EARLY TRANSISTOR AND DIODE HISTORY

AT BELL LABS

Art Uhlir Jr.

Oral History – Art Uhlir Jr.

(Continued)

 

Oral History – Art Uhlir Jr.

(Continued)

 

Please provide additional comments regarding varactor diode development at Bell Labs.

 

Varactor harmonic generators were shown to give enough transmitter power for applications unreachable with existing transistors alone, with ominous portent for Bell hegemony in long-distance communication. Receiver protectors using varactors and PIN diodes were developed for military radars. Diode microwave switches had been made with computer diodes, but the diffused-silicon PIN diodes permitted optimizing the compromise between switching speed and power handling.

 

Already by 1951 (in time to be presented at the 1951 Princeton Inn course) W. T. Read (of dislocation fame) had already analyzed the proposition of using avalanche multiplication to produce substantial microwave power (5 watts at 6 Ghz).  Seven years later we had silicon “dimple diodes” made for high power harmonic generation and I thought to put one in a waveguide and bias it into avalanche. Sure enough, microwave power came out.  Read’s theory was published in the same year. [REF 6].

 

Having a sense that Research might be feeling that I was front-running them a bit too much with quick experiments, I called this observation to their attention. What resulted was a professional quantitative study of the phenomenon as it existed, and a journal article. [REF 4].

 

I wanted to get a spectrum analyzer and put in tuners to get a narrow spectrum and more power..  But all spectrum analyzers were tied up in systems development; new ones were more expensive in 1958 dollars than modern ones are in 2004 dollars.  John Moll left for Stanford and Hewlett Packard and I left for Microwave Associates. So it was left to Barney DeLoach of our old department to get significant power out of fairly ordinary diffused silicon computer diodes, a decade later. [REF 5]. Better heat sinks lead to production IMPATT diodes, but they did not seem to have much appeal in the conventional continuous-wave communications systems which were functioning well with the precise frequencies available from varactor harmonic generators.

 

After your pioneering work with microwave diodes at Bell Labs, you left and joined Microwave Associates. What was your position there?

 

At Microwave Associates I came in as Director of Semiconductor Research in 1958 and worked to proselytize the use of junction diodes in microwave devices. The company's total sales of T-R tubes, waveguide circuits, magnetrons, and point-contact diodes was then about $3 million. They were making more money on point-contact silicon diodes than the total for the company. Their MA 1N23E diodes mixer diodes were envied by Western Electric.

 

 

 

 

Go To Uhlir Oral History, Page 10

 

 

COPYRIGHT © 2004 by Jack Ward.  All Rights Reserved.  http://www.transistormuseum.com/

PAGE 9