PSI - Issue 28

Sahand P. Shamchi et al. / Procedia Structural Integrity 28 (2020) 1664–1672 Sahand Shamchi et al. / Structural Integrity Procedia 00 (2020) 000 – 000

1668

5

(a)

(b) Fig. 3. (a) UD carbon/epoxy sample glued into the steel adapters, (b) the final assembly of the compression setup used in SHPB device.

Prior dynamic compression tests with UD carbon/epoxy laminates series of tests were performed to ensure the validity of the recently developed split Hopkinson compression bar system. These preliminary tests were carried out using glass/epoxy samples, sharing the same geometry as the main specimens. A valid failure mode within the desired gauge length along with a clear record of the stress waves was obtained, confirming the competence of the test apparatus as well as the compression fixture. 4. Results 4.1. Static results All the quasi-static experiments were carried at a nominal strain rate of 0.0016 s -1 . Fig. 4 (a) shows the longitudinal compressive stress-strain behavior of five standard tests for both material arrangements. The mean and standard deviation (STDV) of the strength and the failure strain values are summarized in table 2.Similarly, non-standard quasi static compression tests, using the SHPB compression fixture, were also performed to examine the influence of the test fixture on the results. The overall comparison of both test typologies is presented in Fig. 4 (b). Although the elastic modulus and the failure strain values are quite consistent, a notable scatter has been observed on the average quasi-static compression strength of the standard and non-standard tests. The variability in the data may be ascribed to the sample ’ s failure mode. The predominant failure mode observed in the standard compression tests was in the form of through-thickness shear within the vicinity of the glass/epoxy tabs, while in the case of non standard compression tests, a shear failure was either in the middle of the gauge section or at its very end, which resulted from the specimen´s buckling. An appreciable decline in the elastic modulus of the electrically modified UD carbon/epoxy composite was observed in both test configurations. This can be attributed to the interleaved conductive nylon veil, which resulted in decreasing the average elastic modulus of the laminate from 140 to 116.3 GPa.