ExpandaCore 18 core memory
axelsson at acc.umu.se
Sat Jul 30 03:06:12 CDT 2016
Sometimes you are lucky... A friend got a lot of documentation in a haul
just a few days ago, among it was some documentation about NORD-1 and
the ExpandaCore-18 was among it. So now I have complete schematics and
even some timing diagrams. :-)
For reference I'll include the link to the document page,
Just in case someone would be looking for this document in the future
and finds this post.
He also had a lot of other good stuff for me but not scanned yet, I'll
just have to wait for that. Realistically I don't think I'm ready to add
a hard drive to the machine yet as I'm not sure I can even read the
Den 2016-07-26 kl. 01:15, skrev Göran Axelsson:
> Den 2016-07-25 kl. 14:37, skrev Camiel Vanderhoeven:
>> Op 25 jul. 2016 2:25 p.m. schreef "Göran Axelsson"
>> <axelsson at acc.umu.se>:
>>> My guess so far is that there is a problem with reading and writing to
>> the memory. The problem is that I have no documentation over the memory
>> module except a drawing of the circuitry used to access it. ND bought
>> several different models of core memory for it's early computers and
>> adapted the interface.
>>> So once again I turn to the cctech for help, does anybody have
>> instructions about ExpandaCore 18 from by Cambridge Memories INC,
>> Massachusetts (also known as CMI but probably not the CMI on bitsavers).
>>> So far the only thing I've found was a newsflash in a computer magazine
>> about a sale of memories to another computer maker.
>> I can't help you with that, but is it just the core stack itself you
>> have docs for, or does the CMI part include some driver circuitry? If
>> just the stack, you should be able to do without docs for that. Core
>> doesn't go bad unless it's physically damaged. If it is physically
>> repair can be very difficult, depending on the diameter of the cores.
> Thanks for the answer.
> The ExpandaCore 18 (tm) ;-) is a unit with the driver electronics and
> core memory together. One control board per four memory planes and
> dual ported so a high speed device could write straight into the core
> memory without going through the CPU. In my case it seems like I have
> a packet drive interface that uses the second channel.
> I have put up some more pictures here :
> Maybe I'm jumping to conclusions here but there is one "control board"
> for each four core planes. Maybe it's just a buffer. A problem with
> the core memories is that it sits tight in a crate and I have no
> extender for these cards. At least I should be able to measure the
> signals from the CPU to the control board, and from the control board
> to the memory planes.
>> How are you troubleshooting this? Can you deposit a value in memory from
>> the front panel, then read it back? If that works, but one or more
>> bits are
>> off, there's a good reason to look at the core driver circuitry. If it
>> doesn't work at all, the problem could be anywhere. Try hooking a logic
>> analyzer to the address and data lines, clock, and read/write control
>> for a start.
>> Good luck!
> I can deposit a value in the address register (R) but that is as far
> as I have come right now. I can also look at the other registers. I
> think that the memory content should show up when I enter an address
> but I get back the same value as I entered.
> The operators panel is a bit broken down, there are a couple of bad
> switches but new ones is in the mail. Some keys react on vibrations.
> At the moment I can't do any measurements on the CPU-cards as there
> are no space in that crowded rack. But yesterday I cadded an extender
> for the NORD-1 CPU crate so in a week I'll be able to do some
> measurements on that part at least.
> I got a cheap USB-connected logic analyzer and a digital sampling
> oscilloscope, so with the extender cards I will be able to measure or
> break up and inject any signal I want in the CPU.
> This is too fun to be healthy! :-)
More information about the cctalk