Analysis of a Defected Dissimilar Metal Weld in a PWR Power Plant

During the refueling outage 2000, inspections of the RC-loops of one of the Ringhals PWR-units, Ringhals 4, indicated surface breaking defects in the axial direction of the piping in a dissimilar weld between the Low alloy steel nozzle and the stainless safe end in the hot leg. In addition some indications were found that there were embedded defects in the weld material. These defects were judged as being insignificant to the structural integrity. The welds were inspected in 1993 with the result that no significant indications were found. The weld it self is a double U weld, where the thickness of the material is ideally 79,5 mm. Its is constructed by Inconel 182 weld material. At the nozzle a buttering was applied, also by Inconel 182. The In-service inspection, ISI, of the object indicated four axial defects, 9–16 mm deep. During fabrication, the areas where the defects are found were repaired at least three times, onto a maximum depth of 32 mm. To evaluate the defects, 6 boat samples from the four axial defects were cut from the perimeter and shipped to the hot-cell laboratory for further examination. This examination revealed that the two deep defects had been under sized by the ISI outside the requirement set by the inspection tolerances, while the two shallow defects were over sized, but within the tolerances of the detection system. When studying the safety case it became evident that there were several missing elements in the way this problems is handled with respect to the Swedish safety evaluation code. Among these the most notable at the beginning was the absence of reliable fracture mechanical data such as crack growth laws and fracture toughness at elevated temperature. Both these questions were handled by the project. The fracture mechanical evaluation has focused on a fit for service principal. Thus defects both in the unaffected zones and the disturbed zones, boat sample cutouts, of the weld have been analyzed. With reference to the Swedish safety evaluation system in accordance to the regulatory demands, a safety evaluation was performed using the R6-method. The failure assessment diagram is modified by the addition of the ASME XI safety factors both for limit load analysis and fracture assessment. This results in a very high conservatism since the secondary stresses such as residual stresses are high in the area. In order to quantify this effect an analysis in accordance to ASME IWB-3640, App. C was performed. This analysis provides the decision-makers with a sensitivity study; important to have to value the real risk of any missed defects in the area.