| description abstract | Wear rates (μgm/s) versus rotor speed for carbon samples sliding against smooth and wavy copper rotors (250 μm thick copper sheets were attached to smooth and wavy steel and polycarbonate backings) were identical at some speeds, but at other speeds wear rates for the wavy rotors were almost half those of the smooth rotors. Slider vibrations (periodic, with period set by rotation) perpendicular to the sliding surface were measured and Fourier analyzed. Comparison of vibration spectral amplitudes to spectral amplitudes derived from surface profiles identified vibration modes dynamically enhanced by surface waviness on the wavy rotor. At speeds where wear rates on the wavy rotor were most reduced, amplitudes of certain modes in the vibration spectrum were most enhanced. For all these cases, the product of mode number times speed was nearly constant, suggesting resonance. Contact forces and contact voltage drop (due to a mA current flowing from slider to rotor) were measured and plotted versus time during all experiments. Friction coefficients rapidly varied between 0.1 and 0.4, but averaged 0.2. Traces of friction coefficient versus time for both wavy and smooth rotors were similar, even when wear rates plunged on the wavy rotor. There were no large jumps in the contact voltage drop data, suggesting that the slider never disconnected from any of the rotors. Photoelastic visualizations (Bryant and Lin, 1993) of slider-rotor interfaces revealed concentrated contact on the smooth rotors, but none on the wavy rotors. The absence (induced by vibration) of concentrated contact may have caused differences in wear rates. Appreciable reductions (up to 50 percent) in wear rate are possible by adding small surface waves to a rotor that induce micro-vibrations of the slider-spring-rotor contact system. The effect appears most pronounced at resonance. | |
