13th International Conference on Fracture June 16–21, 2013, Beijing, China -3- Code (IHAPC). The air void of all mixtures was 4 percent. Asphalt binder with penetration grade of 60/70 was utilized for preparation of the cylindrical samples. Table 1. HMA aggregate gradation Requirements Sieve size (mm) Min Max Percent passing 19 100 100 100 12.5 90 100 95 9 67 87 77 4.75 44 74 59 2.36 28 58 43 1.18 20 46 33 0.5 13 34 23 0.3 5 21 13 0.15 4 16 9.5 0.075 2 10 8.4 3. Fracture tests and results Three-point bend fracture tests were conducted on the SCB specimens under pure mode-II loading at -20°C. To achieve this temperature, the SCB specimens were first put into a freezer with the fixed temperature of -20°C for 4 hours. Then the tests were immediately carried out using a universal testing machine and a three-point bend fixture (as shown in Fig. 2). The displacement rate of the upper fixture was set to a constant value of 3 mm/min. In the first stage, the conventional round-tip supports were used to load the specimens. However, some of the tests were not successful. Because, the crack growth initiated from the lower and right hand side fixture and not from the crack tip (see Fig. 3). To avoid crack initiation from undesirable locations, the lower fixtures were modified such that instead of applying the bottom loads in the concentrated points (as was the case in the first set-up) the loads were applied as distributed forces. Several finite element analyses were performed to find the width of distributed load (i.e. the magnitude of b in Fig. 4) in order to provide pure mode-II loading. Fig. 5 shows a typical mesh used in the finite element analyses. The appropriate value for the parameter b was eventually found to be 4 mm. The conventional fixtures used in the first set-up were replaced with the second set-up, and the mode-II fracture tests were repeated. By this modification, the crack extension in all the specimens took place from the crack tip (see Fig. 6). Therefore, the pure mode-II fracture test was successfully simulated experimentally using the SCB specimen. The fracture tests were also performed under pure mode-I loading and at -20°C using the same three-point test set-up. However, for this type of loading, the crack was generated in the middle of the specimen with symmetric loading supports of S1=S2=50 mm (see Fig. 1). Fig. 7 shows sample of the load-load line displacement curves recorded from the mode-I and mode-II fracture tests. In order to increase the reliability of the experimental results, four SCB specimens were tested for each mode of loading. As
RkJQdWJsaXNoZXIy MjM0NDE=