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).
Report comparing the operation of Stewart and Autovac trip valves and detailing improvements made to the Autovac system.
| Identifier | ExFiles\Box 40\4\ Scan137 | |
| Date | 30th October 1925 | |
| Oy1-E-103025 -2- October 30, 1925. half-hearted "trips" in succession. We then looked at the two little trip valves, the "vacuum valve" and the "air valve". In the Stewart the vacuum valve is attached to the operating arm without much slack whereas the air valve has a lot of backlash on the arm. The valves do not move till the device toggles and when this occurs the vacuum valve is almost closed before the air valve opens and the latter is "struck" open by the arm moving at high velocity. On the Autovac both valves move together without much backlash on the arm. Also the toggling device hesitates noticeably on dead centre. Consequently both valves are open together for considerable periods, and valves spend some time in a half-closed and half open position respectively and the engine sucks air, but the vacuum is not completely broken and no "dumping" takes place. We have improved the Autovac noticeably by filing a slot in the arm for the split pin which operates the air valve, delaying its action so that it does not open until the arm is over dead centre. This backlash is about 1/16". We think this is about the limit for the small stroke of the arm on the Autovac. To prove that it is weakening and not strengthening the mixture which causes missing, we find this to be worse on a lean carburetter setting. We have been using a large Stewart tank but will shortly get a small Stewart outfit to fit your tank. If this is equally successful we will send you one to try out. Yours very truly, Maurice Olley MO/E | ||