19 Apr
1999
19 Apr
'99
7:33 a.m.
ISTR that if you had a core set for an 8 that was "mismatched", the hardware
handbook had the procedure for using a scope to "tune" a set into the
appropriate range.
Jay West
19 Apr
19 Apr
7:47 a.m.
At 11:33 AM 4/19/99 -0500, Jay wrote:
ISTR that if you had a core set for an 8 that was
"mismatched", the hardware
handbook had the procedure for using a scope to "tune" a set into the
appropriate range.
Not exactly, this does allow you to "fine tune" a set that already is close
enough, but it doesn't necessarily get you close enough if you weren't
there already. There is a graph in of the hysterisis(sp?) region in the 8/a
book and tuning consists of making sure that the write current and inhibit
current cancel, and that write current from both the x and y drivers will
just flip the bit but not risk flipping other bits.
--Chuck
8:28 a.m.
New subject: Q-Bus and Unibus to ATA info
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 681 bytes
Desc: not available
Url :
http://www.classiccmp.org/pipermail/cctalk/attachments/19990419/f03778d3/at…
3:37 p.m.
Jay wrote:
ISTR that if you had a core set for an 8 that was
"mismatched", the hardware
handbook had the procedure for using a scope to "tune" a set into the
appropriate range.
Chuck wrote:
Not exactly, this does allow you to "fine
tune" a set that already is close
enough, but it doesn't necessarily get you close enough if you weren't
there already. There is a graph in of the hysterisis(sp?) region in the 8/a
book and tuning consists of making sure that the write current and inhibit
current cancel, and that write current from both the x and y drivers will
just flip the bit but not risk flipping other bits.
So you're saying that whenever Field Circus had to replace one of the boards
of a set, they replaced all of them, because there wasn't enough adjustment
range to calibrate them? That doesn't match my recollection; I think it was
commonplace for the H220 to get replaced by itself if it got damaged, and I
think I recall seeing them replace at least one failed G104.
The biggest problem is that the diagnostics can't stress the memory system
as much as data-break peripherals can. Tuning the adjustments for reliable
operation is more difficult than the maintenance manual might lead one to
believe.
9208
days inactive
9208
days old
3 comments
4 participants
participants (4)
-
cmcmanis＠mcmanis.com -
eric＠brouhaha.com -
healyzh＠aracnet.com -
jlwest＠tseinc.com