Prediction of depth of cut for robotic belt grinding

Abstract

Robotic belt grinding systems can be used not only to replace low efficiency, high pollution manual finishing operations but also to improve production rate and manufacturing flexibility, especially for grinding small batches of workpieces with complicated features. The contact wheel is made from soft material with significant elasticity and is tensioned by a grinding belt. Soft contact between the workpiece and contact wheel provides the benefits of high surface quality but reduces the dimensional accuracy of the finished workpiece. This paper analyzes the contact wheel’s deformation caused by belt tension in order to accurately predict the depth of cut. The elastic mechanics based on the power series method is employed to establish and solve the tension model, and the deformation of the contact wheel is obtained. The validity of the analytical model is verified by a finite element software. Then, two modified models of grinding stress distribution are developed, and the distribution of depth of grinding is predicted. Tests are running and showing that the prediction error is less than 3.1% on a given grinding path. An accurate, fast method is thus developed to predict the depth of cut for belt grinding.