Abstract :
The paper is an investigation into the possibilities of using an indirectly heated thermistor as a precise a.c./d.c. transfer device over a wide range of frequencies. The thermal drift inevitable with such a device has been examined and a procedure for its elimination developed, resulting in a very high degree of thermal stability. Experiments made on the reproducibility of the thermistor seem to indicate a complete absence of hysteresis over the working range. It is shown that the device has an extremely high sensitivity and is therefore useful for general laboratory work with simple equipment, although this advantage is somewhat outweighed by the slow response. Experimental results are given for the d.c. reversal errors due to Peltier and Thomson effects; these are found to be quite small. An analysis of the effects of frequency shows that, owing to the large thermal time-constant, the indirectly heated thermistor in its present form can be used at frequencies as low as 0.2 c/s, thus well below the working range of normal measuring devices. The response is very slow owing to this large time-constant, and methods of con struction are suggested which would improve the speed of response at power frequencies and above.
