Pacific Northwest National Laboratory has been conducting studies for the NRC on systems susceptible to pressurized water stress corrosion cracking (PWSCC). Weld overlays with resistant materials are being used to mitigate PWSCC found in reactor components. This study addressed the inspection of weld overlays.Two specimens with a dissimilar metal weld (DMW) between a 516 Grade 70 carbon steel nozzle and a CASS pressurizer (PZR) surge-line pipe segment were implanted with thermal fatigue cracks (TFCs). In the first specimen, four TFCs were implanted in the weld-to-butter region with flaw depths ranging from 13% to 31% through-wall. The second specimen contained five circumferentially oriented IDconnected TFCs (flaw depths ranging from 49.8% to 89.9% through-wall) implanted in the center of the weld during the welding process to minimize implantation artifacts that could show in the ultrasonic data. The specimens were ultrasonically evaluated with phased-array probes having center frequencies of 0.8, 1.0, 1.5, and 2.0 megahertz (MHz). An Alloy 82/182 weld overlay (WOL) was applied and the specimens were ultrasonically re-evaluated for flaw detection and characterization. The Post-WOL flaw depths were approximately 10% to 56% through-wall.The WOL specimens were approximately 53 mm (2.1 in.) thick and with this thicker material suggested that a lower inspection frequency, on the order of 0.8 or 1.0 MHz, was needed to detect and size flaws from the cast austenitic stainless steel (CASS) pipe side of the weld. However, the longer wavelengths of the two lower frequency probes prevented them from detecting the two smallest flaws (10% and 14% through-wall deep). Accordingly, 1.5-and 2.0-MHz probes were then used and it was shown that microstructures in the cast pipe and overlay materials were amenable to higher frequencies as all of the flaws were detected and characterized. It should be cautioned that these results were obtained after removal of the weld crown as it limited access to the flaws and thus prevented full flaw detection.The detected flaws had SNR values of approximately 15 dB which are well above a minimal level of 8 to 10 dB for good detection. The depth sizing errors as calculated RMSEs were within the ASME requirements of 3.2 mm (0.125 in.) for the line scan data and a subset of the raster data where the analyst emphasized tip diffraction sizing. The remaining raster data, based more on specular reflection analyses, were above the ASME required RMSE level. Length sizing RMSEs were generally within the ASME requirements of 19 mm (0.75 in.) except when including the shallow flaws. The presence of a strong weld root signal that was not always separated from the flaw ID corner response contributed to the error in length sizing.This study has shown the effectiveness of ultrasonic examinations of Alloy 82/182 overlaid DMW specimens. Phased-array probes with center frequency in the 0.8-to 1.0-MHz range provide a strong coherent signal but the greater ultrasonic wavelength and larger beam spot size prevent the rel...