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).
Experiments on reducing dielectric loss in coils by improving insulation materials and design.
| Identifier | ExFiles\Box 45\4\ Scan063 | |
| Date | 9th November 1922 | |
| -4- EFCL/T9.11.22. dielectric loss. This applies more particularly to the Elo cases which have so far been standard. It is anticipated that we may effect an improvement either with American bakelite or mouldensite, in fact we already have evidence that mouldensite is definitely better. Another way in which we effected an improvement was a reduction of the number of turns of the secondary with improved insulation. We put down the improvement to the fact that there is more real insulation between the outside of the secondary winding and the inside of the bakelite case, the air space which, from this point of view, being ionised by the H.T. discharge, would act as a conductor. This was proved by the fact that the inside tin foil connected to the H.T. gave the same result. A further experiment has been a 16000 turn coil with this extra insulation (but still allowing about 1/8" air gap) in a mouldensite case, and this has been quite a good improvement as the engine ran with bonnet and radiator shutters closed for 1 3/4 hours before failure occurred. Yet another experiment consists in filling the vacant space with a suitable mixture of paraffin wax and resin. This dielectric will have a smaller inductivity and also a smaller loss, therefore the high potential will not so easily be transferred to the inside of the bakelite case and it would be anticipated that a much reduced loss would occur. See later p. 7 for result. The phenomenon is not readily produced with other makes of coils, e.g. Remy, though it was produced under extreme conditions with a U.S.A. Bosch coil with ballast as Contd. | ||