• Please review our updated Terms and Rules here

Unusual Transistor Computer card

T

tgarnold

Member
Joined
Jan 31, 2022
Messages
29
Does anyone recognize this card? It has unusual gold plated connectors that look more like forked posts. I do think this is a DIY board. The transistors are Fairchild 2N2297 and Motorola 2N2219 from the mid 1960s20231115_225518609_iOS.jpg20231115_225522138_iOS.jpg
 
That actually looks like a really interesting card.

It is likely to be some form of magnetic logic board.

I am intrigued...

Dave
 
So, to add to the intrigue, this is likely a card used by the JPL (Jet Propulsion Lab) in Pasadena, California (predecessor of NASA and now part of NASA). I have a number of oddities that I am trying to wrap my head around. I'll probably post more boards and cordwood units. I think many were purpose built for satellites and missile systems. I have a bunch of electronics from JPL surplus bought in the early 1970s (so prior to that). I'll post a pic of the back as well. It has a single transistor.
 
There are actually a few types of magnetic logic. Two production examples come to mind--and very different from each other. The first is the Univac Solid State mainframes, from a time when transistors were not considered to be reliable devices. See: http://bitsavers.org/pdf/univac/uss/ for a bunch of documentation. The interesting thing about these is that they use majority logic.

Another production device was Eichi Goto's parametron used in NEC's NEAC 1101 system. https://museum.ipsj.or.jp/en/computer/dawn/0017.html

Multi-aperture "hard" cores were also used.

Another more pedestrian use was the magnetic amplifier: https://www.nutsvolts.com/magazine/article/the_magnetic_amplifier Used in very hostile environments, enables a small current or currents to control a large one.
 
Yes, see the documentation cited on the USS. But your article misses an important feature--that of the majority gate. So, a 4 input gate can be configured as "any 2 or more" or "at least 3 with one inhibit"... One interesting aspect of the USS, is that it preserves its state when power is removed.
 
Chuck(G) said:
There are actually a few types of magnetic logic. Two production examples come to mind--and very different from each other. The first is the Univac Solid State mainframes, from a time when transistors were not considered to be reliable devices. See: http://bitsavers.org/pdf/univac/uss/ for a bunch of documentation. The interesting thing about these is that they use majority logic.

Another production device was Eichi Goto's parametron used in NEC's NEAC 1101 system. https://museum.ipsj.or.jp/en/computer/dawn/0017.html

Multi-aperture "hard" cores were also used.

Another more pedestrian use was the magnetic amplifier: https://www.nutsvolts.com/magazine/article/the_magnetic_amplifier Used in very hostile environments, enables a small current or currents to control a large one.
Click to expand...
Nice rabbit hole you sent me down. The USS is both amazing and primitive at the same time.
And it all fits on just 1000 circuit boards.
 
Thanks to all for the great information. I have been researching early ICs (DTL, RTL and TTL) in the JPL data archives. One of the documents I found is called "Circuits and circuit testing for spaceborne redundant digital systems Special technical report no. 3" It covers research in comparing magnetic logic and semiconductor logic. I have several early 1960s ICs from Fairchild, TI, etc. and I didn't realize this is likely one of the magnetic logic devices. Here is a small excerpt with a diagram.

magnetic logic.jpg
 
For those used to today's semiconductor logic, it's important to note that there are no static logic levels--it's all "pulse" or nothing. Each core represents a delay element requiring a clock signal to change state. Essentially, the clock supplies the energy to the system.
 
Here is an excerpt from another JPL article from 1965. There are several more diagrams with logic functions using the 2N2297 transistors.

m logic 2.pngm logic 3.png
 
8820772_orig.jpg

Looks like they plug into this. Copied from DDP-24 https://www.p3oriontopsecret.com/3c-computer-control-company-ddp-24-card-rack.html
 
Finally got the site loaded. Those card sockets makes sense; the D-sub didn't.
So this was in those sub-chasers out of Moffett in the 70s and 80s that rattled my windows when I lived in Sunnyvale? Huh...
 
Last edited:
You must log in or register to reply here.
Back
Top