PSI - Issue 39

R. Yarullin et al. / Procedia Structural Integrity 39 (2022) 364–378 Author name / S ructural Integrity Procedia 00 (2021) 00 –000

376

13

Fig. 15. Crack length b vs cycle curves under (a, b) tension, (c, d) tension/torsion and (e, f) torsion loading conditions.

4.4. Crack growth rate interpretation Crack growth rate diagrams as a function of elastic equivalent SIFs for both alloys and different loading conditions are shown on Fig. 16. As shown in Fig. 16b, a significant impact of varying loading conditions on the fatigue crack growth rate obtained for the break through points (db/dN), is observed for B95AT aluminium alloy. Superimposition of cyclic tension and cyclic torsion leads to lower crack growth rates in comparison to pure tension, while the lowest crack growth rates are obtained for pure torsion loading conditions. On the contrary, the crack growth diagrams for D16T aluminium alloy (Fig. 16a) are barely affected by varying loading conditions. Still in Fig. 16, a comparison is provided between the fatigue crack growth rates determined under complex stress state carried out in the present study with the fatigue crack growth rates determined according to ASTM E647. Indeed, different loading conditions leads to different crack growth curves that are substantially different from fatigue crack growth determined on the standard specimens. It is known that in a structure with surface defects, the crack front can be analyzed considering two specific points: one relates to the intersection of the crack front line with a free surface, and the other is the deepest point of the crack front. The crack growth rate as a function of equivalent SIFs for different crack front points and different loading conditions are presented on Fig. 17. A significant reduction of the crack growth rate is observed in the direction of the