Issue 35

R.A. Cardoso et alii, Frattura ed Integrità Strutturale, 35 (2016) 405-413; DOI: 10.3221/IGF-ESIS.35.46 410 crack initiation path. However, one should notice that the criterion based on Δ σ n is inconsistent, since compressive stresses (usually present during a large portion of the fretting load cycle) do not contribute to crack propagation. Figure 7 : Critical plane model applied in the centre of the process zone for AISI 1034. Fig. 7 illustrates the results obtained by the critical plane model applied in the centre of the process zone ( L/2 ). Note that the maximum shear stress amplitude is reached in two orthogonal planes (-31° and 59°), but the critical plane is the one at -31° once, between these two planes, it experiences the maximum normal stress during a cycle. Therefore, it is clear that the crack initiation path estimated by this method provides a direction, which is opposite to the actual one (30°). Similar kind of inaccuracy to determine crack initiation path from critical plane approach was obtained by Susmel and Taylor [3] for notched components under mixed mode loads. (a) (b) Figure 8 : Results for 35NCD16 (a) the critical direction method; (b) the critical plane model applied in the centre of the process zone. Crack initiation direction for 35NCD16 Unfortunately, the threshold value for the stress intensity factor range, Δ K th , is not available for steel 35NCD16, hence, an estimate based on the work of [17] was done, where the material characteristic length for a steel is given by: 17.1 1 36200            u L (5)