LAUSR

Abstract Gears made of PM steel are of interest for the automotive industry because they can be produced to near net shape with only a few processing steps. Automotive gears experience a complex contact situation with rolling as well as combined rolling and sliding. They also have to be able to withstand high loads and fairly high temperature variations. Earlier work show that the main limiting parameter of the contact fatigue life of PM steel is the porosity. A higher porosity/lower density will decrease the fatigue life of the PM component. In the present study, the pure rolling contact fatigue life of PM steel with different density and surface finish has been investigated. A rolling contact fatigue test rig, where rods of the tested material are mounted between two rolling wheels, was used. Two densities of PM steel, 6.8 and 7.15 g/cm3 and a full density reference steel with two different surface finishes, centerless grinded and superfinished, were tested. The tests were run for a given number of load cycles or until failure (fatigue life). SEM was used to study the surfaces and cross-sections to reveal the mechanisms of crack initiation and propagation. The higher density PM steel (7.15 g/cm3) outperformed the lower density steel (6.18 g/cm3) by a factor of around 4 in fatigue life at the same surface pressure, regardless of surface finish. Cracks are initiated at a depth of around 100 µm. These cracks propagate and eventually they reach the surface, causing surface damage and failure. For the low density PM steel, both sub surface crack initiation and failure occurred earlier (at a lower number of load cycles) than for the higher density PM steel. Severe surface damage or wear were not found until failure occurred. Still, some initial alteration of the surfaces was seen already after 0.5 million load cycles, in the form of removal of the highest asperity peaks on the centerless grinded surfaces, and opening of the surface pores on the superfinished surface. No effect of surface finish was found on the fatigue life. The difference in surface roughness could induce a difference in local stress concentrations at the surface, but in this test the cracks causing fatigue failure are initiated at a depth where the stress distribution is not affected by local surface stress concentrations. This means that for fairly smooth surfaces roughness, the surface of PM steel is not important when it comes to pure rolling fatigue life.