TWI Ltd, Cambridge, UK
Paper presented at the 6th International Pipeline Technology Conference (Rudi's Pipeline Conference), 6-9 October 2013, Ostend, Belgium.
Abstract
Crack Tip Opening Displacement (or CTOD) has been the most widely used fracture toughness parameter within the oil and gas industry for nearly fifty years. Originally developed from research at TWI in the UK during the 1960’s, CTOD was an ideal parameter for characterising the fracture toughness of medium strength carbon manganese steels used in pressure vessels, offshore platforms and pipelines where the application of linear elastic fracture mechanics was insufficient to account for their ductility. Once fracture toughness testing (CTOD testing) became standardised within BS 7448, ASTM E1290, ISO 12135 and ISO 15653, the CTOD concept enjoyed an established international reputation. The development of standardised fitness-for-service assessment procedures, initially through the use of the CTOD design curve, and then to the failure analysis diagram approach described in BS 7910, also allowed CTOD to be used directly to determine tolerable flaw sizes to assess the structural integrity of welds. In more recent times single edge notched tension specimen (SENT) testing has been enthusiastically adopted by the pipeline industry in place of the traditional single edge notched bend (SENB) specimen used for standard CTOD tests. However, currently there is no national standard describing SENT testing, although this is being developed. SENT testing is particularly advantageous when pipeline girth welds are subjected to plastic straining, and a number of assessment procedures based on CTOD have been and are being developed to define strain capacity and flaw acceptance criteria.
Introduction
Crack Tip Opening Displacement (or CTOD) has been the most widely used fracture toughness parameter within the oil and gas industry for nearly fifty years. Originally developed from research at TWI in the UK during the 1960’s, CTOD was an ideal parameter for characterising the fracture toughness of medium strength carbon manganese steels used in the manufacture of pressure vessels where the application of linear elastic fracture mechanics was insufficient to account for their ductility. The development of North Sea oil and gas from the 1970s onwards hastened the application of CTOD testing and analysis concepts for application to the construction of steel jacket production platforms and pipelines. The fracture toughness testing of single edge notched bend specimens (or the ‘CTOD test’ as it is sometimes called) is the standard method to measure it. However, as further progress is made in the development of fracture mechanics, both testing and assessment, the CTOD concept must change and adapt to keep up. To illustrate the story of CTOD over the years we have highlighted some of the key people at TWI who have been involved in this work, however it must be recognised that the research and development of this field has been the collaboration of a much larger group of engineers from many institutions worldwide.
The Origins of the CTOD concept
Fracture mechanics as an engineering discipline was conceived just after the Second World War as a result of the Liberty Ships fractures. Of 2700 ships fabricated using the new welding technology during the war around 400 had fractures, 90 were considered serious and about ten ships fractured completely in half [1]. This failure rate had driven the US Naval Research Labs to research the effect of cracks in steels, and by the 1950s they had developed the linear elastic fracture mechanics (LEFM) description of cracks in brittle materials; work led by Dr George L Irwin [2]. However, the stress-based LEFM did not sufficiently describe the behaviour of more ductile materials, such as medium strength structural steels.
The UK had chosen to begin its own investigations into brittle fracture issues after the war, driven by the UK Admiralty Advisory Committee on Structural Steels, who held conferences at the University of Cambridge in 1945 and 1959 attended by many of those who would become eminent in the fields of structural engineering, metallurgy and fracture mechanics, including George Irwin and Dr Alan Wells from the British Welding Research Association [3]. Alan Wells (Figure 1) had taken a sabbatical at the US Naval Research Lab in 1954 and worked with George Irwin in that time. After his return to the UK and the British Welding Research Association (BWRA), which later became TWI, Alan Wells proposed an alternative model of fracture to LEFM in 1961 [4]. Wells developed the Crack Opening Displacement (COD), later the Crack Tip Opening Displacement (CTOD) model of fracture mechanics from an observation of the movement of the crack faces apart during plastic deformation of notched test pieces. He showed that fracture would take place at a critical value of COD and for calculations below general yield this was proportional to the square of the critical stress intensity factor divided by the yield strength. Furthermore, he showed that the critical value of COD determined in bend specimens and wide plate specimens (representing structural components) of the same thickness were equivalent. Thus he was able to demonstrate transferability of fracture toughness determined from test specimens to other structural geometries. This was to have far reaching implications on the development of fitness-for-service concepts for welded structures for the avoidance of fracture. As a result of this the CTOD parameter was used extensively in the UK for elastic-plastic fracture mechanics (EPFM) analysis of welded structures from the 1960s especially once the development of North Sea oil reserves in the 1970s was driving much of the fracture research at that time [1].
Standard Methods for CTOD Testing
In essence, the fracture toughness test specimen comprises a rectangular bar of material that is notched into the appropriate region (with respect to a welded joint). In the case of the CTOD test, the specimen size is usually representative of the full material thickness. The CTOD test piece originally had a saw cut notch but later used fatigue pre-cracking to produce a sharp notch [5]. The crack mouth is instrumented with a clip gauge to measure the crack mouth opening, and then loaded under quasi-static three point bending to enable a load versus crack mouth opening displacement trace to be plotted (Figures 2 and 3).
As confidence grew in the ability of the small scale CTOD test to predict the fracture conditions of a crack in a full scale structure, the test method became standardised. BSI published a ‘Draft for Development’ (DD19) on applied force for CTOD testing in 1972. This became a standard in 1979 as BS 5762 [6], which described ‘Methods for crack opening displacement (COD) testing’. This was then superseded by BS 7448 Part 1 in 1991 [7] as a ‘Method for determination of KIC, critical CTOD and critical J values of metallic materials’. Part 2 of BS 7448 provided an equivalent method for welds in metallic materials when it was published in 1997 and was largely based on TWI’s experience in testing welds. This was developed further by an ISO committee, with TWI representation by Henryk Pisarski (Figure 4), and in 2010 BS 7448 Part 2 was superseded by BS EN ISO 15653 [8], although Part 1 still remains current for CTOD testing of parent metals.
Once fracture toughness testing became standardised within BS 7448 [7], ASTM E1290 [9], E1820 [10], ISO 12135 [11] and ISO 15653 [8], the CTOD concept enjoyed an established international reputation. CTOD had been established as the fracture toughness parameter for the oil and gas industry so thoroughly, that often the phrase “CTOD test” has been used interchangeably with the more precise “fracture mechanics test” by those in that industry.
The crack tip opening displacement can be conceptually understood as the amount that a crack tip needs to be opened up (or the distance the crack faces need to be moved apart) before unstable propagation of the crack occurs (Figure 6a). However, several alternatives have been put forward as to how to exactly define CTOD, illustrated in Figure 6. For materials with lower ductility the original definition (Figure 6a) is fine, but for more ductile materials, unstable fracture may not occur and the CTOD is determined from the point of maximum load in the test, and CTOD can be considered the opening displacement of the deformed crack at the tip position of the original crack (Figure 6b). A third definition is that most often used when performing numerical models of cracks, which defines the CTOD as the displacement at the points where a 90° angle at the crack tip intersects with the crack sides (Figure 6c).
Fitness-for-Service Assessment
One of the earliest codified applications of the CTOD concept was to provide alternative flaw acceptance criteria to those in Appendix A of API 1104 [15] in the 1970s. The Welding Institute's CTOD design curve was developed by Mike Dawes (Figure 5) and colleagues, at a time when very little guidance was available on the application of elastic-plastic fracture mechanics (EPFM) analyses to common materials, particularly to welded structures with high residual stresses and stress concentrations. The CTOD design curve [16] was intended to provide a logical, simple, and rapid means of determining the allowable crack sizes in welded structures subjected to normal design loads. The importance of CTOD was that it could be used directly to calculate the maximum tolerable flaw size for a given weld.
The CTOD design curve approach was intended to be used as a first, coarse, filter in fitness-for-purpose assessments. The subsequent development of standardised fitness-for-service assessment procedures, initially through the use of the CTOD design curve and then to failure analysis diagram approach, allowed CTOD to be used to more accurately determine tolerable flaw sizes to assess the structural integrity of welds. Published standard fitness-for-service (FFS) assessment procedures also cemented the power of fracture mechanics testing. Successful experience using the CTOD concept to determine tolerable flaw sizes over almost a decade led to the development of a FFS Published Document (PD 6493) of the British Standards Institution. Initially published as PD 6493:1980 [17] ‘Guidance on methods for assessing the acceptability of flaws in fusion welded structures’, the procedure was revised in 1991 and subsequently became a standard, BS 7910 [13], in 1999.
FFS assessment allowed larger flaws to be shown as tolerable, for example in offshore platforms and pipelines, compared to the small flaw sizes permitted by applying ‘workmanship’ flaw limitations imposed by welding standards. At TWI, John Harrison (Figure 7) had performed numerous FFS assessments using both the CTOD design curve and PD6493, and demonstrated the effectiveness of these methods to industry, particularly for oil and gas [18]. He became heavily involved in the standardisation of PD 6493 into BS 7910. These FFS methods were used for a number of offshore installations in the 1970s and 1980s where it had been necessary to demonstrate avoidance of brittle fracture in as-welded joints in thicker section (40-120mm). Similar to pressure vessel practise, highly stressed welds for operation at sub-zero temperatures would normally be subjected to post-weld heat treatment (PWHT) when section thickness exceeds 40mm, and avoiding PWHT needed this kind of rigorous justification. These issues were addressed in greater detail in the UK Department of Energy Guidance Notes on the design, construction and certification of offshore structures, which was first published in 1990 [19]. This document had a large influence on the design of steel jackets operating in the North Sea. The sections dealing with toughness requirements for steels, avoidance of brittle fracture and post weld heat treatment were largely based on work conducted by John Harrison and Henryk Pisarski [20], and utilised concepts described in PD 6493. Where problems had arisen with flaws having been detected and repair strategies needing to be decided, a FFS assessment could justify whether repair without subsequent PWHT was acceptable or not. With the standard methods, performing a FFS assessment became a regular part of the preparation for any new pipeline installation in order to set the fabrication flaw acceptance criteria.
SENT Testing
The desire to extend the application of pipelines to ever more challenging loading conditions during both during installation and operation has been driving the development of fitness for service methods to reduce their over-conservatism while remaining confident that the structure will be safe. Pipelines intended for deeper water, higher pressure, installation methods or upset conditions (e.g. ground movement) involving strains beyond yield impose greater challenges to integrity.
The development of fracture toughness testing had traditionally been through the use of the deeply notched bend (SENB) tests, which are intended to impose a high degree of crack tip constraint, and hence provide a lower bound estimate of fracture toughness. Recognition that flaws in pipeline girth welds are subjected to lower crack tip constraint has led to the development of the single edge notch tension (SENT) test which has lower constraint than the SENB specimen. Collaboration between Henryk Pisarski (Figure 6) at TWI in the UK, and SINTEF and DNV in Norway along with a group in industry partners produced guidance for fracture control for pipeline installation methods introducing cyclic plastic strain, which became DNV’s Recommended Practice DNV-RP-F108 in 2006 [21]. The method intended for pipe installation by methods such as reeling used SENT specimens to measure fracture toughness from a notched specimen whose constraint more closely matched that of a flaw in a girth weld. The method used multiple-specimens to produce a J R-curve which was then used in an assessment procedure based on BS7910, to generate flaw acceptance criteria.
The higher value of fracture toughness that can be obtained from a SENT specimen compared to a SENB specimen has led to a rapid growth of interest in using these specimens for other fracture mechanics assessment and testing. SENT testing has been enthusiastically been adopted by the offshore pipeline industry (eg in DNV-OS-F101) and gradually been accepted by the pipeline industry in general. However, the initial testing procedures for SENT specimens gave the fracture toughness exclusively in terms of J R-curves rather than CTOD.
Although developed for pipeline installation, there is growing evidence that the biaxial loading experience by pipelines during operation may also exhibit similar ductile tearing resistance as the R-curve measured from SENT testing [22], making SENT specimens appropriate for analysis of pipeline operation as well. DNV-OS-F101 standard for submarine pipeline systems [23] describes in its Appendix A a fracture mechanics method to determine the acceptability of flaws. The 2012 version requires the fracture toughness testing to be performed on SENT specimens, and expresses the fracture toughness requirements in terms of J. It allows the fracture toughness to be expressed as CTOD only if the procedure for calculating CTOD is demonstrated to be conservative.
CTOD from SENT tests
Despite its growing popularity, there is currently no standardised procedure for carrying out SENT tests, nor for determining CTOD in a SENT specimen. This gap in CTOD knowledge led to a study by TWI to validate methods for determining CTOD in the SENT specimen [24]. The work compared direct measurements of CTOD made by silicone crack infiltration to a finite element model of the SENT specimen to predict CTOD, and the double clip gauge method to determine CTOD using the equivalent triangles rule. These methods were also compared to CTOD from J equations in recent literature to improve upon the over-conservative equation given in the 2010 version of DNV-OS-F101 [25].
Using silicone rubber crack infiltration allowed direct measurement of CTOD to be made from replicas of the SENT specimen notch, although the method is not practical for routine testing. FEA models also give a reliable way to determine CTOD, but require too much analytical processing to be practical for determining CTOD for every fracture mechanics test. These methods were used to compare the effectiveness of other simpler methods for calculating CTOD.
DNV-OS-F101:2012 [23] now also gives a revised method to calculate CTODmat from Jmat (equation 3), which uses the material yield strength (YS) and an estimate of CTOD to be iteratively improved using the following clauses to determine the parameter X.
CTODmat≤0.1 mm; X = 1.8
0.1 mm < CTODmat< 0.4 mm; X = 1.9 - CTOD mat
CTODmat³0.4 mm; X = 1.5
The intention of this comparison was to provide confidence in the value of CTOD that is determine when using SENT test specimens, so that the validation of the SENT specimen approach keeps pace with the pipeline industry’s need to continue to define fracture toughness in terms of CTOD, while using the most modern test methods.
The Future
Further research into SENT testing with the intention of developing a complete and thorough testing standard is being carried out at TWI as part of a Group Sponsored Project, while research on SENT testing is also active in the US, Brazil and Canada. The authors of this paper are involved in the British Standards committee which will eventually publish the SENT testing standard as BS 8571 in the near future.
Nevertheless, the CTOD fracture toughness parameter is being withdrawn from the fitness for service assessment standard BS7910 in the 2013 version, which will directly use only K to determine the fracture ratio. However, the intuitive understanding of CTOD as fracture toughness by a wide range of industries will mean that it will continue to be used to describe fracture toughness for some time to come. Indeed, there is current discussion on whether J or CTOD are the appropriate fracture characterising parameters when considering assessments in the plastic regime. For example, the strain-based assessment procedure used by ExxonMobil uses CTOD [26]. Pipeline assessment research and development continues towards strain-based methods and the CTOD parameter is well suited for this application.
TWI is proud of its history and involvement in the story of CTOD, and of individual colleagues whose work in those early years allowed CTOD to become such a versatile fracture parameter, well established within the oil and gas industry and beyond. The present is a time of further changes in both fracture mechanics testing, assessment and standardisation, but offers a bright future for further research and development in this field, not just at TWI, but worldwide.
Acknowledgements
The authors are grateful to Dr Mike Dawes for providing a photograph and some biography text for this paper, despite his retirement.
References
[1] Anderson, T L, ‘Fracture Mechanics – Fundamentals and Applications’ Second Edition, CRC Press, 1995.
[2] Irwin G. ‘Analysis of stresses and strains near the end of a crack traversing a plate’, Journal of Applied Mechanics 24, 361–364, 1957.
[3] Knott, J. ‘From CODs to CODES (The Realisation of Fracture Mechanics in the UK)’, Fracture Research in Retrospect – An Anniversary Volume in Honour of George R, Irwin’s 90th Birthday, Edited by H P Rossmanith, Published by A.A.Balkema, Netherlands, ISBN 9054106794, 1997.
[4] Wells, A A. ‘Unstable crack propagation in metals: cleavage and fast fracture’, Proceedings of the crack propagation symposium, Cranfield, UK, Vol. 2, 210, 1961.
[5] Houldcroft, P. ‘Fifty years of service to industry – A brief and occasionally lighthearted history of BWRA and The Welding Institute’, Published by TWI, 1996.
[6] BS 5762:1979 ‘Methods for crack opening displacement (COD) testing’, British Standards Institution, 1979.
[7] BS 7448-1:1991 ‘Fracture mechanics toughness tests: Part 1: Method for determination of Kic, critical CTOD and critical J values of metallic materials’, British Standards Institution, 1991.
[8] BS EN ISO 15653:2010 ‘Metallic materials - Method of test for the determination of quasistatic fracture toughness of welds’, British Standards Institution, 2010.
[9] ASTM E1290-08e1 ‘Standard Test Method for Crack-Tip Opening Displacement (CTOD) Fracture Toughness Measurement’ (Withdrawn 2013), American Society for Testing and Materials, 2008.
[10] ASTM E1820 - 11e1 ‘Standard Test Method for Measurement of Fracture Toughness’, American Society for Testing and Materials, 2011.
[11] ISO 12135:2002 ‘Metallic materials - Unified method of test for the determination of quasistatic fracture toughness’, International Standards Organisation, 2008.
[12] Dawes, M G. 'Elastic-Plastic Fracture Toughness based on COD and J-contour Integral Concepts’, in Elastic-Plastic Fracture, ASTM STP 668, American Society for Testing and Materials, 1979. pp 307-333.
[13] BS 7910:2005 ‘Guide to methods for assessing the acceptability of flaws in metallic structures’, British Standards Institution, 2005.
[14] Harrison, J D. The “State of the Art” in Crack Tip Opening Displacement (CTOD) Testing and Analysis, TWI Members Report 108/1980, April 1980.
[15] API 1104:1973 ‘Standard for welding pipelines and related facilities’, 13th edition (superseded), American Petroleum Institute, 1973.
[16] Burdekin, F M and Dawes, M G ‘Practical use of linear elastic and yielding fracture mechanics with particular reference to pressure vessels, In proceedings, Institution of Mechanical Engineers Conference on 'Practical Application of Fracture Mechanics to Pressure Vessel Technology’, London, 3-5 April 1971, pp 28-37.
[17] PD 6493:1980 ‘Guidance on some methods for the derivation of acceptance levels for defects in fusion welded joints’ (superseded), British Standards Institution, 1980.
[18] Harrison J D, Dawes M G, Archer, G L and Kamath M S. ‘The COD Approach and its Application to Welded Structures’, in Elastic-Plastic Fracture, ASTM STP 668, American Society for Testing and Materials, 1979. pp 606-631.
[19] UK Department of Energy Standard, ‘Offshore installations: guidance on design, construction and certification’. 3rd and 4th Editions. Publ: HMSO, London, 1984 and 1990.
[20] Harrison, J D and Pisarski, H G, ‘Background to new guidance on structural steel and steel construction standards in offshore structures’, Publ. HMSO, London, 1986.
[21] DNV-RP-F108:2006. Recommended Practice F108 ‘Fracture Control for Pipeline Installation Methods Introducing Cyclic Plastic Strain’, January 2006
[22] MacDonald, K A. ‘Fracture and Fatigue of Welded Joints and Structures’, Chapter 2 ‘Constraint fracture mechanics: test methods’ by MacDonald K A, Ostby E & Nyhus, B. Woodhead Publishing Ltd; 2011.
[23] DNV–OS-F101:2012. Offshore Standard F101 ‘Submarine Pipeline Systems’, Det Norske Veritas, August 2012.
[24] Moore, P and Pisarski, H. ‘Validation of Methods to Determine CTOD from SENT specimens’, in proceedings ISOPE-2012, The 22nd International Offshore (Ocean) and Polar Engineering Conference, Rhodes, Greece, June 17−22, 2012.
[25] DNV-OS-F101:2010. Offshore Standard ‘Submarine pipeline systems’ (superseded). Det Norske Veritas, 2010.
[26] Shen G and Tyson W R. ‘Evaluation of CTOD from J-integral for SE(T) specimens’, in Proceedings, Pipeline Technology Conference, Ostend, 12-14 October 2009.
[27] Moreira F and Donato G. ‘Estimation procedures for J and CTOD fracture parameters experimental evaluation using homogeneous and mismatched clamped SE(T) specimens’, in Proceedings ASME 2010 Pressure Vessels and Piping Conference. PVP2010, Bellevue, Washington, USA, July 18-22 2010.
[28] Tang H, Macia M, Minnaar K, Gioielli P, Kibey S and Fairchild D. ‘Development of the SENT test for strain-based design of welded pipelines’, In Proceedings IPC 2010 8th International Pipeline Conference, Calgary, Canada, 27 Sept.-1 Oct.2010.
Appendix – Biographies of TWI Engineers Cited
Dr Alan Wells died at the end of 2005. By the time he formally retired from TWI in the summer of 1988 Alan Wells had notched up 25 years at TWI’s site in Great Abington, the last 11 as Director General. His relationship with TWI, in the guise of the British Welding Research Association, dated back to 1950, a few months after completing his PhD in Soil Mechanics. As a fellow of the Royal Society, Dean of Faculty for four of his years at Queen's University in Belfast, a recipient of the Order of the British Empire and one of the very few non-Americans to have contributed to the US Navy's brittle fracture work, he believed his career peaked relatively early. In 1954 he was able to accept an invitation to work with Dr George Irwin at the Naval Research Laboratory in Washington DC. That placement was key in his role in developing the CTOD concept, but in addition Alan Wells also worked to establish the Wells wide plate test. Wells later spent 13 years as the Head of Civil Engineering at Queen's University Belfast before returning to TWI as Director General. Upon retirement he was involved in several major failure investigations including a fracture under hydrostatic test of one of the Sizewell power station boilers and the Westgate bridge collapse in Australia.
Dr Henryk Pisarski is a TWI Technology Fellow and works in the Structural Integrity Technology Group of TWI. He has been with TWI since 1973. His interests include the application of fracture mechanics based assessment and testing methods to assure the integrity of welded structures, with respect to fracture avoidance and to demonstrate fitness-for-service. He is also involved in the development of strain-based procedures for the assessment of pipeline girth welds. He has managed projects applying fracture mechanics testing and assessment methods to a wide range of engineering structures including ships, offshore structures, subsea components, pipelines and pressure vessels. In addition, he has contributed to standards bodies on fracture toughness testing and flaw assessment (ISO 15653, BS 7448 & BS 7910). He has published a number of papers on these subjects and also carried out expert witness work. Currently, he is the UK delegate to IIW Commission X (Structural performance of welded joints – fracture avoidance) and is contributing to the revision of BS 7910 (flaw assessment).
Dr Mike G Dawes was involved in the development of the CTOD approach to fracture avoidance at TWI from 1968 until he retired in 2000. By that time his CTOD Design Curve and fracture toughness test method relationships were used in all the corresponding BSI, ASTM and ISO standards. Much of his work, including his PhD thesis, was concerned with applications to welded metal structures. This included development of the local compression treatment, which for the first time enabled acceptable shapes of fatigue pre-cracks to be obtained in fracture toughness test specimens extracted from as-welded joints. In 1993, in recognition of his work on international standards, he received the ASTM’s highest award, the Award of Merit, and the title of Honorary Fellow.
Dr John Harrison joined TWI in 1962, when he worked on fatigue of welded structures, concentrating in particular on the significance of weld defects, the topic upon which he obtained his PhD from Cambridge University. He then became responsible for TWI's fracture research, eventually being promoted Head of Engineering Research, covering design engineering, fatigue, fracture and non-destructive testing. In 1985, the Engineering and Materials Group was formed, comprising the Design Engineering, Fatigue, Fracture, Non-Destructive Testing and Materials Departments, and John was appointed Group Manager of a total of 90 staff. John Harrison has published over 50 papers on fatigue and fracture and has been extensively involved in national and international committees. In 1988 he resigned Chairmanship of Commission XIII of the International Institute of Welding which deals with the fatigue behaviour of welded structures and components, a position which he held for 15 years. He retired from TWI in 1998.
An abstract for the paper can be seen below and the full article is available at www.j-pipe-eng.com to subscribers of the Journal of Pipeline Engineering.