Hi gurus!

I have a problem with a dead WD-HDD (specifications see below). When i connect it to the PC it starts to spin for a few times but the PC does not recognise it.

I bougth a donor-hdd at ebay for the repair (specifictions see below).
I allready swapped the PCB but without luck (still the same problem!)
What i found was that both PCBs must be ok because they work both without problems together with the donor-hdd.

I suppose a damaged head!?!

What do you think? Are the HSAs compatible? Can i swap the HSA?
What is the best way to recover my lost holiday photos ??

Specifications of dead HDD:
S/N: WCAS81668767
MDL: WD5000AAJS-00YFA0
DATE: 08 OCT 2007
DCM: DARCHT2AHB
WWN: 50014EE255F0E4D7
PWB: 2060-701477-002 REV A
PCB Sticker: 2061-701447-100 AC XC 4U09 2ER8 E 0004140 8144
(PCB without U12-Rom)

Specifications of donor-hdd:
S/N: WCAS81393827
MDL: WD5000AAJS-00YFA0 (=)
DATE: 04 OCT 2007 (-4days)
DCM: DHRNHT2MBB (=)
WWN: 50014EE255ED98D5 (=)
PWB: 2060-701477-002 REV A (=)
PCB Sticker: 2061-701447-100 04P XF 4U05 0106 B 0001130 8131 (=)
(PCB without U12-Rom)

Thanks for any help and any advice!
Horchi

Does sound like heads issues.

HSA appear compatible, but not swappable by you, with respect.

These are difficult drives even for most "pros", so DIY will be virtually impossible I'm afraid.

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

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

Any idea what a professional solution would cost??

For my potential DIY-solution :
What is the most complicated part while swapping the heads?? How many discs does this drive has?

Your board has a Marvell MCU (88i6745) which is often affected by a fault that mimics head or media damage.

The easiest thing to do is to try a PCB swap. However, you will need to transfer the contents of the MCU's internal flash memory to your replacement PCB. Some board suppliers include such a service for US$10, but there are also cheap commercial software solutions for around US$80 - $150.

See call-forum-administrator-t21565.html

... and wd5000aaks-00tma0-damaged-t19077.html

_________________
A backup a day keeps DR away.

@ Fzabkar:

The fellow said both PCBs work on the donor.

_________________
http://www.datasaversllc.com

So you must know that the data on that disk is below value of a professional recovery assignment for over 500-1000$ (??) - but i dont want to give up and throw my hdd away without trying to fix this problem by myself.

I' m engineer with a good technical aptitude an the possibility to use a clean-room at work. So can somebody give me some advices what i have to follow while changing the HSA?? I want to do it like this guy: http://www.youtube.com/watch?v=Ul97jWv2PKY and I know the risk that further parts of the hdd could destroyed in my experiment...

I want to be sure, that the HSAs are compatible in this special case (although DCMs are a little different:
DARCHT2AHB and DHRNHT2MBB)??

Wich tools can you recommend for this task (merge spreader for heads...)??

I'm deeply grateful for every helpful tip!!

Happy tinkering then.
I think your success chances are over after removing the last screw of the lid, but who dares wins - you never know until you do it . The latest surprise can come out when you open the donor even when it "should have matched".

In my opinion, DIY options are limited to PCB fixes and recovering drives with bad sectors or logical problems.

If you don't care about the data, do whatever you like, because the chances of you succeeding are very poor. There is a greater chance that you will make a recovery harder or impossible, if you should change your mind later. It will likely cost a lot more too.

Mechanical & firmware solutions require (usually expensive) equipment, knowledge, and LOTS of practice. They usually also require compatible donors for parts.

With all of the drives and the different models, there are a lot of "rules" but also a lot of exceptions (I've never seen a global list of the things to match that has been 100% correct). So you have to know what you are doing just to find a compatible heads donor.

With Western Digital, diagnostic gear is used to match heads, because you can't get all of the information from any label on the drive.

Fortunately, for your model drive, the head maps are the same, and heads from a unit with the same model number & etc. made close to the date that yours was manufactured will often work. But . . .

Chances are that the heads will be altered slightly during the removal and reinstallation, or they won't be easily compatible, or there might be media damage, etc. These drives perform a power on self test, and they will spin down if there is a problem reading or writing.

After you swap heads, there is a 98% chance your drive will click and spin down. What then, is the next step?

I know there are people who accuse data recovery pros of being greedy, stingy, etc. However, those people do not have to deal with hard drives that someone else butchered while trying to recover their own data. We see this problem all of the time, and often it is unnecessarily fatal to the data.

A bit of technical knowledge -- without context or the "larger picture" -- is a recipe for disaster when it comes to mechanical repair and data recovery.

You've been advised, so proceed at your own risk.

Good luck.

_________________
http://www.datasaversllc.com

Totally agree.



That is very interesting. I thought they were offering cheap services with pcbs with external rom. If someone offers transfer of internal memory for free or $10, i would really like to meet them. I could even send them a dozen cases we have here that needs this kind of work. What the heck! It's for free!

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

I have a similar problem with a WD5000AAJS. The unit has about 300 GByte of required data with no backup so I need to find a fix. This is being used on a linux installation (just to make things easier). The unit started doing the "click.. click... click" thing on power up, so I looked here to look for some guidance. proceeded as follows:

1. Got a new PCB; PCB was new, in a WD sealed bag; different build spec (slightly) - no platter properties in the flash (dont know how to get the parameters in anyway) but gave it a go anyway. Disk wouldnt spin up...

2. Some advice from here (and from the vendor of teh PCB) was that if the disk is spinning up then going click-click this points to a broken head assembly (ie probably the head preamp).

3. Got a new (to me) WD5000AAJS. Tested on delivery, runs good, dumped 100 GByte data onto it and verified this - all ok. Stripped out the head assembly (and dont start on the "kiss of death when you take the last screw out from the lid": Ive had disks open and spinning and reading and writing happily - as long as you dont spill your cornflakes on them...) and put it into the patient disk. This disk now spins up ok. The click-click-click has changed in nature is now identifiable as (a) move off the park ramp (b) seek to somehwere mid-range in the cylender range (c) seek back to a cylender not far from (b) (d) seek to the middle (e) reset back to the ramp (f) repeat from (a).

4. (As an aside) I put the patient heads into the donor disk and this unit wont spin up, with any of the 3 PCBs I now have. So it looks pretty conlusive that the patient head assembly is knacked.

5. I dont know what tools to use (under Linux if poss) to look at what the unit is doing, nor to try to coax it to look for a servo track and lock to it.

QUESTIONS TO THE COMMUNITY:

I would be very grateful for suggestions as follows (apart from "in the bin" of course):
(A) what is the triple seek sequence, and does this sound familiar?
(B) any idea for tools to analyse what the controller is doing; and
(C) anything else to get this unit up and running again?

Thanks everyone

Nick:

You haven't performed a proper diagnosis. There is a small possibility it could have been a PCB issue, which should be ruled out first by a DIYer. The way to do this has been documented here many times, but you didn't follow it.

You've not only ignored, but discounted advice about the importance of a clean environment.

You don't seem receptive to sound advice and proper procedure . . . Not to be ugly, but what's the point?

Jon

_________________
http://www.datasaversllc.com

I wasn't aware that new PCBs were available for sale, even from WD. Where did you buy it?


So the HDD was spinning up before the PCB swap but not afterwards? Doesn't this suggest that the new PCB may have damaged the preamp? If you place a business card between the HDA and the preamp contacts on the PCB, does the drive spin up then?


... with the new PCB but not the original PCB?


Try the business card trick.

If you could upload a photo of your PCB, perhaps we could test the onboard voltages to determine whether the preamp's negative supply is present, and we could also test whether the original preamp is shorted.

_________________
A backup a day keeps DR away.

I looked here for a procedure to check the PCB, which referred to dumping the trajectory data into the flash on the new board by swapping the flash device itself. Except that in this case the flash is part of the controller chip so the only way to get to it is with a WD software tool to read and write the flash - which I dont have (and, with respect, is somthing that only a disk repair service may have access to as it requires a level of training/learning to use).

I omitted to mention that having tried the new board in the patient, I put the original PCB back into the patient and it resumed its previous symptoms.

This was posted by one of our members on another forum:

"A quick note on WD drives belonging to Tornado family.
Issue with them - PCB running hot and translator generation can mess up.
Drives falling under that category : Tornado, Tornado 2D, Tornado 3D. Most common example in North America : WD5000AAKS-**TMA0, WD5000AAKS-**YGA0
Possible symptoms: Most of the time - partial sector access. Reads fine up to a certain point , then all errors. Can be the front or the back of the drive.
Less common - drive clicks and does not ID at all.
Solution - PCB swap + ROM adaptation. Keep in mind , most of these drives do not have external ROM , however there is an image of it in SA."

There are PCB suppliers who will program a replacement PCB for you, as noted often on HDDGURU. You still don't know if the PCB is OK or not, but now probably the heads and media are not . . .

_________________
http://www.datasaversllc.com

eBay, disk-fixing house, but it was still sealed in the WD bag.



I was converned that the new board was a waste of time - it was a newer fab spec with a couple of extra components. Who knows what these designers do when they iup-issue boards... But I forgot to say that I then put the original PCB back in and it resumed its symptoms. It is obviously difficult with trajectory data (and presumably other data too) written to the platter, that replacing the PCB is rather a "bootstrap" type activity just sufficient to get the data off the patient, is pretty imperfect and most of all unrelaible. For a DIYer I also expect a level of frustration where reads from outer cylenders may work ok but the errors build up as the effects of the poor-match trajectory data starts to bite...



Sorry I didnt do this, but I did try "all boards in both chassis" and the replacement board did seem to work in the patient chassis with the donor heads. The difficulty was that I tried all three PCBs in this arrangement and found that all three cases the heads did differnt things but the spindle motor did spin up in each case. I also tried the patient head assy in the donor chassis and this was toast with all three PCBs. I inferred that the patient head assembly was bust and overloading the controller which never left reset, or never proceeded in execution.



As above, a combination that involved any of the PCBs in the patient chassis with donor head assembly, span the spindle motor but the heads seeked in different ways.



That would be great - except that my camera is somewhere else that i wont be to until next week - so will do that then...

Thanks in the meantime for your help.

Thank you for this.

I had assumed that the new, bag-sealed PCB was good and appears to be proved good on the donor chassis with donor heads. The PCB off the donor disk is known good.

I evaluated the means to get the trajectory data written into the flash on the bagged PCB by a disk-fixer house but it was then pointed out to me that the heads clicking indicated a bust head assembly and not the PCB at all.

I then got a new donor disk and swapped out the heads.... the condition of the platters is obviously now in question but I see no indication (by action of the heads or inspection with a glass) of physical damage.

The donor head assembly is satisfactory.

1. If the PCB is bad, the drive may click (and spin down);

2. If the firmware doesn't match, the drive will click (and spin down)

3. You STILL haven't provided a KNOWN, GOOD PCB with MATCHING FIRMWARE, so you are no closer to knowing WHAT FAILED.

Good luck.

_________________
http://www.datasaversllc.com

honestly without the "WD software" you mentioned to read and write the ROM data in MCU, your efforts are useless, all of them.


Get a way to read and write the internal flash, once you can do that then you can at least TRY to fix this.




I am not appose to DIY (if you were NEVER going to send it to a DR company anyways), but if your going to DIY, then you should really spend a LOT of time reading before you jump to conclusions and perform rash procedures without first following the steps that need to be done beforehand.

@Russ - thanks, except that the starting point is to consider whether the hardware was knacked which now I know which was, which isn't and which is suspect. I fully agree that without software to get into the controller its all a bit useless - and I spent along time reading here and elsewhere looking for clues. But nowhere do I find how to get the WD app for access to the engineering ingress to the controller... including getting in touch with Marvell - which didnt help much. So any ideas?

(and strictly speaking Im not a DIY'er as Ive been a Chartered Engineer developing embedded control hardware and software for over 35 years and my post about working under class 100 conditions in a clean room at a Times 100 Electronics company (and what to do and not do in relation to openeing disks) got misteriously moderated away..)

Very greatful for your help. Thanks for this - havent tried yet (and further issue with using this under Wine / Linux - I'll report back when Ive had a go.

Also looking at http://malthus.zapto.org/viewtopic.php?f=3&t=224 (about getting at the SMART tables) the onward link to
http://www.nazyura.newmail.ru/000006.htm is now bust. Otherwise much useful stuff on malthus.zapto.org - thanks for your support !!