Abstract
An experimental database containing multiaxial fatigue results for C35 steel with induced defects has been previously published [Billaudeau, T., Nadot, Y., and Bezine, G., “Multiaxial Fatigue Limit for Defective Materials: Mechanisms and Experiments,” Acta Mater., Vol. 52, 2004, pp. 3911–3920; Nadot, Y., and Billaudeau, T., “Multiaxial Fatigue Limit Criterion for Defective Materials,” Eng. Fract. Mech., Vol. 73, 2006, pp. 112–133; Gadouini, H., Nadot, Y., and Rebours, C., “Influence of Mean Stress on the Multiaxial Fatigue Behaviour of Defective Materials,” Int. J. Fatigue, Vol. 30, 2008, pp. 1623–1633]. Artificial defects, in a range of 100–900 μm, are introduced at the surface of fatigue samples. The database contains experimental fatigue results for different defect size and geometry, and multiaxial loading cases with different load ratios. In the present study, five models predicting the fatigue limit of materials with defects are compared. Each model is carefully presented, predictions are compared, and input data are discussed. It is shown that it is necessary to take account for the stress distribution around the defect in order to have a good prediction for a multiaxial case.