310 BELL SYSTEM TECHNICAL JOURNAL 



over-all eftkiency better than 45 per cent at a pulling figure of 15 mc/s were 

 made. 



The cathode life in these magnetrons was very short — 10 to 30 hours. 

 Reduction of the severity of the demands made of the cathode was necessary. 

 Strapped magnetrons having twelve resonators and larger cathodes were 

 designed and a few built with promising results. 



The 3 cm. magnetron development program was later broadened by the 

 inclusion of personnel from the NDRC Radiation Laboratories at the 

 Massachusetts Institute of Technology and at Columbia University. Effort 

 on magnetrons having twelve resonators was expanded and prosecuted 

 vigorously along with the work on the eight resonator types. Somewhat 

 later the effort was concentrated solely on the twelve resonator magnetron. 

 The design which was made shortly thereafter became, with only minor 

 changes, the 725A. 



Experience gained with previous magnetrons at all wavelengths enabled 

 one to design a mechanically feasible resonator system with proper strapping 

 to give the desired RF characteristics and tt mode resonant frequency. 

 The interaction space, on the basis of previous work and Hartree's oscilla- 

 tion condition, was designed to meet the specification of 12 kv. and 12 amps, 

 input. The design was complicated both mechanically and electrically by 

 the requirement of interchangeability with the 2J21. Thus good operation 

 at 10 kv. and 10 amps, input was necessary. 



The resonator system of the 725 A is machined into a separable "anode 

 insert" which, when completed, is brazed into position in the so-caplled 

 "shell" carr>'ing the leads and cooling fins. The straps are placed in a 

 channel for shielding from the interaction space and are broken on one end. 



The first cathodes were the ordinary, nickel sleeve type with double car- 

 bonate coating sprayed onto its surface. Not long after the start of the 3 

 cm. magnetron development, it was recognized that the cathode develop- 

 ment problem would be a large part of the over-all project. Indeed, the 

 725A proved to be an excellent magnetron for cathode development work. 

 Many of the cathode improvements and new designs developed for it have 

 been used successfully in other magnetrons. The cathode as finally de- 

 veloped for the 725A was a nickel blank, complete with end disks and heater 

 chamber turned out of nickel rod. Over the cylindrical portion was welded 

 or sintered a fine nickel mesh in the interstices of which the active coating 

 was applied. This is shown in Fig. 79 and will be described more fully 

 with other cathode developments in Section 21. Magnetron Cathodes. 

 Under normal operating conditions the cathode heater is turned off, the 

 necessar}' heat being provided by electron back bombardment. 



Considerable effort was expended in the design of a satisfactory and repro- 



