Hard drive repair (longish)

On 25 Mar 2005 16:48:47 -0800, snipped-for-privacy@aol.com put finger to keyboard and composed:

I used to service the old Control Data BK7 series disc drives. These had removable disc packs with 10 platters. An undetected overnight head crash would leave you with a drive full of metal debris and 20 disintegrated heads. In those days the flying height of a head was about .0001", so the thickness of a fingerprint would have been enough to cause head-disc interference, at least according to CDC's tech notes. I've also seen a Seagate tech note that likened hard disc technology to a Jumbo jet flying at supersonic speed 1m above the ground, counting the blades of grass as they passed underneath.

- Franc Zabkar

--
Please remove one 's' from my address when replying by email.

You could practice first by performing your own vasectomy, and then reversing it. After that you might like to try hard drive servicing without a clean room.

N

try slaving the drive to a a machine with an existing XP or NT-based OS and see if you can read the data

Ken,

I've already tried setting the drive size manually without success. Part of the problem lies with the way LBA-48 seems to work. The drive parameters (C - H - S) are set to the maximum values (65,535 - 16 -

255) and then the drive is apparently queried by the BIOS and reports the full capacity. One of my systems with a bios that supports LBA-48 does not allow me to set parameters manually, the other still errors out.

On the other hand, your idea of using the capacit limit jumper is something I never even thought of. I'll be giving it a try and reporting success or failure.

If all else fails, I've got a few old expansion cards here. I've decided that I will give myself some practice removing and reinstalling SMT ICs on a scrap board before trying the real thing.

Thanks for the advice.

PlainBill

Not a bad analogy. I also like the one: A hard drive head is a little airplane flying around above your data, looking for a place to crash.

PlainBill

I could give it a try before trying anything potentially destructive, but I'm not optomistic. The new drive (and the old hda with the new electronics board) spins up as soon as power is applied. The old drive, (and the new hda with the old electronics board) won't spin up under any circumstances.

PlainBill

This has been reported to work on drives with unreadable sectors. In this case, it's a dead electronics board. Still, it's worth a shot if the board has a crack which will close if cooled down.

PlainBill

Mike, let's see if we can get on the same page here. I have a drive with a bad electronics board - won't spin the drive motor. On that board is a memory device which contains information for the bad drive.

I have a second drive with a good electronics board, and a memory device which contains information for the good drive. If I move the good electronics board to the bad drive, the bad drive now spins up, but is not useable - presumably because the memory device on the good board contains the wrong information.

One possible option is to move the memory device from the bad board to the good board. That way IN THEORY the good board now contains the information for the 'bad' hda. I see two problems with this. First of all, the soldering is at the limit of my ability. Second of all, does the memory device contain information related to the board itself?

Alternatively, if it were possible to read and save the information from each each device, then write it back, I could avoid soldering.

PlainBill

What a clever idea. Unfortunately, I had mine done years ago, so I'll need another subject. I've got my rusty tin snips & oven mits; drop your pants, wise guy.

So how DO data recover services work their magic on disabled/damaged hdd's?

Dan

NSM wrote:

but you originally said that the old drive with the new board on it DOES spin up... if that is so...then even if seen incorrectly in the bios... if you slave it to an existing OS...the data should still be readable

I would start by getting a datasheet for the memory device from the manufacture's website and see what's required to program it. If it's not socketed though I'd be very hesitant to muck with the new drive, I still think it'd be a better idea to find someone who can replace the motor driver chip though.

I've got a couple dead DiamondMax drives right here, they both go click click click and won't read.

instantly.

I opened a 20MB (yes megabyte) drive once simply because it was 12 years old and of zero real value and it continued to work for quite some time though gradually developed some bad sectors. I suspect newer larger capacity drives are a LOT more sensitive though, this thing used an actual stepper motor to position the head assembly. It was fascinating to watch it read and write though, I still have it on display on my bookshelf though a while back I rotated the platter and it ripped one of the heads off the arm, oops.

Of course the "ground" is also quite a lot more level than any field of grass you'd find on earth but it's interesting none the less.

I've

by

run

Forgot about that one, I've had some luck with it, though in this case it's highly unlikely since we know the motor controller is fried. The freezer trick mostly just works with drives that spin up but can't read reliably.

hdd's?

They do something conceptually similar, but they put the platters in their own special (and *very* expensive equipment) in a clean room environment. The read/write heads they use can be manually controlled to read the data, a hard drive's onboard controller and mechanics are just not capable of this.

Remove with a heat gun (mask surrounding components with something). To reinstall, put some liquid flux on the pads, and dip the chip pins in it as well. Put a small dab of hot glue or double sided tape on the bottom of the chip, carefully align it and stick it in place, then form a ball of solder on the end of the iron and carefully drag it across the pins while keeping the ball fed with fresh solder. With some luck the flux wil keep the solder from bridging and the chip will neatly be soldered down. I always thought it'd be super hard to solder this stuff but it turned out it was easier than I thought.

I didn't write this, JURB did.

--
?

Michael A. Terrell
Central Florida

If only you still had the negatives.