25 Nov
2010
25 Nov
'10
4:17 p.m.
Here's another idea that'll get me a place in the Loony Bin.
A little while ago I picked up a couple of Amstrad EME232 double-sided
3-inch floppy drives. Out of the three I have, two have had their
read-write ASIC zapped by someone (in one case myself, in another an
ebay seller) plugging them into a PC power supply to "test". Net result:
the ASIC goes completely deaf.
Let's assume for a minute that the heads in the EME232 drives are
undamaged. That means all the issues are related to defective circuitry
on the PCB. So, how about this for an idea: move the controller chip and
surrounding circuitry (or even the whole board) to the EME232, then wire
it into the existing drive mechanics.
This would -- in theory -- allow the two dead 232s (and possibly the
EME156 single-head drive) to be resurrected.
Anyone care to comment on this idea and its possibility or absurdity?
The other option would be to get a new read-write ASIC for the Amstrad
drives, but those seem to be like hen's teeth. Part number is M51017AP
(made by Panasonic but if memory serves they're not marked with a logo
or manufacturer designation) if anyone wants to check.. :)
Thanks,
--
Phil.
classiccmp at philpem.me.uk
http://www.philpem.me.uk/
25 Nov
25 Nov
8:14 p.m.
controller board swapping used to be a "fairly frequent" (interpret
that at your pleasure) occurrence in the cyber forensic community as a
method of reviving media taken as evidence. Obviously it is rife with
issues and some media will cooperate more than others but in general,
the older media type with less on-board "smarts" will respond fairly
well as long as the board(or chips if that old) are a good match. in
terms of firmware...
steve shumaker.. former (as in a long time ago) computer crime
investigator/ forensics specialist
On 11/25/2010 8:17 AM, Philip Pemberton wrote:
Here's another idea that'll get me a place in
the Loony Bin.
A little while ago I picked up a couple of Amstrad EME232 double-sided
3-inch floppy drives. Out of the three I have, two have had their
read-write ASIC zapped by someone (in one case myself, in another an
ebay seller) plugging them into a PC power supply to "test". Net
result: the ASIC goes completely deaf.
Let's assume for a minute that the heads in the EME232 drives are
undamaged. That means all the issues are related to defective
circuitry on the PCB. So, how about this for an idea: move the
controller chip and surrounding circuitry (or even the whole board) to
the EME232, then wire it into the existing drive mechanics.
This would -- in theory -- allow the two dead 232s (and possibly the
EME156 single-head drive) to be resurrected.
Anyone care to comment on this idea and its possibility or absurdity?
The other option would be to get a new read-write ASIC for the Amstrad
drives, but those seem to be like hen's teeth. Part number is M51017AP
(made by Panasonic but if memory serves they're not marked with a logo
or manufacturer designation) if anyone wants to check.. :)
Thanks,
8:27 p.m.
On 25 Nov 2010 at 12:14, steve shumaker wrote:
controller board swapping used to be a "fairly
frequent" (interpret
that at your pleasure) occurrence in the cyber forensic community as a
method of reviving media taken as evidence. Obviously it is rife with
issues and some media will cooperate more than others but in general,
the older media type with less on-board "smarts" will respond fairly
well as long as the board(or chips if that old) are a good match. in
terms of firmware...
Be careful to swap like-revision boards with like-revision boards if
you do go down this road. Swapping boards in dissimilar revs of Teac
FD-505 drives can lead to magic smoke--take my word for it; I've had
it happen.
--Chuck
8:28 p.m.
You still see HD controller boards for newer drives for sale on ebay, I
didn't think swapping a board on a newer HD would work.
----- Original Message -----
From: "steve shumaker" <shumaker at att.net>
To: "General Discussion: On-Topic and Off-Topic Posts"
<cctalk at classiccmp.org>
Sent: Thursday, November 25, 2010 3:14 PM
Subject: Re: Another Crazy Idea: disc drive controller swapping
controller board swapping used to be a "fairly
frequent" (interpret that
at your pleasure) occurrence in the cyber forensic community as a method
of reviving media taken as evidence. Obviously it is rife with issues and
some media will cooperate more than others but in general, the older media
type with less on-board "smarts" will respond fairly well as long as the
board(or chips if that old) are a good match. in terms of firmware...
steve shumaker.. former (as in a long time ago) computer crime
investigator/ forensics specialist
On 11/25/2010 8:17 AM, Philip Pemberton wrote:
> Here's another idea that'll get me a place in the Loony Bin.
>
> A little while ago I picked up a couple of Amstrad EME232 double-sided
> 3-inch floppy drives. Out of the three I have, two have had their
> read-write ASIC zapped by someone (in one case myself, in another an ebay
> seller) plugging them into a PC power supply to "test". Net result: the
> ASIC goes completely deaf.
>
> Let's assume for a minute that the heads in the EME232 drives are
> undamaged. That means all the issues are related to defective circuitry
> on the PCB. So, how about this for an idea: move the controller chip and
> surrounding circuitry (or even the whole board) to the EME232, then wire
> it into the existing drive mechanics.
>
> This would -- in theory -- allow the two dead 232s (and possibly the
> EME156 single-head drive) to be resurrected.
>
> Anyone care to comment on this idea and its possibility or absurdity?
>
> The other option would be to get a new read-write ASIC for the Amstrad
> drives, but those seem to be like hen's teeth. Part number is M51017AP
> (made by Panasonic but if memory serves they're not marked with a logo or
> manufacturer designation) if anyone wants to check.. :)
>
> Thanks,
10:13 p.m.
Here's another idea that'll get me a place in the Loony Bin.
I suspect you're already there :-). I cerainly should be...
A little while ago I picked up a couple of Amstrad EME232 double-sided
3-inch floppy drives. Out of the three I have, two have had their
read-write ASIC zapped by someone (in one case myself, in another an
ebay seller) plugging them into a PC power supply to "test". Net result:
the ASIC goes completely deaf.
Let's assume for a minute that the heads in the EME232 drives are
undamaged. That means all the issues are related to defective circuitry
on the PCB. So, how about this for an idea: move the controller chip and
surrounding circuitry (or even the whole board) to the EME232, then wire
it into the existing drive mechanics.
Where are you getting this ASIC (or board) from? If it's another simialr
drive, then no problem (although IIRC you have to do the track0 alingment
if you remove the PCB). Also, IIRC, the ASIC is the same between the
signle hardand double head models.
This would -- in theory -- allow the two dead 232s (and possibly the
EME156 single-head drive) to be resurrected.
Where are you going to get all the spares from.
Anyone care to comment on this idea and its possibility or absurdity?
The other option would be to get a new read-write ASIC for the Amstrad
drives, but those seem to be like hen's teeth. Part number is M51017AP
(made by Panasonic but if memory serves they're not marked with a logo
or manufacturer designation) if anyone wants to check.. :)
At one time CPC coudl get most Amstrad spares... But I don;t know if disk
drive parts were amongst them, certainly the service manuals don't seem
to support field repari of thedrives and there re no parts lists...
One other though, which I mentionedsoem weeks ago. The periperhal
componens to the read amplifier seem to be those that I would expect for
an MC3740 IC. I wonder if yo could use that IC and a little external
circuitry to replace the defective read amplifier secion (and the rest of
the ASCI seems fine).
-tony
10:47 p.m.
On 25/11/10 22:13, Tony Duell wrote:
Probably... :)
I was thinking a 3.5in drive (with the DENSITY switch hardwired to 720k)
or a 5.25in 1.2MB drive similarly jumpered.
Reverse engineer the circuitry (i.e. disassemble and trace out on
paper), then etch a new PCB to fit the Amstrad drive shell.
Here's another idea that'll get me a place in the Loony Bin.
I suspect you're already there :-). I cerainly should be... Let's
assume for a minute that the heads in the EME232 drives are
undamaged. That means all the issues are related to defective circuitry
on the PCB. So, how about this for an idea: move the controller chip and
surrounding circuitry (or even the whole board) to the EME232, then wire
it into the existing drive mechanics.
Where are you getting this ASIC (or board) from? If it's another simialr
drive, then no problem (although IIRC you have to do the track0 alingment
if you remove the PCB).
Nope. Track0 sensor is on a separate PCB, in the mechanical bay. You can
swap the PCB with no ill effects on the head alignment.
Also, IIRC, the ASIC is the same between the
signle hardand double head models.
It is. I'll bet it was used on other disc drives as well, though finding
them would be a fun trick. It's a Panasonic chip, so it's probably fair
to assume at least one Panasonic drive used it.
Where are you going to get all the spares from.
Drive mech -- the broken EME232s.
Head amp ASIC etc. -- PC floppy drives bought on ebay. No shortage of
those, and the circuitry on the Panasonic JU247A seems fairly easy to
reverse engineer (single-sided PCB, one ASIC and a stepper controller).
At one time CPC coudl get most Amstrad spares... But I
don;t know if disk
drive parts were amongst them, certainly the service manuals don't seem
to support field repari of thedrives and there re no parts lists...
I don't think they were -- the manuals explicitly state that the disc
drives were non-engineer-serviceable. I'd love to get an alignment disc
for these things...
One other though, which I mentionedsoem weeks ago. The
periperhal
componens to the read amplifier seem to be those that I would expect for
an MC3740 IC. I wonder if yo could use that IC and a little external
circuitry to replace the defective read amplifier secion (and the rest of
the ASCI seems fine).
I take it you hammed the part number, and that you actually meant the
MC3470?
That would probably do it -- the catch, of course, is that the MC3470
seems to be obsolete, and as rare as hen's teeth. Again, unless someone
in the community has squirreled away a rail of them.
If I had a couple of those, the readback problem is solved, and
attention can then be turned to writing and the Shugart interface
(neither of which are especially hard problems to solve with LSTTL).
--
Phil.
classiccmp at philpem.me.uk
http://www.philpem.me.uk/
28 Nov
28 Nov
6:37 p.m.
[Sorry for the delay in replyinjg, I think my ISP has been having mail
problems.]
I was thinking a 3.5in drive (with the DENSITY switch hardwired to 720k)
or a 5.25in 1.2MB drive similarly jumpered.
Right. I might be inclined to try to build the whole thing from scratch
based on the schematics of a 5.25" or 8" drive using SSI/MSI chips. That
would be quite a fun project, I just might give it a go sometime...
As an aside, the drive to avoid for things like this is the origianl SOny
full-ehight 3.5" _single head_ one. It has a head that produces a very
small signal and a step-up transformer (!) on the board before the read
amplifier. No other floppy drive (AFAIK) did it this way. The similar
double-head SOny drives use much more conventional circuitry.
I did...
Let's assume for a minute that the heads in the
EME232 drives are
undamaged. That means all the issues are related to defective circuitry
on the PCB. So, how about this for an idea: move the controller chip and
surrounding circuitry (or even the whole board) to the EME232, then wire
it into the existing drive mechanics.
Where are you getting this ASIC (or board) from? If it's another simialr
drive, then no problem
Reverse engineer the circuitry (i.e. disassemble and trace out on
paper), then etch a new PCB to fit the Amstrad drive shell.
You do realise that schematics or serive manuals exist for some drives.
Teac ones, in particular, if you want to use a floppy drive ASIC as part
of the design.
My experience is that the areas where different floppy drives most differ
is the the motor circuitry, both the spindle motor and the stepper.
Steppes can be common + 3 wires to senquence (3 phase), ditto but with 4
wires, or 2 independant coils (which you have to drive from full-H
drivers). Spindle motors are likely to be brushless things now, but just
how the control circuity is arranged depends a lot on the drive. The
easiest drives to mess about with had a motor that just needs power (12V
or 5V) and has a logic-level (or maybe open-collector-driven) 'run'
input. Older drives may have brush-type DC motors (speed control may be
intenral, on a separate PCB, or on the main logic PCB), or even AC
motors (8" drives).
One other
though, which I mentionedsoem weeks ago. The periperhal
componens to the read amplifier seem to be those that I would expect for
an MC3740 IC. I wonder if yo could use that IC and a little external
circuitry to replace the defective read amplifier secion (and the rest of
the ASCI seems fine).
I take it you hammed the part number, and that you actually meant the
MC3470?
That would probably do it -- the catch, of course, is that the MC3470
seems to be obsolete, and as rare as hen's teeth. Again, unless someone
in the community has squirreled away a rail of them.
This is the first time I've heard the Apple Disk II being describved as
'rare as hen's teeth' :-). YEs, there's one inside...
Since the Amstrad ASIC is working apart from the read chain (or is it the
head switching cirucity?), I wonder if you could keep that for things
like head positioning and kludge on the nevessary bits to fix the broken
part.
-tony
5144
days inactive
5147
days old
6 comments
5 participants
participants (5)
-
ard＠p850ug1.demon.co.uk -
cclist＠sydex.com -
classiccmp＠philpem.me.uk -
shumaker＠att.net -
teoz＠neo.rr.com