After several years in industrial research, Norman Bailey joined The Welding Institute in 1966. He is now a Principal Metallurgist in the Materials Department where he leads a Research Section concerned mainly with the welding of ferritic steels.
His researches into the problems of hydrogen cracking contributed to a method of predicting safe welding procedures now used world wide. He has also led research into solidification cracking during submerged-arc welding and examined the problems of welding very high-strength steels. More recently he has been concerned with understanding how the composition of ferritic steel weld metals can be controlled to achieve consistently the required microstructure, strength and toughness properties in both normal and underwater welding. He has also led projects examining the repair of pressure vessels without subsequent heat treatment, the strain ageing behaviour of the weld metals used for such repairs, and the development of electrodes for wet welding.
He has written numerous reports and papers including: 'The establishment of safe welding procedures for steels'. Weld J 1972 51 (4) 169s-177s, which was awarded the Lincoln Gold Medal of the American Welding Society in 1972 and, with Dr S B Jones, 'The solidification cracking of ferritic steel during submerged-arc welding'. Weld J 1978 57 (8) 217s-231s, which gained the American Welding Society's Spraragen Award in 1979.
Very often welding is the quickest, simplest and cheapest method of repair available. Norman Bailey looks at the factors influencing effective repair.
Welding can be used to repair components which are still in manufacture or which have been in service for a long time. Repair welding techniques may also be used to modify a structure in service. However, repair welding can go badly wrong. An injudicious repair can shorten the life of a component, or even damage it beyond repair. This article is based on part of the national seminar held in Glasgow last year. It considers the principles underlying weld repair, that is to say the questions that need to be considered before embarking on a new repair, and then discusses two more specialised aspects - repair of thick sections without post-weld heat treatment (pwht) and repair under water.
General principles
Although particularly addressed to the problems of repairing steels, much of this section is applicable to other metals.
Is welding the best method of repair?
If repair is called for because a component has a local irregularity or a shallow defect, grinding out any defects and blending to a smooth contour might well be acceptable. It will certainly be preferable if the steel has poor weldability or if fatigue loading is severe. It is often better to erode the so-called factor of safety slightly than to risk putting defects, stress concentrations and residual stresses into a brittle material.
In fact brittle materials - which can include some steels (particularly in thick sections) as well as cast irons - may not be able to withstand the residual stresses imposed by heavy weld repairs, particularly if defects are not all removed, leaving stress concentrations to initiate cracking.
Is the repair really like earlier repairs?
Repairs of one sort may have been routine for many years. It is important, however, to check that the next one is not subtly different. For example, the section thickness may be greater, the steel to be repaired may be different and less weldable, or the restraint higher. If there is any doubt, answer the remaining questions.
What is the composition and weldability of the base metal?
The original drawings will usually give some idea of the steel involved, although the specification limits may then have been less stringent, and the specification may not give enough compositional details to be helpful. If a sulphur-bearing free-machining steel is involved, it could give hot cracking problems during welding.
If there is any doubt about the composition, a chemical analysis should be carried out. Nowadays, a spectrographic analysis can be carried out* for under £50, usually within 24 hours, on any sample a few millimetres thick and large enough to include (after grinding if necessary) a flat surface upon which a circle 15mm diameter can be scribed. It is important to analyse for all elements which may affect weldability (Ni, Cr, Mo, Cu, V, Nb and B) as well as those usually specified (C, S, P, Si and Mn). This is particularly important for engineering steels, such as those in BS 970, in which up to 0.40% Ni, 0.30% Cr and 0.15% Mo are regarded as 'incidental', although these amounts will add 0.03, 0.06 and 0.03, respectively, to the carbon equivalent (CE**) and can seriously impair weldability.
* Contact Norman Jenkins or Sheila Stevens at TWI for details.
** CE = C + Mn/6 + (Cr + Mo + V)/5 + (Ni + Cu)/15
A few pounds spent on analysis could, on the one hand, prevent a valuable component being ruined by ill-prepared repairs or, on the other hand, save money by reducing or avoiding the need for preheat if the composition were leaner than expected.
Once the composition is known, a welding procedure can be devised, either directly [1,2] or via transformation diagrams for the steels [3-6]
What strength is required from the repair?
The higher the yield strength of the repair weld metal, the greater will be the residual stress level on completion of welding, the greater the risk of cracking, the greater the clamping needed to avoid distortion and the more difficult the welding procedure. In any case, the practical limit for the yield strength of conventional steel weld metals is about 1000 N/mm 2.
Can preheat be tolerated?
Not only does a high level of preheat make conditions more arduous for the welder; the parent steel can be damaged if it has been tempered at a low temperature. In other cases the steel being repaired may contain items which are damaged by excessive heating. Preheat levels can be reduced by using consumables of ultra-low hydrogen content or by non-ferritic weld metals. Of these, austenitic electrodes may need some preheat, [7] but the more expensive nickel alloys usually do not. However, the latter may be sensitive to high sulphur and phosphorus contents in the parent steel if diluted into the weld metal.
Can softening or hardening of the HAZ be tolerated?
Softening of the HAZ (heat affected zone) is likely in very high strength steels, particularly if they have been tempered at low temperatures. Such softening cannot be avoided, but its extent can be minimised. Hard HAZs are particularly vulnerable where service conditions can lead to stress corrosion cracking or involve high pressure hydrogen at high temperatures. Solutions containing H 2S (hydrogen sulphide) may demand hardnesses below 248HV (22HRC) [8] although fresh aerated sea water appears to tolerate up to about 450HV. [9] Excessively hard HAZs may, therefore, require PWHT to soften them but, provided cracking has been avoided, a hard HAZ is not necessarily harmful in other respects.
Is PWHT practicable?
Although it may be desirable, PWHT may not be possible for the same reasons that preheating is not possible. Some guidance on repair without PWHT is given later in this article. For large structures, local PWHT may be possible, but care should be taken to abide by the relevant codes, because it is all too easy to introduce new residual stresses by improperly executed PWHT.
Is PWHT necessary?
PWHT may be needed for one of several reasons, and the reason must be known before considering whether it can be avoided. For example, it may be possible to achieve sufficient toughness in an as-welded repair by use of the two-layer technique discussed later, but the degree of softening achieved may not be adequate to resist a service environment which gives a high risk of stress corrosion cracking.
Will the fatigue resistance of the repair be adequate?
If the repair is in an area which is highly stressed by fatigue, and particularly if the attempted repair is of a fatigue crack, inferior fatigue life can be expected unless the weld surface is ground smooth and no surface defects are left. Fillet welds, in which the root cannot be ground smooth, are not tolerable in areas of high fatigue stress.
Will the repair resist its environment?
Besides corrosion, it is important to consider the possibility of stress corrosion, corrosion fatigue, thermal fatigue and oxidation in service.
Corrosion and oxidation resistance usually require that the composition of the filler metal is at least as noble or oxidation resistant as the parent metal. For corrosion fatigue resistance, the repair weld profile may need to be smoothed.
To resist stress corrosion, PWHT may be necessary to restore the correct microstructure, reduce hardness and reduce the residual stress left by the repair.
The risk of thermal fatigue is increased if expansion coefficients of filler and parent metal do not match. A nickel alloy filler may therefore be preferred to the more usual stainless steel type when repairing 'difficult' steels.
Can the repair be inspected and tested?
For onerous service, radiography and/or ultrasonic examination are often desirable, but problems are likely if a stainless steel or nickel alloy filler is used; moreover, such repairs cannot be assessed by magnetic particle inspection. In such cases, it is particularly important to carry out the procedural tests for repairs very critically, to ensure that there are no risks of cracking and no likelihood of serious welder-induced defects.
Indeed, for all repair welds, it is vital to ensure that the welders are properly motivated and carefully supervised.
As-welded repairs
Repair without PWHT is, of course, normal where the original weld was not heat treated, but some alloy steels and many thick sectioned components require PWHT to maintain a reasonable level of toughness, corrosion resistance and so on. However, PWHT of components in service is not always easy or even possible, and local PWHT may give rise to more problems than it solves except in simple structures.
In the case of C-Mn and similar steels, it was realised that the major service problem from an as-welded repair would be the toughness of the coarse grained HAZ, recent developments in consumables having led to weld metals with excellent as-welded toughness at - and some way below - ambient temperature. One way to combat the HAZ problem is to eliminate the coarse grained HAZ, as a refined HAZ microstructure is inevitably tougher. The first practical solution was the so-called half bead technique, [10] where the repair cavity was buttered with a layer of weld metal deposited with fairly small electrodes. Roughly half of this layer was ground off and a second, similar, layer deposited. The coarse grained HAZ of the second layer is intended to lie entirely in the weld metal of the first layer. The refined HAZ of the second layer should then refine the original coarse HAZ in the parent steel of the first layer.
Although adopted by the ASME Boiler Code, [10] the technique is clumsy to operate, as it requires fairly close control of the grinding operation; it is likely to be largely replaced by the more sophisticated two-layer technique. [11-13] This is an adaptation of the method used to weld ½Cr-½ MO-¼ V steel with 2 ¼Cr-1Mo weld metal, [11] where it is also necessary to refine the HAZ to avoid reheat (stress relief) cracking during PWHT.
The technique, illustrated in Figure 1, consists of depositing the first layer of weld metal with small electrodes (2.5 or 3.25mm diameter), carefully controlling arc energy, bead overlap and angle of attack ( Figure 2). A second layer is then deposited, with the same parameters closely controlled, but using larger (4mm diameter) electrodes ( Figure 3). The technique has been developed successfully for C-Mn ( Figure 4) and Mn-Ni-Mo steels and is understood to have been used in practice, including at least one repair on an offshore platform.
For high-temperature steels of the Cr-Mo type, there are complications: the as-welded toughness of available weld metals may be inadequate if a hydrotest is required, and hard zones in HAZ and weld metal may be unsatisfactory for hydrogen service in petrochemical plant. Nevertheless, as-welded repairs on such steels have been used in power station plant [14-15] where hydrotests are not required. The plant can be 'nursed' for a few months, as the service temperatures are high enough to allow tempering of the repair to reduce both hardness and residual stress with a concomitant improvement in the toughness of the repair. Currently a Group Sponsored Project at the Edison Welding Institute and TWI* is studying the as-welded repair of Cr-Mo steels for both petrochemical and power plant repairs. This is investigating TlG and MIG (using solid and flux-cored wires) as well as MMA for repairs, and particular attention is being paid to the possibility of achieving tempering during welding as well as in service. A major problem is that although the Cr-Mo steels in petrochemical plant are frequently used to resist hydrogen in service at high pressures and temperatures, the temperatures are below the creep range where tempering and stress relief are known to occur within a reasonable time scale.
* Project J6085 'Development and assessment of procedures for repair welding of Cr-Mo steels without PWHT'. For details contact the author at TWI or W Bruce at EWI.
Repairs under water
Of the two methods of welding below the surface, wet welding (striking the arc in water) is the easier logistically but the more difficult technically. Not only is the welding arc less stable at depth, but the very high weld hydrogen contents and fast cooling, coupled with the inability to pre- or post-heat, make hydrogen cracking and porosity far worse problems than with dry welding, even at high pressure (hyperbaric welding).
Hyperbaric welding
The normal types of repair welding, MMA, TIG and MIG, can be carried out [16] using pre- and post-heating and even, if necessary, PWHT. However, as the depth increases the arc becomes increasingly constricted; the atmosphere within the habitat is very humid, so that care is needed to keep consumables dry. Welding procedures tend to be conservative and post-heating is often resorted to, as any delay for repair of cracking would be unacceptably expensive.
Because of the high pressure, element transfer is different from surface welding. Carbon and oxygen contents are higher but Mn and Si lower. [17] In addition, the habitat atmosphere (He-O 2 mixtures below about 20m depth) and the narcotic effect of high pressure Ar make He mixtures preferred for gas-shielded welding. The high weld oxygen contents make high weld toughness difficult to achieve, although it is not normally required - seawater temperature rarely falls below 4C.
Wet welding
Wet welding is often thought of as the last resort to patch something up until it can be repaired 'properly'. However, it is used regularly in the North Sea for secondary underwater repairs, such as maintaining electrical continuity and fixing mild steel anode holders to mild steel brackets at depths down to at least 85m. Wet welding at twice that depth should be possible, and the technique has been demonstrated at 100m by welding automatically in water in a hyperbaric chamber. [18]
Welding by MMA is by far the commonest method ( Figure 5), as it obviates the need for wire feeders to operate in water, and shielding gases are not needed. Repairs are often of the patch type, involving fillet welds, so that the special MIG technique, where the shielding gas provides a curtain to keep the water out, [19] is not easy to apply.
