Divergence of Mechanisms and the Effect on the Fatigue LIfe Variablity of Rene' 88 DT

Abstract

In safely and cost-effectively managing aging turbine engine components, it is important to understand the inherent variability in the fatigue behavior of superalloy systems. The variability in fatigue life was investigated in Rene` 88 DT, a superalloy used widely in a number of critical turbine engine components for which low cycle fatigue is typically the life-limiting factor. Loadcontrolled, axial fatigue testing was performed at 593 C and 10 Hz on specimens taken from a pancake forging. A stress-life (S-N) curve was established where numerous tests were performed at single stress levels in order to determine cumulative probability of failure. It is observed that the variability in cycles to failure increases with decreasing stress level. In addition, as the variability in life increases, a separation of individual populations can be identified within the full population of failed specimens, suggesting that different failure mechanisms control the fatigue performance of the separate populations. It is proposed that the application of separate probability functions based upon the divergent failure mechanisms provides more accurate predictions of the variability of fatigue performance and reduces unnecessary conservatism.