DYNAMIC MEASUREMENTS ON ELECTROMAGNETIC DEVICES 1425 



(e) No effect other than one which may be accurately calculated on 

 the dc reading. 



(f) The circuit should not greatly increase the time required for the 

 instrument to reach a steady reading. 



Requirement (c) may call for some comment. It was pointed out earlier 

 that the method requires the time constant of the measuring circuit to 

 be small in comparison with the period of the pulses. A Hlter built in the 

 ordinary way to have high suppression to pulses of period T would not 

 have a time constant small in comparison with this figure. The recjuire- 



Fig. 6 — Vacuum tube low pass filter circuit. 



ment of constant resistance input is a convenient way of expressing the 

 necessity of fulfilling this condition. 



The original work on the system used was done by R. F. Wick, and 

 the featm-es to be described are due to him and E. L. Norton. From left 

 to right in Fig. 6, the elements are as follows: a balanced impedance 

 bridge containing the ammeter in one arm, a blocking condenser, a con- 

 stant current high impedance power supply for the power tube, and a 

 two stage amplifier with an interstage phase adjusting circuit. The three- 

 point double-pole key when in the normal position removes the meter 

 (50 ohms) and substitutes a 50-ohm resistor. When off-normal, the meter 

 is connected in either polarity. 



The operation of the circuit is best understood by assuming the re- 

 actance elements to be omitted from the bridge and the contact on the 

 potentiometer forming one diagonal to be at the lower left. The input to 

 the amplifier is then directly across the line and any feedback is elimi- 



