Hi All,

Firstly, I would like to state that I am completely new to the art of DR. I am from South Africa, so responses and requests for information will be slow.
As such, please bear with me.

On to the issue:
I have the above drive. The drive was working, and plugged into a machine using a "drive walker" as we call it, transferring PDFs for a print run. I managed to damage, while transporting the drive back home, the component marked. Now, when I power up the drive, the second marked component heats up. I would dearly love to recover the data from this drive but if I lose it, so be it. Also, 1.5TB is quite huge here, so its a great loss for me considering the price of drives here. The attached image is one I found on this forum and is exactly the same as mine.

I can post a pic of my board tomorrow, with the 4R7 marked component removed. I have access to an electronic technician with a wide array of tools etc (Hes a SonyCam repairman)... If there are any questions, you can also mail me directly, if you wish, on brendan at adsend dot co dot za

Thanks in advance for ALL those many replies below. I will be documenting any repairwork done and should the drive survive Dexters Lab (Thats his REAL name!), I will definately be posting a how-to afterwards...

Regards,
Brendan.

I think you should contact member Nick_CT he is from your area if i'm not mistaken.

Ask for his advise.
The more you mess with the board the more damage you're doing.

Good luck.

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

From your description it sounds like just the PCB is the problem, straightforward and not expensive at this point. Just be careful who or where you take your drive to in terms of a non-DR place. I could tell you a few nightmare stories, as others could, of drives that go elsewhere first before coming in for DR and during this time things has gone from bad to a lot worse.

I'm based in Cape Town if you require assistance.

_________________
Death is nothing, but to live defeated and inglorious is to die daily.
Data Recovery Cape Town

Nick_CT,

DUDE!!!! Im currently WORKING in Cape Town and will be here for about another month... maybe.

Where exactly are you? Nick_CT,

Check PM.

_________________
Death is nothing, but to live defeated and inglorious is to die daily.
Data Recovery Cape Town

The damaged component is an inductor. It is an important component in either the Vcore or Vio supply for the MCU chip. The chip that is heating up is the motor controller. It contains the MOSFET chopper that energises the coil.

The worst case solution would involve a board swap and NVRAM transfer. The NVRAM chip is an 8-pin component -- it stores drive specific "adaptive" data that must be transferred to the replacement PCB. I have no idea what Nick_CT charges, but a typical firmware transfer or "ROM" transfer costs between US$10 - $20. Some board suppliers will even include it for free.

_________________
A backup a day keeps DR away.

So, Fzabkar, you say I should expect a quote under R200 in a worst case scenario? Ill be popping in to Nick_CT today to have him take a look at the drive...

Ill post here later.
Regards,
Brendan.

No it won't be R200 You are welcome to take it to an electronic repair shop if you like, if you feel that the only problem is an IC that has come loose. I think fzabkar was referring to the cost of transferring the ROM chip? I'll be amazed if you find a data recovery place that will do a job for $10-$20

_________________
Death is nothing, but to live defeated and inglorious is to die daily.
Data Recovery Cape Town

I'm saying that the worst case scenario would probably necessitate a board swap, in which case you wouldn't need to engage a data recovery specialist to do it for you. Paying a DR company to replace a PCB would be like asking a cardiologist to apply a bandaid to a cut finger. That said, it's to be expected that a specialist in any profession will have higher fees. Moreover, I believe that a data recovery company would also backup your data as a matter of course, which is a service that you may not need or want.

If you need someone to do the soldering work for you, then try your local TV repair shop. In fact some DR companies outsource their soldering jobs.

_________________
A backup a day keeps DR away.

Okay, okay... I hear you guys. Thing is, I have about 600GB of Video and about 300 GB of music on this drive. If I can rescue it, Ill be happy, but if I lose it, I really dont care as I have the music backed up and the video is replaceable with 4 hours of work.
So, what I REALLY want is the DRIVE repaired...
that said, I think replacing the board is unnecessary. Im currently trying to source the correct inductor to replace the one I damaged. Unfortunately, Im having a little problem: its quite easy to find a 4.7microHenry inductor to replace the one that was damaged. However, Im not sure if the one that was damaged IS 4.7... It DID have 4R7 printed on it, but Ive already found ANOTHER inductor with 4.7 emblazoned on it but that one was actually WAY different... so Im a little wary of what I put onto the board.
We managed to find a 4.7 (radial) on an old sony board BUT it didnt change the drives behaviour. (no spinning up, overheating chip), so I suspect that either the inductor was faulty or there is something else wrong. Also, we have access to thousands of old camera boards and just need to look. Im sure Ill be able to find a replacement motor controller in there somewhere... If the motor controller is heating up, what else could possibly be wrong? I think we can safely assume that both TVS are intact, since the heating means there IS voltage reaching the controller which means the TVS' are working as normal, or am I wrong? Also, since the controller is heating, doesn't that mean that something AFTER it on the board is faulty, or is my logic like my board: FUKCED?

However, I KNOW that there was nothing else I damaged... The screwdriver bounced DIRECTLY off the inductor... We checked the board for cracks and also checked EVERY track for damaged using a multitester. FLUKE ROCKS! Nothing wrong. Now, if I can get a few pointers, Ill be happy to try ANYTHING. Dexter says that he can move the NVRAM easily.

Nick, I dropped in on Friday but you weren't in. I tried calling you on Saturday, but all I got was "The Subscriber..." story from Voda. I understand from the website that I can expect to pay around R1200 for the repair. If thats the case, I may as well buy a new drive and drop this one in the nearest bin, since I believe I can now get a 2TB drive to replace this one for R700. See... Im not trying to be difficult, but Im not in the market for DATA RECOVERY, Im in the market for a DRIVE REPAIR. Considering this, what can I look at? My options are definately open if the price is right.

As I said, I want this drive repaired... HOWEVER, if repairing the drive is gonna cost me too much, Im not interested. I think Id rather go out and buy a new one and drop this one into a bin somewhere... cheaper and safer.

... and quicker.

4R7 is 4.7

Here is a datasheet for a typical HDD coil:

LPS3015 Series SMT Power Inductors, Coilcraft, 3mm x 3mm x 1.5mm:
http://www.coilcraft.com/pdfs/lps3015.pdf

Note that the DC resistance is very low (0.2 ohms) and the inductance is flat all the way to 10MHz. I believe that switchmode DC-DC converters in recent HDDs have switching frequencies in the MHz range.

BTW, I'm not a data recovery professional, so I can't be certain that a chip transfer is necessary for your Samsung PCB. I have seen threads where some DR specialists say that a chip swap is required, whereas others say that a straight board swap should work. I have also seen a thread where an end user was only successful after transferring the chip.

If you have a multimeter, measure the voltages at the two 4R7 coils. One will be the Vcore supply for the MCU, the other will be Vio, probably +2.5V or +3.3V. Also measure the resistance between each coil and ground.

The 1R2 coil is part of the negative supply for the preamp. Measure the voltage at the anode (non-striped end) of the B240 diode to the left of the coil. I believe you should see -5V.

_________________
A backup a day keeps DR away.

Seems like the best bet here is to just buy a new drive in terms of money, effort and time. Would be a shame to bin the 1.5TB Samsung though, seeing that it's just the PCB at fault.

_________________
Death is nothing, but to live defeated and inglorious is to die daily.
Data Recovery Cape Town

From online research, it looks like most places will charge around R1000 anyway to fix this thing... so yeah, Im gonna buy a new drive. However, Im ALSO gonna repair this one _just_ because the option exists.

Fzabkar, I realise you arent an expert, however, from browsing other threads on this here forum, you seem to have helped many desperate people find their way to retrieving their data...
Ill truly appreciate any help I can get and with a little guidance, my clumsiness and Dextersnimble fingers, Im sure this drive can eventually see some life. Effort is NOT an issue since this is now a project for me. So now its down to scouring the local electronic supply stores for the correct part... I have a feeling this is gonna be a little tough... So far, 2 different suppliers have tried to get me to buy the same radial part that they SAY is the same as the one Im looking for.. but Im very wary of what they are trying to sell: Ive already had bad results with a radial choke that didnt work so Im not gonna get THAT one... However, Ive found THIS one...http://www.mantech.co.za/ProductInfo.aspx?Item=35M0275. Will it work or should I keep looking?

Thanks,
the local 1D 10 T

The datasheet for your component indicates that it has a DC resistance of 4 ohms and a current rating of 280mA. AFAICS, that makes it unsuitable.

http://www.mantech.co.za/datasheets/pro ... WXXXXX.pdf

The part that I have recommended was called out in an application note for a DC-DC converter used in SSD applications. Therefore I suspect that it would also be suitable for HDDs.

I would measure the DC resistance of the second 4R7 coil for comparison purposes.

I suggest you also measure the resistance between the coil and ground. This will tell you if there is a short circuit on the load side. If so, and if the regulator supplies the Vcore, then the MCU will be dead. Otherwise, if the short is on the Vio supply, then it could also be that the SDRAM or flash are dead.

_________________
A backup a day keeps DR away.

Instead of guessing and believing and trying and everything else, the quicker solution beside trashing the drive is to find a suitable PCB and ask the nearest pro (Nick maybe) to turn on the PC3000 UDMA (or equivalent KH/T) and make it work.
But summing up the cost of the PCB (assuming it works as "horror stories" about HDD PCBs are quite common nowadays....) and workmanship it would be comparable to the cost of new HDD.

Moreover, if the "damage" happened when the drive was ON, it would not be surprising to have some other collateral damage requiring complete refurb.

Good luck!

Hey Fzabkar,

Thanks for the input.
Ive measured the resistances and here they are:

At the damaged coil: on the 200 setting, one side gives 46.5 Ohms while the other side is out of range. A continuity check proved positive.

At the second 4R7 coil: both sides give 1.1 on the 200 setting.
at the 1R2 coil: 1.1 on both sides.
HOLD ON! I think my stupid el cheapo meter needs calibration, since when I touch the probes together, it gives a reading of 0.9... Would this then indicate continuity at the 1.1 ohm readings, which would mean those readings indicate ground?

I also noticed that the striped ends of the TVS' arent aligned... its as if one was mounted wrong way around. Is this correct? I shouldnt be questioning this since the drive was INITIALLY working...

Regards
Mudflap.
btw, I hope you guys dont mind... but as I think of more stupid questions, Im gonna ask cos Google doesnt have all the answers...
Now to get back to PCLinuxOS bugzapping for a while

Nice idea.... but this poses 2 challenges: 1. Where does one find a replacement PCB. Funny enough, I tried phonng Samsung who told me to get the data recovered myself and they will replace the drive. No spares. The drive is out of warranty.
2. The cost of REPAIRING the drive is WAY more than the cost of a new drive. Im picking up the new drive at the end of the month. Ordered from work
Regards,
MF

If you really want PCBs you just have to search. Or get a new identical / similar drive and you have a spare HDA just in case the new one gets damaged , be sure to have the PCB made to work with both HDAs.
Just don't expect to have it done for 5 EUR/$ .

Note : it is easier to find a damaged disk with good PCB.

True, true... however, Im from ZA, where the prices are double and the quality is halved (Just look at our broadband!!!)

However, Ive just called a guy who has some dead HDDs lying around. He's gonna check to see if he can find any Samsungs (considering the quality of Samsung, there should be a few!). If he has any, Ill probably poach them off him. Now, I have a question:

Supposing the NVRAM and the MCU are dead, is there any way to get the drive working again? AIUI, the NVRAM is where all the data pertaining to bad sectos and the like is stored. Now, considering the age of the drive (about a year), there probably are a few (yes... heavy use). Is there a way to just reset the drive, clearing a ram chip and making it like new? Obviously, the defective sectors will still be there jst waiting to be picked up gain, but Im not interested in that. Also, will doing this remove any PT info?

Reason Im asking, if the NVRAM is fried, then I HAVE to replace it. This means that either there HAS to be a way for me to reset the drive /RAM for it to work properly. Or is all this worth it?
Dont get me wrong, Im DEFINATELY gonna repair this drive and then give it to someone I dont like, but I have to get it to work first
Actually, to answer my own question: it IS worth it... Ill give this drive to my wife to store all those soppy movies on!!! Oh, the Victory will be SWEET! Ill FINALLY get rid of all those damn flicks that have spoiled so many a saturday afternoon!!!

Regrds,
MF.