PSI - Issue 17

Trevor Sabiston et al. / Procedia Structural Integrity 17 (2019) 666–673 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

671

6

The arrows for the points at 1 million cycles represent that those points are run out points where the samples did not fail. These results have similar trends to those presented in Carvelli et al., 2010, however the stress ratio is different so the stress amplitudes or maximum stresses are not comparable. At 130ºC the stress amplitude is significantly lower than it is at room temperature for a given number of cycles. The slope of the S-N curve is also much lower for the 130 ºC test than it is for the room temperature testing. This is attributed to a softening of the matrix material and increased ductility in the matrix material at elevated temperature as per Charambous et al., 2015. Charambous et al., 2015 also found that there is a reduction in delamination resistance during fatigue loading at elevated temperatures which leads to a reduction in the fatigue life at 130 ºC. This is due to a decrease in the strength of the epoxy matrix material at elevated temperatures. For both temperatures the Basquin equation is a good fit to the experimental data the values of the fit parameters to Equation 1 are given in Tab. 2.

Tab. 2. Parameters for the Basquin equation fits at room temperature and 130 ºC

Parameter

Room Temperature 130 ºC

385.1 -0.027

152.9 -0.018

The relatively low slopes for both the room temperature and 130 ºC testing indicate that the failure of the NCF is a matrix dominated failure mode. For the 130ºC fatigue data, a fatigue failure strength for the NCF may be better for design use as the slope of the S-N curve is very low 6 of the samples did not fail while subjected to the fatigue loading at stress amplitudes within 1 MPa of samples which failed at fewer than 1000 cycles. The observation that Vallons et al. 2009 made regarding the infinite life stress amplitude being half of the UTS for NCF materials was found to not apply under fully reversed loading. However, it appears that this conclusion is valid for the UCS under fully reversed loading as none of the samples failed below a stress amplitude of 192 MPa at room temperature or 96 MPa at 130 ºC.

3.4. Failure Analysis

Typical failure of the fatigue test specimens is shown in Fig. 6. The fracture surface shows inter ply delamination of the NCF to be the primary cause of failure, both sides show this failure indicating that buckling is not present in the NCF material using this specimen geometry.

Fig. 6. Typical post mortem fracture surface of the NCF material subjected to fatigue loading