Hitachi 1TB Sata drive fail

[Mark_67]

Hi Everyone. I am hoping someone can give me some advice on what to do next. My Hitachi 1TB sata drive doesn't power on any longer. This happened after installing a case fan and splicing into the molex connector which supplies power to the drive also. I believe the power/ground wires got mixed up and the drive no longer powers on.

I purchased an identical drive as a donor hoping I could swap the boards and get some of the data back. Unfortunately after swapping the boards the drive still does not power on/ no spinning.

My question is, is it possible that the donor board not powering up the drive is because it is a different version/revision number? I'm hoping this is the case and not a fried motor. Is there any tests I can do to determine if it's the board or the motor? Any help would be greatly appreciated. Thanks in advance.

The May 2010 is the donor drive and put the boards back on their respective drives.

Mark

Attachments

Donor board is the one on the left

Donor is the one on the left

[Mark_67]

I did some research and seen something mentioned about the 12v and 5v diodes. I'm going to purchase a multimeter tomorrow and check these. I hope I didn't do any more damage by testing the donor board on the drive but when I did apply power it didn't even spin up...nothing. btw - if higher res images of the board would help I could take them.

[chinopk]

if the drive did't spin up with other working pcb, may be you should inquire motor instead of diode on pcb..

[pcimage]

Given that you've put a known good PCB on and it still doesn't spin, would indicate some internal problems.

btw You can virtually NEVER expect a straight PCB swap to work on these, but it should at least spinup.

[Mark_67]

Thanks for the replies, I appreciate it. That's what I was afraid of. I was hoping that the different revision/firmware on the donor board somehow stopped it from spinning up. Yes the donor pcb is good, tested when I put it back on the donor drive and it works. I was really hoping the damage only got something on the board. If I remove the 12v and 5v diodes cleanly and test for spin up with a good known power supply, could that damage it further? Any ideas on what I can do myself?

[Mark_67]

If it's an internal motor problem would I be able to tell by reading the voltage at the end of the ribbon cable? Just to see if the proper voltage is going into the casing? I see there is only 4 conductors that go into the casing of the drive. I'm assuming a couple are for data and a couple for power? I have no idea really, just guessing.

[Vulcan]

I see there is only 4 conductors that go into the casing of the drive. I'm assuming a couple are for data and a couple for power? I have no idea really, just guessing.

No, it's not 2 x data & 2 x power to the motor - it's (normally) 3 phases & a common.

With the PCBs removed, you could perform resistance measurements between the contacts on the motor cable, and compare between original & donor drives, if you wanted to do that to get a basic view of motor winding health. There is also another connector under the PCB, which leads to the pre-amp & VCM inside the HDA - I don't recommend probing that one.

However, as pcimage said, the lack of spin-up with the donor PCB on the original drive, shows that there is a problem elsewhere on your original drive (although the original PCB could also be damaged), so as I understand your situation, unfortunately it's "game over" for you attempting DIY.

You could even go down the (risky!) path of comparing voltages generated on the donor drive for the pre-amp, MCU etc., with those present on the original drive and hence find whether the pre-amp is shorting its supply etc., although this requires power to be on, and so one slip with the multimeter probes and you could damage that drive. But whether the problem is the motor or the pre-amp (that one would be my worry) it doesn't make a difference to the decision - working on either one requires opening the clean environment of the HDA, which is outside of what you can sensibly do. In other words, if the data is important, you'll need to use the services of a pro who has the necessary experience & cleanroom / bench.

(I'm not a DR pro, just an electronics engineer, so feel free to wait for a second opinion - but the situation seems clear to me. Sorry for the bad news. There are DR members on the board in Canada, so I'm sure you'll get some recommendations, if you choose to go down that route, as well as those like pcimage elsewhere in the world. Good luck!)

[Mark_67]

Thanks Vulcan, unfortunately my budget is limited these days. I guess I'll have to decide if the data is worth the cost. I'm just guessing right now, that a professional platter and nvram swap is what might get the data off?

Thanks again for the valuable info.

[fzabkar]

If you still have the mis-wired Molex - SATA connector, measure the output voltages at the SATA pins. That will help us determine what happened and where to look for damage. Of course a shorted TVS diode may tell us the same thing.

BTW, you won't need a platter swap.

[poehere]

Like PCimage said if you put on the other PCB board and it did not spin up there is some internal damage to the preamp on this one. Even moving the 12V and 5V will not help out in your case sorry to say this one. You need to get the drive to somene who knows this drive and can work on it becasue there is a reason why your preamp is blown and so you need not only PCB work but also clean room work in changing out the HSA on this one. Sorry but no other way. It it not recommanded that you try and swap out the HSA on this drive. If you do not find the casue of the problems and you change out the HSA you can destroy this one too. So in your case you more than likely have gone as far as you can with a DIY fix on this one. Sorry but if data is important you will need pro help other wise you can buy a new drive and go on and start over.

[Mark_67]

Thanks everyone for your help, I appreciate it. fsabkar, unfortunately I threw out the miswired molex splitter. I'm going to shelf this one for now and save up for some professional work.

[fzabkar]

Thanks everyone for your help, I appreciate it. fsabkar, unfortunately I threw out the miswired molex splitter. I'm going to shelf this one for now and save up for some professional work.

You can still measure the resistances of the TVS diodes on your original board. That information will be helpful.

[poehere]

Hummm even measureing 12V and 5V I do not see how this one can help him out at all. Even if you do not move adaptive from one board to another and you put on the board to the HDD it will spin up. That is if one the preamp is not damaged and two the HDD does not have a spindle motor problem. You can test the PCB on the HDD and if it spins up then you know for a fact it is not the preamp on this one. Then you move on to the next step and that is adaptive that need to be moved over.

But considering PCB will not spin up the HDD for the op how is measuring the TVS going to help him out at all. Even changing them on orignal board can not help him. He more than likely has a problem that is internal to this HDD like preamp, or spindle motor. This one really needs to be opened in a clean room to see why it is not spinning up with the donor pcb board on it.

[Mark_67]

Hi, I had removed the tvs diodes just to see if that was the problem earlier as I didn't have a meter. I figured nothing would happen anyway because the donor board didn't even spin it. Just curious, is the nvram chip the 8 legged one on the board? No matter what, this chip is going to need to be moved to a new board I'm assuming? Would anyone know how many platters this drive has?

[poehere]

Hi, I had removed the tvs diodes just to see if that was the problem earlier as I didn't have a meter. I figured nothing would happen anyway because the donor board didn't even spin it. Just curious, is the nvram chip the 8 legged one on the board? No matter what, this chip is going to need to be moved to a new board I'm assuming? Would anyone know how many platters this drive has?

Hi there if you are thinking of platter swap you are way over your head on this one. Normally platter swaps are the last results in DR. It is not something to just do to see if you can get back data on your drive. Platter swaps take a lot of knowledge and practice to do this one and they are a lot harder than they make them look on U-Tube or where ever else people post videos of these using tin cans cut up with rubber bands on them or what ever to attempt to move their platters from one drive to another. I will say one more time if you value your data you should really find someone to help you on this. You might not have to move platters and it can still be a problem with preamp on this one. But you need proper diagnoses in order to determine this problem on there.

[fzabkar]

But considering PCB will not spin up the HDD for the op how is measuring the TVS going to help him out at all.

I wasn't suggesting that removing a shorted diode will solve the OP's problems. My suggestion was intended as a post mortem examination. AISI, if the preamp is damaged, and if the damage was caused by an external event, then one would expect that the 5V TVS diode would be affected. This is because both the positive and negative supply rails for the preamp are derived from the +5V supply.

[fzabkar]

Hi, I had removed the tvs diodes just to see if that was the problem earlier as I didn't have a meter. I figured nothing would happen anyway because the donor board didn't even spin it. Just curious, is the nvram chip the 8 legged one on the board? No matter what, this chip is going to need to be moved to a new board I'm assuming?

When you get a meter, measure the diodes.

As for the NVRAM, what are the markings on the chip?

FYI, here is my HDD IC database:
http://www.users.on.net/~fzabkar/HDD/HDD_ICs.txt

As for your preamp, I don't know whether the MCU needs to detect the preamp as part of its POST, but if not, then place a thin card between the preamp connector and the PCB. Now see if the drive will spin up.

[Mark_67]

Hi. Poehere, I wasn't thinking platter swap but maybe an HSA swap. Just curious as too how many heads I might be dealing with. The data on the drive is somewhat important but I do have physical printouts of 95% of the records. Re-entering them might take me a long time but it is doable. If I can just get the drive to spin up and read for a short while I was hoping to extract these. Fzabkar, thanks for the database. The markings on the chip are 25FW206 970S.

[fzabkar]

The markings on the chip are 25FW206 970S.

That appears to be a LE25FW206 2Mbit Sanyo serial flash memory.