International Journal of Pressure Vessels and Piping 87 (2010) 127e133

Contents lists available at ScienceDirect

International Journal of Pressure Vessels and Piping

journal homepage: www.elsevier.com/locate/ijpvp

Review

Weld repair practices without post weld heat treatment for ferritic alloys and their consequences on residual stresses: A review
A. Aloraier a, *, A. Al-Mazrouee a, J.W.H. Price b, T. Shehata c
a

Manufacturing Engineering Department, College of Technological Studies, PAAET, Kuwait Mechanical Engineering Department, Monash University, Victoria 3800, Australia c College of Engineering, Taibah University, Taibah, Saudi Arabia
b

a r t i c l e i n f o
Article history: Received 22 October 2008 Received in revised form 31 January 2010 Accepted 15 February 2010 Keywords: Temper bead technique Residual stress Weld repair Welding parameters

a b s t r a c t
The use of the half-bead, temper bead welding (TBW), and cold repair techniques is proving to reduce the cost of repairs and extend the life of aged components in power plants, petrochemical and hydrocarbon processing industries. It has been a signicant area of interest for more than twenty years. A critical factor in this context is residual stress. The presence of residual stresses can lead to cracking which ultimately results in structural failure. This paper reviews the half-bead, TBW, and cold repair technique practices and their consequences on residual stresses within the nuclear, power, renery and petrochemical industries and some of the contributions made by our group of researchers in this area. This paper reviews recent work by the Monash University group. We report our work on TBW residual stresses when measured using neutron diffraction which shows very little reduction in residual stresses over normally completed welds. The use of automatic FCAW has been explored in our group and is reported. 2010 Elsevier Ltd. All rights reserved.

1. Introduction Components operating at high temperature or harsh working conditions are subjected to different failure regimes which require special consideration. In scheduled shutdowns, critical components are inspected and if necessary either replaced or repaired. Generally, the cost effective choice is to extend the life of aged components by repair rather than replacement. Repair can be the only alternative particularly if the required material for repair is not available within outage time constraints. Repair welding maybe carried out using a range of welding techniques depending on the nature of the repair. Commonly, most weld repairs require post weld heat treatment (PWHT) to restore the deteriorated metallurgical and mechanical properties. However, critical issues in the course of PWHT such as time, component geometry and unsupported loading during the time the component is hot, mean that it is not always possible to carry out the full procedure. Duration of hold time during the PWHT cycle can be very long especially for thick walled components which result in the extension of the unit downtime. Complex geometry may cause difculties in reaching the exact region of repair therefore a larger area has to be heated instead. This can lead to distortion particularly when mechanical loads are attached to the structure since

deformation may occur at the elevated temperatures of the heat cycle. Due to the difculties and limitations encountered with PWHT, alternative techniques have been developed. The role of this paper is to review the development of alternative techniques to PWHT. Welding processes, parameters and techniques in the context of TBW and the key aspects of successful weld repair are also considered adding in recent work done by our group at Monash University, Melbourne. 2. A historical review Researchers have shown great interest in studying and evaluating alternative techniques for PWHT. An important workshop is reported in the Welding Research Council Bulletin 412, 1996 [1]. This bulletin established comprehensive guidelines for repair welding pressure vessel and piping systems and associated equipment mainly for the manual metal arc welding process. For the alternative techniques to PWHT half-bead, TBW, and cold repair studies and limitations are given. A more recent survey is by Issler et al., 2004 [2]. 2.1. Half-bead technique The half-bead technique has been used widely by different industries for repairing different alloys such as 2.25Cre1Mo steel. In this technique a cavity is made either by gouging or grinding in

* Corresponding author. E-mail address: as.aloraier@paaet.edu.kw (A. Aloraier). 0308-0161/$ e see front matter 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.ijpvp.2010.02.001

128

A. Aloraier et al. / International Journal of Pressure Vessels and Piping 87 (2010) 127e133

the region required to be repaired. A pre-heat temperature of about 177  C is followed. A stringer bead buttering layer is then deposited using a maximum 3 mm diameter electrodes [3]. The use of small electrodes with low heat input is aimed to reduce the hardness of the HAZ. An overlap of 50% between the adjacent weld beads is maintained to partially temper the previously deposited beads. Half of the rst layer is then removed by grinding from whence the name half bead derives. A new stringer bead layer is placed over the previously machined layer using a larger electrode diameter such as 4.0 mm with higher heat input without weaving to rene the coarse grains of the HAZ [4,5]. Next is to overll the cavity using 4 mm electrodes in order to temper each previous layer. The last layer crown is nally removed by grinding. Inter-pass temperature should always be controlled for a maximum temperature of 230  C until the completion of the weld. A low temperature heat treatment of about 290  C is followed to assist hydrogen diffusion [4,5]. The steps of the half-bead tempering technique are schematically illustrated in Fig. 1. Different studies have been carried out using the half-bead technique. Several studies were conducted to allow the use of half bead welding in the US nuclear industry in the 1970s and 1980s. These studies include: Whitman [6] who reported one of the early practices of using the half-bead tempering technique for intermediate test vessels. Holz [7] discussed the standards and procedures used with half-bead repairs. Wismer and Holz [8] performed successful half-bead welding on cavities placed in three thick-walled pressure vessels. All three reported repairs were in accordance to Section XI of the American Society of Mechanical Engineers (ASME) Boiler and Pressure Vessel Code guidelines. Bryan et al. [9] pressurized to failure a vessel of SA533, grade B, class 1 steel, contained a fatigue-sharpened notch adjacent to a halfbead weld repair that had not been stress relieved. Feldstein [10] discussed a supporting EPRI program. Alberry [11] developed a predictive methodology for the calculation of heat-affected zone hardness distributions in thick-section multi-pass welds in SA 508 Class 2. This methodology was incorporated into a computer model that has been used to develop an alternative repair technique to the ASME XI half-bead technique. Lundin [12] considered programs of PVRC aiming to optimize welding for the fabrication and repair of CreMo and low-alloy steels. These programs led to additional work on improving the utility of low carbon 1.25Cre0.5Mo weld metal by introduction of elements designed to improve elevated temperature properties without impairing weldability potential. Lant et al. [4] appraised the practical weld repair procedures for low alloy ferritic steels and some of the precautions needed.

Although the half-bead technique has proven to be successful as an alternative for PWHT, some problems associated with it need to be taken into consideration. Firstly, precision grinding is needed for the rst layer. The grinding stage is very critical for the repair: the main drawback of the technique being over or under grinding of the rst layer. Over-grinding will result in shallow remaining of the rst layer and thus deposition of the subsequent layer will result in a HAZ with a deeper penetration than intended causing new hard microstructure. On the other hand, under-grinding will result in insufcient penetration and inadequate tempering. The time needed for precision grinding in the half bead technique results in a noticeable delay. 2.2. Temper bead The key difference of TBW from the half bead technique is that the interlayer grinding is excluded in the TBW technique. Also, the heat input of the subsequent layer in the TBW technique is increased in order to able to rene the coarse grain or temper the martensitic/ bainitic microstructure of the HAZ. Fig. 2 shows a schematic illustration of the TBW technique. There is a long history of investigation of TBW. At the same time as the early US work on half bead repairs, the Central Electricity Generating Board (CEGB) in the United Kingdom, established a twolayer renement technique specically to improve the sensitivity of the coarse grain heat affected zone (CGHAZ) in CrMoV steels to stress relief cracking [5,13]. A joint is prepared by making a suitable cavity and a single layer of stringer beads is deposited at low heat input using either 2.5 mm or 3.25 diameter electrodes, which results a narrow HAZ. A minimum 50% overlap with controlled parameter was used to rene the adjacent layer for the stringer bead up to 80% [4,13]. Then, with a higher heat input using 4.0 mm diameter electrode the second layer is deposited then ll-up is completed with the same electrode. Tempering of the HAZ is achieved by controlling the heat input ratios between the rst and second layers using different sizes of electrodes. Pre-heat is maintained at approximately 250  C and PWHT is performed at 700  C [5]. An early study by Feldstein [10] also reported a controlled heat input, multi-pass welding procedure without the necessity for grinding the rst layer. Then, Felix et al. [14] described the development of welding procedures using gas tungsten arc welding (GTAW). Zeemann and Ferreira [15] presented a practical example of an incorrect use of TBW for a welded AISI 4130 oil extraction pipeline, where the last weld bead was not positioned on weld metal but on the HAZ creating a new hard HAZ. Ikeuchi et al. [16] examined the HAZ toughness and microstructure of SQV-2A steel welded by the temper bead technique. In their study, an intercritically reheated coarse-grained HAZ (ICCGHAZ) was simulated. Serious embrittlement was found in the ICCGHAZ. The TBW technique was developed by Ontario Hydro in Canada [17]. In their work, a low heat input of stringer beads are deposited

Fig. 1. Schematic illustration of the half bead tempering technique (a) gouging and grinding, (b) applying the buttering layer, (c) machining half of the buttering layer, (d) applying the tempering layer, (e) lling the rest of the cavity and (f) ground ush the capping.

Fig. 2. Schematic illustration of the TBW technique.

A. Aloraier et al. / International Journal of Pressure Vessels and Piping 87 (2010) 127e133

129

with a minimum 50% bead overlap. Then, with the same overlap percentage, a second layer of stringer beads is deposited. The heat input of the second layer is approximately twice that of the rst layer. Extensive grain renement and tempering of the CGHAZ are aimed out of this procedure. In order to induce the tempering of the HAZ, larger diameter electrode is used with high heat input or weaved beads rather than stringer beads. A pre-heat of 177  C is maintained till the completion of welding. Lastly, to diffuse hydrogen out of the weldment a temperature of 230  C is maintained for 2 h. EPRI studies found that TBW offered an alternative solution to PWHT from a creep, tensile, and toughness standpoint [18]. Gandy et al. [19] discussed industry case studies using conventional PWHT and the TBW technique. Viswanathan et al [20] discussed the results of evaluation performed on service-aged piping using both PWHT and TBW. TBW improved the toughness of the service-aged weldment, while the PWHT lowered the HAZ toughness. More recent work by Viswanathan and Gandy [21] documented the results of an industry survey of weld repair practices and described the results of experimental evaluations performed on service-aged 2.25Cre1Mo steel piping. The overall results of this program provide substantial evidence that service-aged piping systems can be successfully weld repaired with TBW and that life extension by several decades is achievable under the right design and repair conditions. However the review also found that gas tungsten arc weld repairs with PWHT resulted in the best combination of tensile strength, uniform microhardness distribution across the weld, Charpy toughness, and creep rupture life. Another technique which involves a six layer renement where a higher degree of heat input control is used to allow both rening and tempering of the structure employing GTAW and SMAW processes [22] was developed by EPRI in the 1990s. TBW has been expanding to be used in other materials such as chromium-molybdenum steels. TBW has been demonstrated for CeMn steels so that the value of peak hardness was less than 350 HV which is critical for hydrogen cracking [5,13]. The work was also reviewed by Lundin [12] and input from the National Board of Pressure Vessel Inspectors was discussed. Moore [23] investigated weld repair of C-Mo coke drums without PWHT using the SMAW process. The authors extended the work to include as-welded repairs with automatic gas metal-arc welding (GMAW). Automatic GMAW increased productivity and allowed repairs to larger areas of fatigue damage during short-duration turnarounds. Lau et al. [17] summarized the work conducted by Ontario Hydro in non-PWHT weld repair technology with special emphasis on recent developments for CreMo materials. Nawrocki et al. [24] evaluated and compared the stress relief cracking susceptibility of a temper bead weld in HCM2S (a new ferritic steel) and standard 2.25Cre1Mo steel to single pass weld results using Gleeble machine Peddle and Pickles [25] developed TBW and PWHT of 2.25Cre1Mo steel. Both procedures aimed to rene and temper the heat-affected zone (HAZ). The PWHT resulted in the greatest decrease in hardness and also reduced hardness variability throughout the HAZ. Bhaduri et al. [26] investigated repair welding methods for CreMo steels, using 2.25Cre1Mo and 9Cre1Mo materials. Our Monash University research group [27] has, produced an analytical solution which was used to support the use of TBW in repair of damaged structures. A relationship was presented between the depth of penetration as an output and both the traverse speed and the arc length as inputs. We have also shown that while the metallurgical characteristics of TBW may be as good as PWHT welds, the residual stresses are virtually the same for TBW welds as for welds that are NOT heat treated. This is discussed below. The TBW technique has been adopted by different codes and standards such as ASME Boiler and pressure vessel code, Section XI.

However, with the enormous number of studies, some issues need to be taken into account when applying these techniques. A special attention should be given to the toe area in both the half-bead and the TBW techniques. The toe region of a temper bead weldment can have higher hardness because this region does not receive as many thermal cycles as the other regions of the heat affected zones. Therefore, one of the major limitations that needs to be taken into account when performing TBW is toe cracking. Toe cracking is caused by the high-strength weld metal having a higher yield point and tensile strength than the parent materials. When the weld area shrinks on cooling from the welding temperature cracking occurs in the heat-affected zone (HAZ) of the steel because the yield and strength levels of the HAZ are lower than those of the weld metal. The toe region is highly susceptible to fatigue cracking due to the presence of weld defects, stress concentration due to toe geometry and adverse metallurgical conditions such as tensile residual stresses and coarse HAZ microstructure. Moreover, the toe region experiences high dilution and thermal stresses depending upon the type of welding process and procedures adopted during the repair [28]. One method of avoiding toe cracking is to apply a nal temper bead weld to a level above the surface being repaired without contacting the base material but close enough to the edge of the underlying weld bead to ensure tempering of the base material HAZ [3]. Ibarra [29] also proposed this solution by ensuring that the toe of the weld to the base metal is not the last pass and the nal temper bead pass should be ground off. An alternative solution to the ones presented by the previous studies [3,29] is the use of runoff plates which is suggested by [30]. Run-off plates may be tack welded in place at the inside surface of the repair cavity as shown in Fig. 3. Another solution to this problem is to use a low strength weld metal and increase the llet size to meet the weld joint strength requirements. 2.3. Cold repair Cold repair is the repair process which eliminates the need of preheat or PWHT hence time saving in a shutdown repair can be signicant. Cold repair was used early in Russia [22,31e33] then it was developed by the CEGB in the UK to be used with CrMoV steels by employing the SMAW process through the use of nickel-based electrodes which does not require a specialized welding procedure due to its resistance to hydrogen [5]. Typically, in this technique a stringer bead is deposited with 50% overlap on the preceding weld bead using 2.5 mm SMAW electrode. The second layer is deposited with 4 mm electrodes using stringer beads and 50% overlap. Then, the excavation is lled with 4 mm electrodes normally by maximum weave width of 3 times the core wire diameter [34]. This procedure is designed to improve the as-welded HAZ toughness satisfactorily, so that reheat cracking does not occur during start-up, when residual stresses at maximum. The small

Tack weld

Run-off plate

Base Metal

Weld bead
Fig. 3. Schematic illustration of run-off plates.

130

A. Aloraier et al. / International Journal of Pressure Vessels and Piping 87 (2010) 127e133

electrode size and low heat input when using manual metal arc welding produce ne grained heat affected zone structures in ferritic base materials, eliminating the need for controlled deposition techniques or subsequent tempering or renement [35]. Recently, Mitchell [36] reported the use of nickel-based wire using FCAW for cold repair. It was used for CrMoV and 2.25Cre1M components and predicted that the repair applications for cored wires in the power industry will grow signicantly and the use of cored wires will provide clear benets in terms of cost and time [34]. The anisotropic nature of the structure of the nickel-based weld metal complicates the detection of ultrasonic devices of subsurface defects. Also, the difference in the thermal expansion between weld and base metals may lead to cracking and failures. Therefore, this technique is recommended only for repairs of hardenable ferritic components in tightly scheduled plant shutdowns [35]. TBW technique is preferred than cold repairs using nickel-based wires when welding dissimilar materials welds because the later are prone to fatigue cracking during plant cycling [31,32]. 3. TBW and residual stresses In recent research from our group, the neutron diffraction technique was used to investigate the residual stress distributions in carbon steel components with weld repairs [37,38]. Two full penetration weld repairs were made using (a) the stringer bead and (b) the TBW techniques in 25 mm thick plate as shown on Fig. 4.

a
25 mm 600 3 mm 3 mm

The welds were not post weld heat treated. The focus of the measurements is on the values of the sub-surface and throughthickness residual stress near the middle and the toe of the weld. Comparison of the experimental results with the estimates in BS 7910 [39] and R6 [40] (Figs. 5 and 6) show that these codes underestimate the longitudinal surface stress within the weld and HAZ. The transverse surface stresses are by contrast overestimated by both codes. High longitudinal residual stress values have been found in all our work in the longitudinal direction. These stresses normally exceed the uniaxial yield value of the parent and weld metal because of the tensile tri-axiality of the residual stress in the weldment. The highest residual stresses in the transverse direction were found within the weld at approximately 40% for Sample I (Fig. 5b) and 60% for Sample II (Fig. 6b). At the surface, a stress distribution value below the yield was achieved in the longitudinal direction at the toe and the HAZ (Figs. 5a and 6a). Fig. 6 shows comparisons of the through-thickness residual stress distributions at the weld centre line, x 0 mm for TBW and stringer weld. The key issue from the perspective of this paper is that from the point of view of residual stress the temper bead weld appears not to better than the stringer bead weld. Thus we reach the conclusion that while TBW may improve the metallurgical structure compared to stringer bead, the residual stresses are not improved. Note the error bars on the measurements themselves are shown in Fig. 5. The error that may be involved in relation to the exact nature of the welds being tested in this paper is a separate issue. However powerful commonalities are emerging from many measurements being accumulated by many researchers. Examples of these principles are:  The residual stresses in the last bead to solidify are close to the yield condition (unless relieved by distortion of the sample). This means that this location demonstrates high values of tensile tri-axial residual stress and the measured uniaxial longitudinal stress tends to exceed the uniaxial yield stress.  Transverse stresses are lower but the toe of the weld near the HAZ exhibits small scale stress variations due to effects of recrystalisation and rotation of the stress tensor.  Residual stresses drop dramatically on the application of post weld heat treatment. The work in the Monash group has also sought to consider the errors in neutron diffraction and its comparison with theoretical procedures. These issues are considered in detail in Price et al. [41]. 4. The use of automatic FCAW for TBW

b
z x

12

13 7 6 4

14 8 5 3 2 1

11 9

10

35 34 32 33 36 7 8 9 31 18 17 16 6 27 28 29 30 15 5 25 26 14 24 4 23 13 22 Filling layers 3 12 21 Tempering layer 2 20 11 38 37

for TBW

10 19

20

30

Fig. 4. Schematics illustration of (a) preparation for full penetration repairs using (b) Sample I (SW: Stringer Bead Weld) and (c) Sample II (Temper Bead Welded) techniques. The red-doted line represents the line scans for neutron diffraction (ND) measurements. (For interpretation of the references to colour in this gure legend, the reader is referred to the web version of this article).

Welding process can be considered as an important parameter when carrying an alternative technique for PWHT. It has consequences on the speed, quality, preciseness and the cost of repairs and therefore should be considered in this context. SMAW is one of the oldest and still most commonly employed process, which was rst used between 1905 and 1910 [42]. Most of the reported practices and researches of half-bead and TBW techniques used SMAW process. However, there are some disadvantages such as the low deposition rate. Furthermore, since the SMAW is a manual process and a great variation in weldment properties may occur. The gas tungsten arc welding (GTAW) process was used also in carrying some of the repairs that employs half-bead or temper bead techniques. Viswanathan and Gandy [18] reported that GTAW weld repairs to the base metal were generally to outperform SMAW welds. This might be as a result of the lower and better control heat input in GTAW process.

A. Aloraier et al. / International Journal of Pressure Vessels and Piping 87 (2010) 127e133

131

a
1.5

Sample II (TBW )

BS 7910 & R6 Level 1

R6 Level 2

Longitudinal NRS

1.0 0.5 0.0 -0.5 -1.0 -40 -30 -20 -10 0 10 20 30 40 50 60 70 80 90 100

b
Transverse NRS

1.5 1.0 0.5 0.0 -0.5 -40 -30 -20 -10 0 10 20 30 40 50 60 70 80 90 100

Distance from the weld centre line (mm)

Fig. 5. Temper Bead Welding. (a) Longitudinal and (b) transverse Residual stress. Comparison of normalised residual stress (NRS) 1.6 mm below the surface.

Recently other processes such as gas metal arc welding (GMAW), and ux cored arc welding (FCAW) have been used and are showing promising results. The use of both processes has become well established recently in fabrication and repairs due to cost and time effectiveness and reliability [34,36,43,44]. One of the main advantages of GMAW and FCAW processes is the high deposition rate

a
Longitudinal NRS 1.5 1.0 0.5 0.0 -0.5 0

Sample I Sample II

SW x = 0 TBW x = 0 BS 7910 & R6 Level 1

10

15

20

25

b
Transverse NRS 1.0

BS 7910 & R6 Level 2

0.5 R6 0.0

-0.5 0 5 10 15 20 25 Depth (mm)

Fig. 6. TBW compared to stringer welds (no heat treatment) (a) longitudinal and (b) transverse directions. Comparison of through-thickness normalised residual stress (NRS) at the centre line of the weld (x 0) for stringer welded (SW) and temper bead welded (TBW).

which result in a higher welding speed hence saving time and lowering the cost of repair. Another advantage of the GMAW and FCAW is the possibility of automation of these processes that reduce the need for skilled welder and resulting in better control of bead parameters which is vital in carrying the TBW technique [45]. Among the investigations carried out by our group was manipulating the traveling speed, studying overlap between welding beads and changing the shielding gas. With respect to changing traveling speeds different heat inputs were studied, i.e. slow traveling speed results in high heat input and vice versa. Aloraier et al. [46] have varied the traveling speed of a FCAW process in order to obtain different heat inputs between the layers. The fast traveling speed was used to attain small HAZ and the slow traveling speed was then used to temper and rene the HAZ of the previous layer. In automated welding processes, the controlling of the traveling speed is easy; however in manual welding processes the control of the traveling speed becomes very critical since human error may be introduced. In any tempering technique, the overlap between the adjacent beads in each layer must be controlled and is considered a vital parameter for TBW. Most researchers have used 50% bead overlap between the beads because of its practicality. This was consistent that the optimum percentage of bead overlap between the beads in the TBW technique should be within a range between 50% and 70% as reported by Aloraier and et al. [47]. Aloraier [48] have measured the residual stresses using the RS-200 hole drilling method to study the effect of the percentage of bead overlap on residual stresses. It was shown that the 50% bead overlap gave the best results in terms of compressive residual stresses. Another parameter that needs to be considered is the shielding gas [49]. A study has been carried out by our group [50] on the effect of welding process parameters on the welding geometry on steel components. The primary function of the shielding gas is to protect the molten metal from atmospheric nitrogen and oxygen as the weld pool is being formed [51]. In addition, the shielding gas plays an important role in effecting the electrical characteristics of the arc. A sample of the results obtained is as follows.

132

A. Aloraier et al. / International Journal of Pressure Vessels and Piping 87 (2010) 127e133

 The height of the weld bead reaches its maximum value at 25 volts for both gases tested (CO2 and argon).  In the range of 23e25 volts, the size of the HAZ increased when CO2 was used, but it decreased when argon was used. 5. Finite element modeling Computer simulation using nite element (FE) modeling will enhance the understanding of TBW. Researchers have used different FE software to model different welding processes. Many authors have utilized the commercial nite elements code ABAQUS, enhanced with user subroutines, to model weld simulation with great success [52e59]. The nite element code ANSYS was used by several authors [60e64] while others [65e71], have utilized SYSWELD to perform these weld simulations. Although there are numbers of different commercial software shells, researchers did not model the microstructure resulting from the TBW technique. This might be due to the lack of software which are dedicated for welding and particularly which deals with microstructure modeling. However, a package similar to SYSWELD can be a source to obtain this kind of modeling. Recent focus on FE modeling was obvious by the special issue of this Journal which was devoted to round robin activities covering residual stress measurement and residual stress prediction. These activities were aimed to develop experimental and numerical techniques and standards for the reliable characterization of residual stresses in structural welds. However, important areas such as the microstructural modeling still in need to be explored. In recently reported work from our group [41] the use of a neutron beam as a non-destructive method of measuring residual stress due to welding has been explored and compared to a model of the same situation, calculated using the software package SYSWELD. The work demonstrates that the results are in general conformity. In detail, however, there are differences. Some of these differences are presumably due to familiar difculties with theoretical models of complex non-linear material properties and welding parameters. These differences have been minimized by experimenting with varying some of the material properties a little until the results were in best agreement. However, when the cases are compared in detail, new issues emerge. These other issues relate to fundamental difculties in comparing FE results with the neutron beam (or any other) measurement method. Three issues, which were identied in the paper, were as follows.  The difference in the shape and size of the gauge or estimation volumes between the experimental and theoretical techniques.  The difference in the means of averaging or integrating results within these gauge volumes to produce a single output.  Experimental diffraction technique normally measure only direct strain (no shear strains can be measured). When comparing with FEA, these direct strains have to be converted into elements of a six element tensor. To do this fully requires three to six measurements at each point. This is often physically difcult, time consuming and expensive which means that approximate assumptions are often made about the principal directions of the tensor. 6. Conclusions A review of studies on alternative techniques to repair welding using post weld heat treated (PWHT) welding was carried out to assess the advantages, limitation and considerations. The summary of the review can be outlined in the following conclusions.

 The main alternative techniques to PWHT for large welds are termed half-bead, temper bead welding (TBW) and cold repairs. They are said in the literature to be effective in carrying out weld repairs but work continues to be developed which will rene that choice.  Repairs without PWHT are not recommended to be used in highly stressed areas or in components which are susceptible to stress corrosion cracking.  The temper bead welding technique is recommended in when compared to the other two techniques not requiring heat treatment.  New work by our group at Monash University [37] shows that residual stresses are not signicantly affected by the implementation of TBW when compared to stringer bead welding. The advantage in TBW is metallurgical resulting in higher ductility and not in reducing residual stress.  We report our recent work on the use of automatic Gas Metal Arc Welding (GMAW) and ux cored arc welding (FCAW) [46e50].  The understanding of TBW is being improved by computer simulation using nite element modeling. Our recent work has focused on the theoretical limitations of both neutron measurement and nite element modeling in describing residual stress [41].

References
[1] Challenges and solutions in repair welding for power and processing plants. Weld Res Council Bull June 1996;412. Proceedings of a workshop. [2] Issler S, Klenk A, Shibli AA, Williams JA. Weld repair of ferritic welded materials for high temperature application. Int Mater Rev 2004;49(5): 299e324. [3] Doty WD. History and need behind the new NBIC rules on weld repair without PWHT. Weld Res Council Bull 1996;412:3e8. [4] Lant T, Robinson DL, Spafford B, Storesund J. Review of weld repair procedures for low alloy steels designed to minimise the risk of future cracking. Int J Press Vessels Pip 2001;78:813e8. [5] Peddle BE. Carbide development in the heat affected zone of tempered and post weld heat treated 2.25Cre1Mo steel weldments. Kingston, Ontario: Department of Materials and Metallurgical Engineering, Queen's University; 1996. p. 127. [6] Whitman GD. Heavy-section steel technology program. Quarterly progress report for JanuaryeMarch 1977, United States; 1977. p. 118. [7] Holz PPS. Half-bead weld repairs for in-service applications. In: ASME 78-PVP10, Joint ASME/CSME pressure vessels and piping conference, United States; 1978. [8] Wismer SW, Holz PP. Half-bead (temper) repair welding for heavy-section steel technology program vessels. In: Portions of document are illegible, United States; 1978, p. 129. [9] Bryan RH, Iskander SK, Holz PP, Merkle JG, Whitman GD. Test of thick vessel with a aw in residual stress eld. In: US national congress on pressure vessels and piping, United States; 1979. [10] Feldstein JG. Repair welding of heavy-section steel components. In: LWRs. Summary report. Final report, vol. 1, 1984, United States. p. 89. [11] Alberry PJ. Computer model for multipass repair welds in SA508 class 2 alloy. Weld J 1989;68(10):410e7. [12] Lundin CD. Overview of results from PVRC programs on half-bead/temper-bead/ controlled deposition techniques for improvement of fabrication and service performance of CreMo steels. Welding Res Council Bull 1996;412:16e26. [13] Jones RL. Development of two-layer deposition techniques for the manual metal arc welding repair welding of thick CeMn steel plate without postweld heat treatment. Welding Insitute; 1987. p. 31. [14] Felix RD, Bruce WA, Threadgill PL. Development of procedures for hot tap welding onto sour service pipelines. In: Proceedings of the international arctic technology conference; 1991. [15] Zeemann A, Ferreira DA. Temper bead technique a practical situation. Weld World 1993;31(6):412e3. [16] Ikeuchi K, Liao J, Tanabe H, Matsuda F. Effect of temper-bead thermal cycle on toughness of weld ICCGHAZ of low alloy steel SQV-2A. ISIJ Int 1995;35 (10):1203e12. [17] Lau TW, Lau ML, Poon GC. Development of controlled deposition repair welding procedures at Ontario hydro. Weld Res Council Bull 1996;412:35e42. [18] Viswanathan R, Gandy D. Weld repair of aged CreMo steel piping: a review of literature. J Press Vessel Technol 2000;122(1):76e85. [19] Gandy DW, Viswanathan R, Findlan SJ. Performance of weld repairs on service-aged 2.25Cre1Mo girth weldments utilizing conventional postweld

A. Aloraier et al. / International Journal of Pressure Vessels and Piping 87 (2010) 127e133 heat treatment and temper-bead repair techniques. Weld Res Council Bull 1996;412:43e59. Viswanathan R, Gandy D, Findlan S. Performance of repair welds on serviceaged 2-1/4Cre1Mo girth weldments. J Press Vessel Technol, Trans ASME 1997;119(4):414e22. Viswanathan R, Gandy DW. Performance of repair welds on aged CreMo piping girth welds. J Mater Eng Perform 1999;8(5):579e90. Klenk A, Issler S, Shibli IA, Williams JA. Some characteristics of weld repair for creep applicaions. OMMI, Innogy, Operat Eng 2003;2(1):1e32. Moore DE. Weld repair of carbon-moly coke drums without postweld heat treatment. Weld Res Council Bull 1996;412:70e6. Nawrocki JG, DuPont JN, Marder AR. The weldability of a modied 2.25 CreMo Steel (HCM2S). In: Joining of Advanced and Specialty Materials: Proceedings from Materials Solutions Conference '98 on Joining of Advanced and Specialty Materials, Oct 12e15 1998. Rosemont, IL, United States: ASM International; 1998. Peddle BE, Pickles CA. Carbide and hardness development in the heat-affected zone of tempered and postweld heat-treated 2.25Cre1Mo steel weldments. J Mater Eng Perform 2000;9(5):447e88. Bhaduri AK, Rai SK, Gill TPS, Sujith S, Jayakumar T. Evaluation of repair welding procedures for 2.25Cre1Mo and 9Cre1Mo steel welds. Sci Technol Weld Join 2001;6(2):89e93. Ibrahim RN, Shehata T. Use of temper bead welding for the repair of thick walled pressure vessels. Key Eng Mater 2000;183(2):1339e44. Balasubramanian V, Guha B. Establishing criteria for root and toe cracking of load carrying cruciform joints of pressure vessel grade steel. Eng Fail Anal 2004;11(6):967e74. Ibarra SJ. Weld repairs without pwht in the petroleum industry. Weld Res Council Bull 1996;412:12e5. Friedman LM. EWI/TWI controlled deposition repair welding procedure for 1.25Cre0.5Mo and 2.25Cre1Mo Steels. Weld Res Council Bull 1996;412: 27e34. New York, NY, United States. Brett SJ. Repair of CreMoeV components using Inconel 182 electrodes without heat treatment. In: Repair and reclamation. London, England: Welding Institute, Abington, England; 1986. Brett SJ. The long term creep rupture of nickel based cold welds in CrMoV components. In: I Mech E proceedings of the conference on refurbishment and life extension of steam plants; 1987. Brett SJ. Methodology for the use of weld repairs without post-welding heat treatment on creep resisting steels. BALTICA III 1995;2:3. Mitchell KC, Brett SJ. Review of cold weld repair applications. Power Plant: OMMI 2003;2(1):1e9. Shibli A, Gostling J, Starr F. Damage to power plants due to cycling. E.T.D. Ltd; 2001. Mitchell KC. Repair welding in the power industry using cored wires. In: Proceedings of the 1998 1st IEE/IMechE international conference on power station maintenance e protability through reliability, Mar 30eApr 1 1998. Edinburgh, UK: IEE, Stevenage, England; 1998. Price J, Paradowska AM, Finlayson T. Measurement of residual stresses in weld repairs in steels in weld cracking in ferrous alloys. In: Singh R, editor. Boca Raton, USA/Cambridge, UK: CRC Press/Woodhead Publishing Limited; 2008. p. 459e73 [chapter 12]. Paradowska A, Price J, Finlayson T, Rogge R, Blevins R, Donaberger R, et al. Comparison of neutron diffraction residual stress measurements of steel welded repairs with current tness-for-purpose assessments. ASME J Press Vessel Technol 2010;132(1):011502. BS 7910: Guide to methods for assessing the acceptability of aws in metallic structures. London: BSI, British Standards; 2005. Assessment of the integrity of structures containing defects. Gloucester: B.E.G. Ltd; 2001. Report R6, Rev. 4; 2009. Price J, Paradowska A, Joshi S, Finlayson T, Semetay C, Nied H. Residual stresses evaluation in welds by experimental and theoretical approaches; the issue of gauge volume. Int J Mech Sci 2008;50(3):513e21. Cornu J. Fundamentals of fusion welding technology. Advanced welding systems, vol. 1. Kempston, U.K.: Springer-Verlag; 1988. pp. 141. Mitchell KC. Cold weld repairs using ux cored arc welding (FCAW). In: Ninth international conference of creep and fracture of engineering materials. London: Institute of Materials; 2001. Mitchell KC. The applications of ux cored arc welding (FCAW) in UK power plant. OMMI 2003;2(1):1e7. Sato M, Suda K, Nagasaki H. How to weld using ux-cored wires. Weld Int 1997;11(4):264e72. Aloraier A, Ibrahim R, Thomson P. FCAW process to avoid the use of post weld heat treatment. J Press Vessels Pip 2006;83(5):394e8.

133

[20]

[21] [22] [23] [24]

[25]

[26]

[27] [28]

[29] [30]

[31]

[32]

[33] [34] [35] [36]

[37]

[38]

[39] [40] [41]

[42] [43]

[44] [45] [46]

[47] Aloraier AS, Ibrahim RN, Ghojel J. Eliminating post-weld heat treatment in repair welding by temper bead technique: role bead sequence in metallurgical changes. J Mater Process Technol 2004;153e154(1e3):392e400. [48] Aloraier A. Optimization of ux cord arc welding with bead tempering for repair of critical structures without post weld heat treatment in mechanical engineering. Melbourne: Monash University; 2006. [49] Shehata T, Ibrahim RN, Price JWH, Sadek A, Ushio M. Effects of welding parameters of FCAW process and shielding gas type on weld bead geometry and hardness distribution. Trans JWRI 2001;30(2):45e52. [50] Shehata T, Aloraier AS, Ibrahim RN. The effect of shielding gases on the welding geometry of carbon steel using FCAW process. In: Proceedings of the 50th WTIA annual conference, Sydney, Australia; 2002. [51] Liao MT, Chen WJ. Effect of shielding-gas compositions on the microstructure and mechanical properties of stainless steel weldments. Mater Chem Phys 1998;55(2):145e51. [52] Preston R, Smith S, Shercliff H, Withers P. An investigation into the residual stresses in an aluminum 2024 test weld. In: Fracture, fatigue and weld residual stress. ASME PVP, vol. 393; 1999. p. 256e77. [53] Dong P, Hong J, Bynum J, Rogers P. Analysis of residual stresses in AleLi alloy repair welds. In: Approximate methods in the design and analysis of pressure vessels and piping components. ASME PVP, vol. 347; 1997. p. 61e75. [54] Hong JK, Tsai C-L, Dong P. Assessment of numerical procedures for residual stress analysis of multipass welds. Weld J (Miami, Fl) 1998;77(9):372e82. [55] Brown S, Song H. Finite element simulation of welding of large structures. J Eng Ind Trans ASME 1992;114(4):441e51. [56] Dong P, Ghadiali PN, Brust FW. Residual stress analysis of a multi-pass Grith Weld. In: Fracture, fatigue and weld residual stress. ASME PVP, vol. 373; 1998. p. 421e31. [57] Feng Z, Wang XL, Spooner S, Goodwin GM, Maziasz PJ, Hubbard CR, et al. A nite element model for residual stress in repair welds. In: Residual stresses in design, fabrication, assessment and repair. ASME PVP, vol. 327; 1996. p. 119e25. [58] Shan X, Davies CM, Wangsdan T, O'Dowd NP, Nikbin KM. Thermo-mechanical modelling of a single-bead-on-plate weld using the nite element method. Int J Press Vessels Pip 2009;86(1):110e21. [59] Shan XY, Tan MJ, O'Dowd NP. Developing a realistic FE analysis method for the welding of a NET single-bead-on-plate test specimen. J Mater Process Technol 2007;192e193:497e503. [60] Jin X-J, Huo L-X, Zhang Y-F, Bai B-R, Li X-W, Cao J. Three dimensional nite element numerical simulation of residual stresses of all-position welding in duplex stainless steel pipe. Hanjie Xuebao/Trans China Weld Inst 2004;25 (2):52e6. [61] Katsareas, D.E., C. Ohms, and A.G. Youtsos. On the performance of a commercial nite element code in multi-pass welding simulation. in 2004 ASME/JSME Pressure Vessels and Piping Conference, Jul 25e29 2004, . San Diego, CA, United States: American Society of Mechanical Engineers, New York, NY 10016-5990, United States. [62] Lee CK, Candy J, Tan CPH. Measurement and nite element analysis of temperature distribution in arc welding process. Int J Comput Appl Technol 2004;21(4):171e7. [63] Zhu Y-X, Wang Q, Zhao X-R, Sun Q-M. Simulation of welding residual stress based on ANSYS. J Wuhan Univ Technol 2004;26(2):69. [64] Smith MC, Smith AC. NeT bead-on-plate round robin: comparison of transient thermal predictions and measurements. Int J Press Vessels Pip 2009;86 (1):96e109. [65] Chen C-X, Li W-S, Wang Q-P, Gao B-L. Three-dimensional dynamic FEM simulation of temperature distribution for welding process. J Tianjin Univ Sci Technol 2005;38(5):466e70. [66] Junek L, Slovacek M, Magula V, Ochodek V. Residual stress simulation incorporating Weld HAZ microstructure. ASME PVP e Fract Fatigue Weld Resid Stress 1999;393:179e92. [67] Tsirkas SA, Papanikos P, Kermanidis T. Numerical simulation of the laser welding process in butt-joint specimens. J Mater Process Technol 2003;134 (1):59e69. [68] Vincent Y, Jullien JF, Cavallo N, Taleb L, Cano V, Taheri S, et al. On the validation of the models related to the prevision of the HAZ behaviour. ASME PVP e Fract Fatigue Weld Resid Stress 1999;393:193e200. [69] Bate SK, Charles R, Warren A. Finite element analysis of a single bead-on-plate specimen using SYSWELD. Int J Press Vessels Pip 2009;86(1):73e8. [70] Gilles P, El-Ahmar W, Jullien J-F. Robustness analyses of numerical simulation of fusion welding NeT-TG1 application: Single weld-bead-on-plate. Int J Press Vessels Pip 2009;86(1):3e12. [71] Smith MC, Smith AC. NeT bead-on-plate round robin: comparison of residual stress predictions and measurements. Int J Press Vessels Pip 2009;86(1):79e95.