336 BELL SYSTEM TECHNICAL JOURNAL 



support, the other, high. To meet each requirement satisfactorily with 

 the first cathode structure was found ver>^ difficult if not impossible. 



A major step forward was taken in the design of the so-called "soldering 

 iron" cathode in which the heater element is placed in a larger part of the 

 cathode lead, heat being conducted to the cathode surface through a solid 

 rod. The cathode itself is solid, making for greater rigidity and lower 

 thermal impedance. The heater element is placed where it may be made 

 considerably larger and more rugged than if placed inside the cathode, whose 

 diameter is 0.096 in. A view of the 3J21 magnetron cathode structure is 

 to be seen in Fig. 78. The heater is contained within the section of larger 

 diameter immediately adjacent to the cathode. 



The "soldering iron" cathode presented more difficult problems of heat 

 conduction and dissipation than the older design. For an operating tem- 

 perature of 800°C at the cathode surface, it is necessary to heat the heater 

 chamber to about 1000°C, whereas the necessar>' activation temperature of 

 1050°C requires the temperature of the heater chamber to be about 1300°C. 

 As a result, rather careful design to provide the proper balance between 

 heat losses by conduction and radiation was necessary. Radiation losses, 

 which are the more important type, are increased by extending the cathode 

 rod considerably beyond the active surface, as seen in Fig. 78. The cathode 

 should also have good thermal conducting properties in this extension, 

 throughout the main cathode body, heater chamber, and support. Under 

 normal operating conditions the heater is turned off, cathode heat being 

 suppHed solely by back bombardment. 



It is apparent that a cathode design of this type calls for a careful choice of 

 materials. They must be highly refractory as well as of good thermal 

 conductivity and structural strength. Copper, silver, and nickel do not meet 

 all of these requirements. The first cathodes of promise were turned from 

 solid stock of 30-called machinable molybdenum, complete with cathode 

 end disks and heater chamber. The heater end was brazed to a Kovar 

 detail, subsequently welded to the support cone of the type developed for 

 the 4J52 magnetron. However, molybdenum machines poorly. The toler- 

 ances on size did not permit of large scale production of precision molyb- 

 denum parts. Consequently it was proposed to make the cathode of 

 tungsten rod which could be ground to shape even on a production basis. 

 The cathode end disks were to be punched or turned from molybdenum or 

 Kovar and brazed to the tungsten rod. The tungsten rod extends into a 

 hollow molybdenum heater chamber, the dimensions of which are not criti- 

 cal. Inside the heater chamber the heater coil encloses the protruding 

 tungsten rod so that better heat conductivity from the heater to the cathode 

 is provided. The heater is brazed to the cathode at one end in the braze 

 between the tungsten and molybdenum. The other end is spot welded to 



