On the Use of a Stress-Independent Threshold Stress Term in Creep Life Models

The concept of effective stress, where the applied stress is reduced by a resisting or threshold stress, is often used in fatigue and creep strain rate models, but seldom in single-line equation creep life models. Recently, effective stress was used with single-line models to correlate creep life data of two low chromium creep-resistant steels (1Cr-1Mo-0.25V and 2.25Cr-1Mo), which have been used extensively in Rankine cycle power plants. In those models, the threshold stress was independent of applied stress, which causes a prediction of infinite time to rupture when applied stress equals the threshold stress. In this study, the overprediction of rupture time versus the experimental data of a power law model using a stress-independent threshold stress is quantified, and it is shown that other single line creep life models do not need a backstress term to provide equivalent or superior correlations of the same data sets. For the 1Cr-1Mo-0.25V steel, the coefficient of determinations of Larsen-Miller parameter and modified hyperbolic sine models were 13.8% and 6.9% better than the power law model with a backstress term, while the coefficient of determinations was essentially the same for the 2.25Cr-1Mo steel.