Guide: Open hard drive surgery

[Gizmo]

What do you do when you have a dead drive with data that you really, REALLY need to recover? Well, you can either send the drive to a data recovery outfit and spend lottabucks, or you can take the approach of the hardware hacker, and find another drive in your junk pile that is similar enough to swap platters.

Read how I do this on the Front Page:
http://www.aoaforums.com/frontpage/content/view/2490/1/

[Samuknow]

That is pure crazy stuff there mister...Amazing.

I will try thins on a couple of old drives I have. Just to give it a run through...

[Favu]

Now that is an awesome guide right there!!

Maybe the picture thumbnails could be a bit bigger though

top work giz!

[raphael2040]

I have a hard drive that needs recovering.

This seems like a damn good way of doing it! =O

Amazing guide, Mr Gizzy.

[Rondog]

You are definately nuts. I thought hard drives had to be handled in 0 dust environments?

[Aedan]

Only if you want them to last any length of time. Occasionally hard disks do create dust internally (especially after a head crash), so they do have a method of filtering air inside them. The lower the level of dust, the more likely the disk is to survive for any length of time after such surgery.

[Gizmo]

I wouldn't recommend doing this as a plan for extended operation of the drive. This is simply a desperation thing to get your data back.

That being said, I've had two drives in the past that I've performed surgery on that ran for quite a while afterwards (a couple of years). I also had one that ran about 2 days before it had a head crash.

If your environment is reasonably dust free to start with, and you take the precaution of blowing the dust off the platters (like I mention in the artice) when you put the drive back together, you stand a fair chance of having the drive last for quite some time, IMO.

[Samuknow]

So you do not want to rely on this for storage again. Get it running and copy the data to another source....

[Gizmo]

Exactly. Any way you slice it, you HAVE opened the drive, and there is a very REAL risk that you have compromised the operational integrity of the drive.

[Got EpOx]

Brilliant guide Gizmo, makes me wanna try it on my dead drive i have lieing around although i'm not sure that i have a similar Seagate control board. I shall keep my eyes pealed

[danrok]

Just came accross a little tip, for water damaged hard drives.

Originally Posted by BBC article

If you do spill water or a liquid on a hard drive, the temptation is to dry it out on a radiator.

Phil Bridge said: "Don't do this. The hard drive could be damaged more by oxidisation through drying out that from getting wet. We advise you to bag it up and get it to us as soon as possible."

[Daniel ~]

What's great about their radiator? ":O}

[skool h8r]

I'm confused, i thought that the data was stored on the platter. So am i right in thinking you are simply tranfering it to a drive with everything else intact and working, so you can backup?

Still, awesome guide mate!

This is the type of stuff i personally want to see more of, and i know a lot of the modding and tweaking community out there (not just AOA community) feel the same.

[Aedan]

Yup, the data is stored on the platters. Gizmo moved the platters to another drive that had an intact spindle motor.

[Daniel ~]

Ya did us all proud Gizmo!

[aztec61]

gizmo has a lot more patience and skill than I could muster for such a project. An amazing feat of info salvation!!

[benn686]

I'm not sure if I'm the first to try this, but I gave it a whirl!

I accidentally dropped my 400GB Seagate external PATA drive while it was running. Needless to say, it no longer works (I dont even think I hear it spinning). I had a 300GB Seagate drive laying around, they look the same, and even use the same controller board (or at least they show the same PCB model number/rev number).

Anyways, following the guide, I put the 'crashed' platters, which by the way on inspection look fine, they dont appear to have any scratches or anything on them, and I opened the working 300gb seagate drive and replaced its platters with my crashed ones. Unfortunately, even though the platters are definately spinning now, and windows makes the familiar sound when a usb device is plugged in, it never sees the drive.

I'm not sure if its because of my workmanship or if the size difference is the problem.

Also, Im not sure if this is normal, but I had it spin with the top cover off, and it looked like the arm rotated from the center, but stops a little more than half way before reaching the edge. I've never run a working drive with the cover off, but shouldnt the arm rotate the head from the center all the way to the edge?

Was trying to put 400gb platters onto a 300gb drive a doomed theory from the start, or is that I didnt tighten (or tightened too much) a screw here and there??

[Gizmo]

The fact that you are putting 400 GB platters on a 300 GB drive is PROBABLY at the heart of your problem.

[benn686]

Not sure if that was sarcasm, but if it wasnt, the guide shows a 30GB was used in place of the original 40GB, which is the same percentage difference between a 300GB and a 400GB.. why would you expect one to work and not the other?

The number of platters are the same, and internally/externally both seagate drives look identical. Even if somehow the controller couldnt access all 400gb of the new platters, shouldnt at least some of the data (300GB) be visible?

[Gizmo]

No, it wasn't sarcasm. My apologies if it came across that way, as it wasn't intended that way.

In the case of the 30 GiB vs. 40 GiB transplant that I did, I got lucky because both drives used the same platters and heads.

In your case, although the percentage difference in size is the same, there's no guarantee that the 300 GiB and 400 GiB drives use the same media or heads. As I noted in my guide, there's no guarantee that even two drive of the exact same model will have the same heads.

This is why I suspect that your problem comes from the different sizes of the two drives. However, that doesn't mean that I'm right, it's just a guess.

Did you swap the control boards between the two drives? If you've got 400 GiB media mated to a 300 GiB controller, then the 300 GiB controller may not know how to read the 400 GiB media, and even if it does, it may not let you access all of it (or even any of it) because of limitations embedded in the firmware. This would be another reason for things not to work.

Beyond that, you had to open the drive up to do any of this, and there's no guarantee that you didn't finish destroying whatever was damaged, or cause further problems in the process of doing this.

I do crap like this all the time, and I pay meticulous attention to every detail, and <I> think I got lucky with this. As I mentioned several posts back, this is to be viewed as an absolute last-ditch attempt at getting your data back, because the potential for disaster boggles the mind.

I hope I've answered your questions. If you still have questions, I'll do my best to answer them.

Originally Posted by benn686

Not sure if that was sarcasm, but if it wasnt, the guide shows a 30GB was used in place of the original 40GB, which is the same percentage difference between a 300GB and a 400GB.. why would you expect one to work and not the other?

The number of platters are the same, and internally/externally both seagate drives look identical. Even if somehow the controller couldnt access all 400gb of the new platters, shouldnt at least some of the data (300GB) be visible?