13th International Conference on Fracture June 16–21, 2013, Beijing, China -6- microscope, whereas two long distance microscopes sitting on motorized three-axis stages were mounted on the rig of the fatigue testing machine to observe the fatigue crack growth at the edge-notched micro-specimen in horizontal and vertical direction, as shown in Fig. 9. The test procedure itself starts with a scan of the unloaded surface of the micro-specimen. Then the micro-specimen is subjected to fatigue loading with a pre-defined number of load cycles and the surface is scanned again. Repeating this procedure allows monitoring the damage accumulation process on the surface of the micro-specimen and determining the local crack growth rates. 4. Results Under fatigue loading, microcracks were initiated at inclusions in the base material on the surface of smooth specimens. However, these cracks showed only limited crack extension and never caused failure. Instead, one or two edge cracks were initiated in the vicinity of the notch root. A typical example is shown in Figure 10. It can be seen that the crack paths are fairly rough in spite of the fine microstructure and that the averaged crack line is not exactly perpendicular to the loading direction. These effects can be attributed to texture effects in the deformed base material. First of all, the rods used in the forming process possessed a rather pronounced manufacturing texture and showed lines of enhanced chromium content (dark lines in the shaft in Figure 2). These lines are still visible in the flange after the thermo-mechanical forming process. SEM analyses indicated that crack kinking can also be correlated with local bands of pearlite which survived the forming process. Moreover, there are regions in which the grains in the deformed base material are strongly correlated in their crystallographic orientations (e.g. areas highlighted with white dashed lines Figure 11), i.e. the grain structure shown in Figure 6 is replaced by set of subgrains separated by low angle boundaries. A crack passing through such an area is deviated from its original path due to the local anisotropy of the microstructure (yellow dashed line in Figure 11). The growth rate of crack #1 is depicted in Figure 12. Apparently, the crack was slowed down after it had reached a length of about 1mm, and picked up considerable speed at about 1.3mm. Comparison with Figure 10 indicates that these changes in the crack growth rate may be related to the graded microstructure. This conjecture was verified using micro-hardness indentation along the crack faces as shown in Figure 13. Figure 12 indicates that there is a clear correlation between an increase in the hardness (corresponding to the transition region) and a decrease in the crack growth rate, and that the sudden rise in the crack growth rate occurs at the maximum of the hardness curve (martensite).
RkJQdWJsaXNoZXIy MjM0NDE=