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).
'creeping' issue in a gearbox and proposing further tests on spline and serration angles.
| Identifier | ExFiles\Box 6\4\ 04-page131 | |
| Date | 9th May 1928 | |
| To Hs.{Lord Ernest Hives - Chair} from R.{Sir Henry Royce} c. to CJ. Wor.{Arthur Wormald - General Works Manager} c. to BY.{R.W. Bailey - Chief Engineer} X.5310 S E C R E T. R1/149.5.28. INDIA. GEARBOX. X.5310 With reference to the above I understand that one of the angles of splines that we have given has proved satisfactory in preventing the creeping of the 1st. speed. You state that this angle is one in which the driving face is 30° from the radial line. It is understood that you tested this angle previously and it did not give satisfaction. We think that the one that is giving satisfaction is just about 45° rather than 30°. It seems in effect to be about equal to a square shaft, that is, if one carefully measures the angle of the driving surface with a radial line. If this is not so it is a mystery as to why it now stays still. Although we know that it is advisable that the wheel be much larger than the shaft to prevent creeping we are forced to adopt this scheme of sliding the small gears as it so enormously affects the economy of the box, when one has to get in a servo drive. I should therefore like afew more experiments until we know how we can rely upon arranging a box so that the risk of creeping will vanish. It has occurred to me that if we were to chamber a wheel and have 2 rings of serrations, one would probably be less tight than the other, and so the wheel would be disturbed on the shaft so it rolled round, and that is why we have got this particular pattern to stand is that it has a comparatively narrow row of serrations right under the gears, which are bound up with the torque, and there is no tendency to disturb them, as there would be with a long bore of tapering serrations. To clear this matter up a little more I want you to test a shaft with fine 'V' serrations, such as we often use, say about one tenth circumferential pitch, and 30° included angle. I would like you to have one shaft slightly taper in diameter at the first speed end, and another similar shaft as near parallel as we can make it. Put on to these shafts various pinions, various ways round, one pinion being made as nearly parallel as possible, and the other with tapered serrations about equal to the tapered serrations on the shaft, say 1 in 300. If these were run under load in various combinations we might determine in which direction the creep occurred, because we believe that the taper pinion will certainly creep in one direction or another. The test should be made half engaged, contd :- | ||