A Comparison of Residual Stress in Hammer-Peened, Multi-Pass Steel Welds–A514 (S690Q) and S41500

Abstract

As part of a program to assess the in-situ weldability and mechanical performance of the candidate high strength low alloy A514 (S690Q) steel as an alternative to the S41500 martensitic stainless steel for hydro-turbines, three aspects of the welds were studied: residual stress, Charpy toughness and cavitation erosion resistance. The experimental set-up involved robotized gas metal arc welding (GMAW), performed on U-groove and double-V weld preparation cut into 50 and 75 mm-thick steel plates. Half of the welds were robotically hammer-peened after each weld layer, except for the root pass. Strain gauges measured longitudinal and transverse strains during welding and hammer-peening. Once the weld cooled down to room temperature, the strain gauges provided the surface residual stress level at their location. Two-dimensional, sub-surface, longitudinal, residual stress distributions were measured on cut sections with the contour method, using an optical profilometer. The results showed that hammer-peening completely eliminates the near-surface tensile welding residual stress on the A514 steel, whereas on the S41500 steel, the process is less useful due to the already beneficial effect of the low temperature martensitic transformation during weld cooling. Furthermore, hammer-peening the last weld layer confines tensile residual stress inside the weld, while inducing compressive stress at the weld surface. Charpy test results showed that the A514 weld presented better toughness than the S41500 weld and comparable cavitation erosion resistance. Finally, hammer-peening showed a beneficial effect on cavitation resistance of the weld surface.