Switch to full style
Data recovery and disk repair questions and discussions related to old-fashioned SATA, SAS, SCSI, IDE, MFM hard drives - any type of storage device that has moving parts
Post a reply

PCB swap on a WD10EAVS

January 9th, 2018, 4:21

Hi,
I replied to an anoucement on a classified ads website from someone looking for a Western Digital WD10EAVS, so as to recover data from his non-functional unit by attempting a PCB swap. I happen to have a unit of that model, so I could sell it to him (he's in another town, no way to make a quick try beforehand). He is sure that the issue comes from a defective PCB : the HDD doesn't start at all with its native PCB, he tried to swap it with that of a 500GB model, and it did start then, although of course it didn't work properly.
But I've read repeatedly that on most modern HDDs it's not possible to swap the PCB and have it work right away : even if it's a compatible model, there's a chip containing specific calibration data and whatnot (what else exactly ?) only relevant to one particular unit that needs to be unsoldered from the defective PCB and resoldered to the replacement PCB. Is this the case with that particular model ? If so, what exact chip is it ? Is it relatively easy to do with a common soldering iron, or very tricky and requiring specialized tools / experience ? (He seems to be sure that it'll be easy as a pie, but any reasonable grandma would tell ya that a pie ain't that easy ! :))
Then, he says that his HDD's exact model is “WD10EAVS-32D7B1” and the PCB has the code “2061-701590-A00AE. Mine is a “WD10EAVS-00D7B1” and the number on the board is “2060-701590-000 REV A” ; there's also a small label with this code : “2061 - 701590 - A00 AD XW 7C20 4FK2 6 0001380 9381”. Would it be compatible ?
That HDD (mine) is 100% functional, and currently full, so as much as I'd be willing to help the guy, I don't want to go through that hassle if there's zero chance that it will work.
Thanks !


“Grandparents in Hell !... Picture that – picture your grandmother in Hell : baking pies... without an oven...”
George Carlin

Re: PCB swap on a WD10EAVS

January 9th, 2018, 4:32

Some MCUs have internal flash memory. Others have an external serial flash chip at U12. Is your U12 location populated?

Analysis of the "ROM" on a Western Digital ROYL HDD:
http://www.hddoracle.com/viewtopic.php?f=59&t=225

Re: PCB swap on a WD10EAVS

January 9th, 2018, 4:57

Well, what is a MCU ? And what do you mean by U12, is this a code on the PCB ? Let's have a look... I see many E something, a few P something and TV someting, but no U something, if that's what you mean (I haven't unscrewed it to look at the other side). And I couldn't give you more info for that defective HDD which is not mine. Are there that many differences on HDDs sharing the same model number ?

Also, could it be fixable by removing a damaged diode, as I saw in some threads here, including this recent one ?

Re: PCB swap on a WD10EAVS

January 9th, 2018, 5:36

Here is a pic of 701590-001. You need to unscrew your pcb and check the other side.
Mcu is the processor...big chip with an M.
701590.jpg

Re: PCB swap on a WD10EAVS

January 9th, 2018, 6:29

Alright. But what would it entail if there is or is not a ROM in U12 on my PCB ?
And can the owner of the defective HDD attempt something with those diods ?
Thanks for the quick reply !

Re: PCB swap on a WD10EAVS

January 9th, 2018, 6:40

Could You post a picture of the electronic board?
If You do not have external Rom than the rom is located in MCU.
It is possible to recover this rom with PC3000 or rebuild it from SA (using pc3000). Also You can resolder MCU if You have skils :-) Or repair the oryginal electronic board.

Re: PCB swap on a WD10EAVS

January 9th, 2018, 13:24

Your drive belongs to "Kermit" family. All drives of this family have an external ROM chip ( U12). You have two options:
1. To swap the ROM chips by desoldering / resoldering
2. To transfer the ROM content from one PCB to the other by using a program like WDMarvel, WDR, etc.

Re: PCB swap on a WD10EAVS

January 9th, 2018, 13:54

HDD from Inside: Hard Drive Main Parts:
http://hddscan.com/doc/HDD_from_inside.html

Re: PCB swap on a WD10EAVS

January 9th, 2018, 14:16

This one has no ROM at U12:

https://www.harddrivesforsale.com/it/wd10eavs-00d7b1-pcb-2061-701590.html
https://www.harddrivesforsale.com/media/catalog/product/cache/3/image/560x800/b61f6e664c77c06cd32013c973526a8f/2/3/23842a.jpg

Re: PCB swap on a WD10EAVS

January 9th, 2018, 15:40

abolibibelot wrote:Alright. But what would it entail if there is or is not a ROM in U12 on my PCB ?
And can the owner of the defective HDD attempt something with those diods ?
Thanks for the quick reply !

If the drive has ROM chip swap it to the good board...if it doesn't have, swap both diods with the ones from the good board....you can measure them to find the faulty one if know how...or swap both.

Re: PCB swap on a WD10EAVS

January 9th, 2018, 22:41

@COD
Could You post a picture of the electronic board?
If You do not have external Rom than the rom is located in MCU.
It is possible to recover this rom with PC3000 or rebuild it from SA (using pc3000). Also You can resolder MCU if You have skils :-) Or repair the oryginal electronic board.

I can send a picture of my electronic board, but what would be relevant would be a picture of the defective one, right ?
I could ask the owner of the defective HDD to send a picture, and post it, but that would become too time-consuming to help solve an issue for a total stranger with no reward for it.
I don't have access PC3000, and surely neither does the owner of the defective unit.
Repair the original board : how exactly ? Replacing a diod / fuse ? How to methodically identify the defective component(s) ?


@BGMan & fzabkar
Your drive belongs to "Kermit" family. All drives of this family have an external ROM chip ( U12). You have two options:
1. To swap the ROM chips by desoldering / resoldering
2. To transfer the ROM content from one PCB to the other by using a program like WDMarvel, WDR, etc.

This one has no ROM at U12
https://www.harddrivesforsale.com/it/wd ... 01590.html

Well, those are contradictory informations...
But again, the defective one is WD10EAVS-32D7B1, not WD10EAVS-00D7B1 which is my (functional) HDD and could be a potential donor.
Does it change the answer to that question ?

EDIT : Answering my own question, WD10EAVS-32D7B1 PCB has no ROM chip in U12 either :
https://www.harddrivesforsale.com/it/wd ... 00-ad.html


@Spildit
The problem is NOT if you have or not a ROM chip on your PCB. The problem is that if the damaged drive have or have not a ROM chip.

Alright, I got that part by myself, hence my questioning! :)

If the damaged drive doesn't have a ROM chip (external) most likely it will have a copy of the ROM modules on SA. You can place your PCB there and you will gain access to SA with tools like WDMarvel, SeDiv, PC-3000, HRT, whatever and you will read SA and from those modules you re-build ROM (original). Then you write that to your PCB and you are good to go and you will have user area access, assuming the problem is just with the PCB.

What do you mean by “place your PCB there” ? Maybe you mean place the PCB from my (functional) WD10EAVS onto the defective one ?
Again, the owner very probably won't have access to the kind of pro tools you mentioned. He expects this to be a straightforward, plug-and-play procedure. I warned him that it might not be that simple.
How can anything be written to the PCB, if there is no ROM chip ?
And if there's a copy of the ROM modules on the SA, but no ROM chip, then where is the other copy ?

If you have a short diode (TVS) then if you plug the drive to a PSU the PSU will shut down, meaning that on internal PC drives with shorted TVS the system will not even power on. If your system is running fine with the drive plugged to it (to the internal PSU) then most likely the problem is not related to shorted TVS.

Alright, that's definitely a useful piece of information.
So if it is not a short diode, it's necessarily a system area issue ?

Re: PCB swap on a WD10EAVS

January 10th, 2018, 1:17

abolibibelot wrote:But again, the defective one is WD10EAVS-32D7B1, not WD10EAVS-00D7B1 which is my (functional) HDD and could be a potential donor.
Does it change the answer to that question ?

No. The first two characters of the model number suffix represent a customer ID.

According to the following document, "the model number suffix (characters to the right of the dash following the model number) is only for in-house use. Refer to Change Administration’s Model Specification (document number 2096-000101)."

Model Number Format for OEM and Distribution Channels:
http://www.wdc.com/wdproducts/library/F ... 001028.pdf

I haven't seen WD's internal document, but here is an explanation in Russian:
http://www.acelab.ru/dep.pc/doc.pc3000d ... in-N-2.pdf

Here is Google's English translation:
http://translate.google.com/translate?h ... in-N-2.pdf

TVS Diode FAQ :
http://www.hddoracle.com/viewtopic.php?f=100&t=86

Catastrophic failures in Western Digital PCBs:
http://www.hddoracle.com/viewtopic.php?f=100&t=2169

WD's TVS diodes (photo):
http://www.hddoracle.com/download/file. ... &mode=view

newbie info, from and for newbies :) About firmware ...
viewtopic.php?t=6562

Re: PCB swap on a WD10EAVS

January 10th, 2018, 19:48

Maybe we are all answering as if @abolibibelot had both disks with him.

@abolibibelot, if you are just selling the disk to a total stranger, as you commented :

- you can not sell it, if it is ok, and sell it to someone that needs a good disk.

- you can sell it, tell the person that you believe it won´t work, and don´t worry anymore. The problem is of the person that insists in buying the disk.

- you can tell the person to send you his hdd, so that you can try to fix it for him ( for a price, obviously )

Re: PCB swap on a WD10EAVS

January 11th, 2018, 5:55

@rogfanther
Maybe we are all answering as if @abolibibelot had both disks with him.
@abolibibelot, if you are just selling the disk to a total stranger, as you commented :
- you can not sell it, if it is ok, and sell it to someone that needs a good disk.
- you can sell it, tell the person that you believe it won´t work, and don´t worry anymore. The problem is of the person that insists in buying the disk.
- you can tell the person to send you his hdd, so that you can try to fix it for him ( for a price, obviously )

Thanks for summing up what my options are ! :)
I created this thread in part to gather some practical knowledge for myself, having no experience yet with a similar case. (I've never attempted a PCB swap.) Indeed I could have chosen option B, sell it and move on, regardless of the possible outcome, but that would be a waste of a perfectly good HDD if it turns out that it couldn't work for specific reasons (well, actually no : that person said that if the swap doesn't work the one I have could still serve as a backup). That person seems decided to deal with the issue him/herself, so option C is probably not an option... I had not yet decided to sell that one in particular, but I have way too many HDDs, I'm expecting to receive two 8TB units in a few days (WD80EFZX), so I'll surely sell quite a few of the others, might as well sell that one to someone who specifically needs this model, if it is indeed a realistic endeavour to attempt that kind of DIY fix.


@Spildit
If there is no ROM chip there will be ROM flash space embeded on the MCU and you can write the code there. The Vendor Specific Commands to access that space are the same ones used to access the external ROM chip. The drive knows if it have to read from ROM chip (external) or embeded ROM on MCU by resistor setting on PCB.
The copy (backup) will be on SA and the working ROM is embeded on the MCU (inside the MCU).

From what I could gather in the replies I got above, and the links provided, both units do not have an external ROM chip in U12, so, according to what you say here, they must have ROM code embedded in MCU. In short, does that mean that the PCB swap can not work without issuing vendor specific commands ? Would it be wiser for that person to first try to test those diods and remove the potentially shorted ones ? (I provided a link to this thread to the owner of the defective unit, so even if I'm begining to understand some intricacies of such cases, I'm trying to elicit straightforward answers that a person operating at an even lower (!) knowledge level can comprehend... :) )

If the drive doesn't spin at all with the defective PCB there might be other issues like for example a shorted pre-amp inside the drive. A shorted pre-amp will make the drive not to spin. Just because it doesn't spin it's not necessarly a PCB issue. Also if ROM code is damaged drive will not spin. If drive have SA/Firmware issues then it have to SPIN. Because SA is written on platter .... So if it doesn't spin at all it will not read SA at all .... For example imagine a bad S.M.A.R.T. module on SA causing problems. If the drive can't spin at all it will not load that model to start with. Only if the drive is spinning the module can be read and if it's damaged will cause a problem ... The problem is way prior to any SA code on the platter to be loaded....

The owner of the defective drive said that swapping the PCB with that of a 500GB unit was enough to have it start up again (albeit not enough to access the data, obviously). So that would rule out a pre-amp failure, right ? Otherwise, I don't know the likely (external) cause of failure.


If the original PCB is gone / damaged you most likely can't read the ROM on it with firmware tools. If PCB have external ROM chip then there will be no copy of ROM modules on SA. It's not "viable" to read the ROM with exception of some cases that you might get luck to read ROM by TTL ... But if you have external chip + external programmer (cheap) then problem solved ...

Again, hasn't it been established above that both WD10EAVS do not have an external ROM chip in U12 ?
Regardless, what do you mean here by external programmer, and what has to be programmed exactly ? Even if it's cheap it seems way too technical for the average Joe or Jane ! :) Myself, I would dearly try something like that, but again, this person expects this to be merely a plug-and-play procedure, and apparently I was right in suspecting that it would actually be more complicated than that... I'm trying to evaluate the level of that complexity, for something that I most likely won't do myself (on this occasion anyway).

Thanks again.

Re: PCB swap on a WD10EAVS

January 12th, 2018, 11:04

The drives have the rom content either in an external ROM chip, or inside the MCU.

When there is an external ROM chip, it can be swapped from the bad board to the good one.

If the ROM is inside the main MCU, then it has to be copied ( WDR, WdMarvel, HDDSupertool, PC3000, etc ) from the bad board to the good one. Yes, this is done by the software using VSCs. Other ways may exist, but they are more complex and cost more.

Not a thing so easy to your possible customer to do on his own, because if he destroys the chip or other important component in the board, then recovery will be much more difficult and costly.

About the external programmer there is some information here :
http://www.hddoracle.com/viewtopic.php?f=110&t=1738

Re: PCB swap on a WD10EAVS

March 8th, 2018, 13:19

I got an update from the owner of the damaged WD10EAVS : he simply wrote that the PCB swap didn't work – well, as I expected.
I'm asking him questions based on earlier input I got here (did he just swap the boards or attempt something with the TVS diods or the MCU ? what kind of noises did he hear ?...). Depending on the replies I may try to assist him remotely.

The damaged PCB and your PCB might or might not have the external chip. You have to check that out on the PCB. even if they have the same model and PCB version there are the chance that one have external ROM and the other have internal one. The only way to be sure is to check.

You mean, even if the unit I sold didn't have an external ROM chip, if his damaged one does have such a chip, it could be soldered to the replacement board and the assembly would work ? In that case, wouldn't there be a conflict between the ROM information in the transposed chip, and the native ROM inside the MCU (since the replacement board definitely does not have an external chip in U12) ?

Only if you are extremely lucky will the PCBs work without swapping the ROM chip or write the native ROM code on the new PCB as there will be adaptives on ROM modules.

I got that part, from the begining actually, but apparently he didn't, although I stated it repeatedly (that it would be very unlikely to work right away with a simple PCB swap).

Owner can check TVS/DIODES but if the drive is not shorting the PSU and shutting the system down then it will have other issues like open resistor and there will be no short-circuit.

I asked again, but he didn't mention any kind of shutdown.
Question : it only shuts down the PSU if the HDD is connected internally, not if connected through an external enclosure or similar, right ? (Seems obvious but... not much should be taken as “obvious” when it comes to HDDs !)

Pre-amp might still be damaged or heads might still be damaged even if the drive spins with a non-native PCB. What we can assume is that the pre-amp at least is not shorted otherwise it would not make the drive spin.

Head damage would always result in a clicking pattern, right ? Or not necessarily ?

Does the drive clicks ? Does the drive makes a noise of heads reading something ?

I asked, waiting for the reply.

If the ROM is inside MCU there will exist a copy of the ROM on SA (platter) so one would replace the PCB with compatible one and use some sort of WD firmware tool to read ROM copy from SA and write it to PCB.

Probably irrelevant here, but I asked again if there's a chip in U12 on his damaged unit.

If the damaged drive have a PCB without ROM chip (ROM EMBEDED ON MCU) then you could use tools like the following ones to rebuild the ROM by reading the COPY on the SA and write it back to the donnor (new) PCB

Well, before I start to delve deeper, which among those tools can be had for free, if any, even if that's a demo version which would still allow that specific procedure ? Is WDMarvel demo version enough ? From what I could gather, all the others are really expensive tools.

It's very easy if you know what you are doing but can be very complex if you don't understand (fully) what you are doing and you might brick the PCB if you write wrong code to it or if your system does interfeer with the PCB while you are writting ROM (if you don't have hardware based tools like MRT, PC-3000, HRT, whatever ....) This is another risk of software only tools.

Well, that's another situation where it would be adviseable to do a lot of thorough testing before attempting the actual procedure, I guess ! :) What kind of test can I do here ? Take a “guinea pig” WD HDD (I have an empty 250GB IDE one which I found in a damaged standalone video recorder), backup its ROM, then replace it with garbage, then verify that it's no longer functional, then write it back, then verify that it's functional again ?
In any case, I'll remind him that he can bring this to a DR lab, for that kind of failure I suppose that it would be way cheaper than for a mechanical failure (for an electronic failure the one in my area asks 190 to 350€ with the standard 8-15 days delay).
Are there means to prevent the system from interfering with the PCB ? And what exactly do you mean by “interfering” ? By definition ROM should be read-only, right ? (I mean, during the regular operation.)

Re: PCB swap on a WD10EAVS

March 9th, 2018, 9:51

It would be simpler to read ROM with external programmer and write it to the PCB by Firmware tool.

In that case, it would read from the (hypothetical) external ROM chip on PCB A, and write it to the MCU of PCB B (which has no external ROM), correct ?

Don't do it. As soon as you reach the part to "no longer functional" unless it does have extrenal ROM chip that you can re-programm with external programmer or you have a tool that can unbrick the pcb by TTL (and writting correct ROM) you will kill the PCB and you will NOT be able to write ROM back....

Alright, so bad idea, even for testing purposes... Is there something else that can be done to test the tool before a first actual procedure ? (That wouldn't entail purchasing two identical drives, if possible... I do have several pairs of WD HDDs with the same model number – including the two WD6400AAKS mentioned in another thread – but none that are currently “expendable”.)

In this case, I would have to do the procedure twice, ideally : once to (hopefully) make the damaged HDD readable again, and once to write back the original ROM to the replacement PCB, so that the healthy HDD (the one I sold) can be used normally later on. For the second part, it has to be done after the PCB has been put back on its original drive, or it doesn't matter ? In any case, once the correct ROM is written to the PCB, there has to be a power cycle for it to be read and for the HDD to (hopefully) start normally, right ?

Regarding the software tools, WDMarvel's demo version is read-only, so it wouldn't do all that's needed, but there's a one month license for $15, it seems to be the cheapest option for a one-time job (I'll have to be sure that the guy wants me to try the trick – but in any case the acquired knowledge won't be lost, it might prove worthwhile later on). But apparently its scope is more limited than HRT / SeDiv for instance ; which one(s) did you purchase first, and which one(s) would you consider the most powerful and / or most useful in the long run, when dealing with drives from various brands / models ?

EDIT : Oh, and to read from the SA with WDMarvel (or any of these tools), it is not necessary to perform a hot swap ?
In which cases is it necessary to perform a hot swap in order to access the system area ?
Post a reply