A novel acceleration-controlled random vibration fretting test methodology: From classical sinusoidal to Gaussian random excitation

In nuclear power plants, wear of steam generators, fuel rods and assemblies can have serious economic and safety impact. To reproduce realistic oscillatory fretting wear in laboratory, a novel accelerated random vibration fretting wear test methodology, using Gaussian excitations, is proposed. A new experimental device and protocol were also developed, enabling controlled fretting damage under random excitations. The tests were performed on a nickel-based alloy 718 and a proprietary Zr–Nb alloy, using two crossed cylindrical geometries in the fretting wear configuration at high temperature of 315 °C. To generate reference wear rate-work rate relationship, sinusoidal fretting wear tests were carried out in the gross slip regime. True random tests were then performed using a Gaussian acceleration-control strategy. A parametric study on the Power Spectral Density (PSD) defining the acceleration level was investigated beforehand to control the ratio between partial and gross slip domains, through the control of the mean and variance of the displacement Gaussian response. Differences observed between both sinusoidal and random test campaigns, in terms of friction and wear mechanisms are discussed. The dependence of wear rate of Zr–Nb alloy and nickel-based alloy 718 on the accumulated dissipated energy and work rate are established for sinusoidal and random vibrations. •A novel accelerated random vibration fretting wear test methodology is proposed.•A new experimental fretting device and protocol were developed.•The tests were performed on Zr-alloy and alloy 718 at high temperature (315 °C).•True random tests were performed using a Gaussian acceleration-control strategy.•Fretting wear mechanisms are quite different for random and sinusoidal vibrations.