Search Results

The paper gives an overview of the local stress design concept for gears related both to bending fatigue for the gear teeth and rolling contact fatigue on the gear flanks. Test results are presented for notch fatigue and rolling contact fatigue on specimens as well as for sintered gears manufactured from the quenched and tempered sintered steels (Fe-1.5Mo) + 0.5C and (Fe-4.0Ni-0.5Mo)+ 0.5C, both at a density of ρ = 7.1 g/cm3. The transferability of notch fatigue and rolling contact fatigue data from specimens, test rollers and fatigue test bars, to gears is presented on the basis of local properties.

Rolling contact fatigue tests with a sliding of -24% were performed with the mixed alloys Fe+1.5%Cu, carbonitrided, Fe+1.5%Cu + 0.5%C, as-sintered as well as quenched and tempered, diffusion alloyed powder Fe-4.0%Ni-1.5%Cu-0.5%Mo, as-sintered, plasmanitrided, carbonitrided, case hardened and additionally mixed with 0.5%C and prealloyed powders Fe-4.0%Ni-0.5%Mo and Fe-1.5%Mo with addition of Cu and C in different heat treatments. The densities investigated were 6.8, 7.1, 7.4 and 7.8 g/cm3. The tribological environment was the gear-box oil SAE 80 at 80°C. With superposition of sliding the highest rolling contact fatigue strength was achieved with powder forged specimens (density of 7.8 g/cm3). An optimum for the sintered specimens in the range of 6.8 to 7.4 g/cm3 was found at around 7.1 g/cm3 for the most sintered steels independent of the heat treatment. This is due to effects based on open and closed porosity in interaction with the tribological environment.

For the development of new components, design engineers today have access to a broad amount of fatigue data, which were obtained from unnotched and notched specimens. These data can be transformed when the conditions of material, strength, geometry, surface and surface layer and loading mode in the fatigue critical areas are taken into account for constant and variable amplitude loading. The procedure of data transferability is discussed for the example of a randomly loaded truck stub axle where the failure criterion is the first detectable crack, and the local equivalent stress/strain and the maximum stressed/strained material volume are considered. In addition, several problems associated with fatigue life assessment under variable amplitude loading are discussed.