Statistical size effect on multiaxial fatigue strength of notched steel components

Fatigue strength assessment of large-scale structures and components is usually performed on the basis of experimental results from small-scale specimens. A proper transfer from small specimens to large-scale structures requires the consideration of size and stress-gradient induced micro supports effects, which depend on material to loading type. This paper investigates statistical size effects on the fatigue strength of notched electroslag remelted 50CrMo4 steel components under uniaxial and multiaxial loadings. Firstly, stress-based micro support concepts utilizing uniaxial small-scale tests are applied to enable an accurate fatigue design of notched steel parts. Then, the notched large-scale specimens, as representatives for an engineering component, are tested under uniaxial transverse and rotating bending as well as multiaxial loading for validation purpose. Secondly, stress-based methods with highly-stressed surface area and volume are applied in order to consider the fatigue support and statistical size effects. The estimated fatigue strength at the high-cycle regime shows a sound conformity to the large-scale experiments with a maximum deviation of 7.4% in case of the surface, and of 4.3% by the volume approach. The results of the present investigation also reveal that a certain upper threshold value is required for the highly-stressed volume concept in advance in order to ensure a proper fatigue strength assessment. The threshold value is not required for the surface approach and thus, it is found to be more feasible to an engineering purpose, especially in case of surface-initiated fatigue failure modes.