Issue 31

R. Citarella et alii, Frattura ed Integrità Strutturale, 31 (2015) 138-147; DOI: 10.3221/IGF-ESIS.31.11 144 Fig. 9a-b represent the superficial crack growth rate da/dN versus crack length a and versus COD, under pure cyclic tension and combined tension+torsion. It can be seen that there is not a significant reduction of the crack growth rates along the external surface direction when the cyclic Mode III loading is superimposed on the cyclic tension. However, looking at Fig. 7b and considering changes in the general durability of the specimens in pure tension and combined loading (Fig. 8a-b), significant differences in the crack growth rate in the depth direction b under the above types of loading conditions are expected. a) b) Figure 9 : Crack growth rate as a function of superficial crack length (a) and COD (b) . FEM MODEL he crack propagation in the cylindrical specimens N.1 and 2 (Fig. 10a-b), undergoing combined traction-torsion fatigue loads, is simulated by the FEM model shown in Fig. 11. Such model has a length equal to 60 mm, with one end clamped and the other end constrained along the in plane radial directions and loaded along the axial and tangential directions. The mesh is made with 18960 quadratic elements (brick with 20 nodes “C3D20”): such number of elements is nearly constant during the propagation. The whole propagations take nearly one hour calculus on a powerful PC. The initial crack, as indicated by experimental measurements (beach mark technique) has the following sizes: a=0.855 mm, b=0.852 mm, c=3.705 mm. During the propagation, the average crack advance at each step is equal to 0.2 mm. As previously said, the SIF’s along the crack front are calculated by the J-Integral approach. The crack growth rate is calculated by the Paris formula (Eq. 4), whose calibration parameters are shown in Tab. 1. The crack path is calculated using the G-criterion. a) b) Figure 10 : Geometry of specimens N. 1 (a) and N. 2 (b) . T