On thermo-mechanical fatigue in single crystal Ni-base superalloys

Thermo-mechanical fatigue (TMF) is a critical damage process in gas turbine jet engines. Reliable life prediction methodologies are required both for design and life management. Current life estimation approaches are computationally burdensome and/or semi-empirical, significantly limiting their application. Complexity comes about from the multiplicity of damage processes which occur during the simultaneously changing temperatures and loads, characteristic of TMF. Furthermore, these processes interact in ways that are not observed for isothermal LCF. These interactions usually accelerate damage processes and result in significantly reduced TMF life, when compared to other fatigue scenarios. The results of a multiphased approach to life prediction will be presented. In phase I the Neu-Sehitoglu (N-S) cumulative damage model was used to: a) provide initial life predictions and b) identify processes and interactions which most significantly control the life under TMF loading. The N-S model is based on a linear damage summation rule which both explicitly and implicitly includes damage interactions. In phase II a sensitivity analysis incorporating statistical concepts was performed on the N-S model parameters. Specifically a nonlinear optimization was performed to determine optimal parameter values in order to maximize agreement with experimental results (well within 2X). In phase III, informed by phases I and II, a physics-based fatigue/oxidation (also recognizing creep effects) model was developed which correctly predicts effects of frequency, hold-time, temperature, and strain range and oxidation kinetics.