Samsung Spinpoint M8

[logicbox121]

Hi, I have a 2.5" 1TB Samsung hard drive [Model # ST1000LM024] that will not spin up. I know exactly what happened, I plugged in the wrong power supply into the enclosure and I think it blew a diode. There are two similar threads I've found on google that you can look at to get an idea of what I'm trying to do: http://www.techsupportforum.com/forums/f16/help-locating-tvs-diodes-on-my-wd-hard-drive-669949.html and http://www.tomshardware.com/forum/268965-32-samsung-hd103sj. The spinpoint m8 is supposed to have advanced shock protection so I assume that's why it just locked up. However, the drive has no labels on the pcb or guides that show what is what so I can't identify it. Can anyone help with locating the diode I need to remove or whatever other method I can use to get this thing to spin up so I can recover the data?

P.S.- I didn't see a method of attaching photos directly to a post, so I will put up links to them. I labeled the board with all numbers I could find to help.

http://imageshack.us/a/img210/1507/img8844i.jpg
http://imageshack.us/a/img507/7738/img8850m.jpg
http://imageshack.us/a/img841/9015/picture1bjs.jpg
http://imageshack.us/a/img441/3340/presentation1xz.jpg

 

[JustSomeJoe]

Shock Protection only protects against physical shocks, not over/reverse voltage, sounds like you killed it.

 

[logicbox121]

Yes. Shock protection term semantics aside, I'm assuming the fuse or diode that was blown can be cut off to allow power to get to the motor, which is what I'm trying to identify on the board. Obviously this removes any protection from future voltage screw-ups but I am only trying to copy the data off of it.

 

[fzabkar]

I suspect that the component with the "ZA" marking may be a 5V TVS diode. Check it for shorts with your multimeter. If you don't know how to do this, then see my FAQ:

http://www.users.on.net/~fzabkar/HDD/TVS_diode_FAQ.html

Edit:

SD05-7, Diodes Inc, marking ZA, 5V, 350W, UNI-DIRECTIONAL SURFACE MOUNT TVS:
http://www.diodes.com/datasheets/ds31594.pdf

The component to its immediate left looks like it may be in series with the +5V supply. If I'm right, then its resistance should be 0 ohms.

FYI, the other components are ...

MCU - 88i9322-TFJ2
SDRAM - K4H641638Q-LCCC
serial flash memory - 25X40BLIG
motor controller - TLS2602
crystal - "N211" is probably a YWW date code, the frequency is probably 30MHz or so
R220 / 1R00 - 0.22 ohm and 1.0 ohm resistors

A database of HDD IC datasheets is available here:
http://community.wdc.com/t5/Desktop-Portable-Drives/Datasheets-for-HDD-ICs/td-p/523950

The "5172" IC is probably a FETKY (MOSFET + Schottky diode) and together with the two 2R2 coils it probably constitutes the switchmode Vcore supply for the MCU (~ 1.2V).

The 4R7 coil probably generates the 2.5V Vio supply for the MCU, SDRAM, and flash memory.

"Advanced shock protection" is normally provided by a tri-axis accelerometer, at least on Seagate's ASG models. That appears to be what the vacant location in the top right corner is reserved for. There is a standard shock sensor near the motor terminals (white device, angled at 45 degrees), but this is present in all models, with or without advanced shock protection.

BTW, certain numbers are YWW or YYWW (Year / Week) date codes.

207, N211, 1212, 204, 1209

These all suggest that the components were manufactured during the early part of 2012, and that the PCB was assembled in March or April.

 

[logicbox121]

Thanks for the info fzabkar, those are some great resources. I'm going to pick up a multimeter next week just to have, but I got a little tweezer-happy and took the ZA chip off just to try, it didn't do anything. So then I tried the black squre next to it. Also nothing. I had ordered a replacement pcb of the same model and revision, but it didn't work, I suspect the info in the memory is not matching up to the disks. I'm going to bring both pcb boards into an electronics repair shop and have them solder the 8-pin bios chip (25x40BLIG) onto the new pcb. I'm hopeful this will get it working again, I will post the results of this after I have it done.

 

[fzabkar]

Why did you remove the second component? I wanted to confirm its function before we did anything drastic. You should have measured it before doing anything. I suspect that it functions like a fuse, but I'm not certain. If I'm right, then you will need to replace it with a wire link. If you had left it in place, you may have been able to flow a blob of solder over it. Hopefully you didn't discard the two components. We will still need to measure them.

As for the flash memory, some Samsung boards require a chip swap but others don't. I suspect that you may just need to match the firmware version, but I'm not sure.

 

[logicbox121]

Pardon my hastiness. Impatience coupled with the fact that I have another fully functioning pcb led me to judge the risk as low. If need be, those components can always be taken from the second board and placed on the original, doing an even better fix than the solder blob method. The second component next to the ZA actually shattered while removing it, and the fragments were highly magnetic, so perhaps it was a ceramic+iron inductor.

After mentioning simply transplanting the two removed components, I'm now split as to whether I should patch the original or flashing firmware/replacing bios chip. If I can flash firmware, that would be the easiest next step. Otherwise, replacing the two components may not solve the problem if the surge damaged say the motor control or something.

 

[fzabkar]

Agreed, replacing components, if you can, is always best. However, you are trying to recover your valuable data, so caution and patience are advisable. I always try to understand the nature of the fault before proceeding with a repair.

As for the second component, Samsung sometimes uses polyswitches or zero-ohm resistors in that location. Other drive manufacturers use fuses, usually 2A. The failure mechanism is that the diode goes short circuit after experiencing an overvoltage. This short circuit then results in an overcurrent condition which opens the fuse. The solution is to remove the shorted diode and bridge the open fuse.

Good luck.