MEASUREMENT AND CONTROL OF TEMPERATURE 



(d) Orientation; sensitive mercury thermometers will only give repro- 

 ducible readings in one (usually the vertical) position. 



(e) Remote reading; the instrument is not readily adapted to give the 

 temperature of sealed boxes, especially Hght-tight containers. 



(f) Recording; the instrument does not lend itself to automatic con- 

 tinuous recording. 



(g) Most delicate measuring instruments are themselves delicate; glass, 

 and the hazard of metallic mercury released from a broken thermometer 

 are particular problems. 



(h) The mercury-in-glass system is regularly used as a thermostat element ; 

 it is not easy for the amateur to construct such thermostats. 



Other non-electrical thermometers 



Unfortunately for biologists, almost all the research and development 

 in temperature measurement and control has been carried out industrially 

 for temperatures very much higher than any he is likely to encounter. Of 

 industrial types, the only ones he might find used are the bimetal, in which 

 the differential expansion of two metals is magnified to display temperature, 

 and the gas and vapour-pressure thermometers, which use the expansion 

 of vapour sealed in a bellows as the sensitive element. These devices are 

 of no use in sensitive work; they have the advantage only of robustness, 

 large energy of movement and economy. Their mechanical movement 

 is often directly linked with a recording pen in thermographs, where they 

 have application in crude meteorology. 



Electrical methods 



The thermocouple — When two wires of different materials are joined 

 together at either end, giving two junctions, and these are maintained at 

 different temperatures, an e.m.f. is produced which causes a current to 

 flow round the loop {Figure 29.1). It is considered that two processes 



Metal 1 



r.^ V' 



^v. ^ Metal 2 ^ 



Figure 29.1 



contribute to this potential: the Peltier effect, due to the junctions, and 

 the Thomson effect, due to the temperature differences between the ends 

 of the individual wires. The resultant potential, the sum of these two effects, 

 is approximately linearly related to temperature difference, for the pre- 

 dominant Peltier effect is directly proportional to temperature difference, 

 while the Thomson effect is proportional to the difference in the squares 

 of the temperatures. The universality of the device is realized when it is 

 known that thermocouples are regularly used to measure temperatures up to 

 1,600°C in industry; yet they are equally one of the most accurate ways of 

 measuring temperatures in the biological range, and they can follow small 

 changes rapidly and accurately. 

 The advantages of the thermocouple are: the junctions can be minute, 



386 



