Data recovery and disk repair questions and discussions related to old-fashioned SATA, SAS, SCSI, IDE, MFM hard drives - any type of storage device that has moving parts
July 8th, 2013, 6:02
Hi,
this weekend I used the wrong Voltage on my external HD. It was (unknown to me) set to 19V instead of 12V).
Model: Samsung HD103SI (1000GB/R54/32M)
So all I want to do it access the drive, remove the data, and throw the thing away. I have read in a few threads, that I could remove the diodes, plug it into my PC and maybe this would bring the thing back to life.
Is this really true?...I can see that the one diode is light brown. There is also the burnt smell lingering as well. So it´s definately a short circuit. I have also tried to find another HD which has the same Model number and part number and can´t find one anywhere. This is where things start getting complicated. Before I invest 1000´s in data recovery and buying replacement gear that doesn´t match the HD, I just want to ask whether taking the diodes (or just one of them) out and connecting to the pc might work without causing more damage?
your,
Rob
July 8th, 2013, 8:35
Hello robertdbuckley,
please make photo of your PCB and publish it here
BR Jure
July 8th, 2013, 9:04
Hi...as requested, my HD Photo:
- Attachments
-

July 8th, 2013, 9:05
Hi,
You need to explain it a bit more.
You removed the drive from the external casing and connect it to a PC. What happened?
Also, publish the picture of the PCB here.
So we can check the diode's, if has external ROM, etc.
July 8th, 2013, 9:46
islamm3rouf wrote:u can remove all of them
WOOOAAAA there!!!..... before you remove the diodes and re connect it to the PC.... we need to know the circumstances of the failure..... If your PSU is at fault here, you WILL damage the drive with the TVS's removed....
July 8th, 2013, 9:50
robertdbuckley wrote:this weekend I used the wrong Voltage on my external HD. It was (unknown to me) set to 19V instead of 12V).
Sorry.... just read the OP initial post..... If the drive has come from an external casing and the failure is indeed as described, then removing the TVS's should be OK as long as you know that your PC PSU is operating correctly....
Best practice would be to replace the defective TVS's ..just in case...
July 8th, 2013, 10:55
SUCCESS!!
Thank you for the replies. I removed the burned out TVS Diode and the HD started without a problem. I´m copying the data as I write.
What should I do with the damaged HD? I suppose I could solder a new TVS in it´s place could´nt I? I´ve never done it myself, but It seems a shame to throw away an HD when it works ok.
Can you buy these diodes over the internet?
Cheers,
Rop
July 8th, 2013, 11:14
Just as a closing comment. This was a bit of an eye opener for me as this weekend I lost 3 HDs. My Macbook Pro´s Western Digital 500gb started clicking and couldn´t boot, then 2 HDs due to the wrong voltage setting on the adapter.
Mechanical HDs are pretty vulnerable. How are SSDs in comparison? Are they a much more reliable data storage device, or do they also have their weaknesses?
Thanks,
Rob
July 8th, 2013, 11:35
Everything has weaknesses.
Best thing you can do is have backups!
July 8th, 2013, 12:17
robertdbuckley wrote:Mechanical HDs are pretty vulnerable. How are SSDs in comparison? Are they a much more reliable data storage device, or do they also have their weaknesses?
Overall, SSDs are not very reliable at all, however, it all depends on what brand you buy. I would suggest sticking with Intel and Samsung for now. We use a ton of SSDs in servers and everything else is pretty much garbage under heavy loads except these two brands (we use Intel 320, 330, 520 series and Samsung 830 with great success).
July 8th, 2013, 12:36
Once the diode is gone, or remove the drive should work normally as if nothing happened, but it will not have any over voltage protection, the best thing is to put a new one on from another drive and your hard drive is as good as new. Only that component was blown to protect the drive. Perhaps the two other drives you said in your post was damaged the same way...
I'm not sure of SSD's though, too may read's and writes might damage sections of the chip, good old hard drives have proven their worth time and time again.
July 8th, 2013, 15:11
Dmitry Postrigan wrote:Overall, SSDs are not very reliable at all, however, it all depends on what brand you buy.
This article would suggest that OCZ is one brand that should be avoided:
http://www.behardware.com/articles/881- ... tes-7.htmlNote the 30% and 40% failure rates in their Octane and Petrol products (Indilinx controller). Vertex 2 (SandForce controller) is quite bad as well (10% failure rate).
When I tried to explore the reasons for the failure of certain SSDs, my posts at OCZ's forum were at first closed, then deleted. All I asked for were voltage measurements. It seems I hit a nerve. In fact I suspect that OCZ may be operating their NAND flash arrays at 2.8V rather than 3.3V, but I have been unable to receive any feedback other than a single case at this forum. If my suspicions are correct, then I suspect that the SSDs are failing because they don't have enough time to perform critical housekeeping before the NAND supply decays to 2.7V (the lower limit) after a sudden power loss.
July 8th, 2013, 16:21
fzabkar wrote:Dmitry Postrigan wrote:Overall, SSDs are not very reliable at all, however, it all depends on what brand you buy.
This article would suggest that OCZ is one brand that should be avoided:
http://www.behardware.com/articles/881- ... tes-7.htmlNote the 30% and 40% failure rates in their Octane and Petrol products (Indilinx controller). Vertex 2 (SandForce controller) is quite bad as well (10% failure rate).
When I tried to explore the reasons for the failure of certain SSDs, my posts at OCZ's forum were at first closed, then deleted. All I asked for were voltage measurements. It seems I hit a nerve. In fact I suspect that OCZ may be operating their NAND flash arrays at 2.8V rather than 3.3V, but I have been unable to receive any feedback other than a single case at this forum.
OCZ has nothing to do with common failures for SandForce based SSD drives.
All FW updates was dirrectly made by Sandforce team for all vendors who developed SSD on this MCU.
If my suspicions are correct, then I suspect that the SSDs are failing because they don't have enough time to perform critical housekeeping before the NAND supply decays to 2.7V (the lower limit) after a sudden power loss.
Disagree. For most of the nands reducing the voltage affects a lot better on reading and viability.
July 9th, 2013, 3:08
Spildit wrote:I'm glad that you got your data back.
Now if you want to re-use the drive just solder equal diode(s) on the PCB. The easy way to get them are from HDD PCBs that you no longer need (PCBs from damaged drives with good TVS).
They ARE directional.... just observe the correct orientation when replacing....
July 9th, 2013, 5:57
To me, OCZ = gamerz & toy SSD.
Powered by phpBB © phpBB Group.