Seagate ST3500641A 7200.9 500GB Recovery

[71337]

Hi,

I'm trying to recover data from my hard drive, I believe that the issue is corrupt firmware, but first let me explain the symptoms.

Drive was in an external USB caddy and the power supply "popped" and blew a fuse during electrical storm. (yea I know I should have turned things off)

Anyway, the drive spins up OK and sounds normal, but is detected in the BIOS as 128GB. Everything I try, including TestDisk, read the drive as 128GB. In test disk I can manually set the geometry and recover the partition as a single 500GB but it will not write partition data as the geometry does not match. Using PhotoRec I can recover photos from the drive, but only the first 128GB. Drive contains only photos and a few videos.

Thinking the PCB was faulty I obtained a matching replacement, but this caused the drive not to spin up at all. I swapped the CMOS chip from my original PCB to the new one and it spins up again, but still only reports 128GB.

SeaTools and SMART diagnostics all report that the drive is fine.


Does this sound like a corrupt firmware? If so how do I get the correct firmware and flash tool? I tried Seagate customer services who were, lets say, useless.

If not firmware, what else can I try apart from sending it away for recovery?


Drive details:
Barracuda 7200.9
500GB
Ultra ATA

ST3500641A
P/N: 9BD048-520
Firmware: 3.AAE
Datecode: 07085
Site code: AMK

Thanks in advance

Tim

[71337]

* Note, there was no MagicSmoke(TM) when the fuse blew and the pop was from the caddy power supply which now no longer works.

[Vulcan]

Does this sound like a corrupt firmware?

No.

There are some ambiguities in the story so far, but for starters, I assume (though I don't think you've said this), that you have removed the drive from the USB enclosure and are now attaching it directly via ATA. In that case, your "128GB" reported in the BIOS may be due to your BIOS being so old that it doesn't support 48-bit LBAs. Other possibilities also exist...

I'd probably start by using MHDD or HDAT2 (or sg_identify under Linux) to read the Identify Device response from the drive, check the reported drive capacity in that response, check for the existance of an HPA (although there's no good reason why one should exist now, and which would set that specific size), and go from there depending on the results.

P.S. What (exact) OS were you running when you used TestDisc & PhotoRec? I assume that OS was being booted from your usual boot drive in this PC - is that correct? If so, what capacity is that boot drive?

[northwind]

I agree with Vulcan.

That specific size (128GB) implies that we're having a 48-bit LBA issue here (even though it is usually 127GB).

Try the drive on some other machine, perferably a modern one. If you dont have a second machine, and you don't know how/can't/are reluctant etc to flash your BIOS, I'd go ahead and buy a new USB Caddy if I were you. The fact that drive reports original size via USB but not via ATA, means that most probably your BIOS needs to be updated to support new drives.

Hey! Youre lucky!

[71337]

Hi,

Thanks for the replies. Should probably have put the comp spec in. Since the USB caddy blew I have tried on two different machines, one an Intel P4 on an ABIT IC7-Max3 mobo and onboard IDE controller. More recently my new computer, an AMD Phenom II on a ASUS Crosshair Formula IV with add-on PCIe IDE card which works with other IDE drives. BIOS is defiantly not old. I cant remember exactly what brand the card is, but it uses the JMicron chipset. Running Windows 7 64-bit. Haven't used caddy since as the PSU doesn't work. Don't know if I even still have it.

TestDisk was originally run using a boot CD downloaded from here: http://www.sysresccd.org/SystemRescueCd_Homepage

I have also run the Windows version (non 64 bit version). All non-used drives were unplugged to avoid confusion. Boot HDD for the Windows version is a 128GB SSD, and no I'm not getting it confused.


I'll have a look at MHDD and HDAT2 and post up what I find.

Thanks

Tim