Experimental Approach for Estimating Mesh Stiffness in Faulty States of Rotating Gear

Abstract


 
 
 Gear mesh stiffness (GMS) is a principal factor in understanding a dynamic behavior and estimating a health condition of the gear system. Lots of methodologies have been proposed to estimate GMS in normal and abnormal states. However, most of them are performed in an analytical way, therefore experimental studies are limited. Moreover, previous experimental studies have limitations that they were only performed either in a static state or for a normal gear. In this study, we develop a methodology to estimate GMS of a rotating gear in faulty states, root crack and spalling. In the procedures, we employ transmission error (TE) which is defined as the difference between rotation of input and output gear. The methodology proposes the concepts of relative stiffness to remove the effect of low frequency component from shaft motion and variability of individual teeth, and corrected stiffness to exactly estimate GMS of cracked gear. Meanwhile, the study proposes a differentiating algorithm of gear faults between root crack and spalling considering the failure mechanisms of each fault. The developed algorithm is validated measuring the TE from a test-bed of a spur gear. Consequently, the algorithm has differentiated the gear in root crack and surface failure, and estimated the GMS of the gear in faulty states.