Rescuing Data From Newer Drive

[Mac128]

I have a 5 year old Edge DiskGo! drive with an ATA/133 DiamonMax Plus 9, 3.5 Maxtor 120GB mechanism inside. It has always had a problem reliably mounting via Firewire, but USB always mounted trouble-free. After several months of disuse (it has only ever been used as a archive drive), it no longer mounts by any means. It starts up normally and sounds fine, it simply will not mount on any computer via Firewire or USB. This is the first drive I have ever had suddenly stop mounting. Disk Utilities can't mount it. I tired swapping it into a LaCie case, but it only succeeded in mounting in the Finder, but ultimately was inaccessible. Mounting it again failed.

Given that it's not making any unusual noises and otherwise operates normally after previously working fine, what other steps can I take to recover it? Does the freezer method still apply to modern drives (i.e. putting it into the freezer, thawing and hopefully mounting)? Could it simply be the logic board, given that Firewire was always somewhat dicey? If so, how to work around it?

 

[porter]

I'd try plugging it into a Mac with an IDE connector, eg a B+W G3 etc.

 

[Charlieman]

I have a similar cheap enclosure which showed similar symptoms over the weekend. FireWire has never worked reliably on it, so I switched to using a well shielded USB cable almost immediately.

My disk appeared to fail yesterday when I had plugged it into my new Mac mini for the first time -- Finder mounted it sometimes but Disk Utility refused to recognise my partitions. Eventually, I spotted that the USB cable was wrapped around an external power brick, and simply relocating it cured my problem immediately.

 

[Mac128]

Update: I plugged it in today to try some of the suggestions and found it wouldn't spin up at all, then the drive itself began to smoke, or rather a small chip on the logic board caught fire. Clearly there was an issue here and only a matter of time. Well after 20 years, it was bound to happen one day. The question is, what can be done now? It would seem like I could swap the logic board with another, drive of the same model, or crack open the drive and swap it. Either way, I will have to locate another compatible drive. At this point, could a disk recovery service revive the data (which seems likely it should still be perfectly intact), without going through the exact same steps I have outlined?

 

[Christopher]

If you're willing to pay the price of a decent used car you can call DriveSavers....

 

[Charlieman]

Clearly there was an issue here

I despise that euphemistic use of the word "issue". You have a problem -- newspapers have issues...

It would seem like I could swap the logic board with another, drive of the same model

Which is all that a data recovery service would do. Of course, they have controller boards for every drive under the sun and an electronically safe environment, but nothing that you can't create for yourself at this level. Given that you have identified the point of failure so precisely, I agree that it is worth installing a controller board from a matching drive. Unless you can find a case history that identifies that controller boards are freely interchangeable for your model, assume that you need model match and revision match, which may be tricky given the drive's expiry.

I recall that we have been partially down this road on 68KMLA before when discussing OEM versus consumer hard disks. Every modern drive has an EPROM on board that identifies its characteristics, OEM vs consumer, its model number, its serial number etc. Your drive may be beyond revealing that information, but it would still be worth installing your drive in a PC in order to run the DOS-based Seagate/Maxtor diagnostics. It might -- low probability -- help in identifying a donor drive.

 

[tomlee59]

Update: I plugged it in today to try some of the suggestions and found it wouldn't spin up at all, then the drive itself began to smoke, or rather a small chip on the logic board caught fire.

I've encountered two instances (same brand and model of drive) recently where the spindle motor developed a shorted winding, and cooked the motor driver (releasing all of the magic smoke upon which all electronics depends). I hope this is not your problem, for swapping circuit boards does not help in this case. If it was just a spontaneous suicide, swapping boards will solve your problem. There's no simple way of knowing ahead of time, unless you can find/deduce the pinouts for your spindle motor and make a few resistance measurements.

 

[LCGuy]

Which is all that a data recovery service would do. Of course, they have controller boards for every drive under the sun and an electronically safe environment, but nothing that you can't create for yourself at this level. Given that you have identified the point of failure so precisely, I agree that it is worth installing a controller board from a matching drive. Unless you can find a case history that identifies that controller boards are freely interchangeable for your model, assume that you need model match and revision match, which may be tricky given the drive's expiry.

Really? I thought actual, proper data recovery companies (like DriveSavers) actually removed the platters from the drive and retrieved your data from it that way?

 

[MrMacPlus]

Which is all that a data recovery service would do. Of course, they have controller boards for every drive under the sun and an electronically safe environment, but nothing that you can't create for yourself at this level. Given that you have identified the point of failure so precisely, I agree that it is worth installing a controller board from a matching drive. Unless you can find a case history that identifies that controller boards are freely interchangeable for your model, assume that you need model match and revision match, which may be tricky given the drive's expiry.

Really? I thought actual, proper data recovery companies (like DriveSavers) actually removed the platters from the drive and retrieved your data from it that way?

If it was a mechanical failure they would swap platters. It's easier and safer to swap logic boards if it's a logic board fail.

 

[tomlee59]

Which is all that a data recovery service would do. Of course, they have controller boards for every drive under the sun and an electronically safe environment, but nothing that you can't create for yourself at this level. Given that you have identified the point of failure so precisely, I agree that it is worth installing a controller board from a matching drive. Unless you can find a case history that identifies that controller boards are freely interchangeable for your model, assume that you need model match and revision match, which may be tricky given the drive's expiry.

Really? I thought actual, proper data recovery companies (like DriveSavers) actually removed the platters from the drive and retrieved your data from it that way?

What they do depends on what they find. Generally, they open up the drives first, without having applied any power (to prevent doing any more damage). If the heads are damaged, then they will try transplanting the platters into a healthy host. If the heads and platters look ok, then they typically power up and see where things stand. If the drive appears somewhat functional (e.g., the track-following servos work, the read electronics are functioning, and the spindle motor spins at the right speed, etc.) software-based recovery methods might suffice. If not, and if a logic board swap appears sensible, they try that next. And so on. (At least some) data can even be recovered from crashed platters, but the recovery process is much more involved (and expensive).

 

[equill]

... I despise that euphemistic use of the word "issue". You have a problem -- newspapers have issues ...

Bravo! It is what Fowler might have described as a 'vogue word', here deriving from the BigBiz world that ceaselessly seeks to glamourize its tawdry existence, and to give that a significance that it does not have apart from beggaring nations. From BigBiz such metaphors pass to the equally valueless advertising world, and thence by way of the media—all of them—into unthinking governmental and popular acceptance. As you imply, issues come from or are sent from. Issues are not inherently problems, but value-neutral until they prove also to be something else as well as issues.

But one might as well, and not always successfully, inveigh against stakeholders, deliverables, transition (as a verb), leverage (as a verb), robust, and more in current metaphorical abuse by BigBiz. Of course, metaphor has the advantage of being imprecisely defining or definable, or dangerously misleading, in its intended meaning.

de

 

[Charlieman]

What they do depends on what they find. Generally, they open up the drives first, without having applied any power (to prevent doing any more damage).

I'll disagree with Tom's methodology a bit. If the recoverer looks at the controller board and spots a failure point on that, s/he won't open the drive immediately. Opening the drive requires a clean environment which costs money. Logically, s/he would identify the failed component. and work out whether it was on the basic I/O side or the ATA/IDE side. If it is on the ATA/IDE side, swap the controller board. If it is on the I/O side, swap the platters -- expensive.

 

[tomlee59]

It's not actually my methodology. What I described is what a DriveSavers technician told me is their approach. I agree that if they were to see something obviously damaged on the logic board, they'd likely alter their strategy. However, disassembling in a clean room does not involve much overhead for them, so that's actually not as big a consideration as one might think. Their approach is modeled on the Hippocratic oath: First, do no harm. So they do not apply power until they are reasonably certain that doing so will not make things worse.

 

[Flash!]

I paid a little over $1000 to get a drive "recovered" and all they ended up doing was swapping the logic board, something I could have done myself, and was in the procees of doing....i'd purchased an identical drive on ebay. But I ended up sending it to the specialists anyway coz the disk had a client's mission-critical data on it, and I didn't want the responsibility for making it a worse situation. Besides, if it wasn't the logic board I would be up the creek coz I don't have a clean room