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).
Engine testing procedures and carburettor development for improved performance and consumption.
| Identifier | ExFiles\Box 46\2\ Scan114 | |
| Date | 1st April 1922 guessed | |
| contd:- -4- During the same time the Works should run an engine on the test plate with full throttle from 400 or 500 revs. to 3000, and find the petrol consumption and HP. at every 250 revs. and let us know the position of the air valve at each speed after having found the very best size of throat etc. that will force the hand adjustment to come to the same notch for the best power and the best consumption for each speed. This does not seem to have been done with system and precision. The engine is enough to be more silent and powerful with our carburetter than the S.U. Certainly this is my impression of ours. E.{Mr Elliott - Chief Engineer} will let me know the sizes of throat we suggested judging from the 40/50 and the difference in B.H.P. and air valve sizes. It is possible that the 40/50 could do better consumption with slightly larger high speed venturi. The Goshawk engine I think should be taken as 55. B.H.P. against the 40/50 at say 85. B.H.P. but the throat must bear some relation to the extra air valve area as in the 40/50 possibly with the low speed throat added in each case. This sounds very complicated for a thorough investigation but my rough and ready way is to increase the H.S. venturi until it is manifestly too large and then retrace our steps to the correct size which can only be found by experiment. Hs.{Lord Ernest Hives - Chair} now thinks he has had enough of grooved needles for the petrol jets and thinks we should do better with plain cones as in the 40/50. I think they are less expensive and better, and ought not to wear because the measuring part should not touch anywhere; low speed one should just fill the jet and the high speed one have a clearance of say .001" so that it will not stick down. contd:- | ||