MFM/RLL data recovery

[Ryan Underwood]

Replying to a bundle of posts:

> IBM used a twisted cable with their hard disks - in order to avoid having
> to explain "Drive Select" jumpers to their [Computerland level] dealers.
> 
> The hard drive cable twist is NOT the same twist as for floppies!

Oops.  See, I saw a mention of a twisted cable in the Miniscribe manual
which came from Hard Drives International, so I thought I was okay
there.

Anyway, I switched to a straight through connection.  I tried the
Miniscribe drives with the 27X and the ST-238R with both the 27X and the
WD1002SV-SR2.  With the 27X I get no POST/BIOS complaints, but the
system tries to go to ROM BASIC, so I guess it is ignoring the connected
drive.  With the WD1002SV with BIOS disabled and 615/4/26 entered, then
I just get Drive C: error from the BIOS.

> If you are trying to recover files you are probably wasting your time, the 
> XT controllers used unique formatting and you will never read the data off 
> of the drives without using controllers indentical with ones used 
> originally.  It is not good enough to use the same brand or even chipset.

Is that true for MFM drives too?  I know RLL setups were picky about
controller/drive compatibility but I didn't think MFM would be also.

> Another problem I had with early MFM hard drives was they needed to be
> read with the same model of controller that formatted them.

... I guess so.

> For the ST225 look for a DTC 5150 or a western digital 1002-WX1 HD
> controller. These were two of the most common HD controllers for XTs.

Thanks.  I'll ask the owners if these models ring any bells.

> I doubt that you can read them with any of the 16 bit cards.

Are drives formatted with 16 bit cards generally compatible with each
other?

> When I put away any XT drives that I wanted the data off I kept the
> cables and controller card with the drive.

Yup.  I certainly wish that were the case here.

> I have docs for the WD cards. The 27X sounds (without looking) like an RLL 
> controller.

It is.

> Wrong 34-pin cable. Also check the markings on the controllers and the 
> drives very carefully for the pin 1 identification. Perhaps all the solder 
> pads are round, except for one which is deliberately square, (that's pin 1) 
> or maybe it's silk screened on the board/drive.

It is actually silk screened on all these boards.  No problems there.

-- 
Ryan Underwood, <nemesis at icequake.net>