Long time, no post! (For me, anyway.) I recently decided to resurrect
my 3B1, the original Miniscribe drive of which had gotten jostled
while powered off, but with heads un-parked - crash! So I bought an
ST-4096 off eBay and eventually figured out the jumper settings,
formatted it and then installed the Foundation Set. Hurray for 25
year old floppies!
Then I started installing some add-on software, and suddenly the
floppy drive started going "skwee-skwee-skwee". :-/ Now the light
comes on, but nothing else happens.
Anybody have a spare drive?
Cheers -- Peter (once ...harvard!mit-nc!winter!pschmidt)
--
Peter Schmidt | CTO North America | sch...@ipanematech.com
Mob. North America +1 781 883 4818 | Fax +1 781 890 7007
Beyond the Network | www.ipanematech.com
Hmm ... when I had one fail, it was actually dust build-up in
the optical sensors for determining whether the floppy was in place, and
whether the write-enable tab was present.
Pull the drive, hit it with compressed air, and re-install it to
see what that does. (If you don't have compressed air -- take it to a
neighborhood gas station and talk to someone there about using it to
blast the dust out of the drive.)
Or -- it could be a problem which I encountered on my first
Foundation Set. The pressure of the binder had pinched the edge of the
floppy envelope tight enough so the drive could not turn the floppy
inside it. What I did was tell by feel which edge was tightest (likely
the one away from the rings in the ring binder) and using a pair of
sharp scissors, trim off about 1/16" of the edge to free up the floppy
to turn.
That worked fine, and I've re-used those floppies several times
over the years.
> Anybody have a spare drive?
Not at present. Sorry. -- But hopefully the above will help you
to get your drive to work again. Check the floppy itself first, then if
that does not fix it, try the compressed air dust removal.
Good Luck,
DoN.
--
Email: <dnic...@d-and-d.com> | Voice (all times): (703) 938-4564
(too) near Washington D.C. | http://www.d-and-d.com/dnichols/DoN.html
--- Black Holes are where God is dividing by zero ---
--Peter
"DoN. Nichols" <dnic...@d-and-d.com> writes:
--
Also -- remember to check for floppies which can't be easily
turned in their envelopes by gripping by the hub. With floppies which
have been in the AT&T original binders for a long time, that is
certainly a possible problem.
And try one of the earlier floppies in the drive to see whether
it can be read. If so, then the stuck floppy is the most likely
problem.
I conclude that I probabaly really do need a replacement floppy
drive. Next question is, where can I dig one up?
Cheers -- Peter
"DoN. Nichols" <dnic...@d-and-d.com> writes:
> On 2009-09-04, Peter H. Schmidt <sch...@ipanematech.com> wrote:
>> Many thanks for the advice DoN, I'll try the air trick and report back.
>
> Also -- remember to check for floppies which can't be easily
> turned in their envelopes by gripping by the hub. With floppies which
> have been in the AT&T original binders for a long time, that is
> certainly a possible problem.
>
> And try one of the earlier floppies in the drive to see whether
> it can be read. If so, then the stuck floppy is the most likely
> problem.
>
> Good Luck,
> DoN.
--
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=270399458565#ht_955wt_1167
I may put the second in a PC so I can transfer files more easily.
Hmm ... another thing to check is what happens if you power the
floppy from another power supply -- say one for a typical PC. Is your
machine a 3B1 (with the hump under the monitor) or a 7300 (flat top
under the monitor)? The 3B1's power supply has a separate power cable
which goes directly from the power supply to the hard drive, to keep the
current through the system board's connectors to a minimum. IIRC, the
power for the drives to the system board is passed through different
pins on the power supply than that for the memory and that for the rest
of the logic on the board. So different pins can oxidize with different
results.
Two forms of oxidized connector failure -- the actual connector
on the ribbon cable turns brown near one pin and perhaps even burns in
two. The other is overheating of the power supply's pins where they
are soldered on the underside of the board. The connection can
crystalize (a cold solder joint). Sometimes this can be caused by
overheating of the pin by the connector failure discussed first. I have
seen both kinds of failures on 3B1/7300 power supplies, with all kinds of
weird failure modes -- some of which mimic actual drive failures.
So -- double check the power supply connections. First pull the
power connector on the ribbon cable and check it for heat discoloration,
and then remove the four corner screws mounting the power supply and
look at the soldered connections for the power connector (row of
vertical pins).
Use a good contact treatment (Cramolin or DeOxit if you have
them) on the pins and cycle the connector on and off several times to be
sure.
Long ago, I picked up some of the connectors which go on the
ribbon cable from a hamfest (I thought the pin count was right, and it
turned out to be so). I had a connector so badly cooked in the middle
that I had to replace that connector to make a working computer again.
(I also had to re-solder the pin under the power supply to which that
connected to make it reliable.
Only once you have eliminated the power supply considerations do
you want to replace the floppy drive.
Oh yes -- another thing to check -- with no floppy in the drive,
how difficult is it to turn the hub -- usually from the big flywheel
like disc on the bottom. If it moves in a lumpy manner, the bearings
are probably dying. I've had this with an earlier MPI floppy drive, but
not with the drives used in the 3B1.
You may need to look for a similar floppy drive on eBay or some
vendor of old drives, as nothing of the sort seems to be made now.
Ideally, go for the same maker and model number -- and check that any
jumper blocks on the new drive are set just as the old ones were.
O.K. That looks similar enough so it might be right, with
jumpers set identically to the original.
> I may put the second in a PC so I can transfer files more easily.
If the PC is running MS-DOS or Windows, beware that the floppy
formats are different. IBM PC's used 9 sectors per track. The 3B1/7300
used either 8 sectors per track (for the boot floppy and a very few
others), or 10 sectors per track (for most of the OS installation
floppys). So transferring files may be a bit of a problem.
Good Luck,
DoN.
--
FWIW, it is a 3B1, P3....P5 motherboard, used only once or twice by a
little old lady (seriously) and given to me in original packaging with
all original manuals, etc. I even have the original phone cable,
still coiled up!
Drives are en route via UPS Ground, may be a week or so before I can
try them...
"DoN. Nichols" <dnic...@d-and-d.com> writes:
--
It is not identical. It has different jumpers and somewhat different
components, but thank goodness I didn't have to futz with anything.
Have already loaded the Development Set and the R2.0 fixdisk so the
kernel is now 3.51m.
Now, if I can just get kermit on it, then I will be really happy.
Maybe have to configure uucp first, THAT's been a while...
...Peter
sch...@ipanematech.com (Peter H. Schmidt) writes:
--
Peter Schmidt | Technical Director | sch...@ipanematech.com
Mob. North America +1 781 883 4818 | Fax +1 781 419 6527
Beyond the Network | www.ipanematech.com
What's the problem with kermit?
I used to back up my 3B1 via a kermit script at 19200 on a tty link to a
"bigger" computer.
I vaguely recall a uuencoded "bootstrap" version of a minimal kermit that
you first transferred as a captured file on a tty link.
--
Clarence A Dold - Hidden Valley Lake, CA, USA GPS: 38.8,-122.5
> Peter H. Schmidt <sch...@ipanematech.com> wrote:
>> Now, if I can just get kermit on it, then I will be really happy.
>> Maybe have to configure uucp first, THAT's been a while...
>
> What's the problem with kermit?
It's not yet on my machine. :-)
> I used to back up my 3B1 via a kermit script at 19200 on a tty link to a
> "bigger" computer.
Hmm, I like that idea.
> I vaguely recall a uuencoded "bootstrap" version of a minimal kermit that
> you first transferred as a captured file on a tty link.
That would be handy, might be easier than setting up uucp on my RedHat
machine. Guess I have some poking around the Net to do...
...Peter
And i was just going to suggest uucp. :-(
bill
--
Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
bill...@cs.scranton.edu | and a sheep voting on what's for dinner.
University of Scranton |
Scranton, Pennsylvania | #include <std.disclaimer.h>
Great!
> Have already loaded the Development Set and the R2.0 fixdisk so the
> kernel is now 3.51m.
Also great.
> Now, if I can just get kermit on it, then I will be really happy.
> Maybe have to configure uucp first, THAT's been a while...
IIRC, Kermit did not require uucp to be installed first. You
just hang your modem on a serial port, and specify that to kermit when
you start it up.
I originally installed Gkermit on 3B1 by using serial link to PC and
running Xmodem protocol in Windows Hyperterminal. 3B1 should have /usr/
bin/umodem which is compatible with Xmodem protocol.
dennis
I think he was thinking of using uucp to get Kermit onto the machine.
That's waht I did many moons ago.