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 combustion turbulence, comparing various cylinder head, piston, and valve designs.
| Identifier | ExFiles\Box 42\4\ Scan040 | |
| Date | 6th March 1926 guessed | |
| contd :- -2- With a rapid combustion resulting from turbulence it may be expected that the ignition need not be advanced as far as usual. It is said that this is found to be true, and the spark advance may, all other conditions being equal, be considered an excellent index to turbulence. Turbulence caused by the initial intake gas velocity is said to give in the common overhead valve cyl. a general flow as shewn fig 1. LeC. 2211., and to result in the contending currents. The usual L. head engine is thought to make better use of its intake velocity in creating one large general current. The small eddy currents represent loss of turbulence. Ricardo developed a type for making the best use of this flow shewn in fig. (2). LeC. 2211 for which he claims very good results. Figs. (3) and (4) shew the well known Ricardo device for producing turbulence by means of the piston. A general movement of the gases occurs immediately before and during ignition. Fig.4. is the Tyler type. This design was developed it is said by Ricardo after he had observed the effective way in which fig.5. operated with paraffin on his tank engine with relatively high compression without knocking. Ricardo stated that it was almost impossible to make the engine knock on paraffin at compression ratios of 4.75 to 1. Fig.5. is used by Duesenberg and also on the Bugatti aero engine. Fig.6. is a type with which we are familiar, having a venturi between the piston and the Combustion head. Within our own experience we know that the Hispano engine, though apparently a straight-forward overhead valve type, requires comparatively smaller spark advance on either 2 plugs or 1, and leads one to the conclusion that in some way an appreciable degree of turbulence has been achieved. The Expl: Dept: at Derby suggest this is possibly due to the shape of their cams which have a very rapid cut off from full lift to shut. We could try experimentally a similar effect with a flat topped cam and using a masked valve. Our standard valve gear has reached the limit of max. deceleration owing to the question of contd :- | ||