hard drive heart transplant

orubap

n00b
Joined
Nov 3, 2005
Messages
41
A couple of months back my four year old 80gig sata which housed the entirety of mp3 collection died. Of course, I didn't have a backup. For nigh on a year, sorting out a NAS/WHS box was on the to-do list but I never got round to it before it was too late which of course I regret now.

The 60 used gigs on the drive made up around 12000 tracks, a small collection I'm sure by many standards but big enough that I considered a professional data recovery solution. So I contacted ABC Data Recovery and after a few phone calls and email exchanges I sent the drive off with the initial diagnosis of "degraded media" with a quote £269+vat, a good chunk of money but justifiable I thought thinking of all my perfectly formatted id3 tags. I can't recall the exact ins and outs of the conversation but apparently degraded media was a common enough problem and they were confident of a full recovery without even opening the drive.

After they physically inspected it however it turns out that one of the drive heads was damaged, and recovery now would involve acquiring an identical drive, and exchanging parts in a certified clean room, etc. Even if I supplied the second identical drive (I bought them as a pair oh so long ago) still the quote now was near the £500 mark. As much as I missed my music I couldn't justify that expense, so had the drive shipped back. Since then I've spent around £500 on a new machine and £200 turning the old one into a WHS box (better late than never). I now have a couple of terabytes of storage and no music to store :(

Having read THIS I am going to attempt it myself since at this point all I'd be losing is one more 80gig sata which is probably near the end of its life anyway. So besides expressing a long winded morale here of "backup today, not tomorrow" I come to the point of this thread...advice please. I'm thinking I'm pretty much in the same boat as that guy was with one drive that works fine and another with damaged heads, so swapping the platters would be the easiest solution (I imagine swapping the heads be nigh on impossible).

I plan on trying this on the trusty kitchen table with a fan blowing horizontally and slightly up to help reduce dust that may fall on the drive. In the guide linked above you can see the head assembly sits inside a little enclosure which overhangs the platter. However in the image below of a regular desktop hd it looks like the head assembly would be able to move well clear of the platter and I'm hoping my drives follow suit here.
hard_drive_image.jpg


Any thoughts/suggestions on how I should go about this or tools I might need beside torx bits and a needle nose tweezers? The following specs mention besides there being a single platter that there a two heads - does that complicate the job at all, how are they arranged?
specs.jpg


EDIT: Help a bro out, last.fm is a poor substitute for my music :rolleyes: Any percentage of tracks recovered and I'd consider the exercise worthwhile
 
Last edited:
A fan is not going to prevent all the dust from getting into your drive. You think that air blowing over is dust free?

If you still plan on doing this I wish you the best of luck. You are going to need it.
 
Lol, who said the fan would give me a dust free environment? It's an aide nothing more. Have a read on the linked guide or at least on google and you'll see it's not impossible.

I'm well aware my chances of success here are slim, but assuming it fails all I've lost is another old 80gig drive.
 
While I'm sure the dust is very bad for things overall I will say without hesitation its not at all critical for very short term use. I've actually run an older hard drive as the Windows boot drive with no cover on it at all and it ran just fine for about a month. It did not ever fail, it never ran longer as it was just a system I was testing and messing with, the mobo, etc was not even in a case.
 
I don't see how a 100% data recovery is possible if the head had crashed into the platter...
 
Just read the article, the process is about as much of a crapshoot as freezing your drive to unstuck bearings...meaning your mileage may very...a lot.

Good luck, though, if your data is that important and it is mostly recoverable, hopefully this teaches you (and others reading of your unfortunate circumstances) of the importance of backups.
 
I don't see how a 100% data recovery is possible if the head had crashed into the platter...

Initial diagnosis over the phone was degraded media, and full recovery was likely in their opinion. It wasn't until they actually tested the drive that they re-diagnosed it as a faulty head, where even though the cost had near doubled they wouldn't commit themselves to a likelihood of success.

Granted they offered a no-data no-fee service but imagine they had the go ahead and only a 50% recovery was possible - the £500 quote would still be the same.
 
hopefully this teaches ... the importance of backups.

Lesson severely learned bro. I was actually ready to build my WHS box around a year ago but I remember there wre issues with files edited 'on site' getting corrupted which put me off doing anything about it at the time. Still, no excuse for at least not getting a spare drive to copy off to.

Was hoping there be some hdd holistic healers here to part some advice but either way I think I'll try this later on tonight.
 
This. Head crash = you lose something. Quite possibly a lot of somethings.

.



I've opened up a couple hard drives inside a bathroom before. Clean the room, close the door and window, set up a 'table' of sorts to work on and have at it.
 
.



I've opened up a couple hard drives inside a bathroom before. Clean the room, close the door and window, set up a 'table' of sorts to work on and have at it.

This. Just make sure you wear surgical clothing (facemask, gloves, hairnet) to even ensure yourself a better chance at not contaminating your drive.
 
If you are only going to run the drive once, to copy the data off it, then I'd say dust is probably the least of your worries. Keeping the heads from contacting the platters while you are working is probably going to be the hardest part.

I'd rethink the fan idea though, I'd expect it would hurt more by disturbing the dust.
 
You might be able to get most of the data other than the head crashed data.

I find it odd how they can tell if the data is degraded when they never even plugged it in (at least this is what I can tell from your posts). Perhaps they tried to get a little more money from you.

I would do a platter and PCB swap at the same time, perhaps that would give you the most success.
 
TBF that's what they thought from the description I gave them: drive spins, BIOS hangs for a minute while detecting drives, carries on without recognising it etc.

I was wondering about the pcb. I think I'll do the platter first and rehouse it as quickly as possible. If it doesnt boot after that I can swap the pcb out afterwards without exposing the internals again (I'm assuming here).

Kitchen extractor fan is running right now for whatever good it will do before I start the surgery. Plastic gloves, cheap mouth mask, shower cap, and the tools are set to go. The fan idea has been scrapped since I've just found the one I thought I had in my spare room, isn't :rolleyes:
 
TBF that's what they thought from the description I gave them: drive spins, BIOS hangs for a minute while detecting drives, carries on without recognising it etc.

I was wondering about the pcb. I think I'll do the platter first and rehouse it as quickly as possible. If it doesnt boot after that I can swap the pcb out afterwards without exposing the internals again (I'm assuming here).

Kitchen extractor fan is running right now for whatever good it will do before I start the surgery. Plastic gloves, cheap mouth mask, shower cap, and the tools are set to go. The fan idea has been scrapped since I've just found the one I thought I had in my spare room, isn't :rolleyes:

The PCB is a simple swap. I would do it with the head swap to be certain. I wouldn't want the PCB to be corrupt or giving incorrect commands... secondly, I would want the chip to be familiar with the rest of the drive. I don't know too much about the data the PCB retains but from what I do know it should be done at the same time to get the highest chance of success.
 
Defiantly do it, I mean, why not? Worst case you are where you are now, without your music, but both drives are dead, which doesn't seem to be too much of an issue to you.
 
I have swapped platters before. Purchase an identical drive, identical firmware is preferred. Clean the bathroom and make sure you have a hair net, gloves and a mask ready. You should make sure that you have all the tools that might be needed. I have every type of torx available, so I wasnt worried.

Now you have to steam(makes dust settle) the crap out of the shower and put a towel under the door. Dont exit, you will have to stay there and sweat your ass off. Wait for the steam to dissipate and get to work.

I would only swap the platters to the new drive.

Once you are done swapping the platters plug it in and grab all of your data ASAP. The swapped drive has a very good chance of dying quick.

I did this twice. One drive lasted a couple of days and the other 6 months(wasnt used much, crap drive).

If you know someone with a sauna.....that might work better.


Good luck!
 
Mission: Unsuccessful.

Before beginning, I plugged the dead drive into xp to see what happens and the answer was: nothing. The drive spins up but windows refused to acknowledge it. Proceeding with the surgery I changed the platters around without mishap and a very minimal amount of dust and left the pcb's unchanged: plugging the good drive in now with its newly donated platter, xp this time round popped up with the found new hardware balloon :) (<- my face was actually like this), but..... the drive didn't appear in explorer :(.

Checking the Disk Management page the drive appeared there with the unusual label of ST1Disk2Head and had an exclamation mark over it's symbol with the msg unrecognised or unallocated, forgot to screencap. Weirdly the capacity was also reported here as 128GB instead of 80GB. So I thought to myself I'm at least in a better position than before. After that I then swapped the pcb's around: no change, and opened the drive back up checking the head assembly etc before plugging it in a few more times, still no change.

For the hell of it, I took the time to place the platter from the good drive into the dead one, and plugging that into xp, it still wasn't recognised in the slightest. On that note the fact I was dealing with sata here meaning I could hotswap was good as resetting the pc everytime watching the bios would have quickly become tedious. So in the end I managed only to get xp to partially recognise the drive and after an hour of fannying with it I gave up and the drives are just sitting bare now on the kitchen table.

It was actually a fun exercise over all, just a shame I didnt get my data. Anybody got a what.cd invite? No, seriously.

Since I didn't have my camera I only managed to take a few pics with my pda, meaning the quality isn't great but the macro isn't too bad. I'll post them up later with a few pointers for any one else who might try this in the future.
 
Bummer Doc! :(

When you're doing high risk surgeries you lose a paitent once in awhile.

It does sound like a learning experience and maybe you'll do better next time. :)

I'd like to see the pics when you get a chance.
 
for next time watch these videos... just so happened to watch them less than a few weeks ago and I imagine they would have helped you immensely. I learned a TON about recovering hard drives and one day might even try it myself. Good luck again and hopefully... as long as you haven't messed up the platters (turned any in relation to others) might still have a chance.

http://myharddrivedied.com/presentations/video_presentation.html

I suggest the toorcon and advanced data recovery ones...
 
Mission: Unsuccessful.

Before beginning, I plugged the dead drive into xp to see what happens and the answer was: nothing. The drive spins up but windows refused to acknowledge it. Proceeding with the surgery I changed the platters around without mishap and a very minimal amount of dust and left the pcb's unchanged: plugging the good drive in now with its newly donated platter, xp this time round popped up with the found new hardware balloon :) (<- my face was actually like this), but..... the drive didn't appear in explorer :(.

Checking the Disk Management page the drive appeared there with the unusual label of ST1Disk2Head and had an exclamation mark over it's symbol with the msg unrecognised or unallocated, forgot to screencap. Weirdly the capacity was also reported here as 128GB instead of 80GB. So I thought to myself I'm at least in a better position than before. After that I then swapped the pcb's around: no change, and opened the drive back up checking the head assembly etc before plugging it in a few more times, still no change.

For the hell of it, I took the time to place the platter from the good drive into the dead one, and plugging that into xp, it still wasn't recognised in the slightest. On that note the fact I was dealing with sata here meaning I could hotswap was good as resetting the pc everytime watching the bios would have quickly become tedious. So in the end I managed only to get xp to partially recognise the drive and after an hour of fannying with it I gave up and the drives are just sitting bare now on the kitchen table.

It was actually a fun exercise over all, just a shame I didnt get my data. Anybody got a what.cd invite? No, seriously.

Since I didn't have my camera I only managed to take a few pics with my pda, meaning the quality isn't great but the macro isn't too bad. I'll post them up later with a few pointers for any one else who might try this in the future.


ST1Disk2Head is a good sign to have. You need some recovery hardware and software like a PCI3000 with pata/sata adapter and you should be able to reset the drive.

I think you would have avoided this by doing the PCB swap at the same time.


Perhaps give the data recovery guys this information and they may be able to do it for cheaper now.
 
Watched the first couple of vids hmchk, good links, would have been good seeing that beforehand tbh. Plus that guy highlights something I came to realise half way through my transplant: swapping head assemblies may be easier/safer than swapping platters, especially in a multi platter drive. The foil clip contraption is a good idea, I used some needle nose tweezers to accomplish the same by inserting them closed and releasing them to widen as they're slid back over the platter but his technique would free up a hand for sure and would definitely be required on an assembly with more than two heads.
I think you would have avoided this by doing the PCB swap at the same time.

Perhaps give the data recovery guys this information and they may be able to do it for cheaper now.
Perhaps, reason I thought of doing it afterwards is that it seemed to work for the guy in the original article. As for the drive now it's already binned: I went to bed last night and left the drive precariously on top my tower with the cover just resting in place, not screwed down. The cats so fit to knock it off and have a play with the ribbon cable inside, there's a hole in it now where a claw punctured it :rolleyes: Beginning to think I don't deserve any music.


Some thoughts:
Surgical gloves: they keep your greasy skin oils of your shiny platters and are tight enough not to lose dexterity in your fingers. At the least use washing-up-gloves. Also a hair-net/showercap, and mouth mask are advisable.


PIC00011.jpg

On the laptop drive in the original article the heads move well clear of the platter to sit in a plastic housing, however in mine (and I'm am now thinking most regular hdd's) the heads range of movement is between a couple off mm's within reach of the spindle to within a a couple of mm's within reach of the circumference. This means to lift out the platter you have to partially de-construct the head assembly anyway to just get the heads past the circumference of the platter. Since you have to do this, it does in some ways make more sense to just swap head assemblies because as well as saving time it also reduces the contact you have with the platter and of course means your not exposing the under side of the platter or even the ones underneath any more than you have to.


PIC00010.jpg

In this pic you can see this protrusion of the arm being restricted by the metal stopper on the left. This stopper is what restricts the arms range of movement. It's a single screw to remove that metal piece and with that out the way you can now swing the arm past the edge of the platter.


PIC00012.jpg

With that piece off you can now see how the head assembly actually moves. This intrigued me when I saw this: I never really considered how the head moves but sub-consciously I suppose assumed it would be by stepper motor. The end of the arm is sandwiched between two static magnets and actually moves by electromagnetism. I'd have never have thought you could get the precision required here using this method. Anyway, the pivot the arm rotates on is all that actually holds the assembly onto the drive. Unscrewing that and the two screws that hold down the logic board at the end of the ribbon cable is all that's required to remove the whole head assembly. You should be able to achieve this without touching the platters once.


PIC00014.jpg

If the platters have to be replaced, (which would mostly be down to if the motor is dead) then of course putting them back in the right way round is essential. If your methodical in your approach this should be straightforward but accidents do happen and both sides are pretty much indistinguishable from the other. In the pic above (crappy pic, i know), which is a close-up of the platter, you should be able to make out two white dashes etched on the drive (they are visible when the spacer is removed from the spindle), one dash is shorter than the other. On one side they read short-long and on the opposite they read long-short, unsure if this is their intended purpose but it can help to make sure they're re-assembled correctly.


rwhead-1.jpg

Comp Sci students will be familiar with the above pic. Whether you swap platters or heads, either way you need to slide the heads back over the disc. Invariably you need to pry them apart slightly to achieve this and given the heads are positioned on the end of some very flimsy metal alot of care is required here. When the drive spins up these heads fly on a chushion of air so when they're in position they need to be touching the platter. Point being, dont pry them apart too much. If you do notice dust on the platter, turn them (or the whole drive) upside down and tap them firmly to try and get it off. For the stubborn pieces, if you wearing surgical gloves and haven't had them on to long, i.e. the tips of the fingers are still clean you can brush them away.


PIC00018.jpg

Hard drive parts make great cat toys. He really likes ribbon cable.


If I had any advice to give, it would be:
1. Don't put yourself in this position, BACKUP YOUR DATA. I lost seven years of mp3 collecting and as much as I think worst things can happen I'm still pretty pissed with myself. You don't need a raid5 or whs box, just a spare drive to copy your data off to once a month will suffice.
2. If your going to send it off to a data recovery house make sure they have a certified clean room. There's a lot of cowboys I seen advertising, trying their own home made methods and charging professional prices. Beware.
3. Make use of the no-data no-fee service. Send it off for a quote as they'll give you a price to mull over before you decide, and if you do opt for a return so you can attempt it yourself at least they'll be able to give you a (hopefully) correct diagnosis so you know what direction to go in. While the company I chose sent a courier at their expense to pick it up if you do want it returned you most likely will have to pay for the return postage, cost me £17.50. Also, if they can, they will give you an index of the recoverable files before you incur a charge, can be handy.
4. If the motor is gone in your drive then platters will have to be swapped, however if swapping head assemblies is a viable option then go for that instead.
5. I was being serious, any body a got what.cd inite? :rolleyes:
 
Great picture essay; I honestly didn't think it would work given the exact and very, very, very tight mechanical operations and clearance that a hard drive disk entails. Hopefully this serves as a lesson to those to back up their critical data!

Anyways, you'll get your data back in due time, 80GB goes by swimmingly fast nowadays. :)
 
Once you opened that hard drive casing while not in a ISO Class 5+ cleanroom, you already lost any chance to repair the drive. The dust that damages hard drives is not visible without a microscope.
 
People, please watch the videos I listed before stating things such as needing clean rooms, etc. to do any repairs. The video's explain why this is a myth and how drives actually make their own internal dust.

direct link: http://www.youtube.com/watch?v=Tg0Uli2_rwI&feature=related

Also... The point of opening up a drive is to get to the data... not to keep your drive running. The idea is data recovery... NOT drive repair for long term use.

For those that wont watch the video... When a drive spins up the disk, the dust will get thrown right off the platter and usually into special channels that help capture it. Also remember... your reading the data once... not relying on it for long term use.
 
Let me find that video that proves people can actually fly if they use the right kind of duck tape and a copper paperclip. :D
 
Let me find that video that proves people can actually fly if they use the right kind of duck tape and a copper paperclip. :D

Should be interesting as long as its been briefed at some of the most recognized industry experts conferences for multiple years by a recognized professional with a business run off their experiences to back up his credibility.

Let me know when this paperclip, ductape flying contraption expert is invited welcomed at the IEEE Aerospace Conference or some other professional gathering.

While you are looking that up, think about registering for the next SANS data recovery class that this speaker will be teaching. You missed his July class: http://www.sans.org/forensics09_summit/description.php?tid=3032
 
Nothing like a group of people who try to talk a problem to death huh? We've had more than 125 enterprise Seagate drives recovered in a clean room with a success rate of more than 90% performed by Aero Data Recovery, success defined as all data recovered. All done without sitting in a room while someone talks about their theory. :p
 
Yeorwned,

Looks like a good budget rate 1st line software only data recovery company. Glad you had good experiences with them. However, looks like they aren't certified to open up the drives. Instead they refer these drives to more experienced and qualified companies. So you are right... they are busy doing data recovery because they aren't spending time to get educated, trained and certified on being able to do hardware data recovery such as platter or head swaps which is being talked about here.
 
No offense, but ask any real data recovery pro about Scott Moulton, and you'll probably get a big :rolleyes:
 
Back
Top