Dependency of Fatigue Predictions on the Form of the Crack Velocity Equation

A computer search technique was developed to analyze fatigue strength data using both exponential and power law forms of the subcritical crack velocity equation. All crack velocity equations would fit a given set of fatigue data equally well in the data range but failure predictions based on the different crack velocity equations diverge from each other outside the data range. The exponential form of the crack velocity equation best fit both the static and dynamic fatigue data of hot‐pressed Si 3 N 4 and optical glass fibers, whereas the power law form best fit the static and dynamic fatigue data of soda‐lime glass and A1 2 O 3 . To determine the most appropriate crack velocity equation for a given material/environment system, it is recommended that fatigue data be obtained under different loading conditions and the data numerically regressed using the computer search technique with each of the possible crack velocity equations to find which best fits the data.