II 



1314 THE BELL SYSTEM TECHNICAL JOURNAL, NOVEMBER 1956 



Fig. 22 shows the total accelerator and helix interception as functions 

 of collector voltage at various output levels. When there is no rf drive, 

 the intercepted current remains low to a collector voltage of about 200 

 volts at which point it suddenly increases to a high value. This appears 

 to be caused by the phenomenon of space charge blocking. As the col- 

 lector voltage is progressively lowered, the space charge density at the 

 mouth of the collector increases because of the decrease in electron 

 velocity at this point. Increasing the charge density causes the potential 

 depression in the beam to increase until at some collector voltage the 

 potential on the axis is reduced to cathode potential. At collector voltages 

 lower than this, some of the beam is blocked, i.e., it is turned back by the 

 space charge fields. 



When the TWT is operated at appreciable rf output levels, the col- 

 lector voltage must be increased to permit collection of all electrons 

 which have been slowed down by the rf interaction. Unfortunately, some 

 electrons are slowed far more than is the average, so that we must supply 

 to the TWT several times more dc power than we can take from it in the 

 form of rf power. However, as seen from Fig. 22, there is still an apprecia- 

 ble advantage to be gained by operating the collector at lower than helix 

 potential. These curves should not be taken as an accurate measure of 

 the velocity distribution because there are undoubtedly space charge 

 blocking effects which even at higher collector voltages have some in- 

 fluence on the number of electrons returned from the collector. This 

 arises from the fact that the rf interaction causes an axial bunching of 

 the electrons, thereby causing the space charge density in an electron 

 bunch to be much higher than it is in an unmodulated beam. Thus, as a 

 bunch enters the collector, the local space charge density may be high 

 enough to return some electrons. 



IV. PERFORMANCE CHARACTERISTICS 



4.1 Method of Approach 



In this section we will consider the overall rf performance of the 

 Ml 789 and make some comparisons between theory and observed re- 

 sults. The following TWT parameters can be varied: input level; helix 

 voltage; beam current; frequency; and magnetic field. Our approach 

 here will be to first consider the operation of the tube under what might 

 be called nominal conditions. This will be followed by a discussion of the 

 variations in low-level gain and in maximum output over an extended 

 range of beam current, frequency, and magnetic field. By this procedure 

 we are able to obtain a description of tube performance without presenta- 

 tion of a formidable number of curves. Two topics, noise and inter- 



