Few days ago I found my PC with the desktop freezed and with the main hdd making some click sounds.

The hdd makes 2 clicks then it shuts down, and then speed again..2 cliks and so on. If I unplug the sata cable it stops after 5 cycles of 2 clicks. The PC doesn't start at all, and I don't see any bios as long as this hdd stays plugged in the system. So I believe it's a kind of short inside hdd pcb.

I have taken 2 pics. I found 3 Diodes, noted 1-3 in pictures.

click to see the full pcb

sponge marks


D3 shows 78 ohm in one way 1 in other way.

D2 75 ohm in one way and if i switch it on 2k multimeter scale, I get 266 ohm in the other way. so it seems that it's broken.

D1 76 ohm in one way, 1 in the other.

The diodes are marked with "54/59".

I have marked some areas where some oxides can be seen and some hot marks on that sponge. One it's close to D2 and it's seems to be the blue resistor (26K ohm).

Haven't tested the diodes on a greater scale since I don't know if it's a good thing.

Those 3 L's seems to be ok.

Any ideas on what to measure, and what to try?

If the hard drive clicks, the issue could be one of many things. I doubt the issue is a defect in the PCB components. It's can be a bad head, but it can also be lots of other things. (Like firmware issues, dirty connections, faulty read/write channel, bad preamp, or media damage)

Some time the BIOS shows "WDC ROM MODEL-BUCCAN".

see the image

It seems that You have problem with accessing modules in drive SA. I agree with Michael, but I doubt in any dirty connections.
So the structure of all modules in the SA have to be checked first using proffesional tool. There is no point to search defect on PCB.

When this happens the hard drive is coming ready in kernel mode, this happens when the hard drive cannot access the SA at boot time. The data you're seeing there is being pulled from the serial ROM as a backup. As I mentioned before, this could be one of many things. You would need expensive equipment to recover this data if it's not just dirty connectors. So try taking an eraser and cleaning the data contacts on the bottom of the PCB. If that doesn't work I'd recommend professional recovery. But based on the picture of the PCB in OP, it's probably not dirty contacts.

Tried to clean the contacts. No results.

Should I try to find another pcb and sloder the old U12?

However here's another macro of the pcb

This is hard case and has nothing to do with PCB.

If You care about Your data have it tested proffesionaly

If not You can try to access SA using MAVRIK utility (PCB with Marvel processor) from this site:
http://nazyura.hardw.net/000006.htm
But even to try You will need suitable donor drive and dig out what a "hot swap" SA access method is, because Your drive is coming ready in kernel mode.

But mind that each try can make the case worse and worse or even lead to unrecoverable situation
So my suggestion is: search for pro.

The only thing that would rule out would be damage to the PCB in the read/write channel, which is much rarer than the other reasons I mentioned. You can try it, but I wouldn't expect results.

You... haven't worked on many Western Digitals, have you?

_________________
You don't have to backup all of your files, just the ones you want to keep.

Back in business.

I went with it at the begging of the yer at a pro repair center. They called me to tell me about the 24h free diagnostic. I heard nothing new but the 30% chances to be electrical..and the rest to be hardware. After a week I went to take it back, when they stated that there is perhaps something hardware and the repair price tripled. At the end who can check them that it was what they say. They put small money for "electrical problems" to attract people..and then call back to tel..you know, there is something hardware.

So..I found another guy which seems to be affected by the same problem like me. He posted a video showing the clock of dead. From what I've read, the 2 clicks on spin up, it's a common problem on wd, meaning mostly a dead preamp.

I'm in the process of purchasing a donor from ebay. Unfortunately the model is rare and in 2 months I've seen only 2 out there. Finally one is on the road to my home.

Dead hdd
MDL: WD2000JD-60KLB0
DCM: DSCACAJCA

The donor will be the same MDL byt with the DCM DSBHCAJCA

The Latch the Bottom VCM and the Separator are different. Bu I'll need the Headstack and the PCB.

Meanwhile I started to learn about WD internal stuff. I'm coding a windows program (it will run on latest OS) to access the ROM and SA for the moment only for reading. I found some stuff around here from which i could understand how to send some VSC commands to wd hdd's. Don't want to tell you..how many times I rebooted during development , because that hdd became invisible to windows. Really I should get one cheap hdd especially for testings.

I had the hdd sitting on the desktop for the last 3 years. I'll give it a try in a way or other before it goes to the recycle been.

Tried the demo tool called wdr-4.0. It has an option to check the Heads on reading from SA. It gives "broken" on all 4 heads. So, the kernel doesn't even manage to read a single byte from the SA. That's why it shows "WDC ROM MODEL-BUCCAN". The Mod which contains the real model name it's on the SA. The same tool reads and dump fine the ROM which looks good It has the 30, 47, 0A, 0D mods at the end.

Tried also the MAVRIK suite on XP. It doesn't detect nothing. Perhaps it needs dos.

I like to see what's behind the software. To have the full control.

Excellent question.
Esp. on these drives, it is a real PITA.

All this, assuming you will do the head swap flawlessly, which itself is highly doubtful (no offense meant).

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

Well, it's the first time I hear about "head alignment" I've already spent 2 hours reading about this stuff. I'm a guru

I'll ask some local specialists how much will they charge for a head swap since I don't have a clean room and the tools for HSA and H alignment.

After a 6 years univ degree..I'm doing my job for about 35$/hour (when I have clients...which means that I can stay 8h not working). In my work I also use tools which cost may k $. What this "recovery" businesses is doing its...well..let's don't open the discussion. I won't pay more than I earn if they don't show what they've been studying more then me. It seems to be a food chain ..starting from the ones building those real 2$ tools...ending with the fool which pays so much for his stupid data. I bet, they won't pay so much on a brain surgery if that will save their life

More, I will ask to assist on the operation. They say...oooh, I worked 10h on your hdd when in real, he done all that in 1h. As they said..you know..it wasn't an electrical problem...the cost goes 4x higher. Jerks.

There are to few showing real info around here. The large majority teach you not to do it, hoping that they'll catch something. More,what's funny, they hope to catch some info form real pros, which share their knowledge for free.

Regarding my software development I've already implemented the ROM read feature. It runs on windows 7. Only that I don't know when to read the 128kb bios or 196kb version, and who said how many times I can read the rom log. I'll try to implement the write rom option, but I don't know if the command is handled by the MCU from a loader or it's inside the ROM bios code. If it's inside the ROM code, it's clear that If I flash bogus data, the chip is bricked and requires external flashing method. Which I don't own, so during application development I could brick my testing hdd.

Sorry for the delay, I'll have to wait another 2 days before post gets approved
le. oh, it seems that I can post without delay

Just avoid the whole data recovery thing and learn to backup your stuff. This way anybody can become the technician if stuff goes bad.

What gets me riled up is these noobs coming around here. They say they are interested in data recovery operations. No, you're not. If you were, you would have learned the craft a long time ago and thus been able to fix your own problem.

Typically it's the mom'n'pop type that just had a 1TB disk blow up on them.

_________________
On a clear disk you can seek forever.

Your not making any new friends around here with the attitude you have towards DR and the engineers involved within it. You clearly have no clue about what is really involved in DR and (just like all other novices in DR) you want everything for free with your demands and requests.

Quite simply you will NOT get hard earned info for free, you will NOT be able to assist in any DR operation and without professional assistance, like it or not, you will NOT get back your data.

Despite your opinions about professional data recovery and this perception you have of it, without using some professional service you are not going to get anything resolved. And for me, personally, you seem like just the type of customer I would prefer to avoid....

Yeah I realized. Is there a way to delete the offtopic (my opinion at this moment about PR) from previous post? If not, please ignore it, if you still can. Apologize. Here most of the help topics end with stop..you cant...if you want your data seek a pro etc. You don't know what people can if they put in mind that. .

Again sorry for what I've started.

Let's move to the software. If someone knows...the VSC commands are processed by the ROM software...or there's a boot loader inside the MCU?

Spildit 10q for your reply (with small exception ...u being rude ).

When I said "I'm a guru"...it was a joke you didn't catch it, perhaps blinded because of my stupid words in that post which I should kept for myself. ..right.

Next. Yeah, I know that de MCU's OS can lie on a separate rom chip (ex U12) or right inside MCU. What I didn't knew was what "kernel mode" is exactly. The MCU running the code flashed inside ROM (without the mods from the SA), or a special code builtin the MCU. I'm planning to code at the flash routine and I don't want to brick the PCB's ROM during tests, not having an option to reprogram it. So you say it's 100% true that if I flash some bogus data inside the ROM, then I'll have the possibility to re-flash it again with the same VSC commands, using the same method?

Yes I've heard also about running code uploaded directly from PCB's RAM. But this is not an option to the previous problem, if I brick the real ROM, because I don't know at this moment how to code that VSC's. I assume that with RAm loaded OS you can the re-flash the BIOS, if the VCM's are handled by the OS.

"why are you trying to build software". I code because I like coding. I've learned basic stuff during last few days about ata protocol and the controller handling that protocol. More, when I have the code, I can investigate deeper problems occurred, when a standard program outputs you a strange code you don't know what it means. Not telling that I'm allergic to DOS...and the too few app I found runs there.

"So what is the problem with your drive"...the first post explains that.

"trying to pretend to be a pro and superior"...again you are wrong. newer meant that. I'm here to find stuff that I don't know...not what i know. I know that my chances are close to 0 but not for that I'm here.

Can you tell me what the "zone table" or "RAM TABLE" is? I've seen a MHDD script reading that table with a VSC cmd (0x0D)...with a param of 0x06 for marvell chip or 0x0d for royle. I assume it is generated by the OS loaded by MCU, but is that table accessible without the OS accessing the SA?

Ok I found by myself.

It seems that Action Code 0xd offers access to a set of data found in MCU ram..called RAM tables..I assume that the info gets computed by interrogation. The "function codes" (in 512 bytes key) spreads from 0..to n. I found that if the drive can't access the SA...the only function which returns data is 0x01...all others are returning an error in rnsec&rsec which means..perm. ovl. not loaded aka..modules from SA not loaded!!

Well, I'll write some code using VSC's to read the negative tracks for each head. I'll test each head to see if can read the track -1 and if not, I'll see what error it throws. Now, I have the error codes because I've implemented a func for the Table Data reading function I make small steps for putting a diagnostic since I don't have expensive tools, I build what I need. I understood that the MCU doesn't need any SA data (adaptives or what they call it) to access the SA and this is logic. If the heads can't read the SA, then must be a PCB stuff, or the Preamp...or a motor problem (speed stuff etc).

I'm waiting for the donor...so I'm doing what I can till then, research and learning.

I wonder what they didn't made a smart connection around Preamp, to let people getting out only the preamp..not the full HSA



Well so you don't know for sure...if the VSC are handled by the ROM code or the MCU's ..let's say hidden rom...or ...boot loader

There are very good electronic reasons for where the preamp is, and the lack of other connections. Disk drives are designed to be easy & cheap to manufacture (amongst several other relevant factors), not specifically to be easy & cheap for other people to repair