Dynamic analysis of the cutting forces in gear hobbing

Abstract

The work reported in this thesis has been developed to predict and measure the cutting forces in the gear hobbing process. A review of past research in this area has highlighted the need to adopt a different approach to modelling the process in order to predict the cutting forces. The hobbing process has been described using six different co-ordinate systems and the kinematic relationships between these systems established. A single rack profile has been used to represent the profile of a single cutting tooth from the hob which was then extended to simulate the hob itself. When the hob gashes pass through the cutting region surfaces are generated which, if mapped on a regular grid can give the basis to estimate the depth of cut, i.e. the instantaneous chip thickness produced by that particular tooth. The instantaneous cutting forces generated by that tooth then can be estimated by using the concept of a specific cutting force of the workpiece material. The estimation of cutting forces acting on a single tooth space was used to predict the cutting forces produced during machining of a full gear, by assuming that the forces acting in a particular tooth space are equal to those acting on the adjacent tooth space at an equivalent instant in the cutting cycle. In order to validate predicted results, a Churchill PH1612 hobbing machine was retrofitted with a CNC control system at Newcastle University, utilising a programmable multi axis controller (PMAC). A specially made single toothed gear, and a full gear were machined, and cut on this machine, and the cutting forces measured in real time using a 3-axis dynamometer. The force signals produced by the dynamometer were measured utilising a 12-bit ADC card. Code, written in C, was developed to perform the many functions needed for the overall control of the machine, but additionally was used to capture both the cutting forces and axis position data. The results of the simulation and modelling have shown very good agreement with those obtained experimentally.