ICF10B

itself (given by the initiation toughness) is inferred from these tests and summarized in Table 1. The crack propagates across the entire remaining ligament of the test specimen, leaving an intact fiber-bridging zone behind the crack tip. Since the bridging zone is equal to the remaining width of the test sample, this becomes a problem of large scale bridging that is not properly treated using the concepts of linear elastic fracture mechanics [3,4]. Nonetheless, it is possible to calculate a peak load and an associated apparent (or plateau) toughness that can be used to compare samples of identical dimensions. It was observed that the initiation toughness did not show dependence on the fiber orientation, while the plateau toughness did show some dependence on the fiber orientation. This correlates with our understanding that the brittle matrix dominates crack initiation and propagation, but that final fracture occurs only once the bridging fibers have broken or become debonded. It is well known that the stresses exerted by bridging fibers will depend upon the orientation of those fibers, and that aligned fibers are more effective than mis-aligned fibers [5]. -0.2 0 0.2 0.4 0.6 0.8 1 1.2 28.5 29 29.5 30 30.5 31 31.5 32 LOAD (kN) Position (mm) Figure 2: Typical results of compact tension tests. Inset shows the relative fiber orientation for these data. TABLE 1 RESULTS OF COMPACT TENSION TESTS ON THE FAILED WHEEL SAMPLES Wheel Loading Angle Relative Angle Initiation toughness Plateau toughness [°] [°] [Mpa√m] [Mpa√m] Failed wheel 0 6 3.5 (0.6) 6.6 (0.4) 90 6 3.8 (0.2) 7.6 (0.5) 45 6 3.5 (0.2) 5.8 (0.4) Bending tests Stress-displacement curves obtained from 4-point bending tests were compared for the different specimens and fiber orientations (Figure 3). There is a marked reduction in the amount of energy absorbed through cracking for different fiber orientations. In addition, the energy absorption of the failed wheel was lower than that of the exemplar wheels for all fiber orientations. Examination of the failed specimens showed that those samples with fibers in the loading direction failed by breaking the fibers, while those with fibers at 45° to the load failed by delamination of the fibers from the matrix.

RkJQdWJsaXNoZXIy MjM0NDE=