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).
The derivation and methodology of motor torque measurement, including battery conditions and test conclusions.
| Identifier | ExFiles\Box 145\1\ scan0128 | |
| Date | 1st September 1933 guessed | |
| 7. Derivation of Torque. The starting torques at the motor shaft were determined by a bench calibration of the motor at the same temperature and with the first battery used in the tests of Table 11. In this calibration, torques developed by the motor when geared down through a gear ratio 11/120 were measured on a Salter spring balance, and the actual torque at the motor shaft during the tests was then calculated from the known gear ratio of the starter. This method of calibration does not allow for the torque lost in the gearing, but it has the advantage that the motor runs during calibration under conditions which approximate closely to those obtaining when starting the car engine in practice. A curve giving the results of the calibration of the motor, as described above, is attached hereto, and the figures in column 12 of Table 11 have been derived therefrom. Discussion of method. (i) Motor torque. It will be observed that the method of arriving at the motor shaft torque does not properly credit the motor with the torque loss in the gearing. It must be remembered, however, that a similar (though not necessarily equal) loss occurs when the starter is on the car, and on these grounds the figures given are justifiable. (ii) Battery condition. The tests presuppose that the condition of the battery remains substantially unchanged throughout. This assumption is not fully justified, as battery fatigue or change of internal resistance might produce slight alterations in acceleration etc. But any error due to such changes would be small and certainly does not in these tests invalidate the general comparative nature of the results. (iii) Averages. The speeds were taken by a Hasler tachometer and thus were averaged over the starting period. More accurate information could be obtained by reading instantaneous speeds, with corresponding readings of the instantaneous voltages and currents. The apparatus, however, for such accuracy would be exceedingly complicated and expensive, and it is very doubtful whether, for such purposes as the present tests, the results obtained would justify so much elaboration. Conclusion. The readings will be found to be consistent throughout, although some of the results are perhaps unexpected. For example, in every instance the current and torque figures are lower with the clutch in than with the clutch out. This might be taken to mean that when the clutch is out some extra friction loss is introduced, due perhaps to the clutch disengaging mechanism; but this is for the makers of the chassis to explain. There is no doubt as to the facts. We are satisfied, after observing these tests throughout in the closest detail, that the results obtained are representative of practical conditions, and we are glad to remark upon the scrupulous care which the Vacuum Oil Company's engineer took at all times to ensure that the test and running conditions were identical for all the grades of oil tested. | ||