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 clutch performance, discussing deflection tests, design modifications, and a review of performance graphs.
| Identifier | ExFiles\Box 130\3\ scan0197 | |
| Date | 6th March 1941 guessed | |
| Copy for -2- Paragraph "C" - refers to the same check but endeavoring to see how much load it would take to cause a deflection of .005 at each cap screw. This was the equivalent of from 1700 to 2700# load in addition to the pressure springs. Paragraph "E" - refers to acheck on a production clutch on our spinner machine, which is equivalent to having it on the engine flywheel. We believe that your trouble is a combination of several things; one, being initial looseness in the release levers, which we have had some complaint on ourselves and have practically eliminated by a slight change in design, as shown on these inclosed prints. The change consists of eliminating the flat sided pin and the single roller at the yoke position and substituting a round pin with a set of needle rollers. Theoretically, it would appear that this was not sound because of the slight change from horizontal center line that the yoke must undergo due to change in angular position of the levers. Actually it works out very well since we have been using this construction on Cadillac for the past year. Paragraph "F" - is the free play on up and down movement of levers. Numbers 1 and 2 are the extremes of several production assemblies checked, while number 3 is the design described on the prints. Paragraph "D" - refers to a check on these same three assemblies, measuring the throwout of the levers at 3500 R.P.M. This would indicate that if all production assemblies ran as good as Number 1, there would be no need for the change. The new design is consistently low and seems to have completely eliminated the trouble at Cadillac. Of course the smaller this movement is the less initial free play required in the clutch pedal, leaving more travel for clutch release. On their Graph No.2 we are unable to understand how they obtain so much deflection in the cover stamping. On their Graph No.3, comparing this with our inclosed curve, indicates that we actually obtain more centrifugal force than what they calculate. On this point, our curve is based upon actual checks, wherein the levers were mounted freely in a position corresponding to their operating position and the load carefully measured at the inner end at various speeds to hold the levers at the operating position. On their Graph No.1, they indicate only .017 total deflection of the cushioning means under the lining. While we are not sure, we believe that this is the disc design which has cont/d.... | ||