On the repeatability of friction and wear tests for polyimides in a hertzian line contact

Tribological properties for polymers are mainly experimentally determined, while few standards are available and variation in data is often large. Polyimide is slid against steel on a cylinder-on-plate tribotester at 50-200 N and 0.3 m/s. There is a statistical variation of +/- 7% on dynamic friction, +/- 10% on static friction, +/- 8% on wear rates from weight loss and +/- 12% from wear rates from dimensional measurements, or even higher at high normal loads. Different parameters influencing statistical repeatability are discussed. Friction for polymers shows higher variation and wear rates show lower variation than steel/steel pairs due to visco-elastic deformation that has a contribution to friction but reduces stress concentrations. It is experimentally verified that the visco-elastic deformation of polymers in line contact is higher than calculated from theoretical models. The role of recoverable and permanent deformation is verified and there is a difference in deformation at 50-100 N and 150-200 N revealed from static loading tests, creep tests and wear measurements. The limit between running-in and steady-state coincides with the stabilization in contact pressures after 250 mu m diameter reduction. Besides test rig design, a variation in counterface roughness seems the second most important influence.