20 Aug
2008
20 Aug
'08
7:53 p.m.
If you are referring to the schematics for the ASM
(Add-Subtract Mechanism), they are readily available,
although not including the values for the resistors
which determine the logic-gate functions. A scan and my
re-organised versions are accessible at:
http://www.cs.ubc.ca/~hilpert/e/ABC/asm.html
Thanks, that's a really great start. Prior to seeing your reply to my question, I
requested an inter-library loan of the book "The Origins of Digital Computers -
Selected Papers" which contains the paper which is excerpted on your excellent ABC
site. Even though that book now looks like it won't provide the details I'd
expected, since I won't be using the same voltages as the original circuits were
likely using, the exact component values will be different anyway.
I'm not very familiar with vacuum tube circuit design, but I have a PDF copy of the
1952 USAF technical order TO16-1-255 "Basic Theory and Application of Electron
Tubes" which looks like it might be sufficient. I've built two different 12AU7
dual triode vacuum tube stereo headphone amps from plans which use much lower and,
therefore, safer and easier-to-produce voltages (24 - 60V) than were used historically in
such circuits. The required plate voltages for these low-power circuits are produced
using simple voltage multipliers attached to commonly available and inexpensive
low-voltage transformer secondaries. Since, in the case of digital logic circuits
linearity of operation is not a requirement as it is with audio circuits, even lower
voltages might be usable although 24 volts seems to be plenty low enough.
I was doing some calculations and experiments with
vacuum
tubes a couple of years ago to recreate an ASM, to figure out
what might work for the logic-resistor values (not completed).
It would be nice to know what the original values were.
If you are referring to the schematics for the entire
machine, I had the same interest when I was working on writing a
simulator (also on the web pages) to confirm some aspects of the
functioning of the machine. I managed to get the simulator
functioning without them, so didn't feel a dire need to get them,
although it would still be interesting to see them.
Sometime after the sim was accessible on the web, I was contacted
by one of the fellows (Charles Schorb) who had worked on the ABC
reconstruction and I asked him about schematics. His reply:
...
Schematics are a bit hard to come by. They requested
that I return all my schematics after leaving the project for 'my
real job'. Contact Gary Sleege to see if they have lightened up on
this restriction.
...
but I haven't pursued it further.
That's strange that they'd be so restrictive about the distribution of schematics
for a machine that has no current value beyond that of a historical curiosity. I know
that there was a legal battle between Mauchly with his ENIAC and Atanasoff with his ABC,
but that was decided against Mauchly and claimed that his ENIAC plagiarized concepts used
by Atanasoff's ABC which then invalidated Mauchly's earlier patent claim to the
first electronic computer. Since Atanasoff won the case 35 years ago, why would they not
willingly release all schematics now? Strange.
but I suspect
that since school is not in session, I
won't soon get a response.
(Actually, at universities, when school is not in session
can be a good time to contact people as the people are less likely
to be busy with students and teaching responsibilities, at least if
they're not on vacation.)