WD epic fail (ps overvoltage)

[shaun]

Hi all,

I got a drive today with data worth 200-300 EUR on it (namely, a hired typist would hack in the financial data that is newer than the last backup in one work day) and some private photos.

So send-it-to-a-pro is a no-go here, and the drive was left for my practicing.
The motor combo is completely blown, as well as Q6 and L2 which looks like a step down for the Marvell chips core voltage. Bad. The Marvell has a small hole blown into its case as well.
Since the ROM did read without any problems on our programmer, I swapped the ROM onto a very similar PCB (2061-701292-A00 instead of -000).
The motor now spins up, but the drive fails to init (seeks repeatedly)

As many components including the chip containg the data channel have been hit, I'd assume a failed head amp. I'll take the drive to the clean cabinet the next days to have the platters inspected, but what I need to know is whether the head stack from the PCB donor would match. As far as I read, the DCM is not as important as some think...

[drc]

Have fun with the head stack alignment...

[shaun]

You mean the tilting of the head stack axle, which is normally secured in alignment by the screw on the top cover?

[drc]

That's not quite how I'd phrase it, but you're referring to the correct concept.

[shaun]

It may well be that I'm not always using the correct termini, I'm normally working on the electronics side of life and hard disks. The last successful swap I did was on a 20GB WD, maybe this one would be a challenge. But I got it for practicing, so I'll do so

[shaun]

I'm still looking for the criteria to match in order to swap a head stack between WD drives. Can anyone enlighten me?

[drc]

It would probably help for you to post the full model number / DCM from the drives.

[shaun]

Yes sure.
The failed drive (*) is a WD800BB-00JKC0, DCM ESBHNTJCH

The one I'd used as PCB donor so far (suprised that it survived) is a
WD800JB-00JJC0. The PCB is the same for both, so I'd gave it a try.

(*) I opened the drive today in the clean cabinet.
There's a small stain on the top surface, no scratches as far as I could see with the microscope camera. The flex print has melted near the head amp.
Judging from the stain, it has blown up before the drive has spun up, therefore the heads where on the landing zone, so I may have had luck and the EMP has not killed many square millimeters of data

Now, it would also be interesting how to remove the stains. Which liquid would be suitable and available in the required cleanliness, and which kind of swab can be used?

I'm glad that this just educational stuff, the other drive on the table has just a firmware glitch (Seagate 7200.11) which should be much easier to deal with.

[msurgeon]

Now, it would also be interesting how to remove the stains. Which liquid would be suitable and available in the required cleanliness, and which kind of swab can be used?

I

Isopropyl alcohol followed by distilled water. Pardon the indelicacy, but a certain ladies' feminine product makes an excellent swab for the purpose.

[shaun]

Thanks, I should have 0,5l pro analysi isoprop sitting around, should be free from almost any dirt.
Which kind of product for women healthcare do you mean...? When I look around on my wife's shelves in the bathroom I can find some more or less "delicate" materials that could be used as a swab

[msurgeon]

Tampon. Bet you never thought you'd find that in a data recovery toolbox!

[HDD Spaz]

I would like to point out it should be unused. Have I just gone too far?

[N.C.]

Yes sure.
The failed drive (*) is a WD800BB-00JKC0, DCM ESBHNTJCH

The one I'd used as PCB donor so far (suprised that it survived) is a
WD800JB-00JJC0. The PCB is the same for both, so I'd gave it a try.

(*) I opened the drive today in the clean cabinet.
There's a small stain on the top surface, no scratches as far as I could see with the microscope camera. The flex print has melted near the head amp.
Judging from the stain, it has blown up before the drive has spun up, therefore the heads where on the landing zone, so I may have had luck and the EMP has not killed many square millimeters of data

Now, it would also be interesting how to remove the stains. Which liquid would be suitable and available in the required cleanliness, and which kind of swab can be used?

I'm glad that this just educational stuff, the other drive on the table has just a firmware glitch (Seagate 7200.11) which should be much easier to deal with.

I am sorry, to say that, but this drive is more difficult to fix.
You need to do the alignment again.
I can recover for you, but not in this price range, but maybe for 400 EUR....

Regards,
Janos

[shaun]

You are referring to the alignment of the head stack bearing on its axle? I was already thinking about how this would be done best. Nasty idea to secure the bearings in space using the top cover. Rather idiotic I would say...

The drive does not need to be rescued, I'm just practicing on it.
Do you know which criteria to match for the head stack...? No one here did so far as it seems
Luckily, it's just two surfaces in this drive.

Can you point out how you would do the alignment? With the top cover off, I would think, and then attach a shim between the upper bearing and the screw.

[guru]

Just a thought... You put your "ROM" from a 2M cache PCB onto a 8M cache PCB

[guru]

LOL ok I just read "ALL" of the post

[Azurlake]

My experience says that it is pretty important that, in Western Digital drives, the whole model (WD800BB-00JKC0) should match for headstack replacement... I cannot ensure it, though.

[shaun]

Damn...
I will have to double check, but I was almost sure that both PCB had the very same RAM chip.
Would there be a small possibility that WD simply put 8MB cache chips to all boards and let the firmware decide whether the drive actually "has" 2 or 8MB?

But anyway... as the difference between 2 and 8 MB of dynamic RAM is just one address line, and as the controllers are the very same version on both boards (that's for sure), there should be a chance that the unused address line of my 8MB cache is simply ... unused

As it seems I'd need to get a matching drive for the correct heads, so I will probably get an identical PCB as well.

BTW, what's so wrong about "ROM" (you put it in quotes) - although it's a bit stupid to call an erasble and re-writable memory "read only", it's still valid nomenclature to call them Flash ROM.
I prevent using "flash" nowadays, where everybody is "flashing" anything, mainboards, graphics cards, mobile phones, tv sets...

[shaun]

Azurlake,
this sound at least reasonable. I'm still wondering which letters are actually important for the heads. The bay is full of 55JKC0, 63JKC0 and so on. Anyone...?

[scratchy]

Thanks, I should have 0,5l pro analysi isoprop sitting around, should be free from almost any dirt.
Which kind of product for women healthcare do you mean...? When I look around on my wife's shelves in the bathroom I can find some more or less "delicate" materials that could be used as a swab

It is safer not to use them, as when the woman needs them and finds them missing, she will kill you - and I mean that - KILL you.

Use medical grade lint free stuff instead, it is much safer.