ICF10B

ginning. One also might argue with residual stresses stemming from the austenitic-martensitic phase transformation and the corresponding volume increase of 4%. 10000 100000 1000000 10000000 460 480 500 520 540 560 580 600 620 640 660 680 S355MC FB600 TRIP800 DP650 FB650 FB600 FB650 DP650 S355MC TRIP800 σ (MPa) N (cycles) Figure 4: S-N curves of the different steels after 10% pre-straining However, despite the assumed effect of martensite in the finite life range it is remarkable that the fatigue limit seems to correspond to the “high-strength – high-fatigue resistance” relation mentioned above. Since TRIP has the highest yield strength (Table 2) it also shows the highest fatigue limit, see Figure 4. In agreement with the literature [6, 7] the morphology of multiphase steels is responsible for retarding the crack nucleation and thus increasing the fatigue limit. The finer the microstructure the higher is the fatigue limit. The results obtained by the industrially tested wheels according the earlier given procedure until the first crack appears are shown in Table 3. The average fatigue life of the microalloyed grade is about 100.000 cycles. It is evident that for the applied multiaxial state of strain all tested multiphase steels show a clearly better performance, i.e. an up to a twice as high fatigue life. TABLE 3 FATIGUE LIFE OF WHEELS (AVERAGE CYCLES BEFORE CRACK INITIATION) Steel Grade FB650 DP650 TRIP800S355MC Fatigue Life in Cycles 192.700 176.000 205.050 100.000 It is important to point out that the fatigue tests were carried out using paint-baked wheels. The influence of such a paint baking on the fatigue life can exemplarily be seen for DP650 in Figure 5. The whole curve of the bake hardened steel DP650+BH shows higher values for the finite life range as well as for the fatigue limit, which increased about 60MPa compared to its non-baked state. The improvement of the fatigue performance through bake hardening (20 min. at 170°C) is caused by the diffusion of the C atoms to the dislocations created during the (pre-)deformation, resulting in an additional increase of YS, see Table 4. Simultaneously, the tensile strength increases. While the pre-straining accounts for a YS increase of 172MPa, after bake hardening the YS increased another 144MPa (=316-172), see Table 4. This extremely high value might be due to additional precipitation hardening of the ferritic phase by carbides with C stemming from the “C-saturated” brittle martensite before baking. An as such softened martensitic phase could improve the load transfer between the phases and thus contribute to the fatigue resistance increase. To clarify these assumptions, further research is needed. However, it should be paid special attention to the low cycle fatigue behaviour of multiphase steels, which show a large scattering of the results. This scattering is according to Vogt et al. [5] and Akdut [6] related to the phase morphology. Here parameters like grain size, phase size, phase volume fractions, orientation relationships between the phases, local stresses and the orientation of the phases with respect to the macroscopic sample directions play an important role. As said earlier, crack nucleation is generally retarded, but crack propagation might occasionally be enhanced leading to scattering.

RkJQdWJsaXNoZXIy MjM0NDE=