• Please review our updated Terms and Rules here

Data recovery from unknown ST-225 MFM hard disk

P

predator99

Member
Joined
Jun 18, 2017
Messages
45
Hi,

I have bought a scrap lot which contained a Seagate ST-225 MFM hard disk. I am interested which data is stored on the disk - not to find any personal things but maybe some unknown software/device drivers.

I have tried to read some data with different controllers:

National NDC5125
WD1003V-MM2
WD1003-WA2
WD WA6VR (RLL)

But Speedstor and Norton Diskedit dont show any data at all.

I am quite sure the drive is OK. I have tried to low level format the last track (614) only with speedstor (dont expect that any data is there) and it seems to work fine. Afterwards I can perform the read / write tests on this track only without any error.

So what do you think the chances are to read the data from the disk? Do I have to find the exact controller the drive was low-level formatted with? Or is there some compatibility? Maybe that drive was never used..but this is unlikely?
 
There are a lot of different controllers from different vendors; you'll go nuts trying to find the one used for your drive.

There is a "do all" ST506 interface project from David Gesswein that might be what you want--depends on your level of curiosity.
 
predator99 said:
I am quite sure the drive is OK. I have tried to low level format the last track (614) only with speedstor (dont expect that any data is there) and it seems to work fine. Afterwards I can perform the read / write tests on this track only without any error.
Click to expand...
There is always the possibility that track zero is hosed. If that's the case there is little to nothing to be done.
 
Chuck(G) said:
There are a lot of different controllers from different vendors; you'll go nuts trying to find the one used for your drive.

There is a "do all" ST506 interface project from David Gesswein that might be what you want--depends on your level of curiosity.
Click to expand...

Thats a great project. Its a bit expensive for a one off, but as a tool for reading and emulating its the bees knees...
... one other thought, its not been formatted RLL? I know its not an RLL drive but it was something folks tried...
 
Thanks for your replies!

The project from David Gesswein is indeed very interesting and new to me. But for my little task too much effort and too expensive.

As you can see in my list of tested controllers...
National NDC5125
WD1003V-MM2
WD1003-WA2
WD WA6VR (RLL)
...I already tried RLL and non-WD.

8-bit is an idea, I have some of these in spare and will try.

In the mean time I made an experiment with another ST-225: I low-level-formatted the drive in the WD1003-WA2. Then exchanged the controller with the NDC5125. Its a completely different one. Result: Drive is booting without problems...so differences between controllers cannot be that huge? I will also test this drive with a 8-bit controller later...
 
I’ve tried controllers from seagate and dtc in my XT in addition to the original xebec. While I didn’t try every possible combination, at no point was I able to swap controllers without low level formatting again.
 
I gotta hand it to the unintentional security of these older 2-cable drives and controllers. The amount of trouble a drive separated from its controller can cause when trying to read the data back is amazing. You have so many variables beyond the controller BIOS revision that suddenly comes into play such as chip revision and embedded firmware.
 
predator99 said:
In the mean time I made an experiment with another ST-225: I low-level-formatted the drive in the WD1003-WA2. Then exchanged the controller with the NDC5125. Its a completely different one. Result: Drive is booting without problems...so differences between controllers cannot be that huge? I will also test this drive with a 8-bit controller later...
Click to expand...
Actually it isn't very different at all. The NDC5125 is a rebranded WD1003S-WA4.
 
predator99 said:
In the mean time I made an experiment with another ST-225: I low-level-formatted the drive in the WD1003-WA2. Then exchanged the controller with the NDC5125. Its a completely different one. Result: Drive is booting without problems...so differences between controllers cannot be that huge? I will also test this drive with a 8-bit controller later...
Click to expand...

The XT generation didn't know what a hard drive was, so for 8 bit cards the controller manufacturer had to make their own routines - and those do vary between brands.
AT-era MFM cards generally uses the computer's ROM for the hard drive routines - so it's more consistent across controllers.

Another factor is that the low level format data generally never gets updated, and often loses it's strength. So even if you do get a match, you may still get a 1701 error until a fresh low level format has been done. Seems to be a routine maintenance procedure required every quarter century ;) Just something to be aware of. But if you've got more controllers to spare and time, it can't hurt to keep trying.
 
You might put your scope on the output. You should at least see clocks and data at the right speed. That will tell you something.
Dwight
 
SpidersWeb said:
The XT generation didn't know what a hard drive was, so for 8 bit cards the controller manufacturer had to make their own routines - and those do vary between brands.
AT-era MFM cards generally uses the computer's ROM for the hard drive routines - so it's more consistent across controllers.

Another factor is that the low level format data generally never gets updated, and often loses it's strength. So even if you do get a match, you may still get a 1701 error until a fresh low level format has been done. Seems to be a routine maintenance procedure required every quarter century ;) Just something to be aware of. But if you've got more controllers to spare and time, it can't hurt to keep trying.
Click to expand...

But it's the firmware in the controller, not the BIOS code that determines the way the data is recorded (address marks, ECC/CRC, etc.) and there's very little in the way of standards on that count--unlike IBM floppy conventions. Most ST506-interface controllers did agree with the WD standards in the area of ports and registers, so BIOS code is very secondary. But that doesn't solve the mystery of the low-level format.

You can't expect a WD controller to record with the same conventions as an Ultrastor or Adaptec one.
 
Are the 8-bit WD controllers at least consistent with each other? I have three - my WD1002S-WX2 is currently driving my 2 ST-225s but I also have an WD1002A-WX1 and WDXT-GEN2 Plus as spares. I do backup my data on the drives periodically but it would be nice to know a controller swap would be possible without loss.
 
Interesting topic I think...I will continue to test other controllers. Already done:
WD WA6VR (RLL)
Clipboard01.jpg

WD1003-WA2
National NDC5125
WD1003V-MM2
IMG_7780r.jpg

So I dont think the NDC is very similar to the WD even if its a WD clone?

I am going to test some of these with the disk I already formatted in the WD1003-WA2. If it works, the controller is compatible and I can skip to test the disk I like to recover...

IMG_7782r.jpg
IMG_7783r.jpg
 
So testing the XT-cards (in a turbo-XT) was not as esay as expected. Speedstor gives me no option to manually configure a hard disk with most controllers (except the Omti). Therefore speedstor seems to be useless on the other cards if the hard disk is not initialized using the controller BIOS.
Only results are therefore:
- the disk formatted with the 16-bit WD1003-WA2 is not readable in the 8-bit Omti
- all other 8-bit controllers cannot handle disks that are not formatted on the same controller
- none of all the 8-bit cards can read the ST-225 with unknown format
 
You're looking for a needle in a haystack -- and it's a needle that may not even be in there.

E.g., suppose the drive is toast or unformatted (at this point you really don't know that it actually is formatted). If it falls into either of those categories there's no controller that will be able to read it (without first doing a LLF).

I've got some not-so-common controllers if you'd like to spend some more time fishing in virgin waters. Just lemme know.
 
I think he still needs to check for data directly from the drive. There may be noting there to see. In that case it is a lot of work for not.
Dwight
 
If you want to send it to me (MD) and pay return shipping if you want it back I can try reading it with my MFM reader/emulator. Send me a PM.
 
You must log in or register to reply here.
Back
Top