Seagate ST4000DM000 PCB fried

[Cerbero]

Hi there,

I have a Seagate ST4000DM000 who not spin, and no noise. Verifying fzabkar's guide about point tests in this HDD PCB (http://www.users.on.net/~fzabkar/HDD/ST4000DM000_TVS.jpg), the multimeter shows these results, in diode scale:

5V TVS: beeps in both ways
12V TVS: 640 in one way and 1 in other
5V zero-ohm (right): shows 220~250 in both ways
12V zero-ohm (left): beeps in both ways

Looks like the 5V PCB is fried. But I'm confused what I have to do:

1) Only remove the 5V TVS will solve the problem, the drive will back to live?
2) Or, besides remove the 5V TVS diode, I need to link a wire in 5V zero-ohm, like this image (imagehttp://postimg.org/image/4xpnds2h1/)?


Thanks in advance. Regards,

[pepe]

Preamp is still a question, if it is damaged, u can easily damage your data for good.

[fzabkar]

Remove the 5V TVS diode and confirm that the resistance between the two pads is no longer zero ohms. Since the diode was responsible for the failure of the zero-ohm resistor, then a wire link will most probably be OK. If you have the requisite skill, replace the TVS diode, otherwise be careful. Make sure that your PSU is good, and that the voltage is +5V before you plug the drive in again. This is especially important if you have been mixing modular cables between different PSUs.

For the benefit of others who may be watching this thread, if you wish to test the board before installing it on the drive, then upload a detailed photo of the PCB. Sometimes a regulator IC fails when there is an overvoltage on its 5V input.

[mikail13]

Remove the 5V TVS diode and confirm that the resistance between the two pads is no longer zero ohms.
For the benefit of others who may be watching this thread, if you wish to test the board before installing it on the drive, then upload a detailed photo of the PCB. Sometimes a regulator IC fails when there is an overvoltage on its 5V input.

Yes.i have a HD103SJ's board like that.
i 'd remove and changed roasted 5V diode.but pcb is still dead.

Where is the regulator IC?
Can i repair or change that?

[fzabkar]

If your board is like this one ...

http://ep.yimg.com/ay/yhst-144375849714 ... 4301-7.gif

... then the regulators are inside the Dillon SMOOTH chip. Measure the resistance between each coil and ground.

There is one 8-pin chip near the SATA power connector, but I can't see what it does.

[mikail13]

... then the regulators are inside the Dillon SMOOTH chip. Measure the resistance between each coil and ground.

which is?

and how can I?

i have multimeter..

[fzabkar]

If you can't identify the coils, and if you can't provide a photo, and if you don't want to transfer the ROM to another board, then replace the 5V zero-ohm resistors with a fuse or wire link and hope for the best. Of course you would remove the shorted diode and verify that your PSU is supplying the correct voltage.

[pepe]

I think this is about the matter i was talking about. People without proper knowledge trying to apply some (otherwise good) advices from you, then the drive is in a worse condition. Then it is sent to some DR firm who will sweat to get the data back....

[fzabkar]

When I joined this forum 6 years ago, I was telling people that they needn't pay $1K - $2K for a Data Recovery when the solution was as simple as a zero cost DIY Diode Removal. Soon afterwards, I received a PM from a prominent data recovery professional who proposed that, in return for keeping these "secrets" to myself, the pros would look after me. Needless to say, I found this proposal to be abhorrent. Not only did it illuminate the mercenary nature of the business, but it made me wonder why such a trivial problem was treated as if it were an industry busting trade secret.

If the OP wants me to help him, then I would require a photo.

This article identifies the coils that I asked for:

Tutorial - Linear and Switchmode Regulators used in HDDs:
http://www.hddoracle.com/viewtopic.php?f=59&t=231&p=596

Otherwise here is a solution for US$50:
http://www.hdd-parts.com/14043001.html

[mikail13]

Sorry for poor English.

i couldn't remember what the coil's mean. i measure that.there is no problem.

after that i realise i couldn't proper link for 0 ohm resistor. one side was unattached.after proper link my Samsung Hd103SJ Hdd spins.

Thank you for sharing of knowledge
Specially to Fzabkar

Yes.i have a HD103SJ's board like that.
i 'd remove and changed roasted 5V diode.but pcb is still dead.

Where is the regulator IC?
Can i repair or change that?

[pepe]

When I joined this forum 6 years ago, I was telling people that they needn't pay $1K - $2K for a Data Recovery when the solution was as simple as a zero cost DIY Diode Removal.

I agree about the TVS issue more or less. The problem comes when guys come around obviously wanting to get into the DR business to earn a few bucks but know nothing about hard drives not to mention electronics. They can probably solve some simple cases, but with anything beyond their competence they will most probably cause more damage and give back the drive as unrecoverable. They are not users but home made 'DR specialists'. And if you tell them about electronics, which they have no idea about, they will not get wiser. Electronics is something to learn at school for several years to get the idea and the neccessary way of thinking, you can't teach somebody in 2 posts.
If somebody wants to risk his data, it is his business, but if somebody risks others' data, that's something else and we should not assist in that in my oppinion.
Back to the TVS for a moment.
Of cours it does not take much skill to remove a diode and checking a resistor. But it is good to make sure preamp is not damaged in the shock, which is not that trivial process. Most probably it survives thanks to the protection, but if not, it can cause a real disaster, rendering the data into a really bad and unrecoverable state.
So anybody who simply swaps a board without checking the preamp, puts the data to a risk, which is perhaps not very high, but still a risk. It is like russian roulette a bit. Wanna play?
Perhaps u can win once or twice, but if u play it enough times...

[pepe]

So anybody who simply swaps a board without checking the preamp, puts the data to a risk, which is perhaps not very high, but still a risk. It is like russian roulette a bit. Wanna play?
Perhaps u can win once or twice, but if u play it enough times...

This is BS.

If yu SWAP THE PCB (and ROM if necessary) either the Pre-amp is good and the drive will work (if it doesn't have any problem) or pre-amp is already damaged and it will either not work or kill the new PCB.

This is not BS, but i won't argue. Just put 2 pics here from one of my posts in 2010 and let them speak for me.

Attachments

file1.jpg (49.02 KiB) Viewed 15846 times

file.jpg (52.27 KiB) Viewed 15846 times

[marcel86]

If you already knew that the diode is busted then my recommendation is just replace that diode with a new one. If you can’t find the exact same thing just ask the store guy what replacement you can use. And you might want to check your supply too, because that would not happen if you have a good supply in the first place. If your data is really that important then it will be wiser to leave the techie part to the experts.

Regards,
Marcel B.

[Alexii]

In about 6 years of DR iv seen 1 case with a maxtor drive where damaged preamp seemed to wreak havoc with servo tracks. Maybe a handfull of dead PCBs killed by a preamp. All in all preamplifier damage is very rare in our lab. Extremely rare.

[pepe]

In about 6 years of DR iv seen 1 case with a maxtor drive where damaged preamp seemed to wreak havoc with servo tracks. Maybe a handfull of dead PCBs killed by a preamp. All in all preamplifier damage is very rare in our lab. Extremely rare.

I never said it is very frequent, but is it a good attitude to risk user data hoping there is no such damage because we are too lazy to check that crappy preamp? If you can save a case by some tests, why not do it? Would you lay your life into a doctor's hand who's lazy to execute some tests because he's lazy?
You (and your clients even more) were damn lucky your pcbs got fired and the drive not started clicking, otherwise it might have had got nice data patterns...
A professional attitude would involve thinking and measuring, not just 'ah, lets quickly replace that pcb and stuff the money into our pocket'...
Ok, i do not want to convince anybody, just made a side note on the matter.


And please do not send any drives to me, i have enough shit that's been touched by a few dr firms already.

[Cerbero]

Hi there,

I have a Seagate ST4000DM000 who not spin, and no noise. Verifying fzabkar's guide about point tests in this HDD PCB (http://www.users.on.net/~fzabkar/HDD/ST4000DM000_TVS.jpg), the multimeter shows these results, in diode scale:

5V TVS: beeps in both ways
12V TVS: 640 in one way and 1 in other
5V zero-ohm (right): shows 220~250 in both ways
12V zero-ohm (left): beeps in both ways

Looks like the 5V PCB is fried. But I'm confused what I have to do:

1) Only remove the 5V TVS will solve the problem, the drive will back to live?
2) Or, besides remove the 5V TVS diode, I need to link a wire in 5V zero-ohm, like this image (imagehttp://postimg.org/image/4xpnds2h1/)?


Thanks in advance. Regards,

Gentlemen,

Firstly, really sorry for late reply, I had time to verify this issue only yesterday. Anyway, I'm posting the feedback to let you know (and sharing information) who the ohm diode by-bass procedure worked here. The data is safe.

@fzabkar, follow the PCB images with the fix. I don't snip the TVS diode entirely because I will resolder it to send the HDD to RMA service. I have other PCB questions, but I will open a new topic about it.

Thanks for the forum, and to all members for sharing knowledge.

Regards,

Attachments