From: roger.holmes at microspot.co.uk
 To: cctalk at 
classiccmp.org
 Date: Tue, 11 Nov 2008 22:40:11 +0000
 Subject: RE: Who wants to help read a Saturn V core stack?
  From: dwight elvey
  From: ard at p850ug1.demon.co.uk
 Those of you with experience reading cores, we need your help in
 reading out a LVDC core stack. 
 Ouch!. I don't have much experience of reading core, but I've
 worked on
 core memory systems (I guess many others here have too).
 Rememebr that reading core is destructive. Basically, the read
 operation
 is to write a 0 to each core and see if there was a change in
 magnetic
 flux in that core. If there was, it _was_ a 1. If not, it was (and
 still
 is) a 0. Normally you then write the 1 back if appropriate.
 
 Hi
 Reading shouldn't be too difficult. Outputs could be recorded by
 a digital scope. 
 Shouldn't that be one digital scope channel for every bit in the word?
  The rise times of the address and inhibit lines
 Do we know this core memory has "inhibit lines"? My only core
 experience is with 2D planes and I've not heard of inhibit lines,
 maybe they are the magic that allowed 3D stacks to be made. On my
 core, half the current required is supplied on two separate wires, if
 you don't know the current required and you send the critical amount
 down one wire you will clear the entire memory in one go.
  
---snip---
Hi
 Although, not described in most books on cores, the inhibit lines
were quite common. The use made it so that there was greater
margine between the signal level of the address line running through
the row or column bit that were not addressed and the one
bit the that was to be read or written.
 All the bits in a core plane would see the inhibit line. It would be opposite
the direction of the selected address line. If a bit only had one
address line, the inhibit would mostly cancel the field of that
driven address line. This means that a larger current could be
driven through a single address line since it was mostly canceled
by the inhibit line when not crossed.
 When two address lines cross that were driven,
the net effect would be a significantly larger field since that would
see the sum of the three. Think of the inhibit line as kind of
a negative bias.
 I'm not sure when the they started using these but most of the
cores that I've seen have them.
 As for reading one bit at a time, most cores have separate drivers
for each plane so it is not hard to read a single bit at a time without
effecting the other bits in a word. A single digital scope input would be
enough.
Dwight
_________________________________________________________________
Windows Live Hotmail now works up to 70% faster.