Radar started showing some real promise in the late 1930s:
Over the winter of 1936–1937 the Signal Corps Laboratories kept developing new antennas, each one smaller and better than the last. The transmitter antennas enhanced and focused the signal they emitted, while reception antennas became ever more sensitive in picking up faint return signals.
The Signal Corps Laboratories mounted these arrays of metal rods on the chassis developed for the old sound locators, which allowed the antennas to swing and tilt easily to scan the sky. Initially, two receiving antennas were used. A tall narrow one obtained readings for the elevation or height of the aircraft, while a low wide one provided the azimuth or direction of the target. Blair’s engineers also solved the toughest technical challenge of synchronizing the pulses of the transmitter and receiver. The Army Air Corps regularly provided planes for field tests of each new iteration of the equipment.
In May 1937 the Signal Corps Laboratories successfully demonstrated the concept in the field to the secretary of war, senior generals, and several congressmen. The objective of the night time test was to guide a searchlight onto the target so that when the light flicked on the aircraft was already in the beam. The radar set achieved the goal nearly every time, though not entirely on its own as it turned out. Harold A. Zahl, one of the lead civilian scientists working on the project, had noted that one search light in particular was most effective. After the dignitaries departed, he spoke to the corporal in charge. The soldier explained that in most cases he had been able to find the bomber in his binoculars with the aid of a local town’s lights reflecting off the clouds, thus allowing him to precisely direct the searchlight. His purpose had not been to make radar look better, but merely to outdo the aviators in the cat-and-mouse game the two branches habitually played against each other. He allowed, however: “That new secret gadget is all right. Why, every time you fellows turned on the control light it was pretty close to the target — almost as good as my eyes.”
In a used bookstore a couple of years ago, I picked up a copy of an aviation magazine from March 1939. The “Aircraft Radio” column mentions a new invention called a Klystron, used to generate ultra-high-frequency radio waves. Two applications for such frequencies are mentioned: instrument landing systems and the United Airlines-Western Electric Terrain Clearance Indicator. The column mentions that both of these systems had been profiled in earlier issues.
The TCI was a radio altimeter, ie basically radar, albeit radar with a very short range and providing distance but not azimuth information. I would think it would have been an easy mental jump for anyone to go from radio waves for detecting the ground and the distance therefrom, to the use of radio waves for the detection and ranging of aircraft.
Radar detection of aircraft is an easy mental jump, but some of the practicalities of radar detecting distant aircraft, especially from ground stations, are probably not easy to understand unless you just try it and see. Even with a pretty good computer and accurate data on material properties, it’s not so easy to calculate radar cross sections of nontrivial shapes, especially when the wavelength is of the same order of magnitude as the features of the plane or the landscape feature. It’s not ridiculously difficult like various nonlinear partial differential equations that arise in fluid mechanics, but not a calculation that’s easy to do on a napkin either. I think it would be rather tricky to estimate from first principles how much aircraft echoes would stand out from echoes scattered from the landscape.
The practicality of radar reflections for proximity fuzes seems conceptually easy, though. You might not be sure you could make economical rugged electronics that fit in a shell, but you could be pretty sure that if you could make the electronics, the waves should bounce cleanly enough to make a pretty effective weapon.