Hi,

Please go easy I'm green to this. I have a friends HDD that I'm repairing for his Mum purely out of kindness. I have never got so involved in a physical repair of HDD. I normally stick at software based recovery purely as an interest.

A cheap molex connector on a sata usb adapter was reversed so the story goes at his attempt to pull the data. The drive will no longer spin. The data is valuable.

The drive is a Hitatchi 120gb 2.5 SATA HTS541612J9SA00 P/N 0A50686. I have located identical drive and it's on its way, fingers crossed here. It was manufactured in same month of same year (May 2007) and the codes on the pcb board (both) are identical. My plan is to swap the boards but need assistance here please.

I have read around Hitatchi drives and the NVRAM chip which I believe is located at U6 on this board. Given that the board is a great match can I just swap boards or is this NVRAM chip exchange critical for function, like a unique code is to a safe? I ask as I will need to farm this out to a guy who has competence with components/soldering. Could I damage/corrupt the disk if I swapped the board prior to exchanging chip to see if it would work? I can't slip up here.

Another question I have is given the nature of the over voltage causing the malfunction and putting the replacement pcb board to one side for a moment could it be worth bridging the fuse at F1 (I think) just to get data off. Is this likely all that has blown or is this a definite no no? There is no evidence on pcb of charring. A multimeter gave a reading to earth from one side of fuse only, no continuity across it. I'm clueless as whether fuse is on an earth circuit so advice is very much appreciated here..

Thanks

contact pcimage, or look up his DR Business and see if he is close enough to go to or post to.

You MUST transfer the NVRAM, or its contents, from patient to donor.

I would not swap boards without swapping the NVRAM, even for testing purposes. I believe there are certain conditions under which some Hitachi models may rewrite the NVRAM, rendering both patient and donor inoperable.

As for the fuse, I wouldn't bridge it. There is most certainly a fault elsewhere on the board. If you would like to do some simple testing, then upload a detailed photo of the PCB.

_________________
A backup a day keeps DR away.

Hi, Thanks for the replies guys. Really appreciated.

Fzabkar, here are a couple of images that I hope will be ok for diagnostics.

As green to all of this how feasible would it be possible to transfer data as per your last post to NVRAM chip.

'You MUST transfer the NVRAM, or its contents from patient to donor'.

What equip would be needed for this purpose?

As a last resort, this is an idea (only) on the backburner. Would a platter swap be ok from patient to donor in clean area and if so would patients NVRAM have to be swapped to the working pcb also.


This situation is causing angst, sometimes it would be easier just to say no or not offer to help but I'm in deep now and resolution it is hoped is only option and that is thanks to the kindness of you guys too.

Your help is really appreciated.

Kind regards

Soldering equipment, knowledge and skills. But if you mess up the NVRAM, you take a simple low cost recovery and make it a very complex and expensive recovery.

NVRAM is on the PCB and is tied to the data on the platters. That said, if you continue down this path, you are going to destroy any chance of recovery, even for the pros.

If you want your data back, contact PCImage (pcimage.co.uk) before it is too late. The more you try, the lower the odds and the higher your price for professional data recovery services.

_________________
Luke
Recovery Force Donor Inventory: 5,000+ Fully Cataloged HDDs & Growing

Thanks Luke, appreciated.

Have read a fair amount how critical NVRAM is and will not take any chances. I'm in contact with a a competent solderer who would be willing to switch NVRAM over to a working board.

My friend has no funds so I'm working pro bono on this myself. I'm gathering all information at the moment no detrimental action will be taken till sure of way to proceed. I do respect your opinion thanks.

I look forward to hearing from you too Fzabkar.

lcoughey, Hi luke, assuming NVRAM chip is successfully switched if this route is taken. Would you anticipate additional work to the drive or should it spin up and read normally. I use a few pieces of third party software ie Partition Wizard and other Data recovery consoles. Thanks

I couldn't say, with certainty, as I would need a full physical diagnosis of the drive. If the PCB is the point of failure, did it die so quickly that the heads were unable to park and are now stuck to the platters? Did the heads just touch down and scratch the platter or damage themselves? Did the head preamp get damaged too?

_________________
Luke
Recovery Force Donor Inventory: 5,000+ Fully Cataloged HDDs & Growing

Thanks Luke.

IMO the safest DIY approach would to transfer U6 to a compatible donor PCB. Certain PCB suppliers offer a free chip transfer service.

http://www.onepcbsolution.com/
http://www.hdd-parts.com/

Alternatively, your local TV/AV repair shop should be able to transfer this chip for you.

There are some measurements that you could take to assess the extent of the damage. The first would be to measure the resistance of the fuse and a component which looks like it might be a 5V TVS diode. If the diode measures close to 0 ohms, then remove it and retest both the diode and the circuit. If the circuit still measures 0 ohms, then there will be problems elsewhere.

The protection devices on Hitachi HDDs don't appear to do a very good job, so even after removing a shorted TVS diode, there may be additional faults. The next thing I would do would be to determine whether there is any continuity between the +5V pins at the SATA power connector and the V+ / V- pins at the preamp connector. If so, and if the fuse F1 is open circuit, then this would mean that the supply to the preamp is not fused, which would in turn mean that the preamp may be damaged. What resistances do you measure between +5V and V+ and V-?

The next thing I would do would be to measure the resistances between ground and each of the Vcore and Vio test points.

_________________
A backup a day keeps DR away.

I think the Vio regulator is probably wired as follows:



uPA507TE, NEC, P-Channel MOSFET with Schottky Barrier Diode, marking ZA, 20V. 2A:
http://documentation.renesas.com/doc/Do ... V0DS00.pdf
http://pdf.datasheetcatalog.com/datashe ... oac03y.pdf

Edit: The Vio supply is wired to the SDRAM's supply pin, so it would be +3.3V.

K4S641632K, Samsung, 64Mbit, 1M x 16Bit x 4Banks SDRAM, 3.3V:
http://www.datasheet.directory/samsung/ ... K-UC60.pdf

_________________
A backup a day keeps DR away.

Hi Fzabkar,

Thanks for informative reply.

Have located a DMM and have a few measurements for you but need a little clarification on where to take readings from Vcore and Vio components.

I'm new to this and its been a steep learning curve to say the least this past week but I'm in it till the end now.

At this stage I'm trying to preserve donor HD until I know without shadow of a doubt patient PCB is shot and irreparable. Of course I will take your advice and swap over NVRAM chip at U6 if original pcb is not salvageable to donor I have.

Here are some readings for you. (Meter set at 200K)

F1=0.01 ohms
5TVS Diode = 0.0 ohms
Resistances between +5V and V+ and V- = 3 read 0.1 ohms and 1 read OL Please see attachment.

There was no continuity from sata 5v to any of these points so am hoping pre amp is unaffected.

A long shot but could it be just F1 is faulty?

Look forward to hearing from you soon.

Kind regards

Hi Fzabka,

Thanks for reply. As always really appreciated.

No luck with tests/actions outlined before in your message.

Very frustrated. Nvram was switched from patient to correct donor board this eve and drive remained dead, no spinning, nothing.

When PCB placed back on donor drive again nothing.

When donor boards Nvram chip was reinstated to its original board both patient and donor drives fired up. Unsurprisingly patient unrecognised as a drive, donor, perfectly fine again.

I'm thinking NVram chip must have been damaged at initial failure or during transfer...It was connected correctly ie dot to dot.

Where can I go from here? Have put so much time in I feel.we both deserve success.

The safe has the money in, its spinning but have lost the key.

How can we get into this drive?

Have read about Hitachi HD Doctor, and it's ability to rewrite nvrams data from drive but this seems to be specialist equip.

Any thoughts or should I be accepting defeat? Can the chips info be transferred by some sort of reader?

Really not giving up just yet.

Good advices here from fzabkar and fzabkar tehy are very good people in electronics, and HDD DR.

As they said here you need NVRAM exchange, needed soldering station and skill is not work for a novice.

after that maybe can apears other problems as said lcoughey.

If data is important and you are not familiarized with electronics and mecanical failures is better to contact wit a local pro as said here HaQue

good luck!

Happy to take a look, but it won't be free

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

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

Thank you Sean at....

www.pcimage.co.uk

You're a gent. It's winging its way to you as I type. Fingers crossed.

Hi Sean,

www.pcimage.co.uk (very knowledgable and approachable) Highly recommended.

It's winging its way to you now. Thanks for your assistance, fingers crossed.

Thanks to hhdrec et al too. You are right. Once DIY resources and capabilities are used up then it must be left to the pros. Guy's knowing when to stop is critical, a DIY job on something like a drive will occasionally yield success but often it can make matters worse. I pray this is not the case here.

Will let you know how things go...

Regards

Thanks to Sean at PCimage who kindly had a look at drive. He confirmed my worst fears that NVRAM chip was roasted to oblivion by 'overzealous use of heat gun by TV repair man' so data on chip was obliterated.

Position I'm at now is that Sean kindly soldered new NVRAM chip in place but let me know that it will have to have some sort of binary reprogramming which is very involved and specialist.

The drive spins up fine now but is not accessible. No drive letter is visible.

Would anybody have any suggestions please on how I could perhaps learn (or at least attempt) to program such a chip as I still haven't entirely given up on this drive just yet...links, manuals etc.

Sadly, I have since learnt from original posting over a year ago that some data on drive may now be very precious to friend and his family due to tragic circumstances which is why unbeknown to him I have returned to it in a last attempt at recovery.

I have read somewhere about Hitatchi Salvation Doctor being capable of writing to this chip from a file stored on the drive, I don't know how true this is so I'm throwing it out there again to all you geniuses.

Sean, Thank you. PCimage comes highly recommended.

Unfortunatly, that code contains an unknown SA starting point which varies from individual drive to drive. The only option now is to attempt brute force NVRAM rebuilding using every possible SA starting point. Don't waste your money on dreams of Salvation Data tools working, in my experience those tools never are able to do anything useful. I once tried their NVRAM repair, it beeped away for about a month and never accomplished anything but frustration.

You'll need a PC-3000 to pull it off, and even then I don't think it always works. The correct course of action would have been taking it to a professional at the beginning. We do dozens of NVRAM transfers here and have never burned one yet. It's all just a matter of properly controlling the temperature and knowing how to control heat transference.

_________________
Data Medics - Hard Drive, SSD, and RAID Data Recovery Service Company

Common guys, it's old hitachi. I don't see any problems to take copy of nv-ram content from system area.
Oakwood63, I can recover your data (if it's problem in lost nv-ram content) for $100. If it's okay for you - PM me.