•Unilateral sample preparation and control of oxygen did not reduce scatter.•Oxide deposition was observed in some facilities while others reported oxide loss.•Interlaboratory statistics indicate between-laboratory differences were significant.•Temperature distributions within autoclaves proposed as one cause of scatter. Supercritical water-cooled reactor candidate materials are typically corrosion-resistant alloys whose mass changes in supercritical water fluctuate around zero. A previous interlaboratory corrosion experiment exercise revealed large scatter in mass change results between laboratories. Here, we reduced systemic differences between laboratories by unilateral preparation of coupons and unilateral chemical cleaning. Type 310S stainless steel and Alloy 800HT test coupons were exposed for 1000 h to 550 °C, 25 MPa, deaerated water. Average mass loss for 310S and 800HT was 38 ± 26 and 51 ± 31 mg/dm2, respectively. Differences in mass transfer, galvanic and local corrosion, and average coupon temperature may explain the poor reproducibility.