Welding Residual Stress and Multiple Flaw Evaluation for Reactor Pressure Vessel Head Replacement Welds With Alloy 52

Abstract

Small indications were found in one replacement reactor pressure vessel head (RPVH) mock-up being fabricated from Alloy 690 material and compatible weld metals, Alloy 52/152. The mockups were non-destructively examined and the lowest number of cracks found was five and the highest number was 22. There are numerous indications with some of them quite long (50 mm) in length. The source of these weld fabrication cracks is unknown. However, from experience with other difficult to weld materials, the source can range from slag inclusions in the weld metal to hot cracking during the weld deposition process. Hot cracking includes solidification cracking (weld), liquation cracking (HAZ), and ductility dip cracking (DDC). The indications were mostly circumferential in orientation (with respect to the nozzle axis) but some were axial. This paper includes two parts. The first part includes the welding residual stress analysis of RPVH using Alloy 52/152 metal and provides comparison with similar Alloy 82/182 welds. Alloy 82/182 was the material used in the original dissimilar metal welds in these heads. Primary Water Stress Corrosion Cracking (PWSCC) can occur in the primary coolant system when the welds are exposed to water, tensile stress, and temperature (usually higher than 250 C). PWSCC rates are higher in Alloy 82/182 material due to its lower chromium content compared with the replacement alloy. The results for both center hole (0-degree) and side hill (53-degree) nozzles will be discussed. The second part deals with assessment of multiple small cracks in the reactor pressure vessel head penetration nozzles. The finite element alternating method (FEAM) was used for calculating stress intensity factors for cases where multiple cracks exist. More than twenty cracks, which were inserted based on field measurements, are considered in the analyses for both center hole and side hill nozzles. It is observed that the overall stress trends are similar to those without adding cracks. However, cracks introduce more local stress fluctuations. The magnitude of the local fluctuation can be around 100MPa. Limit analysis was also conducted. A new finite element model with a voided-out weld region was used to simulate loss of structural capacity due to multiple flaws. The voided out volume effects on the structural integrity and future performance of RPVH were examined. Discussions based on weld residual stress, multiple flaw analysis and limit analysis conclude the paper.