Hitachi Deskstar HDT722525DLAT80 no spin

[ryvix]

Hi there,

I found this forum and it looks like a good place to start. I have someone whos external hard drive stopped working. They were quoted $2000 from someone and they don't have that. But it has their 2 year olds baby pictures on it they didn't make copies of so they want to recover the data off it.

Someone had them take the actual drive out of the external case and try it in their computer. They said it would power on before they did this and they may have touched the circuit board a little too but now it doesn't power on anymore at all. Now I have it and am wondering what can possibly be done to make it spin up and read the data off. It doesn't spin up in it's case, or if I attach it directly to a power supply in a computer.

I can't see if there are any burnt chips or anything on the drives PCB, either on the front or back of it and nothing smells like its burnt.

I only have the top of the external case and all it has is a Lacie logo and: Design by F.A. Porsche
There is also a sticker on the top side of the back panel: 1046611011364CR 3U2
The PCB inside the external case has this on it:
RD289B-A
M28V0
0639

Top sticker on the drive:
Hitachi Deskstar
Model: HDT722525DLAT80
ATA/IDE
LES

Capacity: 250G
RPM: 7200rpm
P/N: 0A31611
MLC: BA1769
LBA: 488,397,168
CHS: 16383/16/63

0A31611BA17690S6A
S/N: T8CD11RH

White Sticker on bottom edge near IDE cable:
0A29476
BA1588_
Xzn638
15D8

MCU:
Infineon
0A29480
UAB-M3057-HTV
V4 ROM
FR KR SI
EL610157EA7
G0634

White sticker in middle of motor:
13G0832
228
6403
B510WH

Thank you for your time =)

Ryan

[fzabkar]

If you can upload a detailed photo, and if you have a digital multimeter, I can help you with some measurements.

[networks]

Ryan What fzabkar is going toward is getting you to test the TVS diode's with a multimeter. They are power protect diodes that protect 12V and 5V. Just search the forum for TVS Diode fzabkar has some great info here on the subject. Once you have done that let us know how you progress

[ryvix]

Thank you!

I took some pictures but they might not be the best quality possible with this camera.

I got someone to bring a multimeter as well.

I will search for the info and test it, thanks!

Attachments

[ryvix]

And some pictures of the back of the PCB. I tried to take some in different lighting for each.

Attachments

[fzabkar]

Here are the various onboard supplies:

http://www.users.on.net/~fzabkar/HDD/HD ... _NVRAM.jpg
http://www.users.on.net/~fzabkar/HDD/HD ... Preamp.jpg
http://www.users.on.net/~fzabkar/HDD/HD ... io-reg.jpg
http://www.users.on.net/~fzabkar/HDD/HD ... eg-reg.jpg

The first thing I would do is to measure the resistance of the 12V TVS diode (D2) near the power connector. I expect that it's OK -- if it were shorted, then your computer's PSU would have shut down.

I would then measure the resistance between ground and each of the Vio1, Vio2, Vcore, +5V, and -5V supplies. This will tell us if there are any shorted loads.

If there are no shorts, then power up the board on its own and measure the voltages at Vcore, Vio1, Vio2, and -5V. I expect that Vio1 will be +3.3V. I'm guessing that Vio2 will be either +1.8V or +2.5V. Vcore will probably be around +1.5V to +2.1V.

BTW, I'm assuming that the board is the original one. At least the date codes on the stickers and MCU are similar to the date code on the bridge board.

[ryvix]

Hi there,

Sorry for the delay, been travelling around for a few days. Yes I believe it should be the original PCB that came with the drive, but it was purchased as part of the external HD so unless Lacie modified the PCB somehow it should be the original.

For Resistances I set the multimeter to 2000. For voltages I set it to 20.

Resistances:
12V TVS diode (D2) - 178
Vio1 - 359
Vio2 - 358
Vcore - 356
+5V - 1849
-5V - 091

Voltages:
Vcore - 3.29
Vio1 - 3.29
Vio2 - 3.29
-5V - -3.02

It doesn't match your guesses all the time so I hope I did it right. Thanks again for the help!

[fzabkar]

Voltages:
Vcore - 3.29
Vio1 - 3.29
Vio2 - 3.29
-5V - -3.02

It doesn't match your guesses all the time so I hope I did it right.

It looks like I guessed wrong. :-(

Can you measure the voltages at the centre pins of Q1 and Q3? Are they +5V as I suspected? If not, then they could be the Vcore and Vio2 outputs. This would make Q2 the main +3.3V Vio1 regulator, with the remaining voltages being down-converted from this supply.

As for the negative preamp supply, the -3V reading may be correct. I'll be getting some Hitachi PCBs soon (different model) and I'll compare the readings.

[ryvix]

Voltages at center pins:
Q1 - 1.54
Q2 - 5.08
Q3 - 1.58

[ROb]

Change the IC circled and your problem will be solved.

Attachments

hdd_image1.jpg (47.04 KiB) Viewed 17488 times

[drc]

I recommend Luke at Recovery Force for getting a real diagnosis and solution

[fzabkar]

If all else fails, the most expedient DIY solution would be to purchase a compatible PCB and transfer the chip at U5 to your donor. Your local TV/AV repair shop should be able to handle this simple task.

As for the measurements, thanks for your patience. It's now clear that Q2 supplies the +3.3V Vio voltage. This rail supplies the vacant flash memory location at U7, and probably the serial EEPROM at U5. What is the part number of the EEPROM? What is the resistance between its Vdd supply pin and the Vio pin of Q2?

Q3 and Q2 both down-regulate the +3.3V supply. Q3 provides a +1.6V supply (Vcore?) to the MCU. Q1 supplies some other logic with +1.5V. Without knowing their actual values, ISTM that the Vcore voltage is within the ballpark, but I'm not sure of the other rail.

Could you measure the voltage at the supply (bottom left pin) to the SDRAM on the underside of the board? Could you also measure the resistance between this pin and the centre pins of Q1 and Q3? This will tell us which regulator is powering the SDRAM. What is the part number of the SDRAM? Its datasheet will tell us its rated supply voltage.

One final measurement. Could you measure the resistance between the centre pins of Q1 and Q3? This will tell us whether the two rails are shorted to each other.

As for the motor controller, it also controls each of the above regulators. Therefore, even though it has several additional functions, I don't understand why it would be a suspect, given the above information.

[ROb]

The motor controller also controls the regulators. And lot of time same section fails killing the disk and making it a no power case

[fzabkar]

The motor controller also controls the regulators. And lot of time same section fails killing the disk and making it a no power case

But the regulators appear to be working, or are one or more of the supplies out of spec?

[ryvix]

Hi there,

I'm sorry for not responding sooner, but I have been very busy and haven't been able to answer all your questions so I held off until I had some time and could actually see those part numbers.

What is the part number of the EEPROM?

If this is that little black box above U5 then its really hard to see but I think it is:

S93C76
ADVFS
7834

What is the resistance between its Vdd supply pin and the Vio pin of Q2?

I don't know if I did this right but the pin on the top left of the box at Q2 read 001. The pin on the right read 568. The center pin is 591.

Could you measure the voltage at the supply (bottom left pin) to the SDRAM on the underside of the board?

3.29

Could you also measure the resistance between this pin and the centre pins of Q1 and Q3?

This wasn't the easiest thing to do, now I think you are just giving me little challenges =)
Q1 735
Q3 837

What is the part number of the SDRAM?

If I am looking at the right thing:

SAMSUNG 631
K4S641632H-UC60
S6416 ZHR243GFU

Could you measure the resistance between the centre pins of Q1 and Q3?

1114

Thanks again. This is actually kind of fun in a way.

[fzabkar]

Your measurements confirm that Q2 is the pass transistor in a linear +3.3V regulator. The S93C76 NVRAM and K4S641632H SDRAM are both powered from this supply.

K4S641632H-UC60, Samsung, 64Mbit SDRAM, 3.3V:
http://www.datasheetarchive.com/pdf/get ... .pdf&scan=

S-93C76A, CMOS SERIAL E2PROM, 8K-bit, 512 x 16, Seiko Instruments Inc.:
http://datasheet.sii-ic.com/en/microwir ... C76A_E.pdf
http://www.datasheetarchive.com/pdf-dat ... -41509.pdf

As for the other supply rails, I can't see any obvious faults, but my next DIY step would be to purchase a replacement PCB and compare its Vneg, Vcore and Vio voltages against your own. Your last DIY resort would be to transfer the NVRAM chip to your donor PCB. Since its Vdd supply is intact, there is a good chance that this chip is OK.

[ryvix]

You seem to really know your stuff and I really appreciate you taking the time to help, and I am sure the owners of this drive do as well. Thank you =)

I will search for another board somewhere. I was thinking the easiest thing they could do is just buy another identical external hard drive and it hopefully would have the same internal drive with a close match on the board but maybe it would not. Maybe it would make more sense to just search for an identical drive or board some place.

I have one question. You said transfer the NVRAM chip to the donor PCB. Is this because there is some data in that chip which is required for accessing the data on the disks? That would make sense rather than try to track down any broken parts on the old and broken PCB, it would make sense to take that crucial chip and put it on the new PCB and hope it works out.

Thanks again!

[fzabkar]

Yes, the NVRAM stores unique, drive specific "adaptive" information.

[kfdj]

I have the same issue as ryvix a "deathstar" drive that won't spin or respond. I have not had the option to do the testing recommended by fzabkar yet. How did this work out?

[ryvix]

Well, by the sounds of it anything is possible. In my case I found a similar PCB but did not get it to work so I had to give up. I have a feeling it could have been something in the drive itself or something in my case. I sent both PCB back to the owner and I don't know what he did with them. Good luck!