PSI - Issue 61

Yogesh Kumar et al. / Procedia Structural Integrity 61 (2024) 322–330 Y. Kumar et al., / Structural Integrity Procedia 00 (2019) 000 – 000

326

5

The [*MAT_55] material card is a progressive damage model with Chang-Chang failure criteria (Chang and Chang, 1987) for the compression and tensile fiber failure modes, and Tsai-Wu failure criteria (Tsai and Wu, 1971) for tensile and compressive matrix failure modes. The material model incorporates a reduction in the strength parameters once the failure initiates and the degradation are controlled by softening parameters. SOFT, FBRT and YCFAC are the incorporated non-physical parameters in material constitutive equations, and these require calibration for capturing the realistic response of the material. The element deletion is controlled by the threshold failure strain in each failure modes, defining in LS-Dyna environment as DFAILM, DFAILS, DFAILT and DFAILC. The initial magnitudes of these values can be obtained from the ratio of the strength to the modulus of the material in the associated direction. A detailed description on the estimation and robust calibration of these non-physical parameters can be found in a work by the author in (Kumar, Rezasefat and Hogan, 2023). The parameters involved in the [*MAT_55] are summarized in the Table 3. The details of the non-physical parameters have been provided in (Kumar et al., 2024). 4. Results and Discussion 4.1. Experimental results: Global mechanical response and Damage morphology Each laminate design was tested three times at two loading rates to assure the repeatability of the quasi-static compression experiments. Fig. 2 shows the correlation between the compressive strength and maximum failure strain versus the number of plies ( n = 2, 4, and 8) for different laminate design. The quasi-static tests results are reported in Fig. 2a and b at two different loading rates of 4 × 10 -3 and 4 × 10 -5 s -1 , respectively. In both figures, a significant increase in the peak strength with the increase in the thickness of the 90° plies in the sample was observed. For instance, in Fig. 2a, the global in-plane compression strength of the composite increased by 75.4 %, by increasing n from 2 to 6. The effect of the laminate design was more significant for the compressive strength compared to the maximum failure strain. By comparison with the results from two different quasi-static loading rates, a minimal rate sensitivity was observed for the compressive strength which was consistent with previous observations on the studied laminated composites under quasi-static loading regime (Perry and Walley, 2022; Raimondo et al., 2012). Meanwhile, the effect of quasi-static loading rate was more dominant on the failure strain with the change of 13.8 %, 8.85 %, and 7.17 % in 4, 6, and 8 ply samples, as the loading rate increased.

Fig. 2. Quantification of the global in-plane compression strength and failure strain in 4 ply, 6 ply, and 8 ply samples under quasi-static compression loading rates: (a) Strain rate = 4 × 10 -3 s -1 and (b) Strain rate = 4 × 10 -5 s -1 .

The damage morphology of the samples tested under the strain rate 4 × 10 -3 s -1 are illustrated in Fig. 3. The area of the samples with 0° and 90° fiber orientations are masked with blue and orange, respectively for better visualization, as shown in Fig. 3. Different failure modes including interlaminar, intralaminar and fiber kinking have been reported in all the three samples. Consistently, the damage initiated from the interface between the 0° and 90° plies (weakest region with shear dominating stress generation), leading to interlaminar failure (Hartung and Wiedemann, 2013; Pinho et al., 2006). The propagation of the interlaminar failure through the sample transferred the load towards the centrally placed 90° plies where matrix cracking happened and further leading to intralaminar failure. We can evidently observe