Investigation of local tensile strength and ductility properties of an X100 submerged arc seam weld

Abstract

Tensile testing is the most common method to determine the mechanical properties of a weld joint, yet the local tensile properties can be different from the global properties when stress localization occurs. In the current study, Digital Image Correlation was used to explain the difference in local and global tensile behavior in different sections of a submerged arc specimen. Tensile tests are used to compare the base metal and weldment properties of a two-pass submerged-arc welded X100 pipe steel in terms of strength and ductility. The base metal exhibits superior tensile strength compared to specimens machined from the second pass (Pass 2) or the overlap weld region spanning between Passes 1 and 2, corresponding to the reheated heat affected zone. Both cross-weld tensile specimens fractured in the fine grained heat affected zone (FGHAZ). Although the overlapped tensile specimen achieved higher global strain compared to the Pass 2 weld specimen, it sustained lower local strain, which promoted fracture in the HAZ. In addition, the overlapped tensile specimen exhibited a higher degree of strain hardening. The martensite-austenite (MA) microconstituents in the FGHAZ of Pass 2 contained higher carbon than the MA in the overlapped FGHAZ, which led to increased MA hardness as measured by nanoindentation. A higher fraction of MA with a finer spacing in the FGHAZ of the overlapped HAZ zone provides higher strength and strain hardening but lower ductility. The tensile fracture was dominated by void formation from the hard MA/ferrite interfaces in all tensile specimens, and so an increased fraction of MA led to accelerated microvoid coalescence, and reduced ductility in the overlapped HAZ specimens. In-situ strain monitoring technology via Digital Image Correlation was used to identify the potential fracture location during tensile test. X-ray tomography clearly revealed microvoid formation and coalescences of voids.