The Bell System Technical Journal 



Vol. XXV April, 1946 No. 2 



The Magnetron as a Generator of Centimeter Waves 



By J. B. FISK, H. D. HAGSTRUM, and P. L. HARTMAN 



Introduction 



LATE in the summer of 1940, a fire control radar operating at 700 mega- 

 cycles per second was in an advanced state of development at the Bell 

 Telephone Laboratories. The pulse power of this radar was generated by a 

 pair of triodes operating near their upper limit of frequency. Even when 

 driven to the point where tube life was short, the generator produced peak 

 power in each pulse of only two kilowatts, a quantity usable but marginal. 

 Although the triodes employed had not been designed for high voltage pulsed 

 operation, they were the best available. This is an example of how develop- 

 ment of radar in the centimeter wave region was circumscribed by the lack 

 of a generator of adequate power and reasonable life expectancy. More- 

 over, the prospects of improvement of the triode as a power generator at 

 these wavelengths and extension of its use to shorter wavelengths were not 

 bright. Solution of the problem by means of power amplification was 

 remote. A new source of centimeter wave power was urgently needed. 



For the British, who were at war, the problem was even more urgent. 

 They had undertaken a vigorous search for a new t>TDe of generator of 

 sufficiently high power and frequency to make airborne radar practicable 

 in the defense against enemy night bombers. They found a solution in the 

 multiresonator magnetron oscillator, admirably suited to pulsed generation 

 of centimeter waves of high power. 



In the fall of 1940, an early model of this magnetron operating at ten cen- 

 timeters was brought to the United States for examination. The first Ameri- 

 can test of its output power capabilities was made on October 6, 1940 in the 

 Bell Telephone Laboratories' radio laboratory at W'hippany, N. J. This 

 test confirmed British information and demonstrated that a generator now 

 existed which could supply several times the power that our triodes delivered 

 and at a frequency four times as great. The most important restraint on 

 the development of radar in the centimeter wavelength region had now been 

 removed. 



A number of pressing questions remained to be answered, however. 

 Could the new magnetron be reproduced quickly and in quantity? Was its 



167 



