Comparison of three different techniques for measuring the residual stresses in an electron beam-welded plate of WASPALOY

The longitudinal, transverse, and through-thickness (short-transverse) residual stresses in an electron beam-welded plate of WASPALOY, a high-strength nickel-based superalloy, have been characterized using neutron diffraction, X-ray diffraction, and a hole-drilling method. Where possible, the results from the different techniques, and the associated uncertainties, have been compared. For the neutron measurements, the gamma/gamma {111} peak was used for the determination of lattice strains. The X-ray measurements were carried out using Fe K(alpha) radiation, the sin(2) psi technique, and the {311} gamma/gamma composite peak. The Matthar-Soete method was used for the incremental hole-drilling measurements. Unfortunately, due to texture effects, it was not possible to detect the residual stresses within the weld metal by the diffraction-based methods. For the estimation of residual stresses, plane-specific values of the Young's modulus and Poisson's ratio were determined from tensile testpieces using in situ neutron diffractometry. When these data are used, it is found that the neutron, X-ray, and hole-drilling residual stress data are mutually consistent, although the absolute certainties vary with the method employed. The results indicate that, next to the weld, the longitudinal residual stresses approach 1000 MPa and are typically far greater (up to 5 times) than those in the transverse and through-thickness directions. Measurements of the longitudinal strain with distance along the welding direction indicate that the stress state reaches a steady state over the central portion of the plate; for this reason, the majority of the diffraction measurements have been made in the plane perpendicular to the weld at the center of the plate. A simple analysis of the thermal cycles and the extent of plastic deformation induced in the specimen is presented. The plastic "upset zone" has a size which is at least 3 times greater than the cross-sectional area of the weld metal; this suggests that, for accurate analysis of weld-induced distortion, attention should be paid to the evolution of residual stresses in the heat-affected zone as well as the fusion zone.