www.deadharddrive.com

jamiphar · Active contributor Joined: 06 Apr 2011 Posts: 5

I have a Toshiba MK3263GSX 320GB/5400RPM 2.5" SATA drive from a customer's laptop. The laptop's motherboard and hard drive were shorted out when the AC adapter was plugged in (I'm still trying to find the reason).

I've taken a look at the PCB to see if I could figure out what's wrong. I bypassed the "S" fuse and got the drive spinning again, but it still isn't accessible. Are there any other components I could troubleshoot?

I have a working 500GB Toshiba MK5055GSX with a nearly-identical PCB. Would I be able to take the ICs from the shorted PCB and have them work with the donor PCB? I have the necessary equipment for rework/reballing.

Here are photos of the drives:

If anyone can give me some tips on this, I'd really appreciate it!

--James

fzabkar · Deadharddrive regular +3 Joined: 02 Apr 2010 Posts: 210

I'm not a data recovery professional, so I can't give you an accurate answer, but I believe that you may be able to account for any differences by transferring the memory chip at IC602.

BTW, the fuse has a 4A rating:

Littelfuse Surface Mount Fuses, N = 2A, S = 4A:
http://www.farnell.com/datasheets/48294.pdf

See my HDD IC database:
http://www.users.on.net/~fzabkar/HDD/HDD_ICs.txt

I'd also check the onboard DC-DC converters near the TI motor controller. One must be providing the Vcore supply for the MCU. There must also be a +2.5V or +3.3V supply for the SDRAM and EEPROM, and MCU Vio. The preamp (inside the HDA) also requires a DC supply, usually +5V and -5V, although some work off a single supply. If a negative supply is present, then it will incorporate one or two coils, depending on the converter topology.

That said, if a drive spins up, then the supply rails are probably OK. The motor controller monitors each rail and generates a POR if any are out of spec.

jamiphar · Active contributor Joined: 06 Apr 2011 Posts: 5

So, can you think of anything offhand that would damage the drive further if I used the 500GB PCB on the 320GB hard drive with the 320GB IC602 chip? Is this the ROM? Would this take care of any firmware incompatibility?

Also, how should I test the other components you mentioned? Should I hvae the PCB alone connected to a USB-SATA adapter and check for voltages then? Would you be able to identify specifically which components you're referring to (not too many of them are labeled)?

I'm eager to give it a try, but I first want to make sure I won't make things worse.

fzabkar · Deadharddrive regular +3 Joined: 02 Apr 2010 Posts: 210

Yes, even though it isn't one, IC602 is in fact what the data recovery profession refers to as a "ROM". I suspect that any differences will be accounted for by a ROM transfer, but that's a question for others to answer. I know that Hitachi drives will sometimes be corrupted by a board with a non-native ROM (they call it NVRAM), but I don't know about Toshiba.

As for measuring the voltages, I would do it with the board powered up on its own via a lone SATA power connector. You could test the supply pins of the SDRAM and ROM by referring to the datasheets in my database. However, the safest test points are probably the coils, since a slip of the probes will be less likely to short adjacent pins. It would help me if you could scan the board in 1200dpi mode, and upload the area around the TLS25nn motor controller. I would also be grateful if you could upload the complete 1200dpi scans of both sides to a file sharing service, if such file sizes are inappropriate for this board.

jamiphar · Active contributor Joined: 06 Apr 2011 Posts: 5

I actually have scanned the boards, but the SATA connectors keep the board from laying completely flat, so the motor controller area is slightly out of focus. I can give it a shot with a different scanner, though, to see if I can get better results.

jamiphar · Active contributor Joined: 06 Apr 2011 Posts: 5

I just finished swapping IC602, and it worked! I've got the drive backing up via Acronis as I type this. Thanks a lot for the help!

I'm still going to try to get the other PCB working, but that's gravy at this point.

Thanks again!

jamiphar · Active contributor Joined: 06 Apr 2011 Posts: 5

All right, I've removed the SATA connector to get a better image:

There's still a bit of blurriness, but hopefully this will work. I've modified two SATA power cables to hook up to my benchtop power supply, so I can have both boards powered simultaneously and can check for discrepancies with the voltages.

fzabkar · Deadharddrive regular +3 Joined: 02 Apr 2010 Posts: 210

Your latest scan appears to have been unnecessary. Your posted photos were the "small" versions, but I notice that the full sized versions are here:

http://arndtcomputer.com/james/photos/mk3263gsxpcb1.jpg
http://arndtcomputer.com/james/photos/mk5055gsxpcb1.jpg

Sorry to have put you to all that trouble. :-(

Anyway, pin #8 of IC602 would be a Vio supply.

The left pins of the L2 and L3 coils would probably be the Vcore and Vio supplies.

I suspect that FL302 filters the +5V supply for the preamp. I would measure the voltages at each of the pins of the preamp connector, PJ801. The smaller traces would be the signal pins. I suspect that the two leftmost pins would connect to the voice coil.

Is there an SDRAM on the underside of the PCB?

Congratulations on the data recovery, BTW. :-)