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).
Analysis of bearing failures due to adjustment and overloading, with lubricant performance test results.
| Identifier | ExFiles\Box 127\4\ scan0074 | |
| Date | 31th August 1937 guessed | |
| 2. If the complete bearing has failed it is possible that adjustments of the crown wheel have been made which have altered the relative positions of balls and races to such an extent that the load on the bearing was confined to the few balls at the loaded side due to the balls coming down on a smaller dia. of the track. This would soon break up the balls and in turn cause damage to the cage of both bearings. Serious overloading would cause ball failure which would again break up the cage, but this would probably cause trouble on the nearside bearing only. There is only one condition which would cause the cage only to break up. A slack adjustment of these bearings would allow slightly more reduction in effective dia. of the balls taking the load owing to reduced contact angle and increased load concentration. Thus the balls on this side of the bearing would revolve at a different speed from those on the opposite side and considerable wear and eventually collapse of the cage would follow on both bearings. The Pinion Thrust Bearing has a contact angle of 32º and a preload of 200 lbs. per row of balls is obtained by providing approximately .0003 oversize balls. The EP.{G. Eric Platford - Chief Quality Engineer} oils used in this country are not so effective as the American oils. This is due to the fact that the hypoid gear requires a lubricant which will stand much higher loads without seizure. The Alman machine is used extensively in the States for tests but Wakefield's have the only example in this country. Their test on oils obtainable here have shown, about three years ago, that their own oil and Vacuum, give the best figures on this machine and samples which we have tested on a four square rig, on which we can control torque load have verified their results from a practical point of view. The machine consists of a shaft running in a split bearing, the halves of the bearing being loaded to a maximum of 15000 lbs., the shaft driven at a constant speed and torque produced measured on a dial connected with the bearing halves. Wakefield EP{G. Eric Platford - Chief Quality Engineer} gave perfect results up to the capacity of the machine 15000 lbs. Vacuum showed signs of slight "nipping" at 15000 lbs. | ||