Hello,

Because of using Thermaltake PSU's modular Sata cables on a corsair PSU, because the plugs were same and I didn't know this could possibly be an issue if the in and out connectors were the same, one of my hard drives died. Somehow, only one suffered damage, even though I had 3 other drives connected to the same supply. This drive now doesn't turn on at all.

I ordered and received a donor board for the drive (100710248 REV C ) and have been testing various components on each before swapping anything.

A few of the components are different on the boards despite having same revision and model. I am guessing it is normal and varies based on supply and cost differences at time of manufacturing, but I still want to be sure about best course of action for repairing this to recover data at least.

Some of the components I could notice with different numbers on them, which seem relevant:

Old Board:
RAM: SK hynix h5ps5162kfr
BIOS/EEPROM?: 25S81
TVS Diodes?: QA vne 32, kvp 48a (Judging from another post by fzabkar, the latter number is related to manufacturing date and place)

New Board:
RAM: Samsung k4t51163qq
BIOS/EEPROM?: Winbond 25q80Bws14
TVS Diodes?: qa vne 46, 6lf vne 49

I have tried to check the MOSFET or Pre-amp? (marked as question mark in the image, 4403C), and it seems to be fine, but I am not sure if my method is reliable for this case. I tested 4 pins on each side for continuity with each other, and 3 on the other. All the three 0-Ohm resistors marked in red are open, and the diode next to them is shorted.

I tried to search for datasheets of both bios chips, but can't be sure that the ones I found are actually for those chips. But they both seem to be 1.8v 8M-bit serial flash memory, with Winbond supporting Quad SPI while the other only dual SPI (again, can't be sure if datasheet I found for 25S81 is actually for that brand)

Please suggest if I should replace the 4 components that are definitely faulty and marked in red, and perhaps check some more that you can suggest and replace them, OR if I should use the new board and replace the BIOS chip from old board despite the differences in make/model? While I regularly create hobby circuits and sometimes repair devices, this will be my first experience with replacing a SMD with rework station (I will practice on another board first, but any recommendation for temperature setting for BIOS replacement is welcomed)

Thank you very much in advance.

Hi, swap chip rom from old pcb on new, enjoy

If you damage the "ROM", there are only a few people who claim to be able to regenerate it, at very high cost. IMHO the least risky DIY approach (for someone with no soldering experience) would be to remove the shorted 5V TVS diode and flow a blob of solder over the associated zero-ohm resistors.

_________________
A backup a day keeps DR away.

Thank you magnetepazzo and fzabkar.


I have soldering experience but no SMD experience, and the rework station I got is the cheapest one I could get because basically it's just for this repair and I don't trust its temperature setting. I have another old drive board to practice on, but the thermal characteristics of that PCB might be different and give me a wrong idea of which temperature I should use.

The only remaining tidbit I have about only replacing the damaged components is actually another of your post from past, where you had mentioned that in case some other components are damaged other than just TVS diode, then the pre-amp will get damaged too. I am guessing it's the MOSFET I had marked with question mark? Do you have any suggestion to be more confident about the remaining components? Like if there is another test point I should check before connecting the drive.

Preamp damage is common in WD drives (because their "protection" is stupid). Many Hitachi models have the same design philosophy, so they would also be prone to preamp damage. Older Toshiba models aren't much better. Seagate and Samsung models are properly protected, so their preamps tend to survive.

You can test the PCB after your repair by measuring the resistances between ground and each of the coils (3R3, 1R0, 1R0). I expect that the fourth coil (6R8) would be part of the -5V supply for the preamp, so it would normally be grounded.

https://s.yimg.com/aah/yhst-14437584971410/918062911-1.gif

If there are no shorts, then power up the PCB and measure the voltage at each coil. In the case of the 6R8 coil, measure the voltage at the anode of the nearby diode.

_________________
A backup a day keeps DR away.