Dissimilar underwater wet welding of HSLA steels

The high-strength low-alloy S460ML and S460N steels were chosen for underwater wet welding of dissimilar T-joints using covered electrodes. For improving the quality of joints, the temper bead welding (TBW) method was used. The application of TBW in pad welding conditions has been investigated earlier but the possibility of usage of this technique in welded joints was not analyzed. The main aim of the study was to check the influence of TBW on the hardness and structures of the heat-affected zone (HAZ) of dissimilar T-joints made in the underwater conditions. The experiments conducted showed that the technique used can reduce the susceptibility to cold cracking by decreasing the hardness in HAZ, which is a result of changes in its structure. The TBW technique reduced the hardness in the HAZ of the S460N steel by 40–50 HV10 and in S460ML by 80–100 HV10. It was also found that the changes in S460ML and S460N were much different, and therefore, the investigated technique can provide better results in the steel characterized by lower carbon equivalent CeIIW.


Introduction
The number of offshore steel constructions is increasing every year. One of the reasons is the oil and gas sources present under the sea, which are being explored more and more. The constructions working in the water environment can be categorized as ships, bridges, wind turbines, terminals, pipelines, and harbor structures such as quay or marginal wharf [1]. In the offshore constructions, damages due to different factors including corrosion, fatigue, and vessel impact, design faults, mechanical damages, and imperfections in welded joints could be found [2][3][4][5]. It is very important to predict the location of the damages in the constructions. Most of the methods used for the detection of the damages are based on the numerical and computational approaches and are focused on the localized monitoring of strain [6,7]. However, new methods of localization are still developing, for example, a method based on changes in the frequency spectrum [8]. During design processes, offshore structures get their design life, which could often be passed. For example, more than 50% of operational offshore platforms in the Norwegian Continental Shelf and the UK Continental Shelf have exceeded their design life [9]. Engineering is trying to make exploitation time longer, which can reduce the cost of the potential repairs or enable changing the constructions to new ones. This could be done in the stage of design, production, or repair.
A lot of investigations are being carried out on the strengthening, modification, and repair techniques used in offshore constructions. These techniques are classified into five groups: welding, improvement of the weld, clamping, grout filling, and others [10]. Some of them are implemented during the production of the construction as shot peening for treating the multipass welding-inducted residual stresses in offshore wind turbine monopiles [11]. The other welding method used to extend the life of the offshore constructions is welding with the specific sequence [12]. Due to extreme environmental conditions and loading induced upon an offshore construction, the normalized and the thermomechanically treated steels with high yield point are widely used [13]. They allow reducing the weight and size of constructions while maintaining appropriate mechanical properties [14]. The weldability of steels in the air environment is widely investigated. It has been shown that it depends on chemical composition, manufacturing method, and technological, metallurgical, and construction factors, and above all the welding technology [15][16][17]. Much less frequently mentioned is the subject of steels with yield strength higher than 420 MPa welded directly in the water.
The repair technique most often used for offshore constructions is underwater welding, which can improve the quality of joints or repair the gaps in the surfaces. This process is of three main types. The first one is dry welding, in which the welding area and welder are isolated from the environment. This method requires building a special chamber to avoid the surrounding water from contacting the welding zone, which adds to the cost of the process. The pressure inside the chamber could be the same as in the air (isobaric dry welding) or can be due to the depth (hyperbaric dry welding). This method also increases the cost but the quality of the joints could be similar to the joints made in the air [18,19]. The second method of underwater welding is welding with a local dry chamber. In this method, the welder is in the water, but the areas of welding and joint are isolated from the environment by a small chamber, in which the welding gas removes the water outside. This phenomenon produces conditions similar to the hyperbaric dry welding but does not require building any expensive chamber [20,21]. The last is the most popular and the cheapest method of underwater welding which is known as wet welding. This method also does not require building of any special chambers. During wet welding, the welder and the welding area are in direct contact with the surrounding environment throughout the process. The most popular wet welding processes are flux-cored arc welding (FCAW) [22][23][24][25][26] and welding with the use of covered electrodes (MMA) [27][28][29].
Water as a welding environment increases the susceptibility of steel to cold cracking [30,31]. Cold cracks can occur in the welds and in the heat-affected zone (HAZ) of the welded joints even 48 h after welding [32,33]. Welding in the water increases the number of cold cracks due to high cooling rate, which makes the HAZ brittle [22,23,27,30]. This can also generate higher residual stresses after welding. The third reason for cold cracking, which cannot be avoided during wet welding in the water, is the high diffusible hydrogen content of deposited metal. The welding conditions have a high influence on the saturation of the metal structure with hydrogen in each environment [34]. It was proved that water generates much more diffusible hydrogen than during air welding [35]. An additional factor increasing the risk of cracking is the local stress concentration due to imperfections resulting from the instability of the welding arc [36][37][38][39][40].
In the present research, the high-strength low-alloy (HSLA) S460ML and S460N steels were selected as study materials. Both of them are characterized by a high susceptibility to cold cracking in wet welding conditions [27,30]. The cracks occur in the HAZ of these steels which could be the reason for damage of the construction. The previous investigations showed that the use of temper bead welding (TBW) could reduce the susceptibility to cold cracking the S460ML [30] and S460N steels [31]. However, they were carried out in the bead-on-plate conditions, but not in the real welded joints. In this paper, the influence of the application of the TBW technique in the preparation of dissimilar T-joint fillet welds made in the underwater conditions on the susceptibility to cold cracking was investigated. Welding T-joint was chosen because this type of joint is widely used in thermomechanically treated steels [41] and the fillet welds are most commonly used for underwater welding.

Materials and methods
For welding, the S460ML and S460N (12 mm thick) steel plates were chosen as base material (BM). The chemical composition of both BMs was analyzed by the emission spectrometry method and is presented in Table 1. The investigated steels have similar values of yield point (min. 460 MPa). However, they are characterized by a different carbon equivalent (the absolute difference in Ce IIW values of about 0.08%). The ISO 2560-A [42]: E 38 0 R 11 rutile electrodes with a diameter of 4.0 mm were used as a filler material. These electrodes provide good plasticity of the weld, which can minimize the possibility of cold cracking. The chemical composition of filler material is presented in Table 2.
For test three T-joint specimens with fillet welds were prepared. They were made at a depth of 0.2 m (distance from  Fig. 1. During manufacturing of joints, the basic assumption was to simulate the technological and thermal conditions of repair welding of ribs in a freshwater reservoir. The specimens were tacked under water and were able to deform freely, especially the web (S460ML plate, Fig. 1), simulating, among others, the stiffening of structures, which are used in repairs in wet welding conditions. The dimensions of the specimens were smaller than the elements of structures operated under water, which for this element thickness (12 mm) worsened the thermal conditions compared to real ones.
One specimen was welded using the traditional technique (one weld). In two others, TBW technique was used. TBW relies on laying the second bead to the bead welded earlier to providing the local heat treatment of weld and HAZ structures occurred during the first welding. Between two welds, 120 s time was passed. Previous experiments with the investigated steels in the bead-on-plate conditions [30,31] showed that TBW can be an effective method to improve the weldability of steel in the water, but only in the best range of the overlap values between the subsequent beads. This range was estimated as 75-100% for S460ML and 66-100% for S460N [30,31]. In accordance with this information, the welds in the two specimens were welded using the TBW technique with a different pitch (percentages of overlap of the second weld bead onto the first weld) to check the influence of TBW on the properties of the T-joint fillet welds.
The welding parameters were chosen in accordance with the previous investigations [30,31] and are presented in Table 3. Heat input values were calculated without taking into account the thermal efficiency coefficient "k" in accordance with the requirements of EN 1011-1 standard. The coefficient "k" was omitted because its value is unknown in water environment.
All welded joints were tested by visual testing (VT) and penetrant testing (PT). The aim of VT and PT, besides the assessment of the quality of the prepared joints, was to select appropriate places for cutting. Following VT and PT, the selected cross-sections were tested by macroscopic and microscopic testing. During macroscopic testing, the pitch (percentage overlapping of tempering bead on previously laid bead) was calculated from the prepared photos of cross-sections

Non-destructive testing
At first, all six welds were visually tested in accordance with the EN ISO 17637:2011 standard [44]. Then, all of them were subjected to PT in accordance with EN ISO 3452-1:2013-08 standard [45]. Non-destructive tests were performed which showed some imperfections in welds prepared using the TBW technique. The most common imperfections were lack of fusion and undercut ( Fig. 2b and d). These imperfections resulted from temper beads which were laid only for tempering the previous beads. Because the undercuts are classified as imperfect shape and dimension of weld, and do not affect the changes in the structure and properties of joints, all specimens were subjected to further tests. During testing, spatters was found on the surface of the specimens. In all welds, there were areas without any surface imperfection, which allowed the specimens and preparing cross-sections for further examinations. The exemplary results of the non-destructive tests are presented in Fig. 2.

Metallographic testing
Macro-and microscopic investigations were performed in accordance with EN ISO 17639:2013 standard [46]. After cutting, all the cross-sections were ground, polished, and etched with Nital (4%). The main aim of the macroscopic testing was to calculate the pitch in specimens 2 and 3. During macroscopic observations, the imperfection was seen as gas pores in both welds from specimen 1, weld 1 from specimen 2, and weld 2 from specimen 3. All these imperfections were located near the point of contact between the BMs. The used TBW technique did not affect their formation-the second weld did not cover these places. The results of the macroscopic testing are presented in Fig. 3. After macroscopic tests, the microscopic investigations were performed. The weld metal structures in specimen 1 were built of dendrites. In the specimen 1-welded without TBW technique, brittle structure such as low-carbon martensite was found in the HAZ of both steels in each welded joint. Near the fusion line the Widmannstätten structure was observed ( Fig. 4a and b). In addition, in the HAZ of the S460N steel, cracks were found in weld 2, which was welded with a lower heat input (Fig. 4c). These cracks started in HAZ and propagated along approximately 80% of the length of the fusion line in the S460N steel. Cracks were also found in the weld material of the specimens in which the TBW technique was not applied (Fig. 4d). In the case of specimens made using TBW, there was no crack in the weld materials. In weld 1 specimen 2, where the pitch was 0%, the HAZ in S460N was not changed and still presented brittle structures with cracks (Fig. 4e). The pitch was 0%; however, the HAZ from tempering bead refined and tempered the coarse-grained zone of the first bead in the S460ML steel (Fig. 4f). In the specimens with the pitch in the range of 79-100%, the tempering effect was observed in the HAZ of both the investigated materials. With an increase in the pitch, there was a decrease of grain size, which was expected in accordance with previous works [30,31]. In the places where the HAZ of the second bead tempering the weld from the first bead, there was a partial disappearance of the structure of the dendritic base bead and the formation of a ferritic fine-grained structure (Fig. 4g). The structures in the area where the HAZ from tempering bead overlapped the HAZ of the first bead in both steels were characterized by the refined and tempered coarse-grained zone (Fig. 4h). The cracks in the HAZ of the BMs were found only in weld 2 in specimen 3 in the S460N steel (Fig. 4i). In the previous investigations [27,30], it was found that TBW could not repair the microcracks that formed during the welding of the first bead. However, TBW could help to avoid cold cracking that might occur after welding [47,49].

Hardness measurements
The hardness was measured in BM, in each HAZ, and in the weld material for each investigated weld of a specimen. The schema of distribution of hardness points is presented in Fig. 5.
The measurement showed that the S460N steel is characterized by a higher Vickers HV10 hardness than S460ML, as can be observed in the measurements obtained for each specimen. In the specimens in which TBW technique was not applied, the hardness in the HAZ of both the used steels was the highest and exceeded the level of 380 HV10 assumed by It was also observed that the hardness of S460N HAZ was higher than that of S460ML HAZ. In the specimen with the 0% pitch, the heat from the second bead, which was laid in S460N BM, affected the hardness of both steels. Second, HAZ was overlapping first in S460ML, and the hardness in this area decreased by 80-100 HV10 in comparison with the specimens welded without the TBW technique. In S460N HAZ, the hardness decreased by 40-80 HV10. In the next specimens, it was observed that if the pitch between two beads increased, the hardness in HAZ and weld material decreased. The best results for the investigated BMs were observed in the specimens with 100% pitch. The hardness measurements showed that the TBW technique allowed to reduce the hardness HV10 values of joints made in the water to the values lower than in specimens welded in the air [27]. Similar results were observed for welding of HSLA steel butt joints [49]. The reduction of hardness limits one of the factors adding to the initiation of cold cracks, which contributes to a reduction of cracking susceptibility.
The values of hardness in each point are presented in Table 4. The phrase "without TBW" means specimen welded without additional (tempering) bead. The specimen with "0%" pitch was welded with tempering bead. However, in the area of cutting specimen for metallographic testing, the second bead did not lay on the first one (Fig. 3c), but the heat from the second bead affects the previous laid stitch. The hardness distributions for each specimens are presented in Fig. 6.

Conclusions
The results of the experiments conducted in the present study showed that both materials, S460N and S460ML, were characterized by high susceptibility to cold cracking in the HAZ, as was found in the specimens made without the TBW technique. The use of this technique allowed reducing the number of cracks and decreasing the hardness in the HAZ of both steels. The results proved that the Ce IIW values cannot be considered as good weldability indicator for wet welding in water environment. In previous researches [31,32], TBW was used in pad welding conditions. Bead-on plate welding is a relatively simple process which is carried out only with one base material. During T-joint fillet welding, the thermal severity is more complicated, which can result in formation of different brittle structures. The experiments presented in this paper confirmed the effectiveness of used TBW technique in the conditions of preparing welded joints. It is also confirmed that this technique can be used in preparing dissimilar joints. The performed conclusions can be used during repairing of offshore constructions in the cases where used materials are characterized by susceptibility to cold cracking.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.