PSI - Issue 5

A. Eremin et al. / Procedia Structural Integrity 5 (2017) 889–895 A. Eremin et al / Structural Integrity Procedia 00 (2017) 000 – 000

893

5

(a)

(b)

Fig. 3. Crack-tip hysteresis loops for test 1 with single underload.

Another parameter which might be extracted from the “ load-strain ” curves is the loop area. However the level of plastic deformation at the load employed in this experiment is small. This is because area within the loop is also relatively small (~10 -5 kN). However, the calculation error can exceed the latter measurements by an order of magnitude. In doing so, the loop area as an informative parameter might be taken into account but only when higher loads or stress intensity factor are characteristic features of the loading. This will ensure larger material region being involved into plastic deformation.

(a)

(b)

Fig. 4. Maximum of COD (ε max ) after various numbers of cycles after the single underload for measurements (a) at the crack tip, (b) at 630 μm behind the tip.

3.2. Single Overload The four loops of the COD measurements at the crack tip are presented in Fig. 5a: (1), (2) – the two reference cycles; measurements immediately after the overload (3) and after applying 100 “ basic ” cycles (4). The results obtained at subsequent time steps, i.e. after 200, 500 etc. up to 2000 basic cycles after the overload as compared with the reference one, are shown in Fig. 5b. Visual comparison of the loops allows one to conclude that the two reference cycles mostly differ by the COD values. However, their values are much smaller in contrast with ones being measured after the overload. At 200 cycles after the overload and during next 2000 cycles the hysteresis loops look almost similar, except some oscillations in their width. The reason of the latter might be related to errors occurring due to the DIC measurement (because of poor correlation between the images being slightly blurred and badly focused). The COD loops measured at the distance behind the tip are not shown in the paper, since they look very similar and one can hardly distinguish one from another. However, the results of ε max measurement are presented in Fig. 6.