Classically determined effective Delta K fails to quantify crack growth rates

Abstract

Decomposition of the resistance to fatigue crack growth into the intrinsic and extrinsic component is very important for understanding of fatigue failure mechanisms, relation to microstructure and modelling of residual fatigue life. Crack closure for four grades of steel were estimated by the difference between K-max values and the effective Delta K-eff values (measured at the load ratio R = 0.8) corresponding to the same crack growth rate. The results showed that crack closure values obtained by the difference K-max - Delta K-eff were not in agreement with the available crack closure models, both the Newman's model of plasticity-induced closure and the results from finite element analysis. The discrepancies could not be explained by the effect of mean stress, specimen thickness, loading amplitude or T-stress. Therefore, the application of fracture mechanics to fatigue cracks should be revisited. It was pointed out that Delta K-eff may not be a good parameter for quantification of the crack driving force, since the relationship between K-max - K-cl and the cyclic plastic deformation at the crack tip might not be linear.Decomposition of the resistance to fatigue crack growth into the intrinsic and extrinsic component is very important for understanding of fatigue failure mechanisms, relation to microstructure and modelling of residual fatigue life. Crack closure for four grades of steel were estimated by the difference between K-max values and the effective Delta K-eff values (measured at the load ratio R = 0.8) corresponding to the same crack growth rate. The results showed that crack closure values obtained by the difference K-max - Delta K-eff were not in agreement with the available crack closure models, both the Newman's model of plasticity-induced closure and the results from finite element analysis. The discrepancies could not be explained by the effect of mean stress, specimen thickness, loading amplitude or T-stress. Therefore, the application of fracture mechanics to fatigue cracks should be revisited. It was pointed out that Delta K-eff may not be a good parameter for quantification of the crack driving force, since the relationship between K-max - K-cl and the cyclic plastic deformation at the crack tip might not be linear.