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).
Modifications and solutions for carburettor icing issues on starter units.
| Identifier | ExFiles\Box 123\1\ scan0035 | |
| Date | 1st January 1940 | |
| -3- (1) Cont'd.{John DeLooze - Company Secretary} until the annulus is only .010". To make up for this air restriction we drill two .156" dia. holes directly into the mixing chamber beneath the disc. Thus less than a quarter of this air supply is taken past the disc and it is not of major importance if the annulus becomes completely ice bound. The disc should be tapered towards its edge so as to reduce the possibility of fouling the walls,of the mixing chamber due to the reduced clearance when the carburetter has got into a dirty condition. This type of starter has been fitted to 8.B.V. It cured trouble due to freezing in France, and gave no trouble until brought in by Rm.{William Robotham - Chief Engineer} for difficult starting at 55,925 miles - after 30,000 miles running without attention. (2) When freezing up occurs ice frequently forms in the coils of the spring. With the standard annulus of .0325" this accentuates the freezing trouble, but it does not seem to affect matters to such an extent when air is taken through the two 0.156" dia. holes direct to the mixing chamber. Hence a second attempt at a cure has been to remove the spring from the mixing chamber, extend the needle above the disc and attach the spring permanently outside and above the carburetter. In order to reduce the possibility of freezing at the disc which is now of standard dimension it should be built up in a dome shaped manner, thus increasing the mass of metal and reducing the rate of temperature drop at the edge of the disc. Units of this description are fitted to 7.B.V. and 10.B.V. From a starting point of view neither method introduces new troubles which are not already inherently bound up with the standard starter unit. In each case the carburetter behaves as a starter should behave. Silencing of the carburetter is essentially a design problem. Method (1) may prove more awkward from this point of view, but this method certainly has points in its favour. Come what may, it cannot strengthen up beyond a certain point. Method II reduces the possibility of freezing up but should not be regarded as insurance against it. | ||