Dead 2.5" WD800VE PATA - help

[jono-ats]

Well, there is one sure way to rule out so-called "stiction," and it doesn't involve the use of a VOM . . .

[karlshoj]

What could the next step be? try to buy a new PCB or ?

Mounted the PCB again, if I put the ear next to the HDD a single quite click can be heared when power is connected to the drive. Nothing else happens.

[fzabkar]

I would have thought that you would have heard several attempts to spin up, not just the one, if indeed that's what the click was. You could isolate the motor terminals from the PCB with a business card and listen again.

[karlshoj]

Hi,

Now tried to isolate the motor terminals (near the hole in the PCB)
When connecting power to the drive I still hear a very small click, now the motor terminals are isolated it most be something else?
Something at the PCB most prevent the drive to try to spin up?

Thanks, hope you found some new suggestions

[fzabkar]

As you say, if the motor terminals are not making contact with the PCB (ie after isolating connector pins 1-4 of J6), then the click must be coming from somewhere else. AFAICS, the only other source for this noise would be the voice coil. To verify this, you could isolate pins 2 and 4 of preamp connector J1. You could use insulation tape for this purpose.

Also measure the voltages at test points E50, E51, E52, and E53 near the motor terminals on the underside of the PCB. Do this immediately after powering on the drive. I expect that these test points are connected to the motor terminals.

I'd also measure the voltages at E40, E10, E63, and E56 at the preamp end of the PCB. I suspect that E40 may be the negative supply for the preamp. You will now be measuring it under its normal load.

[karlshoj]

Measure this with power connected to the drive and no insulation tape, not much life.

E50, E51, E52, E53 = 0.0 volt
E40 = 0.1 volt
E10 = 0.0 volt
E63 = 0.0 volt
E56 = 5 volt

[karlshoj]

Anything suggestion? or should I try to buy a replacement PCB?

Thanks

[fzabkar]

Sorry, I didn't see your reply.

Anyway, ISTM that the negative supply for the preamp is missing. I would check for a short circuit on this supply by measuring the resistance between E40 and ground, with power removed. Use the 200 ohms range of your meter, not the diode range.

The absence of activity at the motor terminals is consistent with the missing supply. This is because the motor controller chip monitors all the onboard supplies and holds the board in a reset state if any are out of spec.

If there is no short circuit at E40, then you could try a board swap, but you will need to transfer the chip at U12 to your donor.

[karlshoj]

Between E40 and ground is 800 ohm

[fzabkar]

Between E40 and ground is 800 ohm

I presume you performed this resistance measurement while the board was screwed down onto the drive. If so, then this would confirm that the preamp is not shorted.

Just to summarise what you've found so far, it is evident that the SMOOTH chip has some sanity, and that it is correctly generating the Vio and Vcore supplies for the MCU and its peripheral chips. However, if the voltage at E40 (0V) was measured while the preamp was connected, then this would suggest that there is something wrong with the negative supply.

AIUI, the topology of the negative supply is that of an inverting buck-boost converter. Q3 is the chopper MOSFET, D8 is the flywheel diode, and L1 is the inductor. R120 (R270 marking) is probably a current sensing/limiting resistor. Its resistance should be 0.27 ohms.

Other than confirming that neither R120 or L1 is open circuit, and that D8 tests correctly as a diode, I don't know what else to suggest except to troubleshoot this area with an oscilloscope. Resistor R30 appears to be connected to the gate of Q3 (marking STPA), so I would expect to see PWM pulses at this point. These would be coming from the SMOOTH chip and would have a frequency of the order of several hundred kHz, or even as high as a few MHz. The MOSFET is probably a P-channel type, with its source pins connected to the +5V supply. I would confirm that this is the case by measuring the resistance between each of the pins of Q3 and the +5V input.