From the Rolls-Royce experimental archive: a quarter of a million communications from Rolls-Royce, 1906 to 1960's. Documents from the Sir Henry Royce Memorial Foundation (SHRMF).
The assembly methods and testing for cones and gears, comparing pressure and shrinkage processes.
Identifier | ExFiles\Box 136\4\ scan0198 | |
Date | 9th November 1937 | |
-2- cones and gears, and the tests repeated. In spite of the leading chamfer, the gears still tended to shear the surface metal of the cone bores during the assembly operation (the gear surfaces, however, were not damaged) and it would be advisable to radius the leading edge of the gears of future models. From the information obtained to date (see table II) a pressed cone having a nominal diametral interference of .003" with maximum tolerances of .0005 and .0005" is capable of easily transmitting the estimated torque. In no case, however, must the interference be less than .002". The question of producing these details on a production basis will require careful consideration. By changing the method of assembly from pressure to shrinkage, the damage due to shearing would be obviated. Checking the latter method of assembly, it was found necessary to heat the cone to 180°C for a diametral interference of .002", but since this model had previously been pressed on and consequently scored, there was nothing to be gained by checking the release torque. New models are in the course of production in order to check the release torques and assembly temperatures for shrunk cones. A further report on this development will be issued as soon as the tests have been completed. Whilst agreeing that the pressure or shrinkage processes may reduce production costs, we are dubious of the success of these methods of attaching the cone to the gear, it being suspected that, under vibrational conditions, the flange faces will tend to work away from each other. Should this prove to be the case, additional security against such tendency must be provided. Increasing the interference would (within limits) increase the release torque, but would probably also increase the tendency for the flange faces to work apart. HPS{Horace Percy Smith - Experimental Factory Mgr}/W. Bell WHK | ||