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 wear in axle assemblies, focusing on the interaction between gears and bearings due to abrasives and lubrication.
| Identifier | ExFiles\Box 114\1\ scan0085 | |
| Date | 11th January 1937 guessed | |
| To begin with, there is no such thing as pure rolling motion in any of the elements making up an axle assembly, and wear will occur whenever sliding occurs under pressure in the presence of an abrasive. We are not only interested in the wear of the gear teeth but also primarily in the wear in bearings and it is to explain how important the wear in bearings is that I took so much trouble previously to explain the great care taken in the high precision to which bearings are fitted and the gears themselves are mounted. Any wear in the gear teeth or in the bearings starts a vicious circle of wear. It is probably self-evident to you why the gear teeth should wear in the presence of an abrasive. But the wear in ball and roller bearings might be somewhat obscured by the fact that these parts are presumed to operate with pure rolling motion. It is impossible to prevent sliding no matter how geometrically accurate the parts may be made originally, because the parts deform under load and produce zones of sliding under extremely high pressures. There are other zones of considerable sliding under pressure that require non abrasive lubrication to avoid wear. Ball bearings are subject to wear because of pressure and sliding existing between the ball and cage or separator. In figure 19 the point A is a zone of pressure, between ball and cage caused by combined radial and thrust loads on the bearing. Since the relative motion between the two is sliding the presence of an abrasive causes wear and is exhibited primarily in lapping the ball down to a smaller size. I want to mention here that the balls in the bearings as used in motor cars and demanded by specifications, which do not have an accuracy for diameter and sphericity within 25 millionths of an inch are rejected. In taper roller bearings wear caused by abrasive occurs at two points. In figure 20 it will be seen that at "A" the end of the roll contacts a thrust rib on the inner race. This thrust rib is necessary to guide the roller and is subjected to a considerable thrust from the roller. As all of the relative motion between the end of the roller and the rib is sliding the contacting surfaces are subject to wear in the presence of an abrasive. At "B" the surface of the roller contacts the case which maintains the spacing between the rollers and all relative motions between roller and case is sliding. If wear occurs at either point the "stand" of the bearing is decreased which means that the inner race approaches axially further into the outer race thereby developing end-wise clearance. If two sets of bearings are used to oppose thrust in opposite directions the clearance developed produces an end-wise looseness or chuck when the thrusts reverse. The influences of the lubricant on the wear in the axle is best measured by long time tests with a car operating either in normal service or according to a fixed program which includes various types of load conditions which might be met in service, such as would be found in heavy pulling in sand roads and in high speed cruising. Such tests give comparative results. Where it is desired to obtain very accurate results on the comparison of different lubricants as to their effect on wear, severe loadings are not required. Dynamometer tests run at loads required to propel the car at 50 m.p.h. on level roads will give a very accurate comparison of lubricants. In either type of test special care is exercised in the assembly of the gear units to insure that they are as nearly alike as possible and accurate measurements are made on the gears and bearings and on the positions of the gears, backlash, etc., so that both the amount of the wear and the place where it occurs can be accurately determined. It seems to me that under such conditions an axle which has been run for 50,000 miles and which shows half as much wear as another - 10 - | ||