3 Feb
2023
3 Feb
'23
4:40 p.m.
Question: I just used a strong magnet to wipe an old Maxtor MFM drive
(magnet on outside of case). Now the drive will not even seek properly
on start up, just endlessly moves the heads..
Is the drive now toast? Do MFM drives have embedded servo information on
the platter formatted by the factory?
CZ
3 Feb
3 Feb
4:42 p.m.
low level format
On Fri, Feb 3, 2023 at 11:40 AM Chris Zach via cctalk <cctalk(a)classiccmp.org>
wrote:
Question: I just used a strong magnet to wipe an old
Maxtor MFM drive
(magnet on outside of case). Now the drive will not even seek properly
on start up, just endlessly moves the heads..
Is the drive now toast? Do MFM drives have embedded servo information on
the platter formatted by the factory?
CZ
4:44 p.m.
On Fri, Feb 3, 2023 at 4:40 PM Chris Zach via cctalk
<cctalk(a)classiccmp.org> wrote:
Question: I just used a strong magnet to wipe an old Maxtor MFM drive
(magnet on outside of case). Now the drive will not even seek properly
on start up, just endlessly moves the heads..
Is the drive now toast? Do MFM drives have embedded servo information on
the platter formatted by the factory?
I assume by 'MFM' you mean a drive with an interface similar to the ST412.
Embedded servo is rare (unheard-of) on ST412 interfaced drives simply
because the manufacturer has no idea how it will be low-level
formatted and thus where the sector headers will be. So no safe place
for the servo bursts on the data surfaces
But a dedicated servo surface is very common on larger such drives.
That's why you often see an odd number of data heads. There is a head
on each side of each disk, but one is used for the servo.
Sounds like you've wiped that. No sensible way to recover I'm afraid
-tony
4:57 p.m.
Embedded servo is rare (unheard-of) on ST412
interfaced drives simply
because the manufacturer has no idea how it will be low-level
formatted and thus where the sector headers will be. So no safe place
for the servo bursts on the data surfaces
*nod* That's what I would think: MFM should allow you physical access to
the heads to write the format as you see fit. So head positioning should
be mechanical, not servo based.
But a dedicated servo surface is very common on larger
such drives.
That's why you often see an odd number of data heads. There is a head
on each side of each disk, but one is used for the servo.
This is a Maxtor 2190. The RD54 base disk. I've been having major
problems trying to format them on my RQDX3, I'll post that adventure in
a bit. In a nutshell the drives only partially format then error out. On
Dave G's MFM emulator all the tracks appear to be there.
Sounds like you've wiped that. No sensible way to
recover I'm afraid
Yep, with 15 yeads it had a servo platter that is gone. Oh well, it's
securely erased :-) I'll toss the drive, keep the electronics interface
and keep it in mind for the future.
C
5:12 p.m.
On Fri, Feb 3, 2023 at 4:57 PM Chris Zach via cctalk
<cctalk(a)classiccmp.org> wrote:
Yep, with 15 yeads it had a servo platter that is
gone. Oh well, it's
securely erased :-) I'll toss the drive, keep the electronics interface
and keep it in mind for the future.
If you have the tools, it's worth carefully dismantling the dead HDA
taking photos as you go and keeping parts like the heads and
flexiprint. Other bits like the positioner magnet and spinde motor
mght be worth keeping too.
Some months ago I was working on a Toshiba hard drive in a Stride 440
and I wish I'd had some idea of what was on the flexiprint between the
logic board connector and the heads. Of course without a clean room I
didn't dare open the HDA, but had a defective or dismantled one been
available I'd have paid reasonable money for it.
-tony
4 Feb
4 Feb
12:39 p.m.
On 2023-02-03 11:57, Chris Zach via cctalk wrote:
This is a Maxtor 2190. The RD54 base disk. I've
been having major
problems trying to format them on my RQDX3, I'll post that adventure in
a bit. In a nutshell the drives only partially format then error out. On
Dave G's MFM emulator all the tracks appear to be there.
OK, is it a MAXTOR 2190, or a RD54?
If it is really an RD54, it has the weird DEC sectors with the bad block
information.
If it doesn't have the bad blocks information written in the "DEC way"
it won't be recognized as a RD54
5:24 p.m.
Or ST506 ?
On Fri, Feb 3, 2023 at 5:21 PM Alexandre Souza via cctalk <
cctalk(a)classiccmp.org> wrote:
I thoug the right one was st512...can you enlighten me
on this subject
Tony?
Enviado do meu Tele-Movel
I assume by 'MFM' you mean a drive with
an interface similar to the
ST412.
-tony
5:27 p.m.
On Fri, Feb 3, 2023 at 5:21 PM Alexandre Souza
<alexandre.tabajara(a)gmail.com> wrote:
I thoug the right one was st512...can you enlighten me on this subject Tony?
I've never heard it called that.
It's often called 'ST506' but that drive had a few differences from
the later ones. it didn't support buffered seeks AFAIK. The ST412 did
and was the most common of a family of 3 similar drives (ST406, ST412,
ST419) so it tends to be used as the de-facto name of the interface.
-tony
4 Feb
4 Feb
2:26 a.m.
Chris, what kind of magnet did you use?
If it was an electromagnet I could imagine that you caused physical damage
by something heating up sufficiently. If it was a permanent magnet then it
might indeed be servo data which has been erased. Either way I expect you
need a very strong magnetic field to erase a hard drive from the outside.
Tom
On Sat, 4 Feb 2023, 12:40 am Chris Zach via cctalk, <cctalk(a)classiccmp.org>
wrote:
Question: I just used a strong magnet to wipe an old
Maxtor MFM drive
(magnet on outside of case). Now the drive will not even seek properly
on start up, just endlessly moves the heads..
Is the drive now toast? Do MFM drives have embedded servo information on
the platter formatted by the factory?
CZ
2:28 a.m.
Perm magnet. Those neodynium magnets are powerful, keep them away from
disks.
C
On 2/3/2023 9:26 PM, Tom Hunter via cctalk wrote:
Chris, what kind of magnet did you use?
If it was an electromagnet I could imagine that you caused physical damage
by something heating up sufficiently. If it was a permanent magnet then it
might indeed be servo data which has been erased. Either way I expect you
need a very strong magnetic field to erase a hard drive from the outside.
Tom
On Sat, 4 Feb 2023, 12:40 am Chris Zach via cctalk, <cctalk(a)classiccmp.org>
wrote:
> Question: I just used a strong magnet to wipe an old Maxtor MFM drive
> (magnet on outside of case). Now the drive will not even seek properly
> on start up, just endlessly moves the heads..
>
> Is the drive now toast? Do MFM drives have embedded servo information on
> the platter formatted by the factory?
>
> CZ
>
23 Mar
23 Mar
4:22 p.m.
Old MFM/RLL drives with stepper positioners generally have no servo. The same can't be
said of voice-coil positioned drives; they could use any means, including hardware optical
servos, etc. but often used a surface and a head for that purpose... so yes, you can
render one of these drives useless with a magnet if yours has a servo surface. (This is
not the same as embedded servo, which places servo information -with- the data on the same
surfaces.)
4:52 p.m.
Speaking from experience with an old RD54, yep. Put a magnet on the
outside case towards the bottom, spin the drive up and it's gone forever.
On 3/23/2023 12:22 PM, Daniel Daigle via cctalk wrote:
Old MFM/RLL drives with stepper positioners generally
have no servo. The same can't be said of voice-coil positioned drives; they could use
any means, including hardware optical servos, etc. but often used a surface and a head for
that purpose... so yes, you can render one of these drives useless with a magnet if yours
has a servo surface. (This is not the same as embedded servo, which places servo
information -with- the data on the same surfaces.)
24 Mar
24 Mar
6:58 p.m.
Well the data are still there and could be retrieved with a sophisticated servo on data
system and/or a probe head on the data surfaces.
Simpler to hit the spindle motor top dead center with a very large hammer ruining the
bearings and crashing a few heads in the process.
Even then the data are still there so nothing beats a multi-pass full disk wipe
-----Original Message-----
From: Chris Zach [mailto:cz@beaker.crystel.com]
Sent: Thursday, March 23, 2023 9:53 AM
To: cctalk(a)classiccmp.org
Subject: [cctalk] Re: Nuking an MFM drive with a magnet, format/servo gone?
Speaking from experience with an old RD54, yep. Put a magnet on the outside case towards
the bottom, spin the drive up and it's gone forever.
On 3/23/2023 12:22 PM, Daniel Daigle via cctalk wrote:
Old MFM/RLL drives with stepper positioners generally
have no servo.
The same can't be said of voice-coil positioned drives; they could use
any means, including hardware optical servos, etc. but often used a
surface and a head for that purpose... so yes, you can render one of
these drives useless with a magnet if yours has a servo surface. (This
is not the same as embedded servo, which places servo information
-with- the data on the same surfaces.)
7:17 p.m.
On Fri, 24 Mar 2023, Tom Gardner via cctalk wrote:
Well the data are still there and could be retrieved
with a sophisticated servo on data system and/or a probe head on the data surfaces.
Simpler to hit the spindle motor top dead center with a very large hammer ruining the
bearings and crashing a few heads in the process.
Even then the data are still there so nothing beats a multi-pass full disk wipe
* Mr. Drill Press enter the chat *
:)
g.
--
Proud owner of F-15C 80-0007
http://www.f15sim.com - The only one of its kind.
http://www.diy-cockpits.org/coll - Go Collimated or Go Home.
Some people collect things for a hobby. Geeks collect hobbies.
ScarletDME - The red hot Data Management Environment
A Multi-Value database for the masses, not the classes.
http://scarlet.deltasoft.com - Get it _today_!
7:42 p.m.
On 3/24/23 12:17, geneb via cctalk wrote:
On Fri, 24 Mar 2023, Tom Gardner via cctalk wrote:
> Well the data are still there and could be retrieved with a
> sophisticated servo on data system and/or a probe head on the data
> surfaces.
> Simpler to hit the spindle motor top dead center with a very large
> hammer ruining the bearings and crashing a few heads in the process.
> Even then the data are still there so nothing beats a multi-pass full
> disk wipe
Tossing one into a burning slash pile works well, also. Melts that
Zamac, it does. Don't ask me how I know.
--Chuck
7:51 p.m.
On Fri, Mar 24, 2023 at 12:17 PM geneb via cctalk <cctalk(a)classiccmp.org>
wrote:
On Fri, 24 Mar 2023, Tom Gardner via cctalk wrote:
* Mr. Drill Press enter the chat *
The data is still where the holes aren't though ;)
Sellam
Well the data are still there and could be
retrieved with a
sophisticated servo on data system and/or a probe head on the data surfaces.
Simpler to hit the spindle motor top dead center
with a very large
hammer ruining the bearings and crashing a few heads in the process.
Even then the data are still there so nothing
beats a multi-pass full
disk wipe
647
days inactive
696
days old
19 comments
13 participants
participants (13)
-
Adrian Godwin -
Alexandre Souza -
Bill Degnan -
Chris Zach -
Chris Zach -
Chuck Guzis -
Daniel Daigle -
emanuel stiebler -
geneb -
Sellam Abraham -
Tom Gardner -
Tom Hunter -
Tony Duell