EARLY TRANSISTOR HISTORY AT RCA

Norman H. Ditrick

 

Oral History – Norm Ditrick (Continued)

 

There was another engineer, Tracy Kinsel, who was involved in growing the gallium arsenide for us.  These crystals were very hard to grow early on and no one knew how to do it.  Later on, we did have some success using gallium arsenide for diodes and for tunnel diodes.  My main effort with tunnel diodes was with germanium.  We made a quite a few of these and the germanium ones performed pretty well – they were intended for operation up to one GHz, and the intended application was military computers. 

 

There were problems with using tunnel diodes.  Of course, they were two terminal devices and were very hard to use in circuits.  These were difficult devices to make, because the current density was extremely high and so they had to be very tiny – this small size meant they were fragile.  Germanium really was the best material. Silicon was a poor choice because of the ratio of the peak to valley current.  Gallium arsenide gave better performance than germanium, but it was extremely unstable, so it really couldn’t be sold.  I think tunnel diodes were sold mostly in the early 1960s and the high frequency performance was the main selling point.  What killed them was the development by (I believe)  Western Electric of some silicon transistors that were very high frequency devices.  I think these were mesa types with evaporated contacts.  The tunnel diode program was a large one at RCA for a while, and quite a bit of circuitry was developed.  

 

      

 

 

Oral History – Norm Ditrick (Continued)

 

After my tunnel diode work, RCA finally started with silicon transistors, several years behind everyone else.  Silicon requires the use of photo patterning in order to perform very fine work on the top surface.  This was the beginning of photo resist which is now used by everybody.   So beginning in the early 1960s, we started with building individual silicon transistors.  We primarily built MOS devices, and then tried to develop high frequency units and eventually got the frequency up to 800 MHz.  We did this by building tetrodes – devices with two gates.  This was identified as the “3N” series of commercial devices.  These were true tetrodes, with one gate used as a control gate, and the second gate was used to adjust the impedance.  These were very fine devices. 

 

At this same time, we looked at other types of MOS devices, such as high power devices for use in cable TV (CATV) and we also looked at high voltage devices which might be used instead of the remaining vacuum tubes which were still in use at the time.

 

Then we went into integrated circuits, complementary MOS technology, using both N-channel and P-channel on the same chip, which is what is needed to make a good integrated circuit.  I was involved pretty heavily in this, and actually made quite a few trips out to Findlay Ohio  (where the RCA factory was) during the transition to production of some of these devices.             

 

Go To Ditrick Oral History, Page 5

 

COPYRIGHT © 2001 by Jack Ward.  All Rights Reserved.  http://www.transistormuseum.com/

PAGE 4