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).
The causes and characteristics of engine pre-ignition, detonation, and fuel knock.
| Identifier | ExFiles\Box 75\2\ scan0237 | |
| Date | 1st August 1921 guessed | |
| Oy2 - G 1821 Sheet #2. "Pre-ignition", "detonation" and "fuel-knock" are still used over here very loosely to mean what we would call bad chirping, and various people give various explanations, as you will see from the articles in the three S.A.E. Journals of January, February and March, which I have sent you. Woodbury of Duponts (March number) finds detonation produced with difficulty in small cylinders, and delayed by turbulence, but delayed also by increased temperature and occurring (even with acetylene, the worst knocker) only when the mixture was very much over rich. He puts down knock to auto-ignition in remote parts of the cylinder due to the compression wave of the flame travelling forwards and compressing the gas ahead of it to the self-ignition point. This is favored by heat and high pressure and is found to occur over a wide range of mixtures. All of which agrees with practice. I remember that Mr. Royce in 1913 gave me the same explanation, and one knows what a wonderful intuition he has for guessing things right. Horning in June quotes the detonation wave found by others in igniting a stagnant mixture in a tube, but does not actually say that the detonation wave is the cause of "detonation" or "fuel-knock", but that it makes a rough running engine. From your experiments and from the known ability of an air cooled engine, like a Franklin, to keep on running after the switch is cut off, and even to start off of its own accord long after the switch is cut off, one would gather that a change takes place in the gas in a hot cylinder which makes it far more liable to auto-ignite. We tend to do this quite readily. The other day a R-R car was run by mistake with no water in the jacket and had run like this for ten minutes or so before it was discovered. Switching off had no effect, and the car was stopped by the throttle, the bonnet opened and allowed to cool. After ten minutes an attempt was made to get it to the side of the road with the starter, and the darned thing immediately started running again. If the auto-ignition theory is correct, as the cause of knocking, certainly the battery plug, if red hot and above the critical temperature that you discovered, tucked away in a pocket ready for the gas to be compressed over it, would make the knock occur far earlier. And it is reasonable to believe that the plugs were so hot in this engine that as soon as compression commenced (perhaps before) the plug fired the charge. What I wrote the other day, namely that the "heavy ends" of American fuel have a much lower auto-ignition temperature than the average of the gasoline, is confirmed by numerous of the recent writers. See page 218 of March Journal, where Woodbury states | ||