Samsung HD103SJ Logic Board Needed (for 2Tb HDD)

ariberman

Distinguished
Jul 26, 2011
3
0
18,510
Seems the 103SJ is 'fragile' and logic boards blow like fuses! Got a replacement drive from a friend and canabalized the logic board 2 months ago. All was good for about 5 weeks. Now... same problems have returned. Ran out of friends with exact spares!

Who has a board for an HD103SJ Rev. A 2011.02? It looks just like all the other 103SJ boards, but has the SMNT 66 pin dil chip (not the Samsung) next to the 5 and 12 volt diodes.
 
G

Guest

Guest
Hey dude, I'm in a VERY similar problem.
I have a 2009.11 hd103sj (b) and i just can't find any replacements...
I just wasted £40 on a new one, but that is 2011.5, so no luck :(
Sorry i was no help...
 
ariberman, if the "SMNT" chip is actually an ESMT SDRAM, then the board may still be a match.

You can use my notes to confirm whether the two SDRAMs are of equal capacity:
http://www.users.on.net/~fzabkar/HDD/HDD_ICs.txt

If the Marvell MCUs are identical, but you suspect that the boards are not matched, then you may be able to circumvent this by transferring the 8-pin serial flash chip from patient to donor.

This chip is below and between the SDRAM and MCU in the following photo:
http://www.donordrives.com/media/catalog/product/cache/1/image/9df78eab33525d08d6e5fb8d27136e95/s/c/scan-110607-0003.jpg

jamsworld, you didn't mention your drive's symptom. If your drive has a short that shuts down your PSU, then check the two TVS diodes just below the SATA power connector. If either is shorted, then it can simply be removed. However, be aware that by doing so you will have removed the overvoltage protection from the affected supply rail.
 

ariberman

Distinguished
Jul 26, 2011
3
0
18,510
fzabkar,

Thanks for those insights! I'll set to looking at my problem again this weekend.

It would be a wonderful stroke of luck if this situation is just blown diodes. Can these be tested with a digital multimeter set to diode/continuity test while in circuit or do the diodes need to partly removed from the circuit first? (and if so... how is the best way... it's all SMT, but the diodes have pretty big tabs).

A.
 
You can test both diodes in-circuit. They are connected directly across each supply rail. This means that the reading will reflect the resistance of the entire load on each rail rather than just the diode itself. I always suggest that people use the 200 ohms range rather than the diode range since all you are looking for is a short circuit. As long as the diode doesn't test shorted, then it doesn't matter what its actual resistance is.