Abstract

Previously reported stress corrosion cracking (SCC) rates for Alloy 82H gas-tungsten-arc welds tested in 360 C water showed tremendous variability. The excessive data scatter was attributed to the variations in microstructure, mechanical properties and residual stresses that are common in welds. In the current study, however, re-evaluation of the SCC data revealed that the large data scatter was an anomaly due to erroneous crack growth rates inferred from crack mouth opening displacement (CMOD) measurements. Apparently, CMOD measurements provided reasonably accurate SCC rates for some specimens, but grossly overestimated rates in others. The overprediction was associated with large unbroken ligaments that often form in welds in the wake of advancing crack fronts. When ligaments were particularly large, they prevented crack mouth deflection, so apparent crack incubation times (i.e. period of time before crack advance commences) based on CMOD measurements were unrealistically long. During the final states of testing, ligaments began to separate allowing the crack mouth to open rather quickly. This behavior was interpreted as a rapid crack advance, but it actually reflects the ligament separation rate, not the SCC rate. Revised crack growth rates obtained in this study exhibit substantially less scatter than that previously reported. The effects of crackmore » orientation and fatigue flutter loading on SCC rates in 82H welds are also discussed.« less

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call