PSI - Issue 45

James Martin Hughes et al. / Procedia Structural Integrity 45 (2023) 44–51 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

49

6

Measurements of crack closure from crack initiation to failure are only practically possible using the piezoelectric strain gauge due to its increased sensitivity and resolution compared to traditional sensors. In addition, the piezoelectric device can easily be attached to a structure in-situ, enabling field measurements at real-life fatigue frequencies and loading conditions to be analysed. The utilisation of the piezoelectric gauge therefore opens many possibilities for further investigation of crack closure in variable amplitude and random load spectrums, as well as techniques aimed to truncate or compress loading sequences more effectively. 3.3. Comparison with Published Results A number of previous studies have investigated the crack closure phenomenon under constant amplitude loading (Fleck et al., 1983; de Matos and Nowell, 2009; Ashbaugh et al., 1997; Sehitoglu, 1985). Fleck (1983) measured closure loads using a crack mouth opening displacement (CMOD) gauge and a push-rod system. A thumbnail shaped crack was subjected to R = 0.05 cyclic loading. At a/W = 0.4, the plane strain opening load ratio was U strain = 0.85 and the plane stress opening ratio U stress = 0.75. From Fig. 3, the results of this study suggest that the opening load ratio for the same crack length ratio and R = 0 is approximately U 0 = 0.73, which is similar to Fleck (1983). Although the geometry of the crack profile and load ratio is slightly different, the similarity of the opening load ratios provides good confidence in the opening load measurements. Ashbaugh (1997) used finite element (FE) simulations and experiments to investigate crack opening and closure loads on a 12 mm thick (plane strain) CT specimen. Two material hardening models (a linear law and a power law) were incorporated into the FE models, and the results extracted at several stress ratios. The experimental results indicate that for R = 0.1, the normalised load ratio ( P op / P max ) is 0.3 for a/W = 0.25, which matches reasonably well with the results of this investigation. In Ashbaugh (1997) at R = 0.3 and a/W= 0.35, the normalised load ratio was 0.4, which is higher than found in this study (approx. 0.32). Nonetheless, the results gathered in this study show good agreement with the FE results using the linear hardening law. de Matos and Nowell (2009) used a back face strain gauge and CMOD gauge to measure the opening loads of a CT specimen under R = 0.1 constant amplitude loading over a variety of crack lengths. Three different thickness CT specimens (3 mm, 10 mm, and 25 mm) were manufactured from AA6082-T6. Their results showed that, when closure is corrected for, the fatigue crack growth rates of the three different specimens collapse onto a single line. Data presented by de Matos and Nowell (2009) for the 3 mm and 25 mm specimens are presented in Fig. 6, alongside the data gathered in this study for comparison. A good correlation is shown between the de Matos and Nowell (2009) results and the current results. Moreover, the data presented in this paper is much smoother and there are no abrupt jumps, which demonstrates the advantages and advanced capabilities of the piezoelectric strain gauge. In addition, the R = 0.1 results in this study are a subset of the data from a variable amplitude load sequence, which has greater practical applications. Overall, there is good agreement between the two studies.

0.5

3 mm BFSG - de Matos and Nowell (2009) 25 mm BFSG - de Matos and Nowell (2009)

3 mm CMOD - de Matos and Nowell (2009) 25 mm CMOD - de Matos and Nowell (2009)

Current Study

0.4

0.3

P op / P max

0.2

0.1

0

0.2

0.3

0.4

0.5

0.6

Crack Length Ratio, a/W

Fig. 6. Normalised opening load versus crack length ratio for R = 0.1, and comparison with back face strain gauge (BFSG) and crack mouth opening displacement (CMOD) data gathered by de Matos and Nowell (2009) for CT specimens of thicknesses 3 mm and 25 mm.