External WD5000AAKS Won't Spin up

[JMarlow1986]

Hello,
Model is WD500AAKS-22TMA0 WD Caviar SE16
Date 21 Feb 2007
DCM HBNCHV2AAB

PCB Board is 2061-701477-800 AB (sticker)
PWB: 2060-701477-001 Rev A (board itself)


I am having problems with my external WD Hard drive, I plug it in and it won't spin up. I don't hear any clicking or nothing to that effect.
I thought the external case power supply might be bad so i bought a SATA docking station to verify. It Still did not spin up.
After further reading of threads, I did take the pcb board off to check D4 and D3 and R64 and R67.
D4=.631v
D3=.528v
R64=.5 Ohms (auto range)
R67=.6 Ohms (auto range)
Also, visual inspection of the board showed no signs of damage.
Did more reading and confirmation, this board does not have anything populated in the U12 spot, so it tells me the firmware is integrated on the IC chip.
Another post I had read 'fzabkar" suggested testing the following componets, so i did that as well.
Vio to ground=5.96 kohm
VCC to ground =8.40 Mohm
Pin 1 of Ram to ground=2.219 Mohm
R48=.8 Ohms
R45 .5 Ohms
R47 .6 Ohms

l1=.5 Ohms To ground=9.69 Mohms
L2=.5 Ohms To Ground= 162 ohms
L3=.5 Ohms To ground= .5ohms

C1=79.6 Ohms
C2=79.2 Ohms
C3=39.0 KOhms (charging)
C4=39.0 Kohms (charging)
C5= 1 Kohms
C6= 39 K ohms (charging)

I am not sure what they should be measuring for the above components, but they seemed to be on par from another thread I was reading on here. As for what step or steps I need to perform next, i am uncertain. I have also contacted HDD-parts.com and they suggested sending them the board for repair. I just want to make sure I have exhausted all avenues before sending the board out. Thanks in advance for any help/guidance.

 

[fzabkar]

I expect that hdd-parts.com will probably try to transfer the Marvell MCU chip to your replacement PCB.

I would first measure the voltages at L1 and L2, plus the voltage at the anode of the diode near L3. IIRC, these should be +1.2V, +3.3V and -5.0V.

 

[JMarlow1986]

I expect that hdd-parts.com will probably try to transfer the Marvell MCU chip to your replacement PCB.

I would first measure the voltages at L1 and L2, plus the voltage at the anode of the diode near L3. IIRC, these should be +1.2V, +3.3V and -5.0V.

Thanks for the reply,
When hdd emailed me back their response was "For this model hard drive, the board cannot be exchanged with another board. One of the solutions to rescue the drive is to fix the original board."

Also, just before i left it at work today, I did check the motor resistance directly at the leads. A,B, and C, they were all at 2ohms.

As for the new tests you have described, I will check those tomorrow, I have been trying to do it there since I have access to various tools.

 

[fzabkar]

One other solution is to take a compatible replacement PCB and then use a tool such as WDR or WDMarvel to reconstruct the firmware from backup copies in the System Area (SA) on the platters. However, doing something like that requires some experience and practice with donor drives.

The HDD Oracle forum has a lot more information on the subject.

 

[JMarlow1986]

I expect that hdd-parts.com will probably try to transfer the Marvell MCU chip to your replacement PCB.

I would first measure the voltages at L1 and L2, plus the voltage at the anode of the diode near L3. IIRC, these should be +1.2V, +3.3V and -5.0V.

Update 3/4/2014:
tested said points
VIO=3.342V
VCore=1.385V
D2= -5v, However, this was with bare board. I could see this for a few seconds and then it would just go to zero volts. I am assuming that it was looking for the motor and since it wasn't there it turned off? No load to keep it on so to speak?
Also, during the powered test, I probed the contacts at J6,(motor contacts) I would see a voltage from 2-3 volts on the contacts for a quick second then off and on, until that same D2 no longer showed -5v.

Does this mean that everything on the board is ok, but the motor is bad? or could there still be something wrong with the pcb?
Is there a way to power just the motor to see if it spins up?

Thanks

 

[fzabkar]

The voltages appear OK.

AIUI, the motor controller generates the Vcore and Vio voltages. Vcore powers the MCU, while Vio powers the MCU, SDRAM, and U12 (if it exists). If these voltages are OK, the motor controller releases the POK (Power OK) signal to the MCU. The MCU then powers up, copies the code in its internal flash memory to SDRAM, and then transfers control to the RAM code. This code performs a self test (POST) and commands the motor controller to switch on the -5V supply. This voltage powers the preamp inside the HDA. If the preamp is present, the MCU then commands the motor controller to spin up the drive.

ISTM that the drive may have a stiction fault (heads stuck to platters) or a seized spindle motor. The pulsing you are measuring at the motor terminals is "spin buzz", ie the motor controller appears to be trying to kickstart the motor. After a few unsuccessful attempts it gives up and the MCU then switches off the preamp's -5V supply.

You could confirm whether the PCB is at fault simply by swapping it. The replacement PCB probably won't give you access to your data, but it should still spin up the drive.

 

[JMarlow1986]

The voltages appear OK.

AIUI, the motor controller generates the Vcore and Vio voltages. Vcore powers the MCU, while Vio powers the MCU, SDRAM, and U12 (if it exists). If these voltages are OK, the motor controller releases the POK (Power OK) signal to the MCU. The MCU then powers up, copies the code in its internal flash memory to SDRAM, and then transfers control to the RAM code. This code performs a self test (POST) and commands the motor controller to switch on the -5V supply. This voltage powers the preamp inside the HDA. If the preamp is present, the MCU then commands the motor controller to spin up the drive.

ISTM that the drive may have a stiction fault (heads stuck to platters) or a seized spindle motor. The pulsing you are measuring at the motor terminals is "spin buzz", ie the motor controller appears to be trying to kickstart the motor. After a few unsuccessful attempts it gives up and the MCU then switches off the preamp's -5V supply.

You could confirm whether the PCB is at fault simply by swapping it. The replacement PCB probably won't give you access to your data, but it should still spin up the drive.

Thank you for the reply,
I don't have a compatible pcb board laying around, however, I did find another HDD that I know works laying around at work and I applied power to it and took a few measurement to compare to.
One thing I noticed was that with the known good HDD that when I hooked up an O scope to the motor leads I could see the duty cycle, whereas with mine, I saw no duty cycle.
Would this confirm that the smooth controller is not working properly and is probably bad or something else within its circuitry?
If so, I think I am to the point where I have to send it to HDD-parts.com and have them repair the board.

 

[fzabkar]

I haven't scoped a drive with a spin-up problem, so I can't advise you on that score. Normally you should hear faint buzzing sounds if you put your ear against the drive's cover.

If it were my drive, I would try to procure a cheap known good PCB, or a drive with a head fault, for testing purposes before going for an MCU swap.

FWIW, this site has some scope pix:

http://nazyura.hardw.net/000004.htm

FYI here is the datasheet for an early L7250 SMOOTH motor controller:

http://pdf.datasheetcatalog.com/datasheet/SGSThomsonMicroelectronics/mXyuswx.pdf

I have written a tutorial that might interest you:

Measuring VCM and Spindle Motor Current:
http://malthus.zapto.org/viewtopic.php?f=59&t=204&p=484