338 



BELL SYSTEM TECHNICAL JOURNAL 



required by the frequency measuring device was supplied through a 

 tuned buffer amplifier. 



The voltage developed across the tuned circuit of the mixing am- 

 plifier was measured by a conventional form of vacuum tube voltmeter. 

 By setting one attenuator at a very high loss, the magnitude of the sig- 

 nal supplied through the other could be measured, and the process then 

 reversed. If the two signals were adjusted so as to give equal ampli- 

 tudes across the tuned load, then any desired carrier ratio could be 

 obtained by adding a known loss in one attenuator. 



NO.700A 

 OSCILLATOR 



TUNED 



RADIO 

 FREQUENCY 

 AMPLIFIER 



VACUUM TUBE 

 VOLTMETER 



/ 



RADIO' 



FREQUENCY 



ATTENUATORS 



s 



RADIO RECEIVER 

 .(SQUARE LAW 

 /i DETE CTOR) 



RADIO RECEIVER 

 (LINEAR DETECTOR) 



NO.IA 

 FREQUENCY 

 MONITORING 



UNIT 



OS£i) 



RADIO 

 FREQUENCY. 

 AMPLIFIERS 



LOUD 

 SPEAKER 



1 



a 



RADIO 

 FREQUENCY 

 ATTENUATOR 



AUDIO 

 FREQUENCY 



AMPLIFIER 



VOLUME 

 INDICATOR 



Fig. 1 — Schematic circuits of experimental setup. 



The mixing amplifier fed a shielded transmission line which included 

 an adjustable pad. The line supplied energy to either of two radio 

 receivers, one of which contained a square law and the other a linear 

 detector. The output of the receiver was monitored on a loud speaker 

 and also on a volume indicator. Meters were provided for indicating 

 the change in direct current flow in the detector circuit of both receivers. 



In order to study the effects of a noise background, a noise source 

 of constant and controllable level was required. Furthermore, it was 

 desirable that the noise be of a type frequently encountered in practice. 

 The thermal noise generated in a high gain amplifier seemed to be 

 suitable. Consequently, there were connected in cascade two ampli- 

 fiers having a gain of approximately 44 db each, over the entire broad- 



