PSI - Issue 54

Behzad V. Farahani et al. / Procedia Structural Integrity 54 (2024) 638–644 Behzad V. Farahani et al./ Structural Integrity Procedia 00 (2023) 000–000

641

4

the specimen’s both sides till reaching the desired crack length. Nevertheless, Table 3 presents the crack length measured by traveling microscopes at the end of the fatigue test. Due to the purpose of the study, only one side crack length is reported. Regarding the optical experimental setup, it follows the procedure as extensively described in the previously published paper by the authors (Farahani et al., 2022). Therefore, the cracked specimen was statically loaded to reach the peak load and the crack length was measured by DIC and ESPI as reported in Table 3.

Table 3: Fatigue crack results.

Error (%) DIC&REF*

Error (%) ESPI&REF**

Error (%) DIC&ESPI***

a REF (mm) 18.17

a DIC (mm)

a ESPI (mm)

17.89

18.16

2.78

0.07

2.79

 100  a DIC  a REF ) / a REF   100  a ESPI  a REF ) / a REF   100  a DIC  a ESPI ) / a ESPI 

Notice that a REF is the crack length measured by the travelling microscope at the end of fatigue test. Regarding the stress intensity factor, the readers are referred to the published paper (Farahani et al., 2022). A good agreement was verified amongst the crack length measurements acquired by optical experimental studies compared to the reference solution. 3. Results and discussion Considering DIC analysis and Elber’s hypothesis (Elber, 1970, 1971), two virtual vertical clip gauges identified as VCG-1 and VCG-2 with a length of l = 6.4 (mm) were defined on the crack flanks locating at the initial notch region as shown in Figure 2-a). The force-displacement variation was obtained on the defined clip gauges for both loading and unloading schemes as shown in Figure 2-b). Elber further noted that a change in the slope of this curve is due to specimen compliance. Besides, it was inferred that this compliance variation must be due to crack opening and to no other factors such as plasticity. When the specimen is initially loaded, the gauge displacement remains nearly zero indicating the crack is closed at its location. Any gauge displacement in this low-load region is due to material strain between the gauge points, not crack opening. As the crack opens at the gauge location, the displacement initiates to increase at a significant rate, denoted as ‘‘local opening” in Figure 2-b. As the crack opens ahead of the gauge, a gradual compliance change occurs until the crack gets fully opened up to the crack tip (cf. ‘‘fully opened crack” in Figure 2-b). Since the crack is entirely open, loading the specimen further results in a linear relationship between load and gauge displacement with no further compliance change, at least before large-scale plasticity effects begin, (cf. “Widely opened Region” in Figure 2-b). Owing to the aforementioned details, the acquired DIC results demonstrate that the crack started to open at a 3 % of the peak load (300 N) and it reached its full length at 10 % of the peak load (1000 N) then ahead of (1000 N), the crack got stabilized being widely opened.