10 Apr
2007
10 Apr
'07
6:56 p.m.
I suppose in an ideal world I'd assume that
catastrophic failure of the media
could happen very quickly - so priority is to get a snapshot of the disk onto
modern storage before that happens.
The failure mode I've seen is oxide/binder coming off, clogging the head and
carving little concentric rings in the media :-(
Once the buildup starts, S/N ratio goes into the toilet, so the inner tracks
often have errors.
One of the techniques I've thought about to mitigate this is, as you say, just
snapshot all of the data without analysis, to avoid sitting on a track for a
minimal number of rotations, stagger-read tracks, or read them in inner to outer
track order.
There doesn't seem to be any obvious visual indication that the binder will strip
off a disc, oddly enough. You'd think there would be something like physical
discoloration when this was likely to happen.
10 Apr
10 Apr
9:14 p.m.
Al Kossow wrote:
Yes, that's the one. I've not seen it often, but when it does the end result's
not pretty. I had one drive a while back where the upper head got ripped clean
out of the drive when the coating on the disk let go.
I suppose in
an ideal world I'd assume that catastrophic failure of
the media
could happen very quickly - so priority is to get
a snapshot of the
disk onto
modern storage before that happens.
The failure mode I've seen is oxide/binder coming off, clogging the head
and carving little concentric rings in the media :-( Once the buildup starts, S/N ratio goes into the
toilet, so the inner
tracks often have errors.
Soon as I hear that squeaking sound of heads scraping against things that
they're not supposed to, I stop reading - I just didn't get to the drive
mentioned above fast enough :-(
One of the techniques I've thought about to
mitigate this is, as you
say, just
snapshot all of the data without analysis, to avoid sitting on a track
for a
minimal number of rotations, stagger-read tracks, or read them in inner
to outer track order.
Good thinking. I wonder if the surface order matters too - thinking about it,
it seems like the majority of failures I've seen have been on the upper
surface. Possibly something to do with the spring loading arrangement in a
typical drive (I'm talking 5.25" here - I don't think I've seen a
3.5" disk
fail in the same way, and I don't normally touch 8" drives*)
* I've got a pair of Shugart 801's sitting in the garage awaiting cleaning, so
that will change...
Hmm, I wonder if blowing compressed air at the heads when reading would help,
or whether the coating on the disks is sufficiently "sticky" that it'd still
fix itself to the head no matter what?
There doesn't seem to be any obvious visual
indication that the binder
will strip
off a disc, oddly enough. You'd think there would be something like
physical discoloration when this was likely to happen.
I'd noticed that too. Some brands seem to be more prone than others - not
enough to form a useful pattern, but it seems that given a box of disks from
random manufacturers, if one fails then others from the same manufacturer in
that box will fail too. (but disks from the same manufacturer that have been
stored in different conditions may well be fine)
It's irritating. IMHO all the more reason to get everything possible onto
modern storage first, and worry about how to get it *back* onto original media
again at a later date :-)
cheers
Jules
10:30 p.m.
Jules Richardson wrote:
Good thinking. I wonder if the surface order matters
too - thinking
about it, it seems like the majority of failures I've seen have been on
the upper surface. Possibly something to do with the spring loading
arrangement in a typical drive (I'm talking 5.25" here - I don't think
I've seen a 3.5" disk fail in the same way, and I don't normally touch
8" drives*)
... engage brain before posting :) I was thinking of lashing something up so
that the upper head wasn't in contact with the disk at all, reading the bottom
surface, and only then reading the top. Thinking about it some more, the
loading of the head against the surface on both sides is almost certainly
critical, so this wouldn't work :(
11 Apr
11 Apr
1:49 a.m.
Thinking about it some more, the
loading of the head against the surface on both sides is almost certainly
critical, so this wouldn't work :(
I've often wondered if one could make up a drive with less head pressure,
possibly using some additional electronics to clean up the signal - I've
had a few systems come in where people wanted the data recovered, and
they've been suffering from the "circular rings of death". Never explored
the idea further.
Regarding Als idea of reading the disk backward - It would be a fairly
simply matter to make up a version of ImageDisk which reads from the
inside out.
Dave
--
dave06a (at) Dave Dunfield
dunfield (dot) Firmware development services & tools: www.dunfield.com
com Collector of vintage computing equipment:
http://www.classiccmp.org/dunfield/index.html
10 Apr
10 Apr
9:52 p.m.
Al Kossow wrote:
The failure mode I've seen is oxide/binder coming
off, clogging the head
and
carving little concentric rings in the media :-(
If it's a case of the binder absorbing moisture and losing grip, it might be
an idea to try baking the discs for a bit to dry out the binder. The only
problem is, if you get the disc too hot (the words 'Curie temperature' spring
to mind) then the data's toast...
What would be ideal is a floppy drive with a flying (non-contacting) head,
sort of like a hard drive. I doubt it's possible to do that though...
Once the buildup starts, S/N ratio goes into the
toilet, so the inner
tracks
often have errors.
Hm, rig up something to unmount the heads, run some kind of cleaner over them,
then remount them automatically?
I'm probably thinking up overcomplex solutions again.. my design style is
somewhat Rube Goldberg...
One of the techniques I've thought about to
mitigate this is, as you
say, just
snapshot all of the data without analysis, to avoid sitting on a track
for a
minimal number of rotations, stagger-read tracks, or read them in inner
to outer
track order.
Why would reading from inside to outside (track 80 to track 0) help? If the
oxide's loose, won't it build up on the heads (and strip off of the disc)
whichever way you read the disc?
--
Phil. | (\_/) This is Bunny. Copy and paste Bunny
classiccmp at philpem.me.uk | (='.'=) into your signature to help him gain
http://www.philpem.me.uk/ | (")_(") world domination.
6484
days inactive
6485
days old
4 comments
4 participants
participants (4)
-
aek＠bitsavers.org -
classiccmp＠philpem.me.uk -
dave06a＠dunfield.com -
julesrichardsonuk＠yahoo.co.uk