A Comparison of Residual Stresses in Single-Pass and Multipass SA508 Steel Welds

Abstract

The safe operation of a power plant is contingent on the integrity of its welded joints. In turn, the performance of these welded joints is strongly influenced by the residual stresses that remain after fabrication and, where applicable, post-weld heat treatment operations. It is thus critical that careful consideration is given to the influence of weld residual stresses on plant integrity at the design stage, as well as during subsequent service. Since it is generally difficult to measure weld residual stresses in thick-walled components, numerical models are employed to make predictions. Such models, however, will be limited by our understanding of material behaviour, particularly in the case of multipass welds, which involve several complex thermo-mechanical cycles. Here, we report on neutron diffraction measurements, which enable the residual stresses for a single weld bead deposited on to a 20 mm thick SA508 steel plate to be compared with those in an 8-pass groove weld. The same low-carbon (SD3) filler metal was used in the manufacture of both types of specimen. Interestingly, the residual stress distributions for each type of specimen were found to be very similar. This suggests that cyclic hardening effects did not play a decisive role in determining the final residual stress distribution within the multipass weld.