Reply to Fzabkar:



I get such a voltage of 4.59 V only when I plug the PCB into an adapter from SATA to USB. Because if I plug the PCB into the SATA power connector of the PSU, then I get 5.08 V.

In short, the voltage of 4.59 V comes from a USB 2.0 port.

Reply to Fzabkar:



Multimeter in diode/continuity scale:
If I place the black probe in anode of D2 and the red probe in HDA pad #6: there is NO continuity.

The preamp needs two supplies -- a negative -5V supply and a positive supply, either +5V or maybe +3.3V.

Pin #1 appears to be the ground pin. There should be zero resistance between pin #1 and the screw hole.

Pins 4 and 6 should be the positive and negative supplies. One will be wired to the anode of D2. That's the negative supply. The other will be wired to the SATA +5V pins, or perhaps the emitter of Q4 (+3.3V).

At the HDA terminal block there should not be a low resistance between pin #1 and each of pin #4 and pin #6. If there really is 0 ohms between pin #1 and #4, then the preamp is shorted. But, if the preamp is shorted, then the corresponding supply on the PCB would also be shorted, in which case the PCB would not power up.

It seems to me that your measurements are inconsistent with your observations. At least they don't make any sense to me.

_________________
A backup a day keeps DR away.

Fzabkar, excuse me but I was wrong.

Firstly, because in the resistance scale I selected 200 ohms all the times and that's the reason why it was always saying "OL".

And also I wrongly selected the diode scale for measurements.

You are right, my data were wrong.

For example, I have been able to verify that fortunately there is no continuity between HDA pin #1 and HDA pin #4.

I have taken all the measurements again and they are next:

[Note: I have used a better multimeter than before and now the resistance of the new multimeter is only 0.5 ohms (instead of 1.3 ohms)].


1) Pin #1 of J1 is Ground (confirmed).

Multimeter in Ohm scale (selecting 200 ohms):

If I place the tip of the black probe in a screen hole and the tip of the red probe in pad #1 of J1, then there is a resistance of 0.5 ohms. But since the multimeter resistance is 0.5 ohms. Therefore the real resistance of pad #1 of J1 is 0 ohms.


2) Pin #4 of J1 is -5 V: confirmed (but it is working badly)

Multimeter in Ohm scale (selecting 200 ohms):
If I place the tip of the black probe in anode of D2 and the tip of the red probe in pad #4 of J1: I get a resistance of 0.5 ohms. But since the multimeter resistance is 0.5 ohms. Therefore the real resistance is 0 ohms.


3) Anode of diode D2 is working badly, it never gives a reading of -5V

Multimeter in DCV scale (selecting 20 V):
If I place the tip of the black probe in a screw hole as ground and the tip of the red probe in the anode of D2, then I do not get -5V but a reading of 3.24V. (I do NOT get a reading of -5 V, not even just after plugging the PCB into the SATA power conector of the PSU. I always get 3.24 V or around. In fact, after a few minutes, the anode of D2 gives no voltage).



4) Pin #4 of J1 has the same wrong voltage as anode of diode D2

Multimeter in DCV scale (selecting 20 V):
If I place the tip of the black probe in a screw hole as ground and the the tip of red probe in pad #4 of J1, then I do not get -5V but a reading of 3.24V. (I have taken quite a lot of readings and it always says the same).


5) Pin #6 of J1 is +5 V: confirmed.

Multimeter in Ohm scale (selecting 200 ohms):
If I place the tip of the black probe in pad #6 of J1 and the tip of the red probe in pin #8 of the SATA Power connector: I get a resistance of 0.5 ohms. But since the multimeter resistance is 0.5 ohms. Therefore the real resistance is 0 ohms.

Multimeter in DCV scale (selecting 20 V):
If I place the tip of the black probe in a screw hole as ground and the tip of the red probe in pin #6 of J1, I get a reading of 5.09 V.


Now I am going to give you the measurements I have taken to the preamp:




Multimeter in continuity scale:
If I place the tip of the black probe in anode of D2 and the tip of the red probe in HDA pin #4: there is NO continuity. It does not beep at all.

Multimeter in Ohm scale (selecting 200 ohms or 2000 ohms or 20 k-ohms):
If I place the tip of the black probe in anode of D2 and the tip of the red probe in HDA pin #4: I get "OL". (I always get "OL" even if I select 20 k-ohms).




Multimeter in continuity scale:
If I place the tip of the black probe in anode of D2 and the tip of the red probe in HDA pin #6: there is NO continuity. It does not beep at all.




1) Resistance between HDA pin #1 and HDA pin #4:

Multimeter in Ohm scale (selecting 20 k-ohms):

If I place the tip of the black probe in HDA pin #1 and the tip of the red probe in HDA pin #4: I get a reading of 6.80 k-ohms (i.e., 6,800 ohms).


2) Resistance between pad #1 of J1 and pad #4 of J1:

Multimeter in Ohm scale (selecting 20 k-ohms):

If I place the tip of the black probe in pad #1 of J1 and the tip of the red probe in pad #4 of J1: I get a reading of 13.49 k-ohms (i.e., 13,490 ohms).


3) Resistance between HDA pin #1 and HDA pin #6:

Multimeter in Ohm scale (selecting 20 k-ohms):

If I place the tip of the black probe in HDA pin #1 and the tip of the red probe in HDA pin #6: I get a reading of 12.58 k-ohms (i.e., 12,580 ohms).


4) Resistance between pad #1 of J1 and pad #6 of J1:

Multimeter in Ohm scale (selecting 200 k-ohms):

If I place the tip of the black probe in pad #1 of J1 and the tip of the red probe in pad #6 of J1: I get a reading of 47.4 k-ohms (i.e., 47,400 ohms).



Again, regarding resistor R47 (theoretically R100, ie. 0.10 ohms), I have measured it again.

And I got 1,014 ohms (one thousand and forteen ohms!).

I have taken several measurements and I got the same result. Therefore it is clearly damaged.


Also, I get some strange readings in pin #8 of J1:

Multimeter in DCV scale (selecting 20 V)
If I place the tip of the black probe in a screw hole as ground and the tip of the red probe in pin #8 of J1, I get 2.40 V. I have done quite a lot of tests and always it gives me 2.40 V or around.

Is this voltage correct for pin #8 of J1?


In conclusion, it seems that the main problem is the anode of diode D2. Because it is giving an incorrect supply of 3.24 V. And for this reason pin #4 of J1 has this wrong voltage.


Fzabkar, excuse me for my errors.

And many, many thanks for your patience and help.

Finally, I have taken more measurements to diode D2.

Now I have measured its resistance.

Multimeter in Ohm scale (selecting 200 k-ohms):

If I place the tip of the black probe on the anode of D2 and the tip of the red probe on its catode: I get readings between 80 k-ohms and 86 k-ohms.

Thus, for example, a reading says 85.3 k-ohms (ie. 85,300 ohms).

Another reading says 81.7 k-ohms (ie. 81,700 ohms).

And finally, other reading says 83.4 k-ohms (ie. 83,400 ohms).

More measurements to the diode D2.

Now if I connect positive with positive and negative with negative. That is, the red probe to anode and the black probe to catode.

The results are the following:

Multimeter in Ohm scale (selecting 200 ohms):

If I place the tip of the red probe on the anode of D2 and the tip of the black probe on its catode: I get readings between 140 and 145 ohms.

Is this value correct for diode D2?

[Note: Obviously, I already know that current flows from anode (+) to catode (-) in a diode. And that the line engraved on a diode is the catode].

You cannot right test those diodes without removing them off the PCB, you are currently measuring others components resistance present in the circuit around.

I have tried on my PCB, I attach an image that will give you an idea, my multimeter give some junk sometime so pardon me if not exactly right measured, but I have try to no doing error by confirming with the R47 value. Also some values are moving a little. And also my PCB is not identical to your, but this zone yes is visually identical, I have grayed all that change.

I think next, you have no other choice than firstly replace R47 and re-measure those -5V/+5V said by fzabkar, and buy more than one R47 for in case it re-burn just after... I dont know if there is really a culprit about this R47 burning...

Good luck!

Finally I don't know exactly if this D2 cannot be measured on circuit, fzabkar will tell you.

fzabkar say that Q2 can be the culprit of the burned R47.

So I have try to measure it, and wow my cheap multimeter seems doing lot of bullshit, recalibrating it completely change the result and do not correspond to the previous values on lower caliber...

For example, I have retry the D2 to see if I can get your 80k, and maybe yes with the 2000k caliber:
2000k -> 008
200k -> 13.7
20k -> 6.40
2000 -
200 -

So take my measure with pencil...
It seems only my measure inferior to 200 ohms are accurate.

I post my last screen anyway, for you try seeing if your Q2 is damaged.

Diybit, thank you for your pictures with measurements. They are very useful.

I'm not sure they will be useful
Really my multimeter have lose his head... sorry...
And I haven't giant resistance to test it...

Try to do with that...
Look if you find something at about 0ohm where mine are >200.

fzabkar is even saying to retest just by replacing R47.

I repost his text:
R47 does look burnt. That could point to a shorted MOSFET (Q2), shorted turns in inductor L2, or a fault in the SMOOTH chip. The SMOOTH chip controls the voice coil, spindle motor, and the power supplies. Your measurements would indicate that Q2 does not have a Drain-to-Source short, so I would replace R47 and try again, with the board off the drive.

The negative supply has the following topology:


The SMOOTH motor controller sends pulses to the MOSFET causing it be switched on and off at 1MHz frequencies. Energy is transferred to the inductor when the MOSFET is on. When the MOSFET is switched off, the inductor's energy is transferred to the capacitor via the diode. The SMOOTH chip senses Vout and varies the width of the pulses to maintain Vout at the desired level. This is called Pulse Width Modulation (PWM).

WD's circuit has an additional resistance in the ground leg of the inductor. I'm not sure why this is so, but it could be a current sense resistor. Since this resistor is open circuit, and because it shows visible burn marks, it would appear that too much current has passed through the inductor. An inductor has a very low DC resistance (less than 1 ohm), but its impedance at 1MHz frequencies would be high. If the SMOOTH IC were to fail such that its PWM controller were always switched on, this would send a DC current into the inductor, and this could then burn out the resistor. Similarly, if the MOSFET were to fail with a Drain-to-Source short circuit, this would also burn out the resistor. A third failure mode could involve shorted turns in the inductor (something which cannot be detected via a resistance measurement).

The MOSFET pinout would be something like this:

_________________
A backup a day keeps DR away.

It appears that the preamp resistances are OK.

Just FYI, here are some block diagrams of old preamps:

http://www.hddoracle.com/viewtopic.php?f=59&t=229

This one shows the three supply pins (Vcc = +5V, Ground, Vee = -5V):

http://www.hddoracle.com/download/file.php?id=222&mode=view

Notice that the negative supply is used by the write drivers.

_________________
A backup a day keeps DR away.

Fzabkar, thank you for the info.

Now I have measured quite a lot of voltages on the PCB (above all, in the problematic area).

(The black probe in a screw hole and the red probe on the element to be measured)

Next, there is a photo where have I marked the different voltages I have measured.

They are in picture #9.

I have noticed the following:

For C13 and C18. I have measured 0V in both ends of the capacitors. It is strange.

On U1, the voltages were fastly changing all the time and I have measured this:

Sorted from left to right:

The one to the left: 11.53V,
Next to the right: it varies from 2.0V to 1.62V
Next to the right: it varies from 2.47V to 1.10V

Please take a look at picture #9

Thank you

What about D2 fzabkar? For when we try measure the D2 resistance to see if it is shorted on two ways.
If I read right L7250_Negative_Regulator.jpg, and D2 is the Schottky diode "D", so I conclude we can measure it on circuit. Yes?

Oops, or does the signal make the turn through the bottom line and through the SMOOTH?

Oops, or does the MOSFET Q2 is maybe blocking the signal from making the turn? (I do not know this MOSFET logo too, sounds to me that it will maybe pass by the 5k resistance way)

I do not understand those quad too, like "N_FEED", and not well familiar with the triangles that I understand as magic box that contain many electronics components.

Also if you have a good link of documentations about strange logos, like those others than the basics R C D T.

Yes, you should be able to measure D2 in-circuit with your multimeter. It should test like a Schottky diode with a low forward voltage drop.

"N_FEED" is the feedback voltage that enables the SMOOTH IC to sense and control Vout. If the feedback voltage is too low, then the SMOOTH IC widens the pulse. This sends more energy into the inductor. Conversely, if the feedback voltage is too high, then the SMOOTH IC shortens the pulse.

This is the datasheet for the L7250 SMOOTH chip:

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

Current motor controllers have some similarities to this chip, but they also have additional features.

As for the voltage measurements, I don't see any problems other than the negative preamp supply.

_________________
A backup a day keeps DR away.

So as I understand about D2, the current sent by the the multimeter is finally well doing the turn (loop, or maybe I must say detour), as it is blocked by the diode in one way, and so it pass through the SMOOTH and all others, and that's why the resistance grow to a big 80k (or infinite if it cannot make the turn). I'm right?

If D2 was short, the current sent by the the multimeter will take the most short road and just pass through the diode and so get the same resistance in the two ways, about 100 ohm. I'm right?

If I'm previously right, how can you know D1 is not shorted without unsoldering it, where the situation is ambiguous because the detour it done seems having a low resistance and is about identical as if it was shorted?

When you measure the resistance of D1, the meter current flows from the cathode of D1, through the two adjacent inductors, then R45, MCU core, MCU ground, and then returns via the anode of D1 (which is grounded). So you are actually measuring the resistance of all these components in series. R45 is less than 1 ohm, as are the two inductors, so the bulk of the resistance is due to the MCU core, which is around 90 ohms.

_________________
A backup a day keeps DR away.

Yes I get that, all components resistances are added.
What I say is if a shorted D1 give also 90 ohms?

A shorted component would read 0 ohms. A "leaky" diode might read 90 ohms in both directions.

_________________
A backup a day keeps DR away.

Okay! it's finally for I right understand when we can be sure 100% of the diode state without unsoldering it.
It's also to understand for in another situation, like imagine if the "detour" measure 2 ohm instead 90.

And my speak is not to force material32 unsoldering D1

Very thanks for your lessons fzabkar