NATURE OF POWER SATURATION IN TRAVELING WAVE TUBES 853 



pace with the increasing hehx field, so that the net decelerating field 

 still balances to nearly zero. At 4 db below the saturation drive, Fig. 

 7(h), the spur has moved well into the accelerating region, and has been 

 speeded up. The main bunch of electrons is still to the right of the spur, 

 and has been consolidated into a 60° interval. The few electrons in ad- 

 vance of this region evidently no longer find the space charge field suffi- 

 cient to balance the circuit field, and are being decelerated into a second 

 low velocity loop. 



The next three patterns show a continued growth of this second low 

 velocity loop, further consolidation of the 'main bunch', and^the rapid 

 formation of a second bunch in the accelerating field at the end of the 

 spur. It is interesting that at saturation drive, Fig. 7(k) the two bunches 

 are very nearly equal, and in equal and opposite circuit fields, nearly 

 180° apart. The reason for the saturation is that while the main bunch 

 is still giving up energy to the wave, the new one is absorbing energy at 

 an equal rate. The fundamental component of electron current is evi- 

 dently small, and is in quadrature with the circuit field. The current 

 density in the dashed regions is less than 1 per cent of that in the bunches, 

 and probably more than 95 per cent of the electrons are in the two 

 bunches. Two new effects are observable at this level. The second elec- 

 tron bunch has begun to come apart, presumably because of strong lo- 

 calized space charge forces. These forces are also evident in the kink in 

 the velocity pattern drawn by the fast electrons at the same phase as 

 the second bunch. 



Since the majority of the current is in the two bunches at a reduced 

 velocity of 



^^ = -1.1 



2FoC 

 one would expect an output efficiency of 



^ = 2.2C 



The actual measured efficiency 



RF power output 

 DC power input 



was 2.0 C. Under the conditions described, (6) would give 1.4 C. 



At still higher drive levels the pattern continues to develop, electrons 

 from the first bunch falling back into the second, which in turn continues 

 to divide, one part accelerating ahead into a new spur, and the other 



I 



