PSI - Issue 2_B

Richard H. A. et al. / Procedia Structural Integrity 2 (2016) 1821–1828 Richard H. A. and Eberlein A. / Structural Integrity Procedia 00 (2016) 000 – 000

1825

5

The mixed-mode threshold values determined by the procedure described above are illustrated in Figure 3. The threshold values of Al7075-T651 measured by Schirmeisen (2012) in comparison to the threshold value surface of the criterion by R ICHARD show a good congruence for pure mode I-loading, 2D-mixed-mode-loading and for mixed mode II + III-loading with small mode III-ratio. A significant variation, with around a factor of 2.2 above the hypothesis (Richard et al. (2012)), depict the threshold values for pure mode III-loading. The threshold values measured within the experiments by Eberlein (2016) in total are closer to the threshold value surface of the hypothesis. The variation in average is around factor 1.8 above the hypothesis. In addition, a comparison of other materials investigated by Vojtek and Pokluda (2012) as well as by Vojtek et al. (2013) show partially similar threshold value ratios Δ K II,th / Δ K I,th and Δ K III,th / Δ K I,th (see threshold values for austenitic steel in Fig. 3). Ferritic steel exhibits completely other threshold value ratios. Nevertheless, the determined mixed-mode threshold values for different materials in comparison with the criterion by R ICHARD point out that this criterion possesses a widely validity and a generally conservative behaviour. In the literature some experimental determined threshold values for the loading combination of mode I and mode III are given by e.g. (Pook (1985), Pook and Crawford (1991), Yates and Miller (1989), Yoshioka et al. (1984)). The comparison of the characteristic values with values determined by the CTSR-specimen, shown in Figure 4, reveals that the spread of the values given in literature is considerable, which is connected to various applied specimen types. Threshold values determined by the CTSR-specimen and the corresponding loading device are within the spread of the threshold values in Figure 4 and thereby demonstrate its solid performance. 4.2. Threshold values for different materials and specimen types

Fig. 4. Mode I-, mode III- and mixed-mode I + III-threshold values for different materials and specimen types in comparison

4.3. Criteria’s validity by measured crack kinking and crack twisting angles

Based on the measured crack kinking angle φ 0 and crack twisting angle ψ 0 the criteria’s validity predicting the crack growth orientation is examined. Concerning this, Figure 5 shows the measured crack deflection angles for each criterion mentioned herein. The blue dots are measured by Schirmeisen (2012) and are also published by Eberlein (2016). The green dots are values measured within this investigation. The comparison of the crack kinking angle φ 0