EARLY HISTORY OF RADAR — PAGE 319 



the author, working under Young's supervision, the task of develop- 

 ing pulse radar. The author's work on this task was started on March 

 14, 1934.2 



The first step was to develop an indicator to display the outputs 

 of transmitter and receiver. A suitable sweep circuit was built for a 

 commercially produced 5-in. cathode-ray oscilloscope. The next step 

 was development of a pulse transmitter. The transmitter frequency 

 of 60 Mc was chosen because that was the frequency then used in the 

 beat method. Pulse length was slightly under 10 ^usec, and pulse 

 spacing, 100 /^sec., these being chosen as appropriate experimental 

 values. The keyer was an asymmetric multivibrator. The antenna 

 was a single half-wave horizontal doublet with a single resonant re- 

 flector. The pulse power was estimated to be between 100 and 200 w. 

 The first question to be resolved was whether echo pulse energy could 

 be detected during the intervals between transmitted pulses, since 

 synchronous detection, characteristic of the beat method, was known 

 to be more sensitive than asynchronous detection, characteristic of the 

 pulse method. Autocorrelation and crosscorrelation were unheard of 

 in those days, and the trade-off between time and bandwidth disclosed 

 by Hartley,^ as well as the significance of average energy, was not 

 too well understood. The only sure recourse was to try it, and skepti- 

 cism was great. A broad-band high-gain experimental communica- 

 tions receiver was borrowed and connected to a second antenna similar 

 to the transmitting antenna. Coupling between the two antennas was 

 appreciable, and the transmitted pulse caused the receiver to ring for 

 30 to 40 yusec. However, when a small airplane flew across the beam at 

 a distance of about a mile, the received signal caused the receiver out- 

 put following the transmitted pulse to fluctuate violently between zero 

 and saturation. This test was completed in December 1934. Although 

 synchronous detection prevailed owing to the transient ringing of the 

 receiver, the great amplitude of the response left no doubt that an 

 asynchronous detector would also have responded to the reflected 

 pulse. The result was accepted as evidence that echo signals could be 

 detected during the intervals between transmitted pulses, and develop- 

 ment of a superior radar receiver was immediately undertaken. 



Radar imposed four severe requirements on the receiver which were 

 not encountered in conventional receivers of the time. Close proximity 

 of receiver and transmitter subjected the receiver to paralyzing over- 

 load, from which recovery to full sensitivity in the incredibly short 

 time of a few microseconds was mandatory. The first design re- 

 quirement was to eliminate grid blocking. This was achieved by using 



^Sehooley, A. H., Pulse radar history, Proc. Inst. Radio Eng. (Correspondence), vol. 37, 

 p. 405, April 1949. 



3 Hartley, R. V. L., Transmission of information, BeH System Techn. Journ., vol. 7, 

 p. 535, 1928. 



672-174—63 23 



