A STATIC RECORDER 2SS 



the pointer caused by the individual static crashes. The time con- 

 stant RC of the circuit would probably have to be at least five seconds 

 and the speed at which the gain of the set is changed must be cor- 

 respondingly slow. This system will, therefore, react very slowly for 

 great changes in static level. Ordinarily static does not change very 

 fast but if the recorder is working with a directive antenna system 

 that is rotated say 360° in 20 minutes, then a fairly fast working 

 recorder is desirable. The main disadvantage of the continuous sys- 

 tem is that it will be difficult to make the meter give a true indication 

 of the average energy level. 



The method employed is therefore based upon a discontinuous 

 system and will be described in connection with Fig. 2. The rectified 

 output current of the set is integrated over a period of 10 seconds by 

 means of a fluxmeter.^ If, after these ten seconds the fluxmeter 

 deflection is below a certain mark, then the gain of the set is increased 

 one step and, vice versa, if the deflection is above a certain mark the 

 gain is decreased one step. For deflections in between these marks 

 the gain remains unchanged. To change the gain one step and to 

 bring the fluxmeter needle back to zero takes approximately 5 seconds 

 after which the whole process is repeated. The output energy due to 

 static received during ten-second periods is here kept within two 

 definite limits. The gain can be changed only one step after each 

 period, but since each step corresponds to a change of 4 TU (1.58 

 times) in voltage gain it will take only one minute and a quarter for 

 the recorder to adjust itself to a sudden change of 100 times in the 

 energy level of static. 



The Apparatus of the Recorder 



The receiving set is shown schematically in Fig. 2. It is an ordinary 

 double detection set that requires altogether ten tubes, of which the 

 last low frequency amplifier tube must be able to handle 10 watts in 

 order to prevent overloading. The power supply may be rectified A C. 



The gain control is inserted in the first intermediate frequency am- 

 plifier in order to be sure that no tubes are overloaded. The local 

 oscillator shown is used for amplification calibration of the set and 



'•• A galvanometer with negligible restoring torque, whose deflection is proportional 

 to the coulombs sent through it. Its use for the present purpose was suggested by 

 Mr. 1.. J. Sivian of the Bell Telephone Laboratories. He has employed the instru- 

 ment for measurements of rectified speech and noise currents on telephone circuits. 

 The use of a fluxmeter for similar purposes has been indejwndcntly reported by Dr. 

 E. M. Terry, of the University of Wisconsin, at the Dccenibcr 30, 1924, meeting 

 of the U. R. S. I. at Washington, D. C. 



