• Please review our updated Terms and Rules here

Seagate ST-4038 MFM hard Drive Ressurection or Burial?

alan8086

Experienced Member
Joined
Oct 1, 2016
Messages
331
Location
Burnley, UK
Drive 0 Recal Error - the main issue!

Primary Hard Disk Failure - error displayed when the PC goes through its drive detect POST routine.

Hello - I've been getting serious about an ST-4038 mfm drive that has been in my ownership for the last 14 years. Originally purchased off eBay from the University of Central London.

I have it in a Pentium 100MHz box, connected to a Western Digital WD-1003 WA2 controller - all jumper options are at factory defaults.

  • Set to type 22 in the BIOS. All other primary/secondary drives set to 0
  • All on board IDE devices disabled
  • All on board floppy controllers disabled
  • I have the HD jumpered to DS0
  • I have no resistor pack in the termination socket - I have one from a full height floppy but its a 16 pin and the HD requires a 14 pin resistor pack.
  • I am using straight drive cables with no twists
  • floppy drives work fine through the adaptec controller - tried 1.2mb / 1.44mb / 360kb
  • I'm using a DOS 6.22 boot disk
  • I'm using a utility called SGATFMT4 which is a low level format/verify bit of Seagate software meant for these drives.

The drive is working now more than it was - I can hear the head mechanism moving now as it goes through the format/verify process. Initially, the only time there was head movement was at power on. No bleeps from the onboard speaker and no definite POST heat movements and nothing when I was running the SGATFMT4 utility.

That changed last night - on running the format utility, something 'gave' and the head started moving, like it un seized itself. I still get little action at power up. I expected a definite POST head movement routine as can be seen in other youtube videos.

I made my own video here: I keep referring to the drive controller as an Adaptec, ignore me! Its a WD1003WA2


Listening to the drive and noting the displayed errors when its formatting - am I looking at a dead drive or a slowly getting better drive?

Any opinions most welcome :)

I have a second vid - a bit more concise regarding settings etc. Another 15 mins of your life gone! I'll post when its finished uploading.
 
Last edited:
I've tried the same format routine but with ST-4038M selected. The format seems to go a bit quicker and with 'different' head noises. Some slightly alarming ones too. Its the same sector / cylinder/head number as the ST-4038, hopefully no damage being done?!
 
Its ok - I'm used to replying to my own posts - Drive Manager reported a format failure after about 8% I think this drive is toast so I will spare you all the agony of this thread.
 
Sarcastic wink?

Not at all; although I could wait it was an interesting thread and I thought I'd let you know that at least one person was following it. Sorry that it looks like it's toast but there are tools out there that might be able to tell you a little more about what's going on; a horribly wrong interleave for example could really slow things down but that's probably not the issue here.
 
Not at all; although I could wait it was an interesting thread and I thought I'd let you know that at least one person was following it. Sorry that it looks like it's toast but there are tools out there that might be able to tell you a little more about what's going on; a horribly wrong interleave for example could really slow things down but that's probably not the issue here.

Sorry - them voices in ma head again. Can you suggest some tools?
 
SpinRite requires a valid partition to function. It can't format or even access an un{Low Level}formatted drive. For that matter it requires a DOS format, as well. It can only access a drive by its valid drive letter.
 
SpinRite requires a valid partition to function. It can't format or even access an un{Low Level}formatted drive. For that matter it requires a DOS format, as well. It can only access a drive by its valid drive letter.

I thought I remembered something like that... any other suggestions? I use a non-PC OS for this sort of thing.
 
A formatting/diagnostic tool? Have you tried SpeedStor?
The main problem with SpeedStor is that while it can access, diagnose and mark as bad all those unformatted sectors all right it can lead you to believe (inadvertently) that the disk might be OK otherwise. But, for some reason it seems to miss track zero problems which can lead you to go down a rabbit hole with no end. I've had it successfully format a disk's data area too many times only to have it fail at the end and then be unable to write a valid partition in the MBR. I think track zero issues are the major problem with MFM drives.
 
Ah, a possible track zero error then? Speedstor does have a destructive testing option, and I *think* that would test the track the same as the others.

What I would do if a track zero error is suspected, is use the controller's built-in formatter to low level the drive. This usually will NOT verify the media, but all tracks should get formatted. Then use Norton Utilities 4.5 to manually examine the "physical" disk - no partition is needed, it can be all zeros. Scroll down through each sector and note if there are any read errors. If track zero is dead, you should still be able to scroll down through the sectors until you eventually reach something that is readable. If nothing anywhere is readable, then either the drive is totally bad or there was a cabling error.
 
Sometimes you waste far to much time trying to sort out old drives. I know I have but not any more. better of with a xt-ide setup in the long run.
 
Like a good GP I'm referring him to a specialist ;-)
I have a Western Digital diagnostic program that might reveal the issue here. I'm not positive if you actually need a WD controller (which the OP does have) to avail yourself of its usefulness. It claims to be able to indicate a possible track zero problem. I also have one or two LLF programs specifically for WD controllers.
 
Back
Top