Fretting fatigue under variable amplitude loading

Generally, the fretting fatigue problem is treated with constant amplitude loading. However, this condition is not the reality for industrial applications. Although there are few studies on fretting fatigue under variable amplitude loadings, they use a simplistic approach to life estimation (unable to capture the effects of the complex multiaxial and non-proportional stress field that exists under the contact interface) and only the bulk fatigue load has its amplitude alternating. The objective of this work is to propose a numerical methodology for the life assessment of components subjected to fretting fatigue under variable amplitude loading and to study the effects of these loads. For this, experimental tests were carried out by applying H-L (High-Low) and L-H (Low-High) loading blocks to the tangential load, that is, the contact normal load and the amplitude of the bulk fatigue load were kept constant while the tangential load has its alternating amplitude. The numerical methodology proposed for fretting fatigue life assessment includes the wear effect and crack propagation. The multiaxial fatigue parameter SWT, the theory of critical distance TCD, and Miners' linear rule were used to compute the damage in the crack nucleation phase. The node-displacement algorithm based on Archard’s law was applied to account wear. In the crack propagation phase, the CDM (critical direction method) was used to estimate the crack initiation angle considering the worn contact surface, and then the crack is modeled and propagated. The still limited range of experimental data done in this work for fretting fatigue under varying loading blocks seems to indicate that the loading sequence has an effect on the accumulated fatigue damage, mainly to the H-L loading sequences. Based on this observation, a new non-linear damage model was proposed. Using this new damage model, the life estimation results were within a scatter band of 1.2.