Thank you fzabkar

I followed the advice you've mentioned here (and elsewhere) and removed a TVS diode that was playing up, and have now got a resurrected hard drive thanks to you . It had some important data that would otherwise have cost me absurd amounts of money to replace. Let's just say none of the data recovery places in Australia (melbourne) who I contacted wanted to answer my questions about diodes and instead told me my heads were probably broken. I'm sure they knew better.

I'm extremely grateful for everyone who has given their time to people on communities like this, and for sharing their knowledge.

Thanks again.

When you call a DR company and tell them "my hdd is dead" or "my hdd is clicking", then most probably they will GUESS what the problem is. There is no way for them to tell which problem is the one your disk is having, without actual diagnostics of the actual disk.

No offense meant, but i have been accused in the past for long-distance diagnostics (which were ofcourse wrong), for trying to "rip people off" (because i PREDICTED head failure by saying "MOST PROBABLY heads are bad", whilst actual problem was with the fw - as another company has indicated).
The accusation was insulting, wrong, and of course, since then, i REFUSE to tell my opinion on the phone (even if it is too obvious). If I wanted to rip people off by giving fake diagnosis, I would have done it all these years with ease.

Also, if someone calls me and asks me about diodes, zero ohm links etc, I will NOT give them any info too. For two reasons. Reason a) This is my business and i have to protect it
Reason b) There is a chance i give someone guidance about something and they completely screw disk up. Then that someone will pay me a visit and ask for compensation because "i have mis-lead them".
Been there, done that.

PS. Glad you made it work. Be advised that your disk now is not protected from over voltage (TVS 's are there for a reason). Copy your data somewhere else ASAP and have it replaced.

_________________
http://www.northwind.gr
SandForce SSD Recovery
Ransomware Reverse Engineering - NoMoreRansom! partners

The OP clearly said that he mentioned diodes. Presumably he told the DR companies that his drive wasn't spinning, so there would be absolutely no reason for them to conclude that there was a head fault. To me it's a clear case of dishonesty or incompetence.

_________________
A backup a day keeps DR away.

We get dozens of calls from "know it alls" who have read about TVS and such on the Internet calling to tell US what the problem is with their drive. For you to call the company dishonest or incompetent for suggesting that there maybe a head fault rather than give free (possibly incorrect or damaging) advice about fixed a PRESUMED problem, is grossly unfair.

End users also have a tendancy to be "economical with truth" shall we say, fitting the fault they want it to be around the actual facts. How many times have we heard of people wanting to change a PCB on a spinning drive, or swap platters on a clicking one?

Not spinning could be any number of things apart from TVS such as a bad PCB, bad ROM, seized, stiction... I could go on.

It's clear you have a vendetta against all DR companies, but to call a company you don't even know dishonest and incompetent for mis-diagnosing a fault one the telephone without even looking at the drive is downright disrespectful and malicious.

I agree with northwind about not giving free advice fully, as it will often come back and bite you in the ass and provoke "prophets of doom" like you to pop up out of the woodwork and have another spot of "DR pro bashing".

_________________
PC Image Data Recovery
http://www.pcimage.co.uk

New!! HDD-PCB.COM for all your PCB and donor HDD requirements!

In my humble opinion, a burnt diode could mean overload on the drive, and overload could mean preamps and preamps would mean head swap necessary.
Just an example of ONE reason.

OK another example, maybe better: We receive tons of phone calls from people telling us that "their hard drive is BURNT" (meaning it doesn't work).
Please diagnose it.
I'd say 90% of these cases, pcb is ok. So it is NOT burnt.

_________________
http://www.northwind.gr
SandForce SSD Recovery
Ransomware Reverse Engineering - NoMoreRansom! partners

Hi, I know its a pain but I was hoping for some elaboration on this fix.
I have the same HDD as the op, today when I powered up the 5V TVS started smoking. I obviously turned off the power and began searching the tinterweb for a replacement PCB finding this thread on my hunt.

I followed the info and came up with the following:

12V TVS reads 26K Ohms both ways
5V TVS reads 6 Ohms both ways
12V zero ohms reads 5 Ohms both ways
5V zero ohms reads 30 Ohms both ways

Does this shed any light for people with more than half a brain cell (unlike me), or should I be looking for a replacement PCB or better still someone who knows what they are doing???

The 5V TVS diode is shorted, so remove it.

Also add a wire link between the "+" shaped pads to the right of the diode:
http://www.users.on.net/~fzabkar/temp/Z ... _Links.jpg

Make sure your power supply is good because you will now have removed the overvoltage protection from the +5V input.

BTW, your meter's leads and/or selector switch appear to have about 5 ohms of resistance. Touch the probe tips together and you'll see what I mean.

_________________
A backup a day keeps DR away.

A wire link is dangerous. If there is another short circuit on the board there will be fireworks.

_________________
Murphy was an optimist

Datarecovery in Belgium, Holland, France and Germany
Datarecoverytools http://www.drtools.eu

Test the board off the drive.

_________________
A backup a day keeps DR away.

Thanks guys, tried the suggestions and managed to pull all my data off, gonna bin the bloody thing now!!

Time to hunt for a newer HDD for my backups.

Thanks for the info fzabkar, I was in a blind panic yesterday, you were also right about the MM, time for a new set I feel!

To fzabkar;

Thanks for your posts and detailed info how to fix this issue. I have the same drive with the same issue. It went down 2 days ago. I took it to a friend who does PCB work, and he confirmed that your advise was spot on. He took off the diode, cleaned up the socket, and 10 minutes later, the drive spun up and mounted instantly when I turned it on.

He also suggested NOT putting the drive back in the original enclosure or using the same power supply since it probably caused the initial overload to the diode. Fortunately, I picked up an external drive enclosure at a computer store a few years ago, and it's working perfectly.

Hi guys. Signed up just so's i could reply to this post.
My drive (same) suffered the same fate, so thank you for the advice in this post.
I know the initial post is now a bit dated, but would anyone have pic's of exactly what they done?
I purchased a multimeter just to fix this problem, but i am unsure on a couple of things.


"Does your board look like this one?
http://www.ixbt.com/storage/ata2005/sam ... -plata.jpg

If so, set your multimeter to the 200 ohms range and measure the resistance of the two TVS diodes near the power connector. The 12V diode appears to be the vertically mounted part near the screw, and the 5V diode appears to be the horizontally mounted part a little to the right of the connector.

This article should help you identify the components:
http://hddscan.com/doc/HDD_from_inside.html

Confirm that neither of the zero ohm links are open. They should both test as shorts. If either is open, replace it with a 2A picofuse, or bridge it with a wire.

You can remove the shorted TVS diode by snipping it out with side cutters. The drive will work OK without it, albeit without protection on the affected supply rail."

With "testing as shorts" (using chassis as earth), should i not get any reading on multimeter? On the left diode (12v) i get a reading of 0.2 on the "south" side and no reading on the "north" side. The right doide i get readings on both sides (0.2).

With the 2 zero ohm links, the left one i get no reading either end, the right one i only get a reading (0.3) on the "north" side.
Should i be taking out the 12v diode, and bridging the left zero ohm link?
I hope this makes sense as i want to save what data i can from this drive.
Thanks for en excellent post guys. very helpful.
Cheers.

@rick69, the right diode is shorted, so remove it. The right zero-ohm link is open, so bridge it. You can run a wire link between the two cross-shaped pads to the right of it.

BTW, when measuring the resistance of a component, you should place your probes on either side of it.

Good luck.

_________________
A backup a day keeps DR away.

Thanks Fzabkar. Didnt think i'd get a reply, old thread and all..
Here's hoping it works..

Normally I am notified of any reply to subscribed threads by email. However, that facility has been broken for over a month now. I also don't see any private messages. Must be something I'm doing wrong ...

_________________
A backup a day keeps DR away.

Hi I'm having some problems with my HDD Samsung SP2514N (pata).

I accidentally stroked some of the components on the underside of the hard drive with a live wire that us used for case fans (the small 3 pin one) and it turned the pc off. The area touched and caused the mishap was the pins where the ide connector goes.

The hard drive now no longer powers up, no start up sound or nothing. There was no burning or observable damage done to any of the components that I could see, like some of the ppl who have had problems with their drives.

Following the advice here I checked the diodes and the 5v was shorted and the zero ohm link closest to it was open.

I shorted the zero ohm link that was closed with a wire.

I removed the 5v diode that was shorted. However the drive still does not spin up.

-===-

For reference here is a pic posted earlier as a quick reminder of the parts I'm talking about.

http://www.users.on.net/~fzabkar/temp/Z ... _Links.jpg

^ not my pic obviously.

-===-

The 12v diode reads on the multimeter 622 ohms south to north. North to south it's open (no current passes).

The zero ohm link nearest the 12v diode is closed (0 ohms - current passes).
The zero ohm link nearest the 5v is now (0 ohms - current passes) bridged with a wire since it was open before.
The 5v was closed (shorted) but has been removed.

-==-

http://www.users.on.net/~fzabkar/HDD/SP2514N_5V_TVS.jpg
^ not my pic again.
The two components that where above the 5v diode - one is closed (one labelled "inductor?" allows current through unhindered) and the other (smaller one) reads 1368 ohms west to east and 519 ohms east to west.

-==-

Any help would be appreciated, I tried a different pc and different connectors and drive doesn't start spinning/power up.

@royH, ISTM that the damage is likely to be quite serious. I have the same board ... somewhere ... so I'll try to help you with yours. Just give me a little time to find it.

_________________
A backup a day keeps DR away.

Thank you, hopefully it's possible to do something about it.

Happy to take a look for you, and fix for special hddguru rate

_________________
PC Image Data Recovery
http://www.pcimage.co.uk

New!! HDD-PCB.COM for all your PCB and donor HDD requirements!

Sorry, I can't seem to find my board. I know I had one because I uploaded a photo of it here:

http://malthus.zapto.org/viewtopic.php? ... =107&p=125

Here is a photo clip of the Vcore and Vneg (preamp) supply test points:
http://malthus.zapto.org/download/file. ... &mode=view

Here is what I think is the Vio regulator:
http://www.users.on.net/~fzabkar/HDD/SP2514N_Vio.JPG

_________________
A backup a day keeps DR away.