MAGNETRON AS GENERATOR OF CENTIMETER WAVES 275 



In mechanical construction the 700A-D magnetrons involved techniciues 

 like those described above. The input and output leads included copper 

 to glass to tungsten seals much like those in the reproductions of the British 

 magnetron. The end covers were sealed to the resonator body by means 

 of the gold ring technique employed in the British magnetron. 



The 700A-D magnetrons are limited in frequency to the four 10 mc/s 

 bands between 680 and 720 mc/s, respectively. These magnetrons operate 

 at 12 kv. and 8 amps, peak current input at a magnetic field of 650 gauss. 

 Over-all efficiency ranges between 30 and 40%, which is better, as has been 

 explained, than that attained with unstrapped 10 cm. magnetrons. Other 

 data of interest are given in Table I. 



One feature which is immediately apparent from the rated operating 

 conditions of the 700A-D magnetrons is the fact that the ratings are not 

 nearly as high as one might expect from the size of the magnetron. Back 

 bombardment of the cathode at considerably greater input power could 

 easily be handled. The difficulty lay in the fact that it was impossible to 

 drive the magnetrons in the tt mode to much greater currents than the rated 

 currents. If the attempt is made to drive the magnetron harder it either 

 refuses to oscillate at all or oscillates in another mode. This phenomenon 

 has been the single deterrent in the development of higher power mag- 

 netrons at wavelengths greater than 20 cm. It is now recognized as a 

 starting time phenomenon having to do with the rate at which oscillation 

 builds up and the rate at which pulse voltage is applied (see Section 10,6 

 Oscillation Buildup- — Starting). What has been done in studying the 

 phenomenon and in magnetron design to circumvent it will be discussed in 

 some detail in connection with the 5J26, the tunable replacement for the 

 4J21-30 series. 



In quantity production the 700A-D magnetrons presented new problems, 

 all of which arose because of its size. The oxide coated cathode, having a 

 relatively large surface area, gave off a considerable quantity of gas during 

 cathode activation. In as much as the massive copper anode could be out- 

 gassed only by a long baking process at temperatures below the softening 

 point of the glass parts, difficulty with magnetrons "going soft" after seal-ofif 

 was encountered initially. 



The development of the 700A-D magnetrons was carried on simulta- 

 neously with early studies at 10 cm. and with the early attempts to produce 

 power at 3 cm. A number of auxiliar>^ experiments were undertaken which, 

 although they were not a part of the specific magnetron development, con- 

 tributed results of considerable value complementary' to those obtained 

 at the shorter wavelengths. In particular, these experiments had to do 

 with the technicjue of measurement and of magnetron scaling. 



