PSI - Issue 60

210 S.K. Chandra et al. / Procedia Structural Integrity 60 (2024) 203–213 Author name / StructuralIntegrity Procedia 00 (2019) 000 – 000 specimens of the same test material, the corresponding range was 17.5 o −17.72 o . Thus, for the present test material, determined by the optical method is found to be specimen geometry independent, and practically constant under quasi-static ramp rates. 8

Fig. 8. φ−∆a plots for pre-cracked (a) SENT and (b) DENT specimens at different ramp rates 3.3. 5 −∆ plots: corresponding CTOA estimates The 5 −∆ curves obtained from pre-cracked SENT specimens with varying ligament lengths ranging from 12.8 mm to 24.24 mm, shown in Fig. 9(a), de monstrates “peeling - off” of the curves for smaller values from those for the larger values beyond some extent of crack extensions (Heerens and Schoedel, 2009). From the data presented in this figure, ( 5 ) values were determined, Fig. 9(b). The average ( 5 ) over a reasonable extended range of ∆ for which the variation of ( 5 ) is small (designated by ( 5 ) ), could be considered as an approximate measure of . The ∆ at which the curvature in a 5 −∆ curve changes is identified as the maximum crack extension, ∆ for determining ( 5 ) . To improve statistical reliability of the ( 5 ) values computed, instead of Eq. (2), ( 5 ) for a chosen ∆ value was determined by linear least square fit of the set of three 5 ,∆ data: the object point, and the two adjacent points with lower and higher ∆ values. This method, recommended in (Schwalbe et al., 2005; Heerens and Schoedel, 2009) requires 5 ,∆ data with small ∆ intervals. The ( 5 ) values thus obtained attain a steady value, ( 5 ) starting from a higher ( 5 ) value after the initial transient, Fig. 9(b).

Fig. 9. (a) δ 5 −∆a and (b) φ(δ 5 )−∆a plots for pre-cracked SENT specimens with different ligament lengths.