Lo.ii) c.ih'/x'ijxc; c.U'.icrrv oi- ami'i.iiii.rs 



391 



To check wiu'thcr similar results would be obtained at other fre- 

 (jueucies and with reaclixe loads oscillograms were taken with output 

 impedances ha\ing large positi\-e and negative phase angles at fre- 

 quencies of 200 c.p.s. and 1,000 c. p. s. While the width of the ellipse 

 obtained varied greatly as was to be expected, it was found that the 

 points where marked irregularities in the figures occurred were at the 

 same grid excitations as in the case of the figures taken at 1,000 c.p.s. 



Input Voltage > 



Fig. 18— Amplifier No. 4; 1000 c.p.s.; load 7,(X)() ohm positive reactance 



with resistance load. Figures were also taken using the output current 

 to deflect the electron stream magnetically with the same results. 

 Typical oscillograms for these varying conditions are reproduced in 

 Figs. 16, 17 and 18. 



In conclusion, the load carrying capacity of an amplifier may be 

 determined by either method with approximately the same results. 

 An harmonic analysis reveals in detail the amount of each harmonic 

 that is introduced at any load and gives useful data for fimdamcntal 



