In this study, effect of fillet rolling process on fatigue performance of a diesel engine crankshaft was investigated. Crankshafts from two different materials, were studied / ductile cast iron EN-GJS 800-2 and micro-alloyed steel 38MnVS6. Resonance bending fatigue tests were conducted with crankshaft samples. Test plan according to staircase test methodology was used. Statistical analyses were carried out with the test data by Maximum Likelihood Estimation method in order to calculate the fatigue limits and construct the S-N curves based on Random Fatigue Limit (RFL) and Modified Basquin models. Fatigue limit calculations were also conducted by Dixon-Mood method and by Maximum Likelihood Estimation methodology for Normal and Weibull distributions. Fillet rolling process was simulated by computer based analysis in order to calculate the compressive residual stress profile at the fillet region to shed more light on the mechanisms and effect of fillet rolling. Fatigue performances of crankshafts from two types of materials were evaluated both at unrolled and fillet rolled states. Effect of fillet rolling load on fatigue performance was also evaluated with steel crankshafts. It was found that ductile cast iron showed better performance under bending fatigue tests than the steel crankshaft both at the fillet rolled and unrolled conditions. On the other hand, fillet rolling process was found to be more effective on steel crankshaft than ductile cast iron crankshaft in terms of fatigue performance improvement. It was also seen that fatigue limit increases with the fillet rolling load up to a limit where surface quality is deteriorated. Residual stress analysis showed that a higher magnitude of residual stress can develop on steel crankshaft fillet region whereas the effective depth of the residual stress is higher on ductile cast iron crankshaft with the same rolling condition. Residual stress analysis of steel crankshafts rolled at different rolling conditions show that, peak residual stress increase with the increasing rolling load is not significantly high and main effect of increased rolling load is the increased effective depth of residual stresses. The MLE methodology used in statistical analysis of the test data was found to be effective for life regression and fatigue strength distributions analysis. RFL model has provided better life regression analysis and fatigue limit calculations than Modified Basquin model. Dixon-Mood method was found to be overestimating the fatigue limit.
| Identifer | oai:union.ndltd.org:METU/oai:etd.lib.metu.edu.tr:http://etd.lib.metu.edu.tr/upload/12614775/index.pdf |
| Date | 01 September 2012 |
| Creators | Cevik, Gul |
| Contributors | Gurbuz, Riza |
| Publisher | METU |
| Source Sets | Middle East Technical Univ. |
| Language | English |
| Detected Language | English |
| Type | Ph.D. Thesis |
| Format | text/pdf |
| Rights | Access forbidden for 1 year |
Page generated in 0.0019 seconds