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 the design and manufacturing challenges of a bellows thermostat, highlighting issues with adjustment, tolerances, and the lack of a movement stop.
| Identifier | ExFiles\Box 156\1\ scan0070 | |
| Date | 10th April 1935 guessed | |
| - 2 - Messrs. Rolls-Royce Limited. unmixed charges. Having decided which charge is to be employed, the movement obtained for a given temperature rise is determined by the effective area and the spring rate of the bellows employed. While the effective area can be calculated from a knowledge of the bellows dimensions, and the spring rate can be be measured before the bellows is assembled, some tolerance either in temperature or dimensions must be allowed if the completed unit is to pass the final test. The design which you have prepared is entirely lacking in adjustment, that is to say, once the bellows is made and charged it is either right or wrong, and if it is wrong there is no method of correcting it. To produce bellows having an exact mean area and an exact spring value is almost an impossibility by any known method of manufacture, and units of the type under discussion can only be made by trial and error, which you will agree is unsatisfactory from every point of view. The design provides no stop for the bellows movement in the extended condition, and if the units now being returned to you are heated to 93ºC. without the movement being checked, the bellows will be stretched and the temperature at which movement commences will be lowered. Will you please let us know whether a stop is provided to prevent the bellows length exceeding 1.525" and also whether the six units originally despatched to you were allowed to extend beyond this length due to increase in temperature? In the commercial manufacture of thermostats it is the almost invariable practice to arrange for the spring rate of the bellows to be negligible compared with that of the spring which controls its movement. In such a thermostat adjustment of the temperature at which movement commences is made by increasing or decreasing the tension of the controlling spring, and adjustment of the final temperature at which the total required movement occurs is made by increasing or decreasing the rate of the controlling spring by manipulation of the gauge or the number of coils. | ||