Fractal Analysis for Vibrational Signals Created in a Ball-Screw Machine Operating in Short- and Long-Range Tribological Tests

Abstract

In the present study, the vibrational and frictional torque signals acquired from the forward-backward movements of a commercial ball-screw system were considered via mono fractal analysis. The short-range tests were carried out in order to investigate the effects of operating conditions, a nut's inner surface roughness and the applied pretension (preload) on the fractal dimension (Ds) and topothesy (G). The long-range test was conducted to observe the variations of vibrational and frictional torque signals and thus the fractal parameters acquired from the ball-screw operations under the condition of no fresh grease supply during the testing process. The effects of the ball-screw rotational speed and pretension on the G parameter of vibrations were greater than the Ds parameter. In the backward movement, the highest G value always occurred at the highest rotational speed (3000 rpm in this study). The Ds parameter generated in the forward movement by the nut's inner surface before polishing produced a value greater than that by the nut with a polished surface. The G parameter related to vibrational amplitudes showed a value before polishing greater than that after polishing. The unusual vibrational signals are assumed to be related to ball passing behavior. Their experimental frequency was verified to be consistent with the frequency predicted by the ball pass theory. An increase in the rotational speed can bring a significant increase in the number of ball-pass signals. The G parameter and its skewness data, defined for the number distribution function of the G peaks, showed values that in general increased with the test time if the fresh grease was not supplied during the long-range test.