1288 THE BELL SYSTEM TECHNICAL JOURNAL, NOVEMBER 1956 

 II. DESIGN CONSIDERATIONS 



While TWT theory served as a general guide in the development of 

 the MI789, a number of important tube parameters had to be determined 

 either by experimentation or by judgement based on past experience. 

 The most important of these were : 



Saturation power output 12 watts 



Mean helix diameter 90 mils 



7a --^ 1.6 



Magnetic flux density 600 gauss 



Cathode current density '^ 200 ma/cm^ 



These quantities and the requirement of 30-db gain at five watts output 

 largely determined the TWT design. 



The saturation output of 12 watts was found necessary to obtain the 

 desired linearity at five watts output and the 7a value of 1.6 to obtain the 

 flattest frequency response over the desired band. 



The choice of helix diameter and magnetic flux density represented a 

 compromise. For the highest gain per unit length, best efficiency, and 

 lowest operating voltage, a small helix diameter was called for. On the 

 other hand, a large helix diameter was desirable in order to ease the 

 problem of beam focusing and to facilitate the design of a light-weight 

 permanent magnet focusing circuit. In particular, the design of such a 

 circuit can be greatly simplified if the field strength required is less than 

 the coercive force of available magnetic materials. This allows the use of 

 straight bar magnets instead of much heavier horseshoe magnets. More- 

 over, the size and weight of the magnetic circuit is minimized by employ- 

 ing a high energy product material. These considerations led us to choose 

 a flux density of 600 gauss, thereby permitting us to design a magnetic 

 circuit using Alnico bar magnets. 



To obtain long tube life we felt it desirable to limit the helix intercep- 

 tion to about one per cent of the beam current. On the basis of past 

 results we estimated that this could be done with a magnetic flux density 

 2.6 times the Brillouin value for a beam entirely filling the helix. With 

 this restriction, Fig. 2 shows how the TWT design is affected by varying 

 the helix diameter. A choice of 600 gauss is seen to result in a mean helix 

 diameter of 90 mils. 



In the selection of cathode current density, a compromise between 

 long life and ease of focusing had to be made. To obtain long life, the 

 current density should be minimized. However, this calls for a highly 

 convergent gun which in turn complicates the focusing problem. We 

 decided to use a sprayed oxide cathode operating at about 200 ma/cm^. 

 Experience with the Western Electric 41 6B microwave triode had shown 



