On 8/12/2017 8:45 PM, Fred Cisin via cctalk wrote:
On Sat, 12 Aug 2017, Josh Dersch via cctalk wrote:
However, I'm unable to format it. The controller doesn't appear to
actually be laying down a format and attempts to read sectors after
formatting fail with "ID Not Found" errors. I've tried several other
known-working drives (another ST-225, a CDC 94025-51, and an ST-238
with the same results. (And yes, the terminator is present and
accounted for on all drives.) I've run the Model II/12/16/6000
diagnostics and the controller passes with flying colors, as does the
rest of the system.
Before I start digging in deeper to figure out what's going on, I'm
curious if I've missed a step here. I know I'm not the first person
to attempt this, anyone have any ideas?
Long shot:
Is this a system (like the 5160), where there are TWO format programs?
1) a physical low-level format done by special proprietary "non-user"
(unobtanium?) software, followed by
2) a program named "FORMAT" that only sets up the logical structures,
DIRectory, etc., and assumes that the low level format was previously
done
THAT would certainly result in the "logical" format program choking on
"ID not found", when the physical low level format wasn't to it's
liking. (mutual incompatibility between most computers and controllers)
My aplogies if I'm asking granny whether she knows how to suck eggs. . .
No apologies necessary :). The Format I'm running is a low-level format
(I've tried the Xenix "diskutil" program and the format utility included
in the diagnostic routines) so it should be laying out the sectors...
I did some probing of the write logic this evening with an oscilloscope,
armed with the service manual. I found a 74ls54 in the MFM WRITE DATA
+/- signal path that doesn't appear to be passing anything through. I
don't happen to have any spares on hand so it'll have to wait a few days
before I can replace it, but it seems likely that this is the culprit.
I'll keep y'all posted.
Thanks,
Josh
Can you read the drive with the bad bearings (with a flux transition
board?), enough to determine what physical low level format it has?
(any successful track with sector headers)
