PSI - Issue 60

S.K. Chandra et al. / Procedia Structural Integrity 60 (2024) 203–213 Author name / Structural Integrity Procedia 00 (2019) 000 – 000 7 For each choice of , the plot followed the familiar trend: that with increasing crack length CTOA decreases from higher initial values associated with initial crack tip blunting, then as the crack growth stabilizes CTOA reaches a fairly constant value, and with farther crack growth CTOA decreases. The decrease in optically measured CTOA beyond ∆ is because of the “crimping effect” on the crack tip by the developing plastic zone at the specimen boundary on the crack path in SENT specimens or superposition of the plastic zone of the two advancing crack tip in case of DENT specimens. From the trend of all the data presented, the range of ∆ over which the is considered to be constant were identified. Then, for each chosen value, the mean and standard deviation (SD) of the values measured in this stable range were computed. The SD values are also shown in Fig. 5 (a) and 5 (b). Finally, the data for = 1.5 mm were selected because, this choice resulted in least data scatter. 3.2. Effect of notch root radius ( ) and ligament lengths ( ) on − ∆ plots In order to examine the effect of notch tip radius on the optical − ∆ curve, one SENT specimen having notch tip radius, =0.1 mm was tested at a ramp rate 0.006 mm.s -1 and was measured using the optical two-point method taking =1.5 mm. The − ∆ plot of the notched specimen thus obtained was compared with that of a pre-cracked specimen, Fig. 6. The values determined were very similar. The difference was in the transition regimes. For the pre-cracked specimen, the initial transition persisted up to ∆ ≅6 mm, i.e. 6 . For the notched specimen on the other hand, the initial transient regime was much larger, ∆ ≅ 9 . This result shows that with judiciously chosen to ensure a sufficient ∆ range for the critical regime, a notched specimen with a sufficiently small notch root radius can very well be used to determine , thereby significantly reducing the test burden. This conclusion, reached by testing SENT specimens, should in principle extend to DENT specimens also. The − ∆ plots for pre-cracked SENT specimens with varying ligament lengths, , ranging from 12.8 mm to 24.24 mm, did not show any significant differences, and yielded very similar values, Fig. 7.

209

Fig. 6. φ−∆a curves of pre-cracked and notched SENT geometry. Fig. 7. φ−∆a plots of pre-cracked SENT specimens with different l . 3.3. Effect of strain rate and specimen geometry on - ∆ plots The - ∆ curves of SENT and DENT pre-cracked specimens (determined with =1.5 mm) for the three different ramp rates are plotted in Figs. 8 (a) and 8 (b). The plots follow the usual trend: with increasing ∆ , gradually decreased from high initial values, then stabilized to a fairly constant value for a sizable extent of crack growth, and eventually started to decrease due to proximity of the free edge. It is worth noting that (i) for all the specimens, is attained after a transition regime of crack growth ∆ ~ 6 mm = 6 ; and (ii) for the quasi-static strain rates considered, is found in the range of 17.77 o −17.96 o for SENT specimens. For the pre-cracked DENT