The Radar Crystal

1942 · Silicon Microwave Mixer Diode · Work 812 of 1,871

What it was used for

Silicon point-contact cartridge diodes served as the first detector in radar receivers for UHF and SHF. Per the March 1946 article in Radio-Craft the frequencies of the newest radar equipment were beyond what ordinary vacuum diodes could do: tubes were limited by electron transit time, klystrons produced extremely high thermal noise, while in the crystal diode transit time is no factor and noise figures are very low. The silicon crystal diodes developed for radar were designed to operate at 1 or 2 mA. The 1N21 cartridge held an S-shaped point-contact assembly in one end plug and a silicon chip in the other; Western Electric made it from early 1942 for radar systems.

What was new

What was characteristic was not the construction but the material: the point contact of 1906, displaced by valves in the early 1920s, came back because the newest equipment needed it. The new part was purification. Per the Computer History Museum, silicon of 99.999 per cent purity was available by the end of the war, from work mainly at the University of Pennsylvania under Frederick Seitz and at DuPont under MIT's Radiation Laboratory; germanium was worked on mainly at Purdue under Karl Lark-Horovitz. The barb: the whisker stayed vulnerable - bump the diode so the whisker moves and performance degrades or it fails altogether.

What the sources leave open: Quantities remain vague. The Transistor Museum states that production began in early 1942 and was ramped up to over 1,000,000 units by August 1945; the Computer History Museum pages speak only of "millions" of crystal rectifiers. That is not a contradiction but differing precision. None of the sources opened gives a frequency or wavelength figure for the 1N21.

Connections

Every connection names its kind. “Forerunner” on its own would be a claim.

1. Rediscovered inThe Cat's Whisker1906Once amplifying valves took over, the crystal detector remained "primarily as an historical curiosity and hobbyist device". Centimetric radar brought it back: at wavelengths of a tenth of a metre vacuum tube and copper oxide rectifiers were useless and crystal detectors "proved to be the only hope". Millions of silicon diodes such as the 1N21 were built in the 1940s.Source: Transistor Museum: History of Crystal Diodes, Volume 1 · Michael Riordan, Lillian Hoddeson: The origins of the pn junction, IEEE Spectrum, June 1997
2. Follows fromThe Cracked Silicon Rod1940Radar receivers' silicon detectors followed Ohl's silicon work. Every few months Ohl got several pounds of ultrapure du Pont silicon through the Rad Lab; Ohl, Scaff and Theuerer worked out how impurities make n- and p-type silicon; Scaff developed mass production of p-type silicon for Western Electric. Scaff: "We were sending samples to them, as well as samples of our detectors."Source: Michael Riordan, Lillian Hoddeson: The origins of the pn junction, IEEE Spectrum, June 1997 · Computer History Museum, The Silicon Engine: 1941 - Semiconductor diode rectifiers serve in WW II

Further reading

• H.F. Crystal Diodes, Radio-Craft, März 1946 (RF Cafe)
• 1941: Semiconductor diode rectifiers serve in WW II, Computer History Museum
• Transistor Museum, A Brief History of Early Semiconductors (PDF)
• 1N21 Crystal Rectifiers & Related, prc68.com