PSI - Issue 23

Ladislav Poczklán et al. / Procedia Structural Integrity 23 (2019) 269–274 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

273

5

The crack growth kinetics was measured on specimens with artificial crack starter, the hole. It means that neither the crack initiation mechanism nor the fatigue life corresponds to that of smooth specimens. The micrographs of growing fatigue cracks taken by light microscope in all three cycling modes are shown in Fig. 5. It can be seen that in axial mode, two fatigue cracks grow from the hole in direction approximately perpendicular to the specimen axis. In case of multiaxial loading, two fatigue crack initiated, too. Their crack path is inclined to the specimen axis. The angle of inclination was approximately 6 5°. In torsional mode of cycling, there are four cracks with the angle of inclination to the specimen axis around 45°.

The results of the measurement of the crack length during the cycling in all three modes with ε a,eq = 0.44 % are shown in Fig. 6 a). It can be seen that the fastest crack growth is in axial mode and the slowest in the torsional mode. The curve representing multiaxial in phase loading lies between the other two curves. The data were fitted by exponential function – see Eq. 2. = ∙ ∙ (2) A linear dependence between crack length and crack growth rate (Eq. 3) can be derived from Eq. 2. = ∙ ∙ ∙ = ∙ (3) The crack growth coefficient k g corresponds to the relative increment of the crack length for the crack of the length a in one cycle. The coefficient a i represents the fictive crack length whose growth leads to the fatigue life N F (Pol ák and Zezulka, 2005). The values of parameters k g and a i are listed in Tab. 2. It can be seen that the value of k g in pure torsional mode is approximately twelve times lower compared to axial mode. Table 2 The values of parameters a i and k g for three different loading modes with ε a,eq = 0.44 %. a i [μm] k g [-] Axial 18.1 6.72 × 10 -3 . Multiaxial 15.5 3.82 × 10 -3 Torsional 28.0 5.66 × 10 -4 Crack growth rates of all three cycling modes are shown in Fig. 6b). It is obvious that fatigue crack grows significantly slower in torsional cycling mode as compared to the other two modes. On the other hand the highest crack growth rate is detected in case of axial mode. Such a behavior could be caused by phase transf ormation γ → α´. The content of α´ martensite is substantially higher in the specimen cycled in pure torsion and its presence leads to cyclic hardening of 316L steel (Mazánová et al., 2017) . In axial mode, the crack path is perpendicular to the external force, which means that the crack is opened every cycle in pure K I mode. This is not the case of torsional or multiaxial loading where the situation at the crack tip is more complex. Another possible reason may be found in the total crack length – cyclic loading in axial and multiaxial modes led to growth of two cracks, however cycling in pure torsion led to formation of four cracks. It means there is twice larger area of newly created fracture surfaces resulting in need of higher driving force for the crack growth. Fig. 5 Light microscope micrograph of growing crack in three different types of loading a) axial b) multiaxial c) torsional.