I have a wd3000glfs that the owner destroyed by not checking voltage and polarity on modular power supply before plugging in.

It visibly burned the Smooth controller chip as well as burning some surrounding traces.

I obtained the same model drive made in the same factory a mere 4 days later, transferred the U12 chip, but it is not working.

Ive done some Ohm measurements on the head stack pins and found 3 of them to be quite different between the drives. The head stack connector/pins are also noticeably different physically. Not enough to mis land or anything like that, but i can tell it is a different device.

Is there anything further i can move over from the original PCB to the donor PCB to increase compatibility with the head? As far as i can tell the PCBs are identical, at least visually. MCU is identical as well.

I also understand that the power issue could have damaged the original head, and that is why i am seeing the different readings. Does anyone have such a drive they would be willing to perform measurements on?


:EDIT: Just check a bit further and 2 of the 3 pins that don't match up (ohms) originate from the Smooth chip and burned area on the board. I suppose it is safe to say the heads got damaged as well as the PCB.

Disregard, Did head swap from donor drive and image is running.

Head swap on velociraptor? Good skills!

_________________
PC Image Data Recovery
http://www.pcimage.co.uk

New!! HDD-PCB.COM for all your PCB and donor HDD requirements!

The preamp gets its +5V supply from the SATA power connector, and its -5V supply via the SMOOTH IC. If there is an overvoltage on the +5V supply, and if D3 is shorted and R67 is open, then both the preamp and SMOOTH IC become exposed. The usual result is that both are destroyed.

Catastrophic failures in Western Digital PCBs:
http://www.hddoracle.com/viewtopic.php?f=59&t=1119

_________________
A backup a day keeps DR away.

Yep, that is almost exactly what the IC looks like on this board. Of course fuse and diode are fine. DOH!

I suggest to forward that research to WD

_________________
SAN, NAS, RAID, Server, and HDD Data Recovery.

I think i addressed this problem around 2009 or so. I could find that post if i had time...

_________________
Adatmentés - Data recovery

I am actually sending it over to him shortly. So we will see how it goes. I am waiting to hear from the owner to make sure she is ok with the risk of shipping overseas.

SO this copied the first 250GB relatively quickly, but the last 50GB has slowed down to an unreasonable rate. It is currently estimating 100 days to finish.

I am reviewing the data that is being copied and it is copying valid data, but incredibly slowly.

Im leaning towards the heads being worn out, or maybe weak form the start. They came from an eBay donor.
Or
the pending reallocated sectors slowing things down.

Is it possible/advisable to clear the pending sectors?

If it sounds like i am completely wrong, does anyone have any thoughts on how i can get this last 50GB copied?

What are using for cloning?

To find some workaround we should find the problem, which is kinda guessing from this distance.
I would check if the previously reading areas are still fast or they are also slow.

pepe

_________________
Adatmentés - Data recovery

I am using DDrescue, and also tried HDDSuperClone

Yes, i suppose i could have disabled background media scan from the start.

To be honest, i was surprised enough that the head swap worked, i wanted to get data moving over. At 250GB i am pleased with that decision, but i would love to get a complete copy on this.

Would it be advisable to disable media scan at this point, or is the damage done?

Attach the ddrescue log (map) file for analysis. Did you use HDDSuperClone and get the same results where it slows down after 250GB? If so, attach the progress log (project file) from the HDDSuperClone attempt also. I can tell more from a HDDSuperClone log.

_________________
http://www.hddsuperclone.com
Home of HDDSuperClone

Cool, i will post it tomorrow when i am back there.

I ran the first 250gb with ddrescue, let it grind for a few days in DDrescue, and then today moved to HDDSuperClone.
HDDSuperClone seems to actually be worst. It will go into a not responding state every few seconds, but it too is indeed moving data. Like DD, it is absolutely crawling though.

I imported the dd map rather than starting a new clone.

The hddsuperclone log could be less valuable since you imported the ddrescue map file once it hit the slow down and then continued with hddsuperclone. If you would have started with hddsuperclone, then the log would show more (at least to me for analysis). But post both anyway, as it is too hard to tell anything without either of them.

If you are using ddrescue and hddsuperclone for the cloning which indicates that you do not have professional hardware, this question just has to be asked. Did you do the head swap in a clean room / laminar flow workbench environment?

_________________
http://www.hddsuperclone.com
Home of HDDSuperClone

ok i will have them tomorrow for you.

You are correct in assuming i do not dedicated hardware for this.

The head swap was done in reasonable clean room, but by no means an actual clean room. So yes there is every possibility that there was a bowling ball at around the 250GB mark that boned me on this. That being said, i am observing it reading valid data in the HDDSuperClone via the preview window, so I'm skeptical that the heads or media is physically damaged.

Its odd because its not even trying sectors multiple times or anything like that, its just slow as hell.

This is being done for a friend, and the data is not of great value. More of a convenience to not have to reinstall OS, programs etc.

why not?

_________________
HddSurgery - Professional Data Recovery Tools