NEGATIVE FEEDBACK 



425 



input from the noise generator was obtained. In the region where the 

 effect of receiver tube noise was evident the assumption of a linear rela- 

 tionship between disturbance level and output noise permitted correc- 

 tion of the curve so that equivalent disturbance levels could be related 

 to any setting of the noise attenuator, or to the receiver output noise 

 level without feedback. 



The signal-to-noise ratios at the output of the receiver without feed- 

 back are considerably higher than the corresponding ratios of carrier 

 and disturbance levels existing at the amplitude detectors. This is the 



70 



Z 60 



"D 50 



O 



LJ 



O 40 



30 



8 10 12 14 16 18 

 FEEDBACK IN DECIBELS 



20 22 24 26 



Fig. 12 — Composite of data from Figs. 9 to 11 expressing ratios of harmonic levels to 

 fundamental level. Curves (a) from Fig. 9, (6) from Fig. 10, and (c) from Fig. 1 1 . 



result of two factors. The intermediate-frequency wave is modulated 

 to the extent of i 7 kilocycles while the range of frequencies appearing 

 at the output terminals of the receiver is limited to 4 kilocycles. 

 Hence at the output of the balanced detector the noise level in the 

 absence of modulation is 9.6 decibels below that which would be ob- 

 served at the output of an amplitude-modulation system. Further- 

 more the admittance characteristic of the complete intermediate-fre- 

 quency system is such that the effective disturbing voltage delivered to 



