THE MEASUREMENT OF CURRENT 47 



as possible without touching it, is a heater wire through which 

 the current to be measured is sent. On the passage of the cur- 

 rent the lower thermal junction is warmed by radiation and 

 convection; consequently a direct current, due to the thermo- 

 electric action, flows around the loop which is thus deflected. 

 The heater filament is straight or else bent back and forth form- 

 ing a grid. Its inductance is therefore very small and the 

 instrument is consequently adapted for the measurement of 

 alternating currents of high periodicity, the deflection, which is 

 read by the mirror and scale method, being practically propor- 

 tional to the mean square value of the current through the heater. 

 The damping is due to currents induced in the loop as it moves in 

 the field and the electrical constants are such that critical damp- 

 ing is attained. The period of the instrument is 3 or 4 seconds. 



As with any thermo-electric device constancy of zero reading 

 depends on uniformity of temperature. Sudden fluctuations 

 in room temperature should be avoided. Slow variations which 

 give time for the temperatures of the hot and cold junctions to 

 equalize are not nearly as important. To assist in maintaining 

 constant temperature conditions, the working parts of the instru- 

 ment are enclosed in a heavy gun-metal case, the front of which, 

 E, may be removed when it is necessary to inspect or adjust the 

 instrument. The zero should be read after each observation. 



Interchangeable heaters are used. They are of various 

 resistances depending on the sensitivity required. Those having 

 a resistance below 4 ohms are made of wire, while those above 

 this value consist of a deposit of platinum on quartz, made into 

 the form of a grid. 



The sensitivity attained, as given by the makers of the instru- 

 ments, the Cambridge Scientific Instrument Co., is shown in 

 the following table. 



The instrument may be calibrated with direct currents and 

 then used on alternating-current circuits. It is more sensitive 

 than the electrodynamometer, not subject to errors due to in- 

 ductance or capacity, and at high frequencies does not disturb 

 the circuit conditions as much as the dynamometer. 



The sensitivity may be controlled to a certain extent by ad- 

 justing the proximity of the heater to the hot junction. This 

 is done by turning the ebonite milled head, F. Great care is 



