Sign in with
Sign up | Sign in
Your question
Solved

HDD recovery - motor wont spin

Last response: in Storage
Share
July 24, 2012 10:57:20 PM

Hi all
I have dead HDD - pcb is burn (phisical damage) probebly connected both power ports??
I got the same HDD model - tested it and it is working - took the pcb and assembled it to the dead hdd,
But the motor wont spin up
Any idea what to do next?
Thank you!
a b G Storage
July 25, 2012 1:10:34 AM

throw the bad hard disk away and use your new drive.
m
0
l
July 26, 2012 2:58:25 AM

I forgot the main issue, Sorry :D  I need the info from the dead HDD...
m
0
l
Related resources
a b G Storage
July 26, 2012 2:51:31 PM

about the only think you can do now is move the platters in a clean room environment to the new hard disk.
m
0
l
July 27, 2012 4:47:57 AM

Any way to simulate clean room in home environment?
I was thinking about aquarium with suction

What do you say?
m
0
l
a c 316 G Storage
July 30, 2012 4:07:15 AM

Please do NOT open your drive.

Instead I suggest you first upload a photo of the burnt PCB. Sometimes there is an easy DIY fix. A model number would help a lot, too.
m
0
l
July 30, 2012 7:16:10 PM

Thank you for your help

you can see the damage on the left hand side bottom.
Its a Hitachi Deskstar 7K160 (HDS721616PLA380) 160GB 7200RPM Hard Drive
m
0
l

Best solution

a c 316 G Storage
July 30, 2012 11:33:16 PM

Let me first draw your attention to the motor connector at the top of the photo. When removing the flex lead, you need to release the two locking tabs by pulling them away from the connector.

If you decide to replace the PCB, you will need to transfer the 8-pin NVRAM chip at U5, or its contents, from patient to donor. This chip stores unique, drive specific information.

As for the damage to your PCB, it appears that there has been an overvoltage on the +5V supply. This has taken out (shorted?) the 5V TVS diode at D1. Someone (you?) has removed it. This was the DIY fix I was referring to. However, in your case there appears to be a series resistor (R5) which probably acts like a fuse. I suspect it is open circuit.

You can see it a little better in this photo:
http://i.ebayimg.com/00/s/MTIwMFgxNjAw/$(KGrHqR,!pIE8VbQtv7QBPUeRlTBwQ~~60_3.JPG

I would clean up the charred area using a toothbrush and isopropyl alcohol (audio/video tape head cleaner). Then I would measure the resistance of R5 with a multimeter. Also measure the resistances between the +5V pins at the SATA and PATA power connectors, and the "+5" test points adjacent to the middle 100uH coil (L3, 100 MJ). This will tell us if there is continuity between the power connectors and the +5V logic on the PCB. If R5 is open, then you could flow a blob of solder over it, or you could run a jumper wire from the power connector to the +5 test points.

See http://www.users.on.net/~fzabkar/HDD/TVS_diode_FAQ.html

... and http://www.users.on.net/~fzabkar/HDD (photo clips)

If you get this far and the drive still doesn't spin, then it could be that the preamp is damaged, in which case the head stack will need to be replaced. That's not a DIY job. Nevertheless we could still take further measurements to narrow down the fault, if you feel capable.

Here are the power connector pinouts:
http://pinouts.ru/Power/sata-power_pinout.shtml
http://pinouts.ru/Power/BigPower_pinout.shtml
Share
July 31, 2012 5:19:21 PM

Thank you very much for the informative answer and you’re willing to help
The drive contains photos and document of my young brother,
He is the one to blame :p  I think he contacted both power ports at the same time...
The data on the drive is not critical information, but it is not pleasant to lose 8-9 years of pictures, video, documents emails etc...
He was not aware of the amount he had in his HDD until it crashed... and he didnt have a backup (I know he is an idiot..)

If I understand your answer correctly I have few options
Replace R5 on the old PCB and other components on this area
or
Extract the NVRAM chip

Seems like replacing the NVRAM is the best way for me – since I have a new PCB,
And seems like other components are destroyed on the charred area - I am uploading close up image.
Any other part to replace from the “patient” pcb?

Thank you again!

m
0
l
August 8, 2012 1:48:51 PM

Anyone,
Any other part to replace from the “patient” pcb beside the NVRAM?
Thanks!
m
0
l
a c 316 G Storage
August 10, 2012 9:27:51 AM

Sorry, I thought you had solved your problem.

See the following PCB matching guides:
http://www.donordrives.com/hard-drive-pcb-donor-swap-ma...
http://www.donordrives.com/blog/pcbswapguide

BTW, here is a much better view of your board:
http://forum.hddguru.com/download/file.php?id=2404&t=1&...

It appears that resistor R5 feeds some other area of +5V logic (preamp?) rather than the 5V TVS diode, so maybe it is OK. Did you test it? It should measure 0.45 ohms. If it is open circuit, then you will need to do some additional troubleshooting before replacing it.

Did you remove diode D4, or did someone else do that?

Did you fix your PSU before reapplying power? If not, then you may have done more serious damage to the PCB, including the NVRAM.

Are you capable of using a multimeter?
m
0
l
August 14, 2012 4:15:51 AM

I don't have electronics diploma but I am capable of using a multimeter for most of the things (open circuit voltage capacity current etc)

I can solder stuff to the PCB (even if the NVRM legs are tiny…)

But I need precise instructions on how to attack this problem.
D4 diode blow up probably.
R5 resistor is 1.1 oham on both PCB boards.
m
0
l
a c 316 G Storage
August 14, 2012 7:17:36 AM

R5 is OK. The extra 0.6 ohm resistance would be accounted for by your meter leads.

I would power up the original board on its own and measure the following voltages:

- +5V test point to the left of the 100MJ coil (L3)
- +12V test point below the +5V test point
- transistor Q3 to the right of R5
- transistor "C6052" at the top left corner
- transistor Q1 to the right of the motor connector
- voltages at coils both 100MJ coils (L1 and L3)

The above voltages will be the onboard power supplies. Be very careful, as a slip of your probes will result in serious damage.
m
0
l
August 15, 2012 1:31:51 AM

Just want to make sure, you want me to connect the damaged PCB to the sata power?
m
0
l
a c 316 G Storage
August 15, 2012 2:17:23 AM

Yes, that's right. You don't need to connect the SATA data cable.

Did you actually try to spin up the drive with the damaged PCB in its present state?

In most cases the TVS diode protects the rest of the board, so removing the shorted diode is usually enough to get the drive going again. Judging from the uncertainty in your replies, I'm guessing that somebody else removed the diode. You obviously know how to use a multimeter, so you would have been able to confirm the resistance of the diode, had you done it yourself. I've asked you whether your board was subjected to the same overvoltage after removing the diode. If so, then you may be wasting your time.

m
0
l
August 15, 2012 4:11:09 AM

>>> Did you try to spin up the drive with the damaged PCB in its present state?
It wont spin

>>>I'm guessing that somebody else removed the diode
No one removed anything, it was exploded.

Q3 transistor has just 2 legs connected to pcb reading is 640 ohm
same for C6052

>>> Q1
regarding to Q1, Iam not sure how to test this - I tried to follow this instructions on how to test the transistor video here :
http://www.youtube.com/watch?v=8VK_4gZU__I
I put my multimeter on beep mode, red probe on left leg of the transistor, black probe on the middle or right leg should beep, but it wont beep - but the same thing on the good PCB.

>>>voltages at coils both 100MJ coils (L1 and L3)
its 0 - no voltage.
On the good HDD that I got from ebay the voltage is 3.7v for L3 and 3.2 for L1
I rechecked that the power adapter gets 12.41 and 5.14
m
0
l
a c 316 G Storage
August 15, 2012 10:59:20 PM

With respect, you seem to have missed the point of the tests. I'm asking you to measure the voltages at the designated test points. I'm not asking you to test the transistors, I need the voltages at each of their pins. BTW, the metal tab (collector) is the middle pin.

First of all, the fact that the diode exploded suggests that it may have gone open circuit, in which case the overvoltage that killed it may have gone on to damage the motor controller chip. That's the large square IC in the top left corner. This IC controls the spindle motor, voice coil positioner (VCM), and the onboard supply voltages. The motor controller IC uses the +12V supply for the spindle motor and VCM, and the +5V supply for its digital logic. The fact that the overvoltage hit the +5V supply would suggest that the motor controller may be brain dead. If so, then the onboard supplies would most likely be absent, in which case the main "infineon" MCU would not be getting its Vcore and Vio voltages. These are the voltages that I'm asking you to measure. For example, the transistor at the top left should measure +3.9V, +5V, and +3.3V, IIRC.

The two coils together are part of a switchmode Cuk converter. This converter supplies -5V to the preamplifier on the head stack inside the drive. The same preamp also gets +5V from SATA power. That's why I'm concerned about it. IIRC, there should be +5V on the top coil, and -5V on the bottom one, but these numbers may be for a different model.

Did you confirm whether +12V and +5V were present at the test points to the left of the top coil?

BTW, one would normally expect that the drive should at least spin up with the replacement PCB, even with non-native NVRAM. That said, several data recovery professionals (at HDD Guru) have also reported similar strange no-spin behaviour, but nobody has offered an explanation (perhaps because they want to keep it secret).

Anyway, at least you have a good board to compare voltage measurements, so that should help us immensely. If the board turns out not to be salvageable, then you need to at least confirm that there is no short circuit on the supply to the NVRAM, otherwise this chip may be dead and your DIY chances would be nil. Whatever you do, you must not damage this IC.

Best of luck.
m
0
l
August 22, 2012 6:40:50 PM

fzabkar, I invited my friend to come help me with the measurements
I will post the result
Thank you for your patients.
m
0
l
January 25, 2013 1:37:50 PM

Hello, fzakbar

i have HITACHI DESKSTAR model HDS721616PLA380, 160 gb, it was working few das before and accidently it fell down from a small height and its stops detecting since then. i really need to recover the data from the HDD, PLS HELP
m
0
l
!