24 Nov
2015
24 Nov
'15
8:45 p.m.
I have an aftermarket hard drive system for the TRS-80 Color Computer. It includes a
Miniscribe 3425 5.25" hard drive and a Xebec S1410 SASI disk controller. This
Miniscribe hard drive has a stepper motor positioner, and Miniscribe actually sprang for
an optical track 0 sensor... unlike the 3.5" Miniscribe 20M drive in my first Amiga
hard drive system, which simply slammed the stepper against a hard stop to find track 0.
I'd like to image this drive with the MFM Reader/Emulator card if I can get it to
work. Actually getting the whole system working with one of my CoCos would be even cooler,
of course. I have no idea what might be on the hard drive; it was an eBay purchase, back
when eBay and I were still on speaking terms.
The hard drive is blinking an error code on its LED, reporting that it cannot cover the
track 0 sensor. Measuring the sensor pins with a DMM while the drive is powered makes me
believe that the optical sensor itself is working correctly. Its output disappears into a
Miniscribe-marked, presumably custom IC, so I don't have high hopes of debugging this
further within the bounds of my gumption.
Interestingly, the interruptor on the external stepper shaft appears to cover the sensor
on the final step against a hard stop, so it's not clear why this drive even needs the
sensor.
Does anybody have a donor Miniscribe 3425 available for sale or trade? The heads can be
rattling around in drifts of aluminum filings for all I care, as long as the logic board
is somewhat likely to be functioning. Or maybe has one sitting around that they might like
to loan to me to image for them, and they don't mind if I risk board swaps after
making a valiant effort to get the data off their drive for them first?
So far, I've only used my new MFM Reader/Emulator to image and emulate a Tandon TM503
in a Tandy Fifteen Meg Disk System, and it worked wonderfully in that role.
--
Mark J. Blair, NF6X <nf6x at nf6x.net>
http://www.nf6x.net/
24 Nov
24 Nov
8:48 p.m.
I forgot to mention that I got the LED error code table from a technical manual for later
Miniscribe drives, so it's possible that the blink pattern means something different
on the 3425. I think it's likely to be correct, though, given the way the stepper is
behaving at power up.
--
Mark J. Blair, NF6X <nf6x at nf6x.net>
http://www.nf6x.net/
9 p.m.
Hmm, now this is interesting. It occurred to me to try tacking a bit of aluminum foil tape
onto the interruptor blade to make it cover the optical sensor earlier before hitting the
hard stop. That made the drive seek differently, and then report either "Track Zero
sensor misadjusted" or "Seek error during burn-in or recal". Maybe the
sensor is just knocked slightly out of adjustment? I'm hesitant to move it if
there's any chance that it's in the correct location, but maybe this is worth
experimenting with.
--
Mark J. Blair, NF6X <nf6x at nf6x.net>
http://www.nf6x.net/
9:37 p.m.
On 11/24/2015 08:45 PM, Mark J. Blair wrote:
The hard drive is blinking an error code on its LED,
reporting that
it cannot cover the track 0 sensor. Measuring the sensor pins with a
DMM while the drive is powered makes me believe that the optical
sensor itself is working correctly. Its output disappears into a
Miniscribe-marked, presumably custom IC, so I don't have high hopes
of debugging this further within the bounds of my gumption.
Maybe it's trying to tell you that it's really a brick? :)
(For those who recall the Miniscribe debacle).
--Chuck
9:45 p.m.
On Nov 24, 2015, at 21:37, Chuck Guzis <cclist at
sydex.com> wrote:
Maybe it's trying to tell you that it's really a brick? :)
(For those who recall the Miniscribe debacle).
LOL!
After my experience with that 20M SCSI drive on my Amiga, I'm tempted to say that
they're all bricks. Some of them just pretend to be hard drives for a while first. ;)
Anyway, I have it happily spinning up now after moving the interruptor vane on the outer
stepper motor shaft a bit. It seems that the drive expects to see the track zero sensor
trip some distance before the hard stop, but not too far before. Whether the drive's
idea of where track zero is vs. where it actually is on the media remains to be seen, but
I'll be setting up the reader shortly. I think there may be a chance of success since,
if I'm guessing correctly, track location and spacing on these drives is determined by
the stepper motor, and the track zero sensor should just need to be within a half track or
so of actual track zero position. Of course, all bets are off if the internal positioner
roller shifted on the stepper's internal shaft, rather than just the interruptor
shifting on the external shaft. Crossing my fingers...
--
Mark J. Blair, NF6X <nf6x at nf6x.net>
http://www.nf6x.net/
10:11 p.m.
On Tue, Nov 24, 2015 at 9:45 PM, Mark J. Blair <nf6x at nf6x.net> wrote:
Anyway, I have it happily spinning up now after moving the interruptor vane on the outer
stepper motor shaft a bit. It seems that the drive expects to see the track zero sensor
trip some distance before the hard stop, but not too far before. Whether the drive's
idea of where track zero is vs. where it actually is on the media remains to be seen, but
I'll be setting up the reader shortly. I think there may be a chance of success since,
if I'm guessing correctly, track location and spacing on these drives is determined by
the stepper motor, and the track zero sensor should just need to be within a half track or
so of actual track zero position. Of course, all bets are off if the internal positioner
roller shifted on the stepper's internal shaft, rather than just the interruptor
shifting on the external shaft. Crossing my fingers...
I did some support work for a product that used a 3.5-inch 20MB
Miniscribe drive back in the mid-80's which had a failure mode where
the track zero flag would rotate slightly out of position on the
shaft. When that happened the drive would flash a specific error code
pattern on the diagnostic LED.
You could recover from that specific failure by loosening the set
screw on the flag just enough that it could be rotated without
rotating the shaft, then powering up the drive with an otherwise
unused manufacturing / test jumper pair shorted. When powered up in
that mode the LED would turn on / off just as you rotated the flag
back into the correct position in the opto sensor. When you got it
into the position where it was just at the on / off threshold you
would then carefully tighten down the set screw on the flag without
rotating the shaft. Then power cycle the drive without the test jumper
in place and hopefully all would be well.
There were specific torque limit screw drivers and a Loctite product
for the screw that were supposed to be used by the service people when
they were doing this, but that was to return the drive to service, not
just a one time data recovery. That was almost 30 years ago and I
don't remember all of the details now.
10:19 p.m.
On Nov 24, 2015, at 22:11, Glen Slick <glen.slick
at gmail.com> wrote:
I did some support work for a product that used a 3.5-inch 20MB
Miniscribe drive back in the mid-80's which had a failure mode where
the track zero flag would rotate slightly out of position on the
shaft. [...]
Very interesting!
I have the drive spinning up happily, though track alignment is still a crap shoot.
I've tried dumping it with the MFM Reader/Emulator, and its software could not make
sense of the track format or CRC. I dumped a raw MFM transitions file for more analysis,
since the Xebec controller might just use a track format that the emulator software
doesn't currently support. I tried analyzing the disk on several cylinders, in case
the tracks line up with steps but the track zero sensing is offset by some number of
tracks.
I think I'll try reading the drive with its original controller next, but first I need
to do some work on the chassis. There's lots of electrical tape falling off of sloppy
wire splices in there, as well as signs of rodent nibbling on some wires. I don't know
if its power supply works yet, and I'll do some repair work on the chassis wiring
before I try smoke testing it.
--
Mark J. Blair, NF6X <nf6x at nf6x.net>
http://www.nf6x.net/
25 Nov
25 Nov
4:17 a.m.
On 11/25/2015 12:19 AM, Mark J. Blair wrote:
I have the drive spinning up happily, though track
alignment is still a
crap shoot. I've tried dumping it with the MFM Reader/Emulator, and its
software could not make sense of the track format or CRC. I dumped a raw
MFM transitions file for more analysis, since the Xebec controller might
just use a track format that the emulator software doesn't currently
support.
I *think* the track format for the S1410 is documented in the board manual,
so if you can get raw track data then there's a chance that you might be
able to make sense of it.
8:09 a.m.
On Nov 25, 2015, at 04:17, Jules Richardson
<jules.richardson99 at gmail.com> wrote:
I *think* the track format for the S1410 is documented in the board manual, so if you can
get raw track data then there's a chance that you might be able to make sense of it.
The sector format is described. but not in a great deal of detail. David Gesswein has
provided some hints to help me begin the challenge of decoding things. It appears that the
drive is now seeking off by about 4 tracks, and I think I should be able to use diagnostic
output from the MFM Reader/Emulator to get the track 0 sensor readjusted by trial and
error.
--
Mark J. Blair, NF6X <nf6x at nf6x.net>
http://www.nf6x.net/
24 Nov
24 Nov
10:22 p.m.
By the way, I found a stamp marking on the inside of the drive's front bezel which I
think represents the expected attitude of a Miniscribe drive. :-p
https://twitter.com/nf6x/status/669368721094238211
--
Mark J. Blair, NF6X <nf6x at nf6x.net>
http://www.nf6x.net/
11:43 p.m.
No luck with the hard drive system on my CoCo 2 or 3, but I don't know if I have it
set up right. There's one cartridge with the SASI interface (presumably...), and
another with an Owl Ware Hard Disk BASIC 3 ROM and floppy controller. I can make the hard
disk LED blink with commands like DRIVE P (park hard drive), but the BASIC just returns
I/O errors. Might be hardware, might be PEBCAK.
--
Mark J. Blair, NF6X <nf6x at nf6x.net>
http://www.nf6x.net/
3097
days inactive
3097
days old
10 comments
4 participants
participants (4)
-
cclist＠sydex.com -
glen.slick＠gmail.com -
jules.richardson99＠gmail.com -
nf6x＠nf6x.net