Seagate ST38410A 8.4GB IDE Hard Drive, please?

[roy bates]

good luck i hope it all goes well.

[prowler]

Thanks, Roy and Steve.

I stripped the old drive apart today.

These so-called Medallist Seagate drives are fitted with a rubber protective jacket, which makes them more difficult to get at but, after a bit of thought, I found a way to remove it without risk of tearing it:
I inserted a plastic ruler between the jacket and the top of the drive mechanism to break the bond which had formed between them and removed it by carefully slipping the corners from one side and pulling it away towards the other. I am confident that it will be possible to replace it again intact.

Next, I removed the three Torx (T-8) screws securing the pcb and immediately the board came loose.

Two sets of contacts protruding from the drive mechanism make electrical connection with the pcb via pads in two clusters on the underside of the board - one of three pads, designated J7, and another of eighteen, designated J1 - when the pcb is screwed down.

If the drive electronics have failed, there is no visual indication of it, but that's not unusual.

It just remains now to strip the new drive, swap the pcbs and, if I'm lucky, salvage the lost data, and I'm hoping to get this done tomorrow with photographs.

[prowler]

When this drive failed, it wouldn't even spin up after restarting the machine.

However, when fitted with the new pcb it does spin up, but it still fails to initialize, and no movement of the drive heads is evident.

I am beginning to think that both the drive mechanism and electronics may be damaged, so I'm reluctant to try the new drive mechanism fitted with the original pcb in case it too should suffer damage.

If it makes any difference, I did notice that the original drive and new drives have different firmware versions (the original has version 3.03 firmware and the new one version 3.32), otherwise they appear to be practically identical.

I bought another of these drives from eBay last night after I had discovered how easily they can be disassembled, and it was dispatched this morning, so I will soon have another pcb to try with it.

If the content of the original drive is lost, then I will recreate it as far as possible on one of the new drives and mirror that with the other drive to prevent this happening again. After all, these drives now cost an order of magnitude less than the original drive did ten years ago when it was new.

Whatever the outcome, I will still upload some pictures if anyone is interested.

Edit: If the original drive mechanism is damaged, it hasn't had a detrimrental effect on the new pcb, because the new drive still works fine when fitted with it. (I have been taking anti-static precautions throughout. )

[OddbOd]

Originally Posted by prowler

If it makes any difference, I did notice that the original drive and new drives have different firmware versions (the original has version 3.03 firmware and the new one version 3.32), otherwise they appear to be practically identical.

The fact you didn't realise that this can mean the difference between success and total failure is the reason why you should stop immediately and seek advice from someone who really knows what they are doing! Read this "success" story of near failure and think about the fact that there are precious few cautionary tales along the lines of "I fucked up my drive because I thought I knew what I was doing".

If you choose to take my advice and the drive contents are really important to you (which I suspect is not the case otherwise you would simply replace it and restore from backups) then get in touch with Scott Moulton at myharddrivedied.com, a guy who is helpful and very well informed when it comes media recovery.

The tone of this post may seem harsh but previous posters have quite innocently given you poor advice to the point where I had to step in. Best of luck and please remember The Tao Of Backup.

[prowler]

Originally Posted by OddbOd

The fact you didn't realise that this can mean the difference between success and total failure is the reason why you should stop immediately and seek advice from someone who really knows what they are doing! Read this "success" story of near failure and think about the fact that there are precious few cautionary tales along the lines of "I fucked up my drive because I thought I knew what I was doing".

Hi OddbOd, and thankyou very much for your reply!

It might not look like it occasionally, but - trust me on this - I do know what I'm doing.

If the different firmware revisions have caused by first attempt to fail, I am quite sure that it won't mean catastrophic failure in the event of a subsequent attempt with another pcb.

Thanks for the link to the success story, it looks very interesting!

Originally Posted by OddbOd

If you choose to take my advice and the drive contents are really important to you (which I suspect is not the case otherwise you would simply replace it and restore from backups) then get in touch with Scott Moulton at myharddrivedied.com, a guy who is helpful and very well informed when it comes media recovery.

As you have guessed, the contents are not of the utmost importance to me. This is chiefly because I do have backups, but they're not bang up to date.

Hands-on help from a third-party is out of the question, because the drive contains confidential material.

Originally Posted by OddbOd

The tone of this post may seem harsh but previous posters have quite innocently given you poor advice to the point where I had to step in. Best of luck and please remember The Tao Of Backup.

Not at all, mate. Your observations are always welcome.

[roy bates]

will you be trying it again prowler?

[prowler]

Yes, mate. I've bought another Seagate ST38410A drive from eBay, which should arrive on Thursday or Friday this week, and then I'll have another go.

I have no idea what caused this drive to fail.

I got no warning whatsoever. One moment I was editing a text file with Notepad in Windows, and then, when I tried to save the file, I got a blue screen with the message that Windows could not write to the hard drive.

I assumed that there was insufficient space left on the drive to save the file, so I reset the machine with a boot disk in the floppy drive intent on deleting one or more files to make room, but the Primary Hard Drive failed when it tried to reboot!

When it spins up on the new pcb I tried it with yesterday, it makes no unusual noises, but there is no evidence of head movement at any time prior to the "Primary Master Hard Disk Fail" message after the POST, but the new drive mechanism still works fine when fitted with it.

I know that the original pcb is fried because the drive doesn't spin up when fitted with that board. However, if the drive mechanism is damaged too, then that will be the first time I have encountered that problem, so I am reluctant to give it up without at least one more try.

[roy bates]

even with different firmwares you should at least hear the heads move when the motor starts up.
maybe the heads are stuck.

[prowler]

Originally Posted by roy bates

even with different firmwares you should at least hear the heads move when the motor starts up.

That's what I was thinking, tbh.

Originally Posted by roy bates

maybe the heads are stuck.

Maybe they are, but then I can't imagine what might have caused it, unless the drive electronics failure - as well as interrupting the motor supply - also sent the heads slamming against the end stop in an attempt to park them and they are now stuck there.

Thinking these sort of things is what's stopping me from trying the new drive mechanism with the old pcb.

If the drive I'm expecting in the next couple of days doesn't provide a solution, then I might try to unstick the heads by gently shocking the drive in a direction parallel with the platters.

Would you happen to know if the drive heads park towards the centre or edge of the platters? Though I suppose they could be stuck anywhere, really.

Why don't hard drives have transparent lids?

[OddbOd]

Originally Posted by roy bates

maybe the heads are stuck.

A drive with "stuck" heads won't spin at all, this may just be a difference in terminology though as the term normally refers to stiction, I'm sure this is what prowler means and it simply doesn't happen with modern drives.

Originally Posted by prowler

Why don't hard drives have transparent lids?

Demonstration models used at trade shows do have transparent covers, they are usually made out of a special type of coated plastic that disperses static charges. The reason you can't buy them is because it adds to the manufacturing cost with no tangible benefits. Western Digital used to make the Raptor X but stopped when their customers finally realised that it was stupid idea that they wouldn't pay for.

Originally Posted by prowler

Maybe they are, but then I can't imagine what might have caused it, unless the drive electronics failure - as well as interrupting the motor supply - also sent the heads slamming against the end stop in an attempt to park them and they are now stuck there.

You really need to do some reading as your ideas about hard drives are straight out of the 1970's whereas the drive in question was manufactured no earlier than 2000.

http://en.wikipedia.org/wiki/Hard_disk_drive_failure
http://www.techarp.com/showarticle.aspx?artno=84&pgno=0

[prowler]

Originally Posted by OddbOd

You really need to do some reading as your ideas about hard drives are straight out of the 1970's whereas the drive in question was manufactured no earlier than 2000.

http://en.wikipedia.org/wiki/Hard_disk_drive_failure
http://www.techarp.com/showarticle.aspx?artno=84&pgno=0

Thanks for the links, mate. There's certainly some interesting reading there.

Hey, I'm not looking to become a hard drive guru here, you know. If I can't find a simple DIY fix for this drive in order to recover the data, I'll just replace it with the new units(s) instead.

[roy bates]

nah, im not talking about stiction some drives heads get stuck on the stopper ive seen this before,some even have a mechanical arm that releases the heads after the motor reaches speed ive seen that fail.

i all honesty,prowler wants to investigate the drive himself and fairplay to him. it is after all his drive and he really has nothing to loose but data on a already faulty drive.
say for instance,prowler wanted to take the top of the drive to have a look inside,peaple could say stop there dont do that but he may go ahead and have a look anyway it is his drive.

[prowler]

Originally Posted by roy bates

i all honesty,prowler wants to investigate the drive himself and fairplay to him. it is after all his drive and he really has nothing to loose but data on a already faulty drive.

Absolutely right!

Originally Posted by roy bates

say for instance,prowler wanted to take the top of the drive to have a look inside,peaple could say stop there dont do that but he may go ahead and have a look anyway it is his drive.

I'll probably give it up if the next drive electronics board doesn't get it working.

Eventually, I will open it to satisfy my curiosity, but I'll leave it until I have recreated the lost data as far as possible just in case I get a brainwave in the meantime. After all, once I open the drive the data will be gone for good.

[Amigan123]

You could try Spinrite it takes a long time to work but I have managed to resurrect some drives with it.

https://www.grc.com/spinrite.htm

Spinrite Demo on Youtube
[ Show youtube player ]

More info at link below
http://www.diydatarecovery.nl/DiskPatch.htm

[prowler]

Originally Posted by Amigan123

You could try Spinrite it takes a long time to work but I have managed to resurrect some drives with it.

Thanks for the suggestion, Amigan123, but the operating system won't detect the drive unless it initializes at boot time, which it steadfastly refuses to do since it failed - even with a fully working drive electronics board attached.

The other hard drive I bought from eBay arrived today and, again, it's a good one, but it has the same firmware revision as the last - v3.32 - so I've decided not to try it on the faulty drive mechanism.

Instead, I've decided to restore the partial backups to it as a first step towards recreating the content of the original, faulty drive.

If I do have any success with the old drive by trying to shock the heads into life, then I'll certainly consider Spinrite as a means of recovering the lost data if necessary.

[Amigan123]

Oh I see the actual drive wont detect.

As a last resort If you just want to recover the data old drives will still operate if you open them them up,
obviously not recommended as they are made in clean room and some dust will get in.

You could even swap the platters over and put them into the known working drive using gloves, if it exactly the same model it should work.

[prowler]

Originally Posted by Amigan123

As a last resort If you just want to recover the data old drives will still operate if you open them them up, obviously not recommended as they are made in clean room.

You could even swap the platters over and put them into the known working drive using gloves, if it exactly the same model it should work.

I may give that a try, but only as a last resort. I'm happy with the two replacement drives I've bought, as they seem to be as near as possible a matched pair and both are newer than the original drive.

Therefore, if I decide to try and swap platters or disk heads with another drive mechanism, it will have to be from yet another similar drive unit.

I've already had a look for a suitable one on eBay, but there isn't one I'm willing to buy at the moment...


Edit: On the other hand, OddbOd supplied a link to these 'myths' above:

1. You can fix hard disk drives by swapping their damaged PCBs,
2. You can rescue your data from a dead hard disk drive by moving its platters to an identical "donor" hard disk drive and
3. A dead hard drive can be revived by smacking it on the side when it spins up.

Certainly the first method didn't work in this case, probably for the reason given, so it's likely that the second method wouldn't be successful either.

However, there's no harm in trying the third method, I suppose, strictly as a last resort, though I'm not going to try shocking the heads back to life with the power applied.

But first I must discover what is preventing the drive from spinning up when it's attached to the old pcb, and possibly fixing that will actually fix the drive!

[prowler]

I had a few spare moments earlier this evening, so I tried to get my original, faulty hard drive mechanism working with the pcb from the second replacement unit I bought from eBay, but the result was exactly the same as with the first replacement pcb I fitted.

Both replacement hard drives are now reassembled and fully working. When I've recreated the content of the original drive on one of these units as far as I am able, I'm going to mirror it with the other one!

Unfortunately, the original drive mechanism will have to be be destroyed.

[LocalH]

Originally Posted by prowler

Unfortunately, the original drive mechanism will have to be be destroyed.

Once resigned to that fact, I hope you at least have some fun destroying it. It can be quite cathartic :P

[prowler]

Originally Posted by LocalH

Once resigned to that fact, I hope you at least have some fun destroying it. It can be quite cathartic :P

Well, it looks like I'm certainly going to have some fun here one way or the other. I've just had the secondary hard drive fail on the same machine! I couldn't believe it!

I do have one good thing to report, though: I had just completed backing it up! How lucky is that? However, having returned the machine back to standard configuration, that hard drive failed when I next booted it up.

It could be that the secondary hard drive controller has failed; the DVD-ROM drive on that channel wasn't detected either. But I'm inclined to think it might be the PSU. It's difficult to test all the supply voltages with the machine fully assembled in its desktop case, so I've decided that tomorrow I will strip the machine completely apart, give everything a good clean and then test the PSU to start with, repair or replace it if necessary and then refit everything back in the case, powering it up and checking each component as I go.

This is my main emulation machine, so it's vital that I get it fully working again soon. I have a great selection of spares, including two identical motherboards, so I'm hoping it's only a matter of time and patience.