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).
Fuel systems, induction pipes, and comparative power tests.
| Identifier | ExFiles\Box 75\2\ scan0249 | |
| Date | 8th August 1921 guessed | |
| Oy2 - G 8821 Sheet #7. Possibly the scheme of just blowing out of the inlet bends such liquid fuel as consents to go out, will seem very crude to us in a few years time. It looks as though a serious development of this scheme would take the form of a whirl-chamber in the supply pipe, to separate out the liquid fuel, or even revolving paddles driven by a small series motor to form a definite and positive mechanical separator in the inlet system. Also boiling the fuel may not be the right thing to do with it. Several schemes are in use for pumping the fuel from the gas tank to the carburetter, and some such scheme might be used to pick up the liquid fuel accumulated from the induction pipe in cold starting and put it back in the tank or in the carburetter after the engine had had a chance to get warm enough to use it. (5) Power Tests with Induction Pipes: Full throttle tests on the dynamometer will not show up all the merits or otherwise of any induction pipe scheme heated or otherwise. Dickinson and Sparrow (January Journal) fixed a flywheel on the dynamometer so as to take acceleration tests under approximate road conditions, and can detect "loading-up" in this way. Nelson has taken a curve of "horsepower used" on the level road and constructs fuel consumption and power curves on the dynamometer by setting the throttle to simulate road conditions. It has been recently shown in the Bureau of Mines' tests at Pittsburgh in connection with the new Hudson vehicular tunnel for New York (see Journal for April) that nearly all cars and trucks with average carburetter settings behave fairly well as regards economical use of fuel, when pulling full throttle even as low as 15 m.p.h., but are riotously extravagant at light loads (idling, level running, or slight down grades). Because of the difficulty of getting really good low speed running on the level on our cars, as compared with many American cars (Locomobile, National, even Hup, Reo or Chevrolet) and because of the missing which seems always to occur when idling at low speeds, I have got the idea that possibly our part throttle distribution is really very far from good. | ||