8 Detecting Fatigue Crack Closure and Crack Growth Delays After an Overload Using DIC Measurements 61 Fig. 8.6 COD measurements from DIC analysis at different crack lengths (the curves of right side were translated in y-direction for better visibility) Fig. 8.7 Strain measurements from strain gage placed at back face specimen at different crack lengths (the curves were translated in y-direction for better visibility) explore the post-OL FCG data after it reached its maximum delay, COD measurements at different positions along the crack flanks in points A, C, D, E and F and the corresponding full-field strain map were studied. Figure 8.8 corresponds to the point before overloading, and shows that the crack peels open from the mouth to the tip as the applied load increases. Figure 8.9 illustrates the change in COD values after the OL. Due to crack tip blunting, the crack remains fully open and there is no evidence of crack closure. Figures 8.8b and 8.9b show the corresponding strain tip fields. The data points around the crack faces and tip are excluded from the analysis to avoid their intrinsically higher noise level. Figure 8.10 correspond to point C in Fig. 8.3, where the FCG rate reached its maximum delay. Note that the crack flanks near the tip remain completely closed during a significant portion of the load cycle. According to Elber, this behavior is explained by the larger reaction of the elastic residual ligament over pzOL, as shown in the crack tip strain field of the Fig. 8.10b. Figure 8.11 corresponds to point D, about 7 mm from the OL application point, just outside the theoretical pzOL, where the FCG rate da/dNshows some recovery in Fig. 8.3. The COD values at points near the crack tip show a nonlinear behavior along the crack flanks. It can be seen that up to a load0.6 Pmax the portion of the crack surface near the tip (at 1, 2, and 3 mm from the crack tip) is partially open whereas the portion near the OL location (at 4 and 5 mm from the tip) remains closed. Closure near the OL location is also observed in Figs. 8.12 and 8.13, which correspond to points E and F in Fig. 8.3, where the crack tip has propagated up to about 9 and 11.5 mm from the OL application point, respectively. COD values decrease near the OL location at 5, 6 and 7 mm from the tip in Fig. 8.12a and, at 8 and 9 mm from the tip in Fig. 8.13a. It is worth mentioning that, Figs. 8.8, 8.9, 8.10 and 8.11 used the same stereo camera configuration, whereas in Figs. 8.12 and
RkJQdWJsaXNoZXIy MTMzNzEzMQ==