability methods addresses the primary challenge associated with quantifying risk for new pipelines as industry failure statistics are not directly applicable to modern pipeline designs, materials, and operating practices. In the pipeline industry, reliability models exist for the most significant threats, including third-party damage, internal corrosion and external corrosion. In addition, geotechnical threats can be characterized in terms of expected magnitude and associated frequency of occurrence, thereby enabling pipeline reliability to be established for each geo-hazard.

Consequence scoring was based on modeling full bore rupture spill scenarios and determining whether these spills would potentially intersect identified “consequence areas”.

Over the course of the application and hearing process two SQRA’s were undertaken. Following the filing of the first SQRA, additional measures were included in the pipeline design to reduce the frequency of failure and to reduce potential consequences. This resulted in the calculated overall risk being reduced by a factor of 84%, primarily due to increases in wall thickness resulting in a reduction in the likelihood of 3rd party damage and in a reduction of consequence by an increased number of valves.

Commentary by Dr. Valentin Fuster
2014;():V003T12A028. doi:10.1115/IPC2014-33694.

This paper presents the severe and often times game-changing limitations associated with our observational learning mode and the ineffective knowledge base that results from it. More specifically, this paper stands our current approach of pipeline risk assessment on its head, describing how “rare-events” data, as they relate to significant pipeline failures, are simply not being addressed by our probabilistic risk analyses and modern statistical approaches. Further, risk blindness has been shown to consistently minimize the perception of high-consequence risk in the corporate world. Touching on themes from Nassim Taleb’s book, The Black Swan [1], this paper is about those failure events that lie outside the realm of regular possibility. Such events would never have been convincingly predicted prior to their occurrence, and yet they carry an extreme impact. Note that just because a failure was catastrophic does not necessarily mean it was a Black Swan. For example the financial melt down of 1982 was a catastrophe, but it was not a black swan. There have been numerous pipeline failures over the years that were quite catastrophic, but which were not Black Swans as referenced in this paper. They were neglect. It is important not to confuse the two.

Topics: Pipelines , Failure
Commentary by Dr. Valentin Fuster
2014;():V003T12A029. doi:10.1115/IPC2014-33705.

To the end of 2012, Enbridge Pipelines employed an in-house developed indexed or relative risk assessment algorithm to model its liquid pipeline system. Using this model, Enbridge was able to identify risk control or treatment projects (e.g. valve placement) that could mitigate identified high risk areas. A changing understanding of the threats faced by a liquid pipeline system and their consequences meant that the model changed year over year making it difficult to demonstrate risk reduction accomplished on an annual basis using a relative scoring system.

As the development of risk management evolved within the company, the expectations on the model also evolved and significantly increased. For example, questions were being asked such as “what risk is acceptable and what risk is not acceptable?”, “what is the true risk of failure for a given pipe section that considers the likelihood of all threats applicable to the pipeline”, and “is enough being done to reduce these risks to acceptable levels?”

To this end, starting in 2012 and continuing through to the end of 2013, Enbridge Pipelines developed a quantitative mainline risk assessment model. This tool quantifies both threat likelihood and consequence and offers advantages over the indexed risk assessment model in the following areas:

• Models likely worst case (P90) rupture scenarios

• Enables independent evaluation of threats and consequences in order to understand the drivers

• Produces risk assessment results in uniform units for all consequence criteria and in terms of frequencies of failure for likelihood

• Aggregates likelihood and consequence at varying levels of granularity

• Uses the risk appetite of the organization and its quantification allows for the setting of defined high, medium, and low risk targets

• Quantifies the amount of risk in dollars/year facilitating cost-benefit analyses of mitigation efforts and risk reduction activities

• Grounds risk assessment results on changes in product volume-out and receptor sensitivity

• Balances between complexity and utility by using enough information and data granularity to capture all factors that have a meaningful impact on risk

Development and implementation of the quantitative mainline risk assessment tool has had a number of challenges and hurdles. This paper provides an overview of the approach used by Enbridge to develop its quantitative mainline risk assessment model and examines the challenges, learnings and successes that have been achieved in its implementation.

Commentary by Dr. Valentin Fuster
2014;():V003T12A030. doi:10.1115/IPC2014-33728.

This is a report of the calculation of the probability of brittle fracture and crack arrest for a series of X42 A-series and B-series pipelines. This paper provides the probabilistic analysis to determine the probability distribution of crack propagation velocities using the material resistance developed from Charpy and Drop Weight Tear Tests (DWTT).

Commentary by Dr. Valentin Fuster

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In