Hi

Samsung SP2014N hdd PCB burned.
Red circle in attached image is burned part.
so i try replace it.
but, when i power on it is burn again.

what can i do for this symptom?

i don't have same HDD.
i need a way fix PCB.

thanks.

It's better to check motor, preamplifier and vcm then use a new pcb and 'make it work'... The circuit at this stage is rather complicated.

I've just been through a reverse engineering process on a similar Samsung PCB with another forum member. If you have a multimeter and some time, then I can try to help you make sense of that area of the circuit. At face value, it doesn't seem very complicated.

Before you installed the replacement component, did you determine if it was a transistor, and whether it was NPN or PNP? If you have another spare, then measure the resistances between each pair of pins using the diode range of your meter.

I would remove the faulty transistor (?) and power up the board on its own. Then measure the voltages at each diode and coil, at the supply pins (1 & 27) of the SDRAM (datasheet below), and at the other transistor. Also measure the voltages at the unoccupied pads of the faulty transistor.

I suspect that the 4R7 coil is part of the Vcore regulator (+2.1V ?). The rightmost 100R coil is probably the -5V supply for the preamp.

The 5V and 12V TVS diodes have small filter inductors on their load sides. Confirm that the voltages on either side of each inductor are as expected.

K4S641632H-UC60, Samsung, 64Mbit SDRAM:
http://www.datasheetarchive.com/pdf/get ... .pdf&scan=

_________________
A backup a day keeps DR away.

And during the "reverse engineering" you didn't discover what transistor is, right ?

Thank God it's Christmas

I didn't need to. The problem was elsewhere. :mrgreen:

BTW, the OP in that case repaired his PCB and recovered all his data.

_________________
A backup a day keeps DR away.

Bullshit.

The circuit around the faulty component has nothing to do with the preamp. The preamp stuff (+/-5V supply and communications with MCU) is elsewhere. As for the other two, at least one of them is probably bullshit, too.

_________________
A backup a day keeps DR away.

My arse. When you have time and a real drive at hand - not a picture - do some experiments and simulate what can be the problem - there are also at least 1 to 3 parts that can be failing on pcb alone in addition to hda parts to make that symptom (big clue !!). And rebuild the schematic...

You really should see a proctologist in the new year. I don't think we could stand another year of your self indulgent stronzate.

Either you're ignorant, or you're deliberately obfuscating this thread. Once again, the faulty component has no bearing on the preamp. That's just bullshit.

As for a physical PCB, that would be preferable, but we can't always have what we want. Remote reverse engineering, across two hemispheres, is far from ideal, especially when the user is a complete novice, but it's possible. Furthermore, even in the absence of a resolution, the principles involved in the process may be instructive on their own.

_________________
A backup a day keeps DR away.

You should see a tech , instead - but a very good one. And read less 007 stories - there are better ways to ask for info, as people are not so fool. Get a SP similar structured series from Ebay , open it and ruin it in a certain way (hint - 2 steps) , connect pcb and see the failure mechanism. Or work with power...
Times are changed since in-line tube TV and Shugart floppy drives, dude. And the Internet is no longer a substitute for real life.
Merry Xmas anyway ! At least today forget circuits and shut down the PC.

Blah blah blah.

Tell me, do you really have a schematic for the OP's drive, or are you just blowing it out your arse?

If you do have a schematic, then it should be easy for you to confirm that the preamp is in no way related to the faulty component. If we were to give you the benefit of the doubt and assume that you are in fact a competent tech, despite the complete absence of any body of work in support of this tenuous assumption, do you still maintain that any of the three components (preamp, VCM, spindle motor) could be responsible for the failure?


And modems ...

http://dilbert.com/dyn/str_strip/000000 ... .print.gif


Unfortunately the reality of life is just too horrible to complicate. I suppose I could venture into DR (Data Recovery, not Diode Removal), but this requires the sort of drive and motivation that I find difficult to muster.

The formula for success seems simple enough, though, ie ...

DR wannabe + PC3K + manual = HDD Guru


And a Merry New Year to you.


I wish I could.

I need to fix my mother's computer, my garage door controller keeps losing its memory, my parrot chewed through my speaker cable and defecated through a vent hole in my modem (ADSL, not dialup), the stove needs new cooktops, the phone needs a new battery, my backup hard drive no longer shows up in BIOS (no referrals, please), my GPS has power problems, etcetera, etcetera.

_________________
A backup a day keeps DR away.

I confirm that if it is not one part - or max. 3 in worst case - on the pcb that failed on the very beginning , the problem is now on the HDA. Simple test - get a new pcb, that likely won't work per se, and see if the 'transistor' or the combo chip goes tits up. I see hundreds of these drives every month. If I wasn't so sure , I wouldn't offer service even without original pcb's or after partial tinkering. Of course not for free.
I can't give any other kind of information about this case.
P.S. On a stock of more than 100x 6y drives there were 2 that broke every pcb . Quick check of electrical parameters at HDA connections showed no significative difference from a good drive... and only dissection determined the problem - just for fun.

I'm convinced that's galactic bullshit, but I'm reserving my absolute judgment until I receive feedback from the OP, or from one of your more honest and knowledgeable colleagues.

I feel certain that the part is an NPN transistor. I'm convinced that it has nothing to do with any component inside the HDA. Instead it is most probably configured as a linear regulator, possibly to provide the +3.3V supply for the SDRAM (or maybe that's the function of the other transistor), or a second Vio rail for the MCU. The only other remote possibility I can think of is some obscure emergency retract function.

As for the HDA components, the VCM is driven from two pairs of MOSFETs in an H-bridge configuration. The spindle motor would be driven from 3 pairs of MOSFETs, one pair for each of the 3 phases. Each of these MOSFET drivers is internal to the combo chip. The preamp gets its +5V supply directly from the SATA connector via a filter inductor, and its -5V supply from a DC-DC converter elsewhere on the board. It communicates directly with the read channel in the MCU. I'm convinced that there is absolutely nothing in any of these circuits that ties them to the failed transistor.

As for seeing "hundreds of these drives every month" (yeah, right :lol: ), it is clear to me that you could see a million of them and still not understand how they work. In fact I suspect you see about as many of these drives as you see stiction faults in laptops. :roll:

_________________
A backup a day keeps DR away.

You must have a burr under your saddle . Stop thinking and talking like you have supernatural powers : God DO EXIST but IT'S NOT YOU, dude.

Read CAREFULLY AND SLOWLY and with a pair of glasses if it is not enough what I wrote :

I confirm that if it is not one part - or max. 3 in worst case - on the pcb that failed on the very beginning , the problem is now on the HDA. Simple test - get a new pcb, that likely won't work per se, and see if the 'transistor' or the combo chip goes tits up.

I know the hope is to make loose the patience and give an explained detailed info so it can be used and spread elsewhere , God (not you) knows the purpose or the intention, but, again, doesn't work anymore. Times are changed.

About the rest, there always will be attempts to put people under bad light , with me the effect is always like a fart on the moon / outer space. Live with it.

Hi BlackST! now understand more about "Lubranizzato" ... eheheh sorry for the offtopic guys but i couldn't resist
Bye
Luca

_________________
Data Recovery pro in Italy
www.ultrarecovery.com

It seems that your self indulgent tripe has impressed a clueless sycophant, but the rest of us need facts.

Speaking of which, "kope", from a previous thread, has kindly taken some additional measurements for us. He has a Samsung SP2514N which has a similar PCB. He confirms that the tab and centre pin of the NPN transistor near the motor controller both connect to +5V (1.6 ohms), and that the emitter (third pin from left, tab facing upwards) connects to Vdd (pin 1) of the SDRAM (1.6 ohms). Since the SDRAM is specified for 3.3V operation, it is evident that the transistor is configured as a 3.3V linear regulator, as I surmised.

Therefore, your assertion that any of three components inside the HDA could be responsible for this malfunction, or be damaged as a consequence of it, is stronzata galattica. That you would continue to assert this despite my corrections damns you as a black stain on this forum and on the data recovery community in general.

_________________
A backup a day keeps DR away.

It's only your brain that need a help:

Quote :

I confirm that if it is not one part - or max. 3 in worst case - on the pcb that failed on the very beginning

... means that SOMETHING in the PCB at the very beginning could have made the damage. That could be something beneath the "transistor" .If the damage went beyond (overload the power supply rail that has tried to compensate) it could have fried something else too.
OR reverse cause : abnormal sink from a problem on HDA parts (there are 3) cause crap power supply to rise power / spike and then the damage. Beside the Calypso I have 2 Poseidon chassis where the problem is in the HDA and made same damage on 2 PCB (one time the transistor, another one the combo chip+transistor) using a common power supply (the fault, if it was used another safer PS would have spared the PCBs, but it happened pre-screening a box of HDDs before refurb attempt so it was a quick way to give them power). Having seen it on Maxtor and on same series drives (both used) , to me it can perfectly be. The Maxtors were opened for curiosity (not yet the Samsungs), the motor was not stuck and apparently there was no visible damage to VCM and nothing strange came from preamp measurement (quick check). Nevertheless, they fucked two 7250 and 1 7250 + U500 + some SMD resistors almost immediately (the drive tried to start). Classified as "rotture di coglioni" exactly like other things and moved on.

If you can't understand and read between the lines, it's not my fault and if you don't get the info you want and it drives you mad, neither. I can't care even less about the cause or post-mortem diagnose, but what if the OP put another PCB and had on same environment the same damage or combo chip or something else fried ? Not all the people have dozens of PCBs eventually to waste or can get easily the chips and have what it takes to repair a PCB easily.

Finally , about the rest , why worry ? This forum and the data recovery community in general think the exact contrary (not DR that may be confused with diode removal), like it or not. Cruel world and forum, this one ! You'll have better luck next life Meanwhile, think about other stains. The ones on the modem.

I've always known you were a bullshit artist, and this thread is the proof I've been waiting for. Until now your chest beating has always involved vague claims, supposedly subject to the DR code of silence. But in this case your bullshit has been quite specific and able to be scrutinised.

The OP's drive appears to have a short on the 3.3V supply. This would suggest that either the SDRAM or MCU, or both, are damaged. It could also be that the motor controller is driving the regulator transistor full on. However, there is no relationship between any component on the +3.3V supply rail and any component within the HDA. If the OP had performed the measurements I requested in my first post, then your nonsense would have been exposed at the outset.


Even your diehard sycophants wouldn't dare risk trashing their reputation by defending your tripe. You may be able to fool some of them, but you can't fool all of them.


Bird faeces are easy to clean. It's the shit that comes down the phone line that's the real problem. :mrgreen:

_________________
A backup a day keeps DR away.

Have the decency to shut up and get some Valium, at your age is dangerous.
Recommended by sycophants and WHO, otherwise we'll loose an income and a fun source.

Your previous insinuation that someone would lie in order to elicit a response with the correct information puts your own responses into perspective. In fact I recall that one of your own colleagues accused certain unnamed data recovery individuals of intentionally disseminating false information.

Furthermore, this thread demonstrates just how much data recovery professionals know about hard drive electronics. It is significant to note that none of your colleagues offered the correct explanation. Instead that was left up to an outsider working from photographs, and a novice who is still coming to terms with a multimeter.


I'm always happy to provide entertainment. Glad you enjoyed it. :mrgreen:

_________________
A backup a day keeps DR away.