A B S T R A C T Many multiaxial fatigue parameters are available to estimate the onset of the crack initiation in mechanical components subjected to fretting fatigue conditions. This study focuses on the ability of the cracking energy density parameter to predict the fretting fatigue life. To estimate this, obtained predictions are compared with those given by another commonly used parameter, namely the Smith-Watson-Topper parameter. First, fretting fatigue experiments were achieved using a new designed setup with a mono-contact configuration (aluminum/steel). A finite element model was used to compute the stress/strain fields around the contact zone. Because of the high stress gradient in the contact zone, the estimates of the two fatigue parameters, when using the mechanical quantities on the most loaded material point, lead to too conservative predicted values. It was therefore necessary to define a kind of process zone in which crack nucleation takes place. Using an averaging method to compute stress and strain in this zone, the obtained results show a quite good agreement between the estimated fatigue life and experimental data, whatever the used fatigue indicator. The size of this zone was then compared with the material microstructure and seems to be in the same magnitude that the average grain size.Keywords fretting fatigue; life prediction; cracking energy density; volumetric approach.
N O M E N C L A T U R Ea = contact width A = composite compliance E = Young modulus f = yield function F 1 , F 2 = load cells J 2 = second invariant of deviatoric stress tensor k = yield stress K = relative curvature p = cumulative plastic strain rate p(x) = Hertz pressure p 0 = peak contact pressure P l = applied load per unit length Q = tangential force R = load ratio S ij = deviatoric stress tensor X ij = backstress tensor X ef = effective distance H, γ = hardening parameters of Armstrong-Frederick constitutive law σ f , ε f , c, b = material constants ε max = maximum normal strain Correspondence: A. Amrouche.