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).
Typed memorandum discussing design improvements and specifications for a governor system, including pistons, weights, and springs.
| Identifier | ExFiles\Box 37\1\ scan 303 | |
| Date | 4th May 1926 | |
| Contd. -3- RL/M4.5.26. Relay centre piston is better light but its gravity could be used if not exceeding 250 governor revs. The governor weights we should like heavier (but lead is too plastic) and we should like more flexible governor springs, judging from 7-EX, as it behaves. We therefore suggest ultimately changing the arrangement of springs (see sketch). This reduces their effect by half and seems in every way preferable (less stress and larger gauge wire). If the weight of the central piston valve were not reduced owing to it working better in bronze (which we anticipate) we can use this weight instead of the 8° initial set on the governor springs. (D) At present the return springing is double that required and has 100% rating so that at full advance the springing is 4 times the required amount. Initial advance is undesirable except to ensure no relay lag - i.e. governor movement without corresponding advance, so the 2 or 3° in graph 1. is ample. Return spring over relay should be only 6 lbs. in retard, and lowest rating possible (not exceeding 50% = 9 lbs. max.) Spring at magneto in retard 2 lbs. and 4 lbs. or less fully advanced, i.e. put on with 1 1/4" stretch, longest spring possible, fine wire, i.stress say 70,000) large diameter. We do not agree with over-springing for safety or other reasons, and the divided springing seems best. This suggested springing gives over 50% margin to return from fully advanced firing with a reasonable margin at zero. Such spring allows the motor to work with the lowest oil pressure. This is necessary to keep the leaks down. FINALLY. Your graph 1. is nearly all we desire: you will see this by my graph. Your graph is not good. Contd. | ||