ICF13A

13th International Conference on Fracture June 16–21, 2013, Beijing, China -8- Figure 6. Threshold curves according to Döker and Murakami. As expected, the data points for R = 0.1 and 0.5 lie close to the approximation curve. For R = -1, the data points calculated from the inclusion sizes are slightly above the calculated threshold value. The values considering the ODA sizes are around 7.3 MPa√m and with this lie well above the approximation curve. This difference indicates that, for R = -1, crack closure effects, which are not considered in the applied fracture mechanics approach, are an important issue. Concerning the relatively high ductility of the investigated 12% Cr steel, crack closure effects due to residual stresses near the crack tip are probably much more pronounced than in the high-strength 100Cr6 examined in [29]. The same conclusions can be drawn from the comparison of the individually calculated ΔK values with the ΔKth-R curve proposed by Murakami (blue dashed line in Fig. 6), which is determined according to Eq. (4) with the material hardness HV, the√area parameter and the material constant α. (4) 4. Summary 1. No frequency effect was found for fatigue tests with zero mean stress in a frequency range above four orders of magnitude. 2. In the LCF/HCF-regime, specimens either failed by cracks initiated at PSBs or surface inclusions. Above 2·107 - 4·107 load cycles, fish-eye cracks were found originating from oxide inclusions. The inclusion diameters ranged from 16 µm up to 45 µm and follow a normal extreme value distribution. The inclusion size has a decisive influence on fatigue life as shown by the modified S-N curve proposed by Murakami [9]. The α-parameter describing the load ratio dependency was determined to 0.546 and was in the vicinity of α values measured in recent investigations at other high-strength steels. Thus, the √area-concept is well applicable for this steel in a large range of load ratios (-1 < R < 0.5). 3. For R = 0.7, the fatigue fracture surfaces are similar to quasi-static tensile fractures with high deformation and sample elongation due to extremely pronounced cyclic creep. R > 0 R < 0 ΔKth (Murakami) ΔKth (Döker) Experimental data (inclusion sizes) Experimental data (ODA sizes)

RkJQdWJsaXNoZXIy MjM0NDE=