Load Interaction Effects on Fatigue Crack Growth
Abstract
The fatigue crack propagation rate can be either increased or decreased by the previous load history (overload, block loading, different load ratio, etc). Currently, these load sequence effects can be explained either by using crack closure or internal stress concepts. They are studied in Part I and II of the dissertation accordingly.
In Part I the last 35 years of research in the crack closure area were carefully reviewed. A new Quadrature (Q) method for crack closure estimation, based on integration rather than differentiation of the load-displacement data, was developed and compared to the ‘best’ methods from the literature. The new method was able to reduce the scatter in the opening load estimations to a negligible level, but does not collapse the results for different load ratios (0.1 and 0.9). In Part II a general relationship between fatigue crack growth rate (da/dN) and the two-parameter (ΔKtip and tipKmax) crack driving force was derived using fundamental fatigue (ε-N curve) properties. Based on this analysis, a new way of representing the da/dN data by means of the crack propagation (CP) table was proposed. In order to make the CP table sensitive to the load history effects, it was scaled using the applied and internal stresses and the corresponding stress intensity factors, characteristic for the crack tip. Two methods for calculating the internal stress intensity factors were developed, adopting the weight function and the new clamping force concepts accordingly.
Finally, the CP table at the crack tip was successfully used together with the two-parameter crack driving force equation to predict da/dN for different load ratios, block loading and a single overload. Calculation of the crack closure was not needed in order to predict the experimental data accurately.