Lazarus?

February 5, 2008 12:25 PM

I still have hope for my dead HDD. How do I bring it back to life?

The other day, in an attempt to connect an external CD drive to my iBook, I decided to rip open my LaCie.

When it came to reconnecting the HDD to the LaCie hardware, I was stupid enough to do this with the power flowing. I was also stupid enough to try and connect the molex the wrong way round. A few sparks and a terrible odour later, I plugged the LaCie into the iBook. ...Not a whimper from the LaCie.

I took the HDD out of the chassis and connected it directly to my PC. Still dead. I know little about HDD repair but as I can't afford to send it to a specialist, it's DIY all the way.

I assumed that the problem was with the PCB. Luckily I was able to track down a board that matched the model of my HDD: Seagate Barracuda, model: ST3250823A, p/n: 9Y7283-511, firmware: 3.03.

The new board arrived today. I swapped it for the old one, reconnected everything, and plugged it in.

It spins up fine, which is an improvement at least. But after that, nothing. It doesn't do the clicky, clicky thing and neither will it mount. I've tried other HDDs in the LaCie chassis and they work fine. I connected the seagate to my PC, it spun up but still no clicky, clicky. I assume the problem is with the board... not an exact match? But, as I said, I know very little about HDD repair and so I'm worried that the problem is worse than I expected.

The new PCB was rather pricey but as a last resort, I'm considering taking a 12w soldering iron and swapping the suspect component. (I've located the component, the crack and the smell give it away, but I'm not sure what it is. It's a small black square with two thick legs at either end. The one on the old PCB is marked: QE R524, and the new one is marked: AE GP519 -- are these resistors? Do they share the same value?)

So, where do I go from here? Is the problem with the PCB? Will I see my work again? Is no clicky, clicky a bad sign?

posted by popcassady to computers & internet (9 answers total)

No clicky clicky is a bad sign. The read heads aren't moving so even if the thing is spinning, no information is being read off the platters. If you're really adept, you could remove the platters (with some gloves and a pretty spotless dust free environment) and swap them with a working HDD of the same exact make/model. Even then that's a crap shoot. Otherwise, if you can't send it to a data recovery service (they might not be any help either, but usually they give free estimates) then you might have to cut your losses and move on.
posted by Burhanistan at 12:35 PM on February 5, 2008

SpinRite is the best at this. It's the best money you can spend on fixing/maintaining a Hard Drive.
posted by TheDukeofLancaster at 12:47 PM on February 5, 2008

BTW I'm no expert, but Spinrite will only work if your read heads will move across the disk, if those are physically "fried" (toast, destroyed, furbared, whatever you want to call it) the only way to recover data is to take the disk into a clean room and open it up, put the actual disks in another hard drive and read the data. This is done by VERY expensive labs.
posted by TheDukeofLancaster at 12:51 PM on February 5, 2008

It doesn't sound like the circuitry (such as it is) inside the hard drive enclosure is the problem (assuming that is what you mean by PCB) - you mention that other drives work in the enclosure, so I think you have damaged the disk. I use a Mac, Windows PCs, and Linux, and I have to admit that when I have a problem with an external drive on a Mac I can usually fix it on a Windows PC. I went through something similar recently with a spare drive I was trying to resurrect as a Mac peripheral and found some useful utilities on the manufacturer's website - so I would check on the Seagate site and run the diagnostics on a Windows machine. I'm afraid you have probably destroyed the disk however.
posted by thomas144 at 1:11 PM on February 5, 2008

Granted this is way over my head, but is this a case where the old hard drive in the freezer trick would work?
posted by piedmont at 1:33 PM on February 5, 2008

It's very likely that the component that is burned and cracked is not the cause of the board failure. Something else, may have caused that to burn up, likely a diode or a resistor. So unless you have the troubleshooting equipment and the know what each test point is supposed to be (from the service manual) you are unlikely to fix it by swapping out the component.

I doubt the information on your hard drive is lost forever however. I have a feeling that you have the wrong board for the drive. Also, examine all components not on the PCB to ensure that you didn't burn those out either.
posted by bigmusic at 1:58 PM on February 5, 2008

Joy! I fixed it!

I went for broke and swapped the component -- I tried everything else, including the diagnosis tools (thanks for the pointer thomas144) -- my gut feeling was that the board was incompatible.

£40 for a small black rectangle that's probably no more than a common all garden resistor!

At least I've learnt my lesson. Come pay-day I'm going straight out to buy a dedicated backup drive.
posted by popcassady at 2:33 PM on February 5, 2008

At least I've learnt my lesson. Come pay-day I'm going straight out to buy a dedicated backup drive.

Oh come on! Where is your sense of adventure? What are the odds of it happening again? :-P
posted by ian1977 at 2:42 PM on February 5, 2008

How many times am I gonna have to jump in on a data recovery thread on AskMe to say that SPINRITE IS (mostly) SNAKE OIL?

As much as I love Leo Laporte, his incessant hyping of Steve-Hypemonger-Gibson's stuff does a real disservice to laypeople who are looking to "salvage" thier dead harddrives. I too once thought that SpinRite was "the best", until I learned that pretty much every reputable data recovery geek out there does *not* recommend it, because it can cause much more damage to a dying harddrive if it's used as the first line of attack.
posted by melorama at 10:11 PM on February 5, 2008 [1 favorite]