Experimental verification of the mechanisms causing stick-slip motion originating from relative deceleration

In literature, the stick-slip phenomenon is usually explained from an initial stick-phase. The stability conditions with respect to stick-slip (the occurrence or not) are also derived by checking the reoccurrence of a relative standstill (stick-phase) after the slip-phase that has been following an initial stick-phase. However, stick-slip can also arise from an initial situation of pure slip. In this paper, it is experimentally shown that friction induced tangential oscillations during stick-slip can merge into stick-slip during an impressed deceleration, if their amplitude is sufficiently large. Owing to the interaction between friction phenomena and the dynamic behaviour of the mechanical system of which the friction interface is a part, the (negative) friction force-relative speed gradient (dF/d nu(rel)), which gives rise to the tangential oscillations and has to be considered for the study of stick-slip, must be derived from changes of the relative speed with the same time-constant as the mechanical system. It is also experimentally shown that the risk of stick-slip during deceleration is enhanced by an increase of the time-length of deceleration and by an increase of the tangential stiffness of the mechanical system. The correlation of the experimental results with previous theoretical analysis is clarified. As a result, some widely accepted models for the friction as a function of the relative speed have to be put in question. It is also shown that the major problem in determining the gradient dF/d nu(rel) is the inadequateness of the classical friction test rigs. (C) 2000 Elsevier Science S.A. All rights reserved.