Rolls-Royce Archives
         « Prev  Box Series  Next »        

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).
Technical discussion on the causes of failure in ball bearings used in motor car wheels.

Identifier  ExFiles\Box 20\7\  Scan015
Date  6th April 1912
  
634
THE AUTOCAR, April 6th, 1912.

Causes of Failure in Ball Bearings.

A contribution by Mr. John V.{VIENNA} Pugh to the Discussion on the Paper read by Mr. G.{Mr Griffiths - Chief Accountant / Mr Gnapp} F.{Mr Friese} Barrett on the above subject before the Institution of Automobile Engineers, on Wednesday, February 14th. The paper was published in 'The Autocar' of February 10th. Mr. Barrett's reply to Mr. Pugh will be found on the next page.

With much of Mr. Barrett's paper I find myself in disagreement. He is proposing to the automobile engineers of this country a somewhat complex and costly system of ball bearings as far as the wheels of cars are concerned. Most of the points apply chiefly to the front wheel bearings, and so I will confine my remarks to these.

One of the oldest and simplest forms of employing journal bearings in the wheels of motor cars is illustrated in fig. 1. The inner elements of each of the two bearings are clamped on to the axle by a nut and a distance piece, the distance piece being slightly longer than the distance between the two shoulders of the hub. The inner bearing takes the thrust from the outside, which is always the greater, and the outer bearing deals with the thrust in the other direction. The large washer is placed between the nut and the outer bearing to prevent the wheel coming off in the case of a bearing failure. It will be observed that the arrangement consists of very few parts, and that so long as the nut is kept in position (and it is fixed by the well-understood method of a split pin) the hub cannot possibly come off its axle. The method of keeping dirt and water out of the bearings at the inner end is extremely elementary, but in later examples of this type it appears to have been carefully thought out, and from such hubs of this type as have come under my observation, and from enquiries that I have made, this arrangement gives excellent results.

I give below what I think is a pretty accurate list of over twenty firms who are using this arrangement.

Fig. 1.

Argyll, Austro-Daimler, Belsize, Benz, Berliet, Brasier, B.S.A., Deasy, Delage, Germain, Mercedes, Pierce Arrow, Darracq, S.A.V.A., Sheffield-Simplex, Singer, Spyker, Stoewer, Sunbeam, Renault, Standard, Vauxhall, Vinot, Vulcan, and Calthorpe.

Mr. Barrett, in his paper, condemns this system entirely. I should be the last to urge that the Institution of Automobile Engineers should accept as final the practice of any large number of firms, however eminent, but I do think that we should at least carefully examine the criticisms brought against this practice, and the more complex arrangements which the writer of the paper proposes to substitute for it.

The author condemns this arrangement because—
(1.) Journal bearings are subject to thrust.
(2.) Neither of the revolving elements is clamped.
(3.) The bearings are not, in his view, sufficiently well protected from water and mud.

Taking these points in order:
(1.) The charge that journal bearings are subjected to thrust is not strictly true. The action at the tyre of the largest lateral force that can be applied to a car changes the pressure on the bearing from the vertical line by probably less than 20° (see my article in The Automobile Engineer of October, 1910). No sound theoretical reason has been advanced to show that a journal bearing is unable to withstand a force at 20° to the vertical, or even very much more. There is a line of high speed drills made by Messrs. Alfred Herbert, Ltd., which use journal bearings to transmit the whole of the drill pressure, and I cannot ascertain that these bearings have given the slightest trouble in use.

I note that it has sometimes been urged that a journal bearing can take direct thrust or direct journal load, but that it cannot take the two together. I believe there is not the slightest ground for this view, which I imagine is

Fig. 2.

based on tests of single bearings in which the load and thrust have been so applied as to produce a twist—that is to say, to throw the axes of the two bearing elements out of line with each other. The outer element approximates too closely to a spherical belt for it to be able to stand a twist of this kind, but where a pair of bearings is used it is quite obvious that there is no tendency whatever to displace the axes of the two elements of either bearing, for the other bearing takes care of this.

(2.) Bearings fitting loosely into their housings will certainly 'creep,' but the movement is pure rolling with no slipping whatever, and the degree of looseness that is met with in practice is not one which would introduce hammering effects, so that with ordinarily hard hubs there is not the slightest reason to expect that this pure rolling will produce serious wear in the hub during the life of the car. It should be remembered that the movement of the balls in their races is by no means pure rolling, and this circumstance, coupled with the enormously greater area of contact between the bearing and its housing, will far more than counterbalance the greater hardness of the balls and ball races. On the other hand, there is no reason why loose fits should be employed between the bearings and the housings.

(3.) I quite admit that the arrangement for the protection of the bearings appears extremely crude, and it is certainly extraordinarily simple and cheap to make. I think it would be useful if enquiries could be made on a wider scale than can be made by a single individual, as to the success of this method of keeping dirt and water out of the bearings.

There is another series of bearing arrangements, as in fig. 2, in which the cap is screwed on to, or into the end of the hub to keep the dirt out. From a great variety of experiences I am confident that this practice is not effective unless it is supplemented with the device used in fig. 1 and shown in fig. 2 by dotted lines. I think, probably, the double arrangement is better than the single canister lid{A. J. Lidsey}, but have not been able to prove this; it has, however, one advantage over fig. 1, and that is that in addition to keeping water out it can keep in a large quantity of an oil or fairly fluid grease. With this style of bearing, the thrust is dealt with by the outer or the inner bearing as in fig. 1, or one of the bearings is clamped, in which case it is the usual, but not the invariable, practice to make the one bearing take the thrust in both directions. This series of bearings is also condemned by the author of the paper.

Now with regard to the designs, the use of which the author advocates. These appear to be embodied in his fig. 25, which is provided with two journal bearings, the revolving elements of which are clamped while the stationary elements are free to slide the one on a stub end and the other on a sleeve on the stub end. The first criticism is, of course, that it is more complicated, more costly and heavier. The second, that the space taken up by the sleeve on which the outer of the two journal bearings is mounted, takes
  
  


Copyright Sustain 2025, All Rights Reserved.    whatever is rightly done, however humble, is noble
An unhandled error has occurred. Reload 🗙