Boron trifluoride and its derivatives: by Harold S. Booth and Donald R. Martin. 315 pages, illustrations, diagrams, 14 × 22 cm. New York, John Wiley & Sons, Inc., 1949. Price, $5.00

Abstract

A calculation model of grinding temperature during full tooth groove form grinding (FTGFG) was established through the simplification of the three-dimensional (3D) heat conduction process and based on the theoretical analysis of the transient moving heat source temperature field. The proposed model calculated the grinding temperature distribution of the gear surface and also described the coupling effect of two adjacent grinding heat sources at the junction of the tooth groove. Based on the grinding principle and the contact geometry relationship between the formed grinding wheel and the workpiece, the distributions of grinding parameters, grinding wheel geometry parameters, the actual contact arc length between the grinding wheel and the workpiece, tangential grinding force, and the heat partition to the workpiece were analyzed. A 3D non-uniform heat flux distribution model was also developed based on the actual tangential grinding force distribution, the linear velocity distribution of the grinding wheel, the heat partition ratio along the cross-section of the tooth groove, and the triangular heat flux model in the direction of the contact arc for the contact zone during FTGFG. Finally, a test method for form grinding temperature measurement was designed, and the theoretical model was verified by experimental results. Test results revealed that the temperature distribution on the tooth groove surface during form grinding was non-uniform. Due to the coupling effect of two adjacent heat sources, the temperature increased sharply, and a steep peak was formed at the flange where the bottom plane and the transition arc intersected or the transition arc and the involute intersected. A comparison between experimental and theoretical results on the ground surface and in the transverse section revealed that the analytical model had better prediction accuracy and that the relative error was within 15 %.