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).
Carburettor servicing, fuel distribution problems, and related test methodologies.
| Identifier | ExFiles\Box 147\5\ scan0131 | |
| Date | 8th July 1937 | |
| -2- Rm{William Robotham - Chief Engineer}/CFSll/R.8.7.37. There is no obvious reason why one should be better than the other, and the reason for Vauxhall's choice lies in the fact that Zeniths have the more comprehensive Continental servicing system. We can say, therefore, that the Stromberg carburetter can be adjusted to give any degree of lean mixture for part throttle requirements: the degree of "leanness" being controlled by the capability of the other factors to efficiently cope with it (distribution, spark voltage etc.) Distribution. Our investigations of distribution problems have in recent months been carried out on a "Cambridge Exhaust Gas Analyser". This instrument will give reliable percentage variations in the region of maximum power mixture strengths, but readings are off the calibrated scale for maximum economy mixture strengths. Also, in the case of siamesed exhaust ports, it is impossible to guarantee the recording of a sample of any individual cylinder due to interference from the second exhaust valve in the siamese port. The method adopted by General Motors Research Dept. gets over both these difficulties by the Spark plug temperature method, temperatures being measured by Thermocouples which pass down the centre electrode. At the present time, the demands of an air silencer design are controlling our distribution results by causing a variation of 30% in mixture strength between front and rear cylinders. We know that running with any air silencer on the road, therefore, gives good distribution. We know also that further improvement results from balancing projections that affect the air flow through the carburetter. The degree of "weakening off" allowed by this better distribution is carried in stages until just before the engine starts to miss at light loads full throttle. In the past we have carried out tests under this condition, and the result has been a very small gain. We are now of the opinion, however, in view of more recent results, that there is a definite reason for this small gain from "better distribution" as follows:- | ||