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).
Comparing the performance of Stewart, Autovac, and Kingston fuel systems, detailing issues with misfiring.
| Identifier | ExFiles\Box 40\4\ Scan172 | |
| Date | 22th December 1925 | |
| -2- Within the next few days we will ship to Derby the Stewart parts which have been in use about 2000 miles on the Phantom at Springfield. These parts can be applied directly to the tank made by R.R. We should be greatly obliged if Derby will arrange to test this device accurately, as they have done the Autovac, and let us know the results. One of the new Stewart tanks, with the modifications listed above, will be sent as soon as samples are received. The difference in action to which we attribute the success of the Stewart tank is that the toggling action is quick and positive, as compared with a slow and hesitating action on the Autovac. Thus we find when the car is idling that the Autovac will operate partially several times before it finally trips over and drops the fuel into the lower tank. Under these conditions the air and vacuum valves are both partly open and rich mixture is drawn from the vacuum tank into the induction pipe. Under similar conditions the Stewart operates instantly and no richening of the inlet mixture is noticeable, or at the most the engine may change its beat for two explosions. We cabled Derby (Attention Mr. Hives) on November 30th:- "Autovac misfiring due to hesitating toggle action Stewart tank fits same flange overcomes misfiring almost entirely." The difference in the mechanism will be evident from an inspection of the sample. Kingston system. We would like to add that we have been discussing an alternative system, the Kingston "Oil-Vac", with the makers. A sample tank was shipped to Derby attention Mr. Hives on Dec. 13th. This system uses a bell syphon to give an intermittent action, thus dispensing with most of the mechanical parts of the Stewart, but having no means of closing off the suction valve at the moment of discharge. They therefore must use a source of vacuum of limited "capacity" to avoid violent agitation of the fuel in the upper tank at the moment of discharge. | ||