Talk Of Building A Computer...

From: William Donzelli <william_at_ans.net>
Date: Thu Nov 20 09:45:01 1997

> >century, transistors weren't avaible... vaccum tubes... huge ones, but

There were some rather small Western Electric tubes from the 1920s and
30s.

> I think there is something even more fundamental here. Valves
> (thermionic, in tubes) have quite different behaviour to trannies.
>
> A JFET behaves fairly like a triode, but designs that use pentodes and
> nonodes and things as multi-input gates are going to be very difficult
> to translate.

All true, but additionally, tubes really do not like doing digital work at
all. If they are in a cutoff condition (no electron flow from cathode to
plate), a charged cloud of electrons forms around the cathode. This cloud
greatly increases the time it takes for the tube to turn on. This is
called "sleeping sickness".

> Of course by the 1960s there were somve very nice valves around that
> weren't available to the 1940s computer pioneers - the 7586 nuvistor
> springs to mind: a very nice triode in a metal can about 1 inch tall
> including pins, and less than half an inch in diameter. Can't remember
> the spec, though. Such devices could make a valve machine quite a bit
> smaller than Colossus, Eniac, Edsac, etc.

The 7586, as well as the other nuvistors, also suffer from the sickness.
Special "computer only" tubes were developed in the 1950s - I can not
recall numbers, but they are similar to their analog cousins.

> Or if you want to be way out, what about tubes with several valves in?
> Things like double diode triodes are quite common, and someone even put
> most of a radio set into one tube (passive components and all).

Tried and died. The Germans tried it with the rather expensive Loewe
tubes, then we made the 6N6G, then the Zahl tube for the AN/TPS-1. The
idea is not new, just not popular.

William Donzelli
william_at_ans.net
Received on Thu Nov 20 1997 - 09:45:01 GMT

This archive was generated by hypermail 2.3.0 : Fri Oct 10 2014 - 23:30:35 BST