Experimental evaluation of multiple-component residual stress distribution in a dissimilar metal repair joint of ferritic/martensitic steel using the asymmetric-cut and the symmetric-cut contour method

Abstract

A thick multi-pass butt-welded ferritic/martensitic steel (P91) joint is prepared and the local region of the weld is repaired with an ENiCrFe-3 nickel-based alloy using the manual metal arc welding method. The repair weld including part of the base metal is then subjected to ultrasonic impact treatment (UIT). A three-cut contour method, including two asymmetric cuts and one symmetric cut, is developed in the present study to obtain the two-dimensional residual stress distributions at different locations, and the stress release after multiple cuts is considered to get the original stress distribution before cuts. The measurement procedure is introduced in detail. The longitudinal stresses in the repair weld and the initial weld of the P91 steel joint, as well as the transverse stress at the weld center, are finally obtained. The effects of the dissimilar metal repair weld at the local region on the longitudinal and transverse welding residual stresses are investigated. In addition, the applicability of UIT to mitigate the surface stress in nickel-based alloy repair weld is analyzed. The results show that the transverse and longitudinal stresses in the nickel-based alloy repair weld are both tensile stresses, the maximum longitudinal stress is close to the yield strength of the B91 weld metal and occurs in the heat-affected zone of the repair weld located in the initial weld. Repair welding causes the internal transverse compressive stress region to be narrower than in the original weld. UIT can be used to introduce compressive stress to the surface layer of the nickel-based alloy repair weld.