Microstructural and mechanical performance of underwater wet welded S355 steel

https://doi.org/10.1016/j.jmatprotec.2016.07.039Get rights and content

Abstract

Underwater wet shielded metal-arc welding using three heat inputs were investigated. The columnar microstructures in weld metal contain large amounts of ferrite with second phase aligned and grain boundary ferrite, and some polygonal ferrite and acicular ferrite. The coarse grain heat affected zone (CGHAZ) is dominated by lath martensite and some bainite. The microstructure evolution of the reheated columnar region and CGHAZ is discussed to explain the beneficial effect of temper bead. The unaltered CGHAZ in the top layer is the most preferential sites for cold cracks due to the lack of tempering. Although large heat input can reduce the hardness and the susceptibility of cold cracking, it also degrades the mechanical performance of weld metal owing to the microstructure coarsening and less reheated regions.

Introduction

Rowe and Liu (2001) showed that underwater wet shielded metal-arc welding had been widely applied for many years in the repair of offshore platforms, especially for the water depth less than 60 m. The shortened repair duration is the major advantage of this technique. A number of experimental studies have been published in recent years based on the metallurgical investigation and the mechanical properties using various electrodes under different water depths. Guo et al. (2015) added nickel to the electrode covering and showed that 2.5% (wt.) nickel could obtain the optimal mechanical properties. Rowe et al. (2002) found additions of titanium and boron effectively refined the ferrite grain size of the as-deposited microstructures and reduced the oxygen content in the weld metal. Santos et al. (2012) developed an oxyrutile electrode for wet welding combining the good operability and the low diffusible hydrogen content. Teran et al. (2014) investigated the mechanical properties and the structural integrity of T-welded connections repaired by grinding and wet welding, and a U-shape profile in the bottom of the grinding was recommended to obtain the weld with less defects of pores and slags, compared with a rectangular profile. However, there is little information available in literature about the effect of welding heat input on the microstructure and mechanical performance.

Underwater wet welds are usually related with the high cooling rate, quenched microstructure and arc instability resulting from the presence of water around the electric arc. Although numerous studies have focused on the mitigation of the detrimental fast cooling rate, it is still one of the major problems restricting the application of this technique. Preheat treatment and post-weld heat treatment are common methods to reduce the cooling rate in air welding, whereas it is almost impossible to perform in underwater environment. Therefore, some simple and practical techniques have been explored. Temper bead welding (TBW) technique is the most widely used technique without any complicated underwater equipment. ASME Section IX defined that a temper bead is one that is placed at a specific location at the surface of a weld to modify the metallurgical properties of the heat affected zone (HAZ) or previously deposited weld metal (WM). Fydrych et al. (2013) reported that the maximum hardness in HAZ of S355J2G3 steel underwater wet joints was 400 HV10 and could be reduced to 350 HV10 by using TBW. Fukuda et al. (2009) developed the ambient temperature TBW technique using underwater laser bead welding with low heat input to mitigate the degradation of toughness of low alloy steel in pressurized water reactors. They found that the highest value of hardness was reduced to 300 HV after 6-layer tempering from previous 500 HV and the results of Charpy impact tests exceeded those of base metal. Another widely used technique is half-bead technique similar to TBW. The top half of the first layer is removed by careful grinding with the second layer deposited on the ground surface to temper the first layer. This continues until the required fill is achieved and then the final layer is ground off. The major difficulty of this technique is precise grinding as pointed out by Lant et al. (2001). Toe cracking is a major problem in underwater wet welding using both the half-bead and TBW techniques, as the toe region does not receive as many thermal cycles as other regions. Ibarra (1996) suggested that run-off plates could be tack welded above the plate close to the edge of groove, on which weld beads were deposited to temper the toe region.

In this work, the effects of welding heat input on microstructures and mechanical properties are investigated and the microstructure evolution in multi-layer welding is characterized to give the metallographic evidence of the application of TBW, half bead welding and run-off plates.

Section snippets

Materials and experimental procedure

Normalized 12.7 mm-thick S355 steel plate with the microstructure of polygonal ferrite and pearlite is chosen as base metal (BM) because of its wide application in marine structures. The carbon equivalent (CE) is 0.39 according to IIW. The filled material is a commercial E7014 rutile underwater electrode with a diameter of 3.2 mm and a waterproofing of wax. The covering is mainly composed of iron powder and titania. The content of diffusible hydrogen (welded at 0.5 m water depth) is 46.72 mL/100 g

Weld bead geometry

Fig. 3 shows the weld geometry parameters together with the cross-sectional observations of bead-on-plate welds. No welding defect is found by naked eyes. The weld width and the weld penetration increase remarkably with increasing welding heat input just as one expects, while weld reinforcement seems to increase slightly. It indicates that the welding heat input has little effect on the weld reinforcement owing to the insufficient time for molten weld metal to spread out resulting from the fast

Discussion

The inferior mechanical properties in underwater wet joints discussed above are related to the brittle microstructures and the occurrence of hydrogen induced cracks (HIC). Fig. 12 depicts the two typical cracks in HAZ and WM, respectively. Fig. 12a shows a hydrogen induced crack in UACGHAZ of the cap bead where the coarsest martensite forms. The crack tends to propagate perpendicular to the direction of the transverse tensile residual stress. Thus, UACGHAZ of the top layer turns out to be the

Conclusions

On the basis of all the obtained results, the following conclusions can be drawn:

  • 1

    The columnar microstructures in WM consist of large amounts of FS (A) and GBF, and small amounts of PF and AF. As the heat input increases, the relative percentage of FS (A) and AF decreases with increasing GBF and PF. In CGHAZ, the microstructure is dominated by lath martensite, although some upper bainite arise at large welding heat input.

  • 2

    The unaltered CGHAZ in the top layer is the most preferential sites for

Acknowledgments

The authors acknowledge the financial support of marine scientific research project (Investigation on the key technology of underwater welding and inspection processes – Developing special machines and processes for underwater local dry and wet welding) sponsored by Offshore Oil Engineering Co., Ltd. Great thanks are given to Jun Cao and Wei Xu from Offshore Oil Engineering Co., Ltd.

References (18)

There are more references available in the full text version of this article.

Cited by (71)

View all citing articles on Scopus
View full text