Experimental and Numerical Determination of the Local S-N Curve of Shot-Peened Steel Gears
Mechanical design of components can be performed according to different criteria, aiming to prevent unexpected failures and to maintain structure weight at low values. When loads are strongly variable with sporadic peaks in intensities, “safe life” criterion is applied. In particular this criterion prevents failures in components expressively designed for a limited number of load cycles and according to very strict weight specifics. A typical application of this methodology is on competition cars, where many components must be designed for a life limited to the only race duration. The use of such criterion is often expensive, since it requires the knowledge of local Wöhler S-N curve. The purpose of this paper is to determine, by experimental and numerical methods, the S-N curve at the tooth root of a gear in the transmission of a Formulai competition car. This component (Fig. 1) is carburized, hardened and shot-peened with peen diameter of 0.3 mm (S110) and shot operating pressure of 4 bar. Parameters of the analysed spur gear are modulus, 4 mm, pressure angle, 20°, face width, 5.5 mm, number of teeth, 22; moreover a correction (correction factor = 0.96) is applied. Fatigue tests for the evaluation of the fatigue limit and the determination of the S-N curve were performed on a Rumul resonant machine (Fig. 2), able to work at a frequency of 108 Hz. A device had to be used to correctly refer the gear to the testing machine, in order to reproduce work load distribution, as described by Olmi et al. in . Such device is able to apply a bending fatigue load just to a couple of teeth: on any gear five couples of teeth can be tested.