EXPERIMENTAL STUDY OF THE WELD RESIDUAL STRESS IN MANUALLY AND MECHANICALLY FABRICATED DISSIMILAR METAL WELD

The dissimilar metal welds between the Inconel 600 flow element and the SA-106 Grade B carbon pipe with Alloy 82 or Alloy 182 filler material of some CANDU designs have been identified as being susceptible to Primary Water Stress Corrosion Cracking (PWSCC). Initiation and growth of PWSCC in a Dissimilar Metal Weld (DMW) are driven primarily by Welding Residual Stresses (WRS). The present paper focuses on the experimental study of weld residual stress distribution in manually and mechanically fabricated DMWs with emphasis on the effect of repair. A series of DMW samples are firstly fabricated in accordance with the original welding procedures for those DMWs in the field, which were fabricated in 1970s and 1980s. Multiple thermocouples were used to record the temperature evolution during the entire welding process. These samples were then examined by ASME qualified personnel in accordance with the requirements for Class 1 weld in Article 9 of Section V of ASME BVPC using Visual Testing (VT) and Radiography Testing (RT). Repair was then performed in some samples, and further NDE examinations were performed. The qualified samples (with and without repair) were finally subject to destructive weld residual stress measurement using contour method. It is observed that weld repair dramatically changes the distribution of weld residuals tress. The use of a constant through-thickness WRS of 60,000 psi (415 MPa) is justified as the bounding case.