Crack initiation: the choice of tests available to calibrate Dang Van's criterion

Abstract

ABSTRACT Dang Van has proposed a method for predicting fatigue life under high cycle fatigue conditions, where the time for a crack to propagate to the critical length is negligible compared to the time it takes for it to initiate. His argument is that if shakedown is achieved at the grain level, a crack will never initiate. Since the exact stress state at the grain level is very difficult to quantify, the determination of the fatigue limit is achieved using calibration experiments. These experiments need not be restricted to bending and torsion as is common practice. Indeed, any form of specimen geometry and loading path may be used to estimate fatigue life. The comparison of many such tests can not only lead to a more accurate definition of the fatigue limit, but also indicates the suitability of the criterion. Moreover, by ensuring that calibration and specimen data lie close together on the fatigue locus, a better prediction of initiation conditions is achieved.