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 memorandum discussing the stress calculations for bolts with forced-on collars.
| Identifier | WestWitteringFiles\K\October1923\ Scan133 | |
| Date | 1st October 1923 | |
| [Handwritten] R T H C TO BY FROM R.{Sir Henry Royce} c. to HS.{Lord Ernest Hives - Chair} [Stamped] ORIGINAL RE. BOLTS WITH FORCED ON COLLARS. [Handwritten] K1334 In reply to your BY1/P21023., it is assumed to begin with that a short length of the bolt stem is left proud for the reception of the collar so that this latter can be forced into its position with great pressure, leaving the metal of the collar stressed up to say 20 tons per square inch. The next point is the co-efficient of friction. Mr. Royce is of the opinion that this would be 100% under the conditions given. The sectional of the collar on one side is .062 X .6 = .0372 per sq. in. Tension in the collar is 20 X .0372 = 744 tons. Multiply by 2 π for total load on bolt stem, 744 X 2 π = 4.68 tons. The end pull on the bolt to move this collar would be 4.68 tons. Section area at bottom of bolt thread = .0767 Stress in bolt = 4.68 / .0767 = 61 tons per sq. in. Thus if the collar is stressed to 20 tons per sq. in. it is necessary to stress the bolt to 60 tons per sq. in. to cause movement. This would not be strictly correct as no account has been taken of the torsion of keeping the bolt from turning. However the collar stress could be put up still higher. Perhaps you would try some experiments with these collars to see how much it takes to pull them off with an ordinary nut and spanner with different stresses in the collar. E.{Mr Elliott - Chief Engineer} | ||