Issue 35

E. Giner et alii, Frattura ed Integrità Strutturale, 35 (2016) 285-294; DOI: 10.3221/IGF-ESIS.35.33 289 The specimen material is an aluminium alloy 7075-T6, with E = 72GPa,  = 0.3. For some of the cases analyzed, the indenter material is changed according to Tab. 2 included in the results section. The application of the linear regime is deemed valid, due to the very small edge radius of the indenter and the relative high yield stress of the aluminium alloy. As a consequence, the existing plasticity is very localized and a small scale yielding assumption can be applied, analogous to the small scale yielding assumption admitted in LEFM around the crack tip. This is confirmed by the observation of the tested specimens, which showed no macroscopic evidence of plasticity (see micrographs in Fig. 1 in which crack faces match very well each other and a view of the specimen contact surface in Fig. 7 of our previous work [7]). The loading sequence is represented in Fig. 6 for one of the examples analyzed (case 3, see results section), where four load steps have been considered in the analysis. Due to the non-linearity of the contact problem, loads were applied in sufficiently small time increments. At time t = 2.00 the maximum  Bulk is being applied, which produces a clear opening of the crack. When the bulk load is decreased in the first half of step 3 (2.00 < t < 2.50), mode I is reduced and a clear mixed mode condition appears, which has been observed through FE analyses. Note that the vertical load due to the indenter is kept constant during the cycle and mode II increases its dominance over mode I as  Bulk is reduced. At time t = 2.50 crack face contact is produced and a mode II condition is present at the crack tip. At time t = 3.00 the bulk load is completely reversed (since the stress ratio in case 3 is R =-1) and the load is transmitted through the crack faces. The end of the contact zone acts now as a strong stress raiser, as the specimen is compressed against the contact corner. Results in the following section are presented for the load step 4 (3.00 < t < 4.00), when ‘‘shakedown’’ of the numerical model response is produced). It has been verified that the stress states at t = 3.50 and t = 4.00 are very similar to those at t = 2.50 and t = 2.00, respectively. Figure 5 : Left, model geometry. Right, complete contact testing rig, showing the contact elements. Figure 6 : Loads applied to the numerical model for one of the cases studied (case 3). Evolution with time.