[ 713 ] 



LXVI. On Direct Heading Resistance- Thermometers, with a 

 Note on Composite Thermocouples, B// Albert Campbell, 

 B.A.* 



H^HE measurement of high temperatures is a matter of 



JL growing importance both in scientific and technical 

 work, and thanks mainly to Prof. Callendar's work on the 

 subject, platinum resistance-thermometers at present hold 

 the foremost place for accuracy in such measurements. In 

 their mo>t usual form, however, resistance-thermometers do 

 not indicate the temperature directly, but it has to be deduced 

 from the reading by means of a formula, table, or curve. 

 In order to get rid of this trouble. I have recently invented 

 two simple arrangements, by either of which the reading of 

 a re-istance-box, in connexion with a resistance-thermometer, 

 gives directly the actual temperature to a good degree of 

 accuracy. 



I shall designate the two methods : — 



(A) The Rectifying Shunt Method, 

 and (B) The Rectifying Loop Method. 



I have named them thus, because the result of using the 

 device is to reduce to a straight line the curve connecting 

 the box-reading with the temperature. I shall first describe 

 (A), which is the simpler but the less accurate of the two. 



(A) The Rectifying Shunt Method. 



In this method the arrangement is extremely simple, and 

 consists merely of the addition of an appropriate shunt to 

 one of the arms of the Wheatstone's bridge used to measure 

 the resistance. Before I show how this shunt is applied, let 

 us consider the connexion between resistance and temperature. 

 For platinum, the formulae ordinarily used are equivalent to 

 the following 



where R and R^are the resistances at 0° C. and t respectively, 

 and a and (3 are the constants. As to actual values, a. and j3 

 vary slightly in different specimens, a being of the order of 

 0*003, while £ is near 0-0000005. This parabolic formula 

 has been found to hold with considerable accuracy over a 

 wide range of temperature (up to 1000° (J.) f. For sim- 

 plicity we may for the present make E equal to 1 ohm. 



* Communicated by the Physical Society : read March 10, 190& 

 . example, .J. A. Barker, Phil. Trans, vol. cciii. p. 343. 



