12 P. A. Lara et al. Table 2.1 Test matrix for HF pulse loading conditions Fig. 2.7 Crack length for varying load (left) and frequency (Right) For this matrix, the HF pulse was only applied at the peak of the ordinary wave of 5 Hz; hence the phase shift was held constant as well as the log decrement of the function was maintained at 0.21. Introduction of other factors would increase the matrix size to a 33 design resulting in 27 treatments and are recommended for future experiments. 2.4 Experimental Results When plotting the data, as shown in Fig. 2.7, the initial results were the crack length for specimens where the load is varied while the frequency is held constant (left) and specimens where the frequency is held constant and the load is varied (right). One can see that when comparing the number of cycles it takes to grow a crack to a nominal length of 1.5 mm, the varying load (left) has a noticeable effect on the crack growth. It takes 20–30 K cycles in delay depending on the load. One can see that the number of cycles it takes for the crack to grow 0.5 mm on the baseline is substantially smaller that the cycles needed for the specimens that had an HF pulse applied. When looking at the varying frequency (right), one can see that a lower frequency has a higher effect for the same max load of 2300 N. When looking diagonally across the matrix, one can see in Fig. 2.7 that the lowest block with the lowest load and frequency behaves very similarly to the baseline, but other blocks begin to deviate as the load and frequency of the HF pulse are increased. Vic2D DIC was utilized to evaluate the strain fields ahead of the crack before and after the HF pulse was applied. Figure 2.8 shows the various stages of the strain changes surrounding the crack; in addition, the complete crack path is also shown in relation to the strain zones. At 40 K cycles, the crack is moving in a downward direction, and upon application of the HF
RkJQdWJsaXNoZXIy MTMzNzEzMQ==