Joining requirements for oil and gas work - yesterday and tomorrow

TWI Bulletin, January/February 1994

Richard Dolby is Director, Research and Technology of TWI. He received his first degree and doctorate from Cambridge University. Following employment at Alcan Industries and General Electric Company, he joined The Welding Institute in 1966 and spent 14 years in the Materials Department, becoming Head of that Department in 1978. His major interest during that time was in the welding metallurgy of ferritic steels, specialising in lamellar tearing and reheat cracking, and in HAZ and weld metal toughness problems. He is the author of over 50 technical papers and for two of these won the Pfeil prize of the Metals Society in 1972 and the Sir William Larke Medal of The Welding Institute in 1982. He was awarded The Welding Institute's Brooker medal in 1990 for distinguished technical contributions to the profession. He became a Fellow of The Welding Institute and of The Institute of Metals in 1977 and was elected to the Fellowship of Engineering in 1987.

Speaking at a recent oil and gas conference, TWI's Director of Research and Technology Richard Dolby looks at the joining trends of recent years and the joining demands, both on and offshore, which can be expected of the later nineties.

'One of the tasks of any manager helping to run an R&D company, whether a stand alone organisation or part of a larger group, is to show that R&D has tangible benefits to the financial bottom line. Judging from the shrinkage of many corporate laboratories in large international companies at the moment, this is no easy task.

However, in setting about justifying R&D, the starting point is an assessment of successes and achievements. From the customer viewpoint one very important factor to be considered, apart from the investment made, is the 'time to market'. In other words, the time it takes to get the benefits of the R&D into the market place, achieve competitive advantage and make money.

This presentation summarises a number of R&D successes in joining and related technologies relevant to the oil and gas industry over the last 20 years or so, and looks at the time it has taken for the market to perceive the benefits. It goes on to mention some of the R&D challenges still facing us.

We should start with a knowledge of the key technologies involved, see Fig.1. An organisation like TWI spends just over half of its R&D effort on joining processes and just under half on design, engineering, structural integrity and materials behaviour. Why are we doing R&D? The answer is that our aims are very clearly to:

• Reduce costs;
• Improve safety and reliability;
• Improve product quality;
• Produce options for successful innovation.

Figure 1 Key technologies in joining R&D.

Most of these are the drivers for technological change in each industrial sector.

In picking out R&D successes for the oil and gas industry, this paper highlights six. For each the milestones are shown in the figures and the challenges will be identified.

R&D successes

Avoidance of lamellar tearing

The TT test was built into the UK Offshore Guidance Notes issued in 1977 and that year also saw the emergence of special plate such as 'Hyzed' with greatly improved TT ductility. Since then TT testing in plate procurement has regularly been carried out. Steel cleanliness has dramatically improved from many suppliers. It has now reached the point where, for plate procurement for node areas, TT testing has been abandoned for some projects, full confidence being placed in the accreditation of the steel making process route of certain suppliers. This perhaps is the last milestone of this success story. Figure 2 shows that first acceptance of the results from the start of R&D took eight years.

Figure 2 Avoidance of lamellar tearing - milestones.

Weldable duplex stainless steels

The early sixties saw the emergence of several steel grades with a two phase structure having improved resistance to chloride SCC (stress corrosion cracking) and the problem is to maintain corrosion resistance and toughness in the weld region. Intensive research on weldability can be traced back to the early seventies ( Fig.3) and the first results showed the materials to be generally weldable. It was not until the mid-seventies that the importance of the ferrite:austenite phase balance in the HAZ became regarded as crucial to service performance and steels then emerged alloyed with nitrogen to improve the HAZ problems.

Figure 3 Weldable duplex stainless steels - milestones.

In 1978 we saw the first subsea pipeline in duplex steel in Europe. Five years later, the oil and gas industry gave general acceptance to duplex steels for production platforms and other installations both to save weight, because of their useful high strength, but also to handle CO ₂ containing mixes.

The next milestone was the mid-eighties when the superduplex steels emerged, primarily to improve corrosion resistance and seawater handling, but they also had improved strength compared with the normal duplex grades. However, welding was not without its problems initially, and weldability of the super duplexes is still being studied. The conditions leading to satisfactory welds and appropriate consumable design are now much better understood, particularly the concept of the safe 'welding window' which develops the optimum microstructure without embrittling intermetallics.

Here it took about five years from the start of welding R&D into duplex steels to first acceptance in the offshore industry. However, there is still no comprehensive international guide on welding of duplex steels and cost is still a major consideration.

Challenges remaining:

• Optimisation of welding procedures for super duplex grades;
• Cheaper weldable duplex grades.

Very low hydrogen MMA electrodes

Figure 4 MMA electrodes giving very low hydrogen levels - milestones.

At that time hydrogen levels in basic low hydrogen electrodes were typically around 5-6 ml/100g but only after high temperature baking of electrodes. Levels much lower than these were achievable by electrode designers at the time, but at the expense of electrode handleability and fragility of their coatings. Some further milestones in achieving low levels with good handleability are shown in Fig.4. The current very low levels of less than 3 ml/100g have had a substantial and beneficial effect on electrode management in offshore fabrication yards. The improvements in electrode coating formulation in electrode manufacturing and in electrode packaging have eliminated the need for baking of electrodes and the requirement for heated welder quivers. Quality assurance has been simplified and cost savings of around 50% of total consumable costs have been achieved overall.

In addition it has been possible to eliminate preheat altogether in some situations, with further substantial cost savings being made. It is the installation of preheat equipment which is the major cost element to the fabricator rather than the absolute level of preheat temperature.

A better system is still needed to predict weld metal hydrogen cracking and further substantial costs could be saved if we can reduce the 48 hour inspection delay normally involved in fabrication.

Challenges remaining:

• Reduction of waiting period prior to inspection;
• Improved system for predicting weld metal cracking.

Avoiding fracture in welded joints

The milestones for this subject ( Fig.5) stretch back many decades, but from the viewpoint of offshore and arctic engineering, we shall start in the mid-sixties. Linear elastic fracture mechanics (LEFM) had wide acceptance in the nuclear and aerospace fields, but only after the concept of CTOD had been proposed in 1961, and a further seven years R&D had elapsed, did the first proposal for assessing defect significance in welded structural steels emerge. This was in 1968 and the approach began to be used on an individual company basis from about 1970. Gradual unification of the K and CTOD test methods took place, and the way was clear for the first publication of a document from any standards organisation on defect acceptance in C-Mn steels in 1980; the first edition of PD6493.

Figure 5 Avoiding fracture in welded joints - milestones.

It took another decade before there was a formal unification of K, J and CTOD test methods and the publication of the revision to PD6493 in 1991. More recently the software package PC6493 has simplified application of the standard.

This has been a major success story for the oil and gas industry. Despite the ongoing debate about the fiddly nature of FM (fracture mechanics) tests, the need for a simpler approach and the fact that FM testing can be on the critical path in the fabrication of jackets and engineering structures, it has had enormous benefits to the financial bottom line of many organisations. Many oil and gas companies now use the approach in the design concept phase, for problems during fabrication and for problems in service. To give one example, the acceptance of defects in brace to chord connections during one Forties platform construction in 1974, using the CTOD design curve approach, saved £60m, by eliminating steel replacement and repair costs and floating out the jacket on time.

The industry has learned gradually over 20 years to live with defects, and confidence in present day analyses such as PD6493 is increasing.

What are the challenges left? Two major areas are 'constraint' and 'risk'. Methods for predicting effects of constraint on toughness will lead to the long sought-after simplification of FM testing techniques, and incorporation of probabilistic methods of analysis in PD6493 will allow some of the current conservatism to be appreciated and reduced where required.

Challenges remaining:

• Prediction of effects of constraint on cleavage fracture;
• Probabilistic assessments.

Design against fatigue

Figure 6 Design against fatigue of tubular joints - milestones.

The DEn Offshore Guidance Notes issued in 1984 gave much improved design advice on fatigue and included, for the first time, rules to cover the thickness effect. A new phase of R&D started in the UK in the same year and another review of European experience took place in Delft in 1987. Industry now finds itself in a position where there has been significant increase in knowledge arising from many R&D programmes in Europe and elsewhere since the current Guidance Notes were issued in the mid-eighties, but there are few documents keeping pace with this information.

Among the R&D challenges in the fatigue area in the next decade are development of a generalised methodology for predicting hot spot stress ranges by finite element methods, the behaviour of internally stiffened tubulars, and the wider use of high strength steels in fatigue sensitive areas. There is also an urgent need for an up-to-date single international design guide which incorporates generally accepted knowledge gathered in the last 10 years.

Challenges remaining:

• Generalised prediction system for accurate hot spot stresses;
• Behaviour of stiffened tubulars;
• Corrosion fatigue of high strength steels (500-700MPa);
• Single international design code incorporating new knowledge.

Avoidance of post-weld heat treatment

Figure 7 shows the milestones in terms of improving weld metal and HAZ toughness. MMA weld metal Cv and CTOD properties have significantly increased over 30 years, through a progressive understanding of the controlling chemical and metallurgical factors. Plotting the improvements in HAZ toughness is more difficult. There were great difficulties in measuring the toughness accurately until the late 80s, however, Fig.7b shows how the CE (carbon equivalent) has decreased dramatically over 20 years. CE can be regarded as a useful measure of HAZ toughness for the same steel type, and HAZ toughness undoubtedly increased and contributed to the present day position. Now joints of up to 100mm thickness are used in the as-welded condition offshore, where operating temperatures are -10°C minimum. We have successfully got away from the long standing 40-50mm thickness limits for PWHT. 

Figure 7 Avoidance of PWHT - milestones: a) MMA weld metal, as-welded toughness; b) HAZ as-welded toughness, as indicated by CEC carbon equivalent.

Of course a decision not to use PWHT also depends crucially on having the analytical methods such as in PD6493 mentioned earlier, where the combined effects of stress, any residual stress and toughness can be quantitatively assessed.

Where next as regards toughness challenges? Undoubtedly there has to be R&D into better materials for arctic temperatures, both steels and consumables and, as a second challenge, development of steels and consumables for welding up to 10 kJ/mm. Much has already been done, but reliably controlling the HAZ toughness is very difficult in tonnage production and much more work is needed on this aspect. The industry is somewhat divided on this issue, with some fabricators considering that heat inputs up to 5 kJ/mm are quite adequate and that current steels need no further development in this respect.

Challenges remaining:

• Good toughness at -30 to -50°C (arctic);
• Steels and consumables for welding at 10 kJ/mm.

Summary

What are the lessons? Perhaps there is already a clear message. The summary on 'time to market' or, indeed, 'time to money' shown in Fig.8 makes it very obvious that most R&D success stories in materials and welding and, indeed, in many other fields, have evolved over a 5-10 year period. Wide acceptance of new technologies in welding related fields takes 10 years, and whilst we may well ask ourselves 'why so long?' and 'can we not shorten the timescales?', the oil and gas industry operates a properly cautious approach to innovation and new thinking. Usually, the industry moves forward by consensus and materials and fabrication codes improve incrementally, each new step requiring careful validation on an international basis. Of course, validation can take years in view of the complicated and lengthy testing required for the engineering structures.

History shows us that the timescales for innovation in this industry are lengthy. There has to be a good deal of 'patient' money invested in the necessary R&D to each successful innovation. Many of us realise this but perhaps our message is not clear yet to Governments and the Boards of our major oil and gas companies.

Finally, there are three welding process innovations which were effectively born in 1993. I cannot predict accurately when these will have acceptance in the market place, but I am confident they will all make it.

The first is low pressure EB welding. Conventional in-vacuum welding usually requires expensive vacuum chambers. It has now been shown that with specially developed EB guns it is possible to produce narrow beams at around 1mbar, a pressure easily reached in a few minutes by simple mechanical pumps and chambers. I believe this development to be a major breakthrough which will radically alter the economics of EB for tubular manufacture.

The second innovation is a friction process known as 'stir' welding. This word essentially describes the mechanism. It involves a rotating mechanical tool which pierces a hole at the joint and is then drawn along it. The metal is mechanically stirred; there is no melting and a solid phase joint is formed. This looks to be an excellent process for aluminium sheet and plate up to 12mm thickness or more, and could have application for topside structures.

The final innovation is also a friction process - 'pillar' welding. A rotating consumable fed into a machined hole creates a solid phase weld as the shear zone progressively moves up the hole. This looks to be ideal for filling holes or repairing cracks and could easily work sub-sea.

All three innovations could reduce fabrication costs significantly.

As this is an era where cost reduction initiatives are well in place in all parts of industry, there will be many quick successes in this endeavour particularly in the field of reducing documentation, certification, and so on. But new cost effective engineering developments involving new materials, structural designs, new joining and inspection approaches, require longer timescales. Ours is not a short term game and realistically we need to think in periods of at least five years. Success stories will continue to emerge and we should continue to document them and learn the lessons'.