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).
Test results on shrinkage temperatures and release loads for a Bentley 50 rear axle retaining ring.
| Identifier | ExFiles\Box 111\4\ scan0049 | |
| Date | 8th August 1938 | |
| 802 To Dr. [redacted] from HPS.{Horace Percy Smith - Experimental Factory Mgr} c. Mr. [redacted] c. Rm.{William Robotham - Chief Engineer} c. Dn/Hdy.{William Hardy} F5H HPS.{Horace Percy Smith - Experimental Factory Mgr}2/VS.{J. Vickers}8.8.38. BENTLEY 50 REAR AXLE - PD.90. Various particulars relative to shrinkage temperatures, release loads, dimensions after shrinkage and after separation have been obtained in connection with the present B.50 rear axle scheme whereby a retaining ring is shrunk on to the axle shaft. The rings were heated to varying temperatures lying between 600 and 800°F.{Mr Friese}, and after cooling in air were pressed off dummy shafts by an Electraulic Press. The following results were obtained :- Diametral Interference. Shrinkage Temperature. Increase of External dia. of Collar on Shaft. Release Load. Bore Increase after release. .002" 300°C. .002" 3 tons Nil. .004" 350°C. .004" 5 " Nil. .006" 400°C. .005" 8 " .001". .008" could not be shrunk on at 400°C. With an interference of .004 the time margin was ample to allow for traversing the full length of the axle shaft, but with an interference of .006 at the maximum temperature the shrinkage operation could only just be effected. Higher temperatures were not attempted to avoid affecting material. From the above results it would appear that the .004 diametral interference is the most suitable, having a reasonable time margin for assembly within the recommended temperature range, a release load conforming to the minimum pressing load required by PD.90 and no permanent set after removal. A further test was carried out using a dummy shaft and a .004 interference collar, allowing the ring to cool off without the aid of clamps. It was thus shown, since there was a gap of .0015 to .002 between the collar and the shoulder of the shaft (representing the ball race), that the collar tended to move away from the bearing due to contraction whilst cooling. continued. | ||