120 



BELL SYSTEM TECUM CAL JOURNAL 



sideband potential as a function of one of the modulating potentials 

 with the other as parameter, as shown in the dotted lines of Fig. 5. 

 The experimental data are plotted as the full lined curves and appear 



28 

 26 

 24 

 22 

 20- 



16 14 12 10 8 6 



10 12 14 16 18 20 



Fig. 4 



to be in good agreement with the theory, the divergence being due 

 presumably to the incomplete suppression of the positive lobe in the 

 experimental set-up. The measurements were carried out with the 

 current analyzer as described in connection with Fig. 3, grid current 

 being measured as a function of the two input amplitudes. The 

 sideband grid potential was then determined by multiplying the 

 sideband current by the external grid resistance. 



Possibly the most striking thing shown by Fig. 5 is that the sideband 

 amplitude is independent of the larger of the two inputs, when the 

 ratio of one input to the other is made sufficiently great. Hence we 

 must provide sufficient carrier amplitude to insure that the resultant 

 sideband shall be linearly proportional to the impressed signal up to its 

 greatest value so that good quality of speech transmission may be 

 assured. 



Our earlier analysis using eq. 5 to represent the grid current charac- 

 teristic led to a sideband amplitude proportional to the product of the 

 two inputs whereas in this case in which the positive lobe is completely 



