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).
Letter discussing battery ignition condenser tests and comparing test procedures.
| Identifier | ExFiles\Box 34\1\ Scan290 | |
| Date | 6th September 1921 | |
| AMERICAN BOSCH MAGNETO CORPORATION Springfield, Mass. Sept. 6, 1921. Rolls-Royce of America, Inc. Springfield, Mass. Re. Battery Ignition Condenser Tests. Att. Mr. Olley. Gentlemen:- I have at hand your letter of August 17th, addressed to Mr. Rosner, relative to tests on our battery ignition condensers. The tests which you outlined are quite different from, and in some respects are more severe than, those which we make. Our standard breakdown tests are 1000 volts d.c. for three minutes before assembly and for one minute after assembly. The insulation resistance at 500 volts d.c. must be greater than 1.4 megohm at room temperature before assembly and .7 megohm after assembly. Compared with these tests your breakdown test of 450 volts, 40 cycles for 10 minutes gives a maximum voltage of only 636 volts, assuming a sine wave. It does, however, heat the dielectric because of the hysteresis or absorption losses in the paper. These losses are small in the mica condenser and, therefore, cause no heating. With any paper condenser the loss is very much greater. When the varnished paper is heated we have found that the capacity, losses and leakage all increase. The leakage does not increase to such an extent that at 170 or 180°F the condenser is by any means inoperative. A laboratory test at 2000 cycles per second (which we consider near the operating frequency) indicated that the capacity of one particular sample increased from .32 mf. at 70°F to .4 mf. at 190°F.{Mr Friese} At the same time the phase difference increased from 2° to 7.5°, indicating that the loss was greater at the higher temperature. At 190°F.{Mr Friese} the insulation or leakage resistance when the condenser is heated would, of course, prevent its holding a charge as specified in your test. It should be noted that a high temperature test should be applied only when the condenser is assembled in the housing. If the condenser is heated before being assembled it will loosen slightly and thereby lose some of its capacity. We have found by experiment that the condenser operates satisfactorily at high temperatures -- partly because of the increase in the effective capacity mentioned above and partly because a shunt resistance of 1000 ohms may be used in parallel with a contd. | ||