A Parametric Study of Rm/T and Weld Repair Effects on Through-Wall Axial and Hoop WRS Distributions in BWR Nozzle Dissimilar Metal Welds

Abstract

Weld residual stress (WRS) distributions are an important input into fracture mechanics evaluations necessary to determine the residual lives of dissimilar metal welds (DMWs). Since the DMW geometry and the presence or absence, size, and location of weld repairs is nozzle specific, finite element WRS analysis is often used to predict through-wall weld residual stress distributions. It is important to note that despite small differences in plant specific geometry or weld location specific weld repair geometry there are substantial similarities between the configurations that have been evaluated in the numerous weld specific finite element WRS analyses documented in the literature. Important insight can be gained from parametric studies of simplified geometries in order to understand the significance of different parameters on the resulting WRS distributions. The results of such studies can allow engineers to focus resources on refining accuracy of critical inputs and to support simplified model development suitable for incorporation into design and fitness for service codes. This paper documents the results of various studies performed to validate the ability to use a simplified pipe-to-pipe model for simulating relative effects on through-wall WRS distributions of pipe and weld repair geometry, investigate the effect of pipe mean radius to wall thickness ratio, weld repair depth (ID and OD), and weld repair sequence. Fifteen cases are analyzed. The dimensions selected for each case span a range of wall thickness, Rm/t and depth of repair values representative of typical Boiling Water Reactor (BWR) nozzle DMWs. The results are used as input into a simplified WRS model presented in a separate paper [17].