WAVE PICTURE OF MICROWAVE TUBES 1361 



drift tube. At the point where the noise current is a minimum the voltage 

 is "jumped" to the hehx voltage. A second drift tube follows, so that 

 there is a critical distance between the jump and the helLx. 



The effect of this "voltage jump" gun is to deamplify the component 

 of the space-charge wa^'es which is associated with the noise current at 

 the current maximum. In space-charge-wave amplifier terms, this com- 

 ponent sets up the decreasing wave only. Thus, in the second drift tube 

 the ratio | ?max/?min I is smaller than in the first. 



By usmg a single velocity jump, traveling-wave tubes with noise 

 figures around 8 db have been made. 



The use of more velocity jumps has been proposed. It can be shown, 

 however, that as z'max is deamplified, imia must be amplified. This sets a 

 theoretical limit of around 6 db to the noise figure attainable by means 

 of space-charge-wave deamplification alone. 



iNO.I-H"" NO. 2- 



DRIFT TUBES 



Fig. 15 — When a two-potential or "velocity jump" gun is used, the noise figure 

 can be reduced by space-charge-wave deamplification of the noise on the electron 

 stream. 



NOISE CANCELLATION 



It would be highly desirable to build a travelmg-wave tube such that 

 the electromagnetic input would excite an increasing wave, but the noise 

 in the electron stream would excite only some combination of the 

 decreasmg and the unattenuated waves. If we succeeded in this, the 

 noise introduced by the tube could be made as small relative to the 

 signal as desired, merely by making the tube long enough. Can we ac- 

 complish this by means of some special structure near the input end of 

 the tube? 



We can represent the noise on the electron stream at some reference 

 point by means of a velocity fluctuation v and a current fluctuation i; we 

 have seen that neither can be zero. Because the system is linear, super- 

 position applies, and the amplitudes of the growing, attenuated and 

 unattenuated waves which are excited are the sums of the amplitudes 

 excited by i and v independently. 



Suppose that v = 0. Then the beam carries no power. Thus, i cannot 



