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 calibration of speed and mileage indicators, considering accuracy at different speeds.
| Identifier | ExFiles\Box 59\2\ Scan128 | |
| Date | 23th March 1931 guessed | |
| -8- (infinitely slow mathematically), the mileage record is correct, therefore also the speed indications at slow speeds will be correct. As the instrument is supposed to be internally correct both the mileage and speed indications will fall away with increasing speed according to the parabolic law. It is possible, by increasing the gear ratio of the drive, to make both functions come correct at some one speed of our choice. From the speed indicating point of view we should want this one speed to be fairly high in order that the speed indications at high speed would not be too low by an undesired amount. From the mile counting point of view this would not be very satisfactory as at all speeds lower than this somewhat high speed, mileages would be counted too high. As it is very desirable for mileages to be correctly counted we should, from the mile counting point of view, want this speed to be at the more moderate speed of say 45 miles per hour. This is probably a good speed to select to render the mile counting, on the average, correct. The mile counting below 45 miles per hour would be a little in excess, but above 45 miles per hour would be in defect by greater amounts and it is considered that on the average the speed chosen would be a suitable speed at which the mile counting should be correct. Owing to the fact that this chosen speed would not be suitable for correctness of the speed indicating portion, the argument for the scale contraction to correct the speed indicating portion of the instrument is, in this case, enhanced by the fact that the instrument contains a mile counter whose correctness is desired. It will be seen therefore, that a speed scale contraction on the basis of 3% at least, at 90 miles per hour is desirable. Having effected such a contraction, speed indications will now be correct at every portion of the scale but mile indications will not be affected and therefore we have rendered the instrument internally incorrect except at one speed i.e. at the low speeds the mile indications are high in relation to the speed indications and at high speeds vice versa. There has been, however, a very effective argument for doing this. There is also some amount of argument for over-contraction of the scale by which the speed indications at high | ||