Problem with a HDD PCB

I have measured the resistance of inductor L2.

And the readings tell me that it is 0.4 ohms. (And since the multimeter resistance is 0.5 ohms. Therefore I come to the conclusion that inductor L2 is shorted. However there is a voltage of 3.20V at both ends of inductor L2).

I have taken several measurements to inductor L2 in order for me to be totally sure. And I get the same results.

Another thing I have noticed is that capacitor C39 has the same resistance in both directions. I get a reading of 4.42 k-ohms in one direction. And in the reverse direction I get a reading of 4.40 k-ohms. (Of course I have taken several measurements to capacitor C39 and I get very similar readings and they are almost the same in both directions).

Does this mean that capacitor C39 is damaged?

By the way, I want to thank Diybit for his pictures. They have proven to be useful for me. Because I have compared his measurements with mine. And this is the way how I have noticed these 2 problems with my PCB.

The capacitor is OK.

The normal reading for a good inductor is less than 1 ohm. An inductor with shorted turns would also register less than 1 ohm. Therefore your multimeter reading is inconclusive.

Since my resistance multimeter is 0.5 ohms. Therefore, the real resistance of inductor L2 is practically 0 ohms.

Fzabkar, if this is inconclusive. Then, how can I know for sure whether inductor is in good condition or not?

material32 wrote:Since my resistance multimeter is 0.5 ohms. Therefore, the real resistance of inductor L2 is practically 0 ohms.

Fzabkar, if this is inconclusive. Then, how can I know for sure whether inductor is in good condition or not?

You need to measure the inductance of the inductor. Some multimeters have such a feature.

Would be this tool useful for testing inductor?
https://www.ebay.com/itm/184230177899

I don't have it, but it would be interesting to try.

I have exactly the same ring tester, but in a different box. I only use it for testing large coils and transformers. It doesn't provide an inductance reading, but simply tests for shorted turns. An inductor that has shorted turns won't "ring" when it is excited by a tuned circuit.

I have a multimeter with an inductance and capacitance test function for smaller components. However, its range doesn't go low enough for this particular case.

This Samsung example is a similar circuit:

https://forum.hddguru.com/viewtopic.php?f=1&t=31014

In this case there is no resistor in the ground leg of the inductor, so the MOSFET burns up. I have seen several similar threads. In each case, after the MOSFET is replaced, it just burns up again. I don't know the root cause of the problem, but I suspect that the coil is bad, or the motor controller is outputting DC.

Before spending more money to try to repair the HDD, I have taken more measurements to the preamplifier chip of the heads.

I have made a test to all the HDA pins in order for me to be totally sure whether or not the preamp is damaged.

The following are the results that I have gotten when measuring the resistance of the preamp.

It seems that there is a problem with HDA pin #9 and also with HDA pin #15.

Here are all of my measurements in detail:

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

I place the tip of the black probe in HDA pin #1 and the tip of the red probe in the other different pins of HDA (of course, individually).

HDA pins #1 and #2: 11.90 k-ohms (ie. 11,900 ohms)

HDA pins #1 and #3: 23.1 k-ohms (ie. 23,100 ohms).

[Note: For the just mentioned measurement I have selected 200 k-ohms, instead of 20 k-ohms as usually].

HDA pins #1 and #4: 6.79 k-ohms (ie. 6,790 ohms)

HDA pins #1 and #5: 11.22 k-ohms (ie. 11,220 ohms)

HDA pins #1 and #6: 12.59 k-ohms (ie. 12,590 ohms)

HDA pins #1 and #7: 11.25 k-ohms (ie. 11,250 ohms)

HDA pins #1 and #8: 12.59 k-ohms (ie. 12,590 ohms)

HDA pins #1 and #9: 0.5 ohms

[Note: It is necessary to take into account that the multimeter resistance is 0.5 ohms. Therefore the real resistance between HDA pin #1 and pin #9 is 0 ohms].

HDA pins #1 and #10: 12.04 k-ohms (ie. 12,040 ohms)

HDA pins #1 and #11: 11.80 k-ohms (ie. 11,800 ohms)

HDA pins #1 and #12: 12.60 k-ohms (ie. 12,600 ohms)

HDA pins #1 and #13: 11.80 k-ohms (ie. 11,800 ohms)

HDA pins #1 and #14: 12.03 k-ohms (ie. 12,030 ohms)

HDA pins #1 and #15: 0.5 ohms

[Note: It is necessary to take into account that the multimeter resistance is 0.5 ohms. Therefore the real resistance between HDA pin #1 and pin #15 is 0 ohms].

HDA pins #1 and #16: 12.00 k-ohms (ie. 12,000 ohms)

HDA pins #1 and #17: even when selecting the highest scale of resistance (i.e. 2000 k-ohms). It gives no reading or "OL".

HDA pins #1 and #18: 10.82 k-ohms (ie. 10,820 ohms)

HDA pins #1 and #19: even when selecting the highest scale of resistance (i.e. 2000 k-ohms). It gives no reading or "OL".

HDA pins #1 and #20: 12.01 k-ohms (ie. 12,010 ohms)

The problematic HDA pins are number 9 and number 15. Because both of them give a reading of 0.5 ohms.

Are any of them shorted?

If so, what is the probably cause of this disfunction?

Again, HDA pins #17 and #19 give a reading of OL (even when selecting the highest scale: i.e. 2000 k-ohms).

Is this normal?

Or does it mean that there is a problem?

I would be grateful if Fzabkar could give me his opinion about all the above measurements.

@material32, you are "over-thinking" this problem. The only reason I asked you to measure the preamp resistances was to check if there was any obvious damage resulting from the failure of the negative supply. To this end we only really needed to measure the supply resistances.

In any case, pins 5 & 7, and 11 & 13, are the read and write channel differential pairs. Since they are balanced, the resistances of each member of each pair should be the same, which in fact they are.

Pins 17 and 19 are the voice coil pads. There should be an infinite resistance between these pins and ground, and a very low resistance between the two pins, probably of the order of a few ohms.

Pins 9 and 15 appear to be grounded inside the HDA. I don't know any more than that.

Thank you, Fzabkar, for your reply.

Then, you mean that all the resistance values of the preamp chip are correct.

Do you mean that?

I don't see any problems, but I don't know the functions of many of the pins, so I can't be 100% certain.

I haven't success to repair my multimeter... and it's even worst now... I need a new one...
So I cannot compare for you those 19 pins, sorry...

Reply to Diybit:

I am sorry to hear this.

Don't worry about your being unable to take the measurements.

Anyway, thank you for your good intentions.

I have already ordered an HDD with the same model of PCB. And with the same capacity: 500 GB. But it has a different firmware.

The PCB is exactly the following: 2060-701444-004 REV A

While I am waiting, I have tried to do a backup of the ROM which is embedded in the Marvell MCU of my damaged PCB.

I have used the WDR Demo v2.1 in order to this.

Of course, I have carefully followed the instructions given in the HDD Oracle Forum.

Here:

http://www.hddoracle.com/viewtopic.php?f=86&t=1127

http://www.hddoracle.com/viewtopic.php?f=32&t=33

I have placed the damaged PCB (without the HDD) into a SATA power connector of the PSU. And also I have connected the SATA data cable between the SATA port of my motherboard and the SATA port of the PCB. But unfortunately, I have NOT been successful in doing the backup.

I have taken several screenshots in order for you to see the results.

Please, see "Picture #10 - Trying to backup ROM with WDR v2.1.jpg" that I am enclosing with this message.

Again, I have accidentally broken a pin of the jumper plug J6 of the HDD PCB.

And now, when I try to measure the voltages on the four pins of J6. There is no voltages at all. (Of course, prior to this, I connected the PCB with the SATA power connector of the PSU).

I keep the broken part of the pin, but the problem is that it is very small and therefore it is very difficult to solder.

I have thought about glueing both pieces together. But I am afraid that if I do so, there will be no continuity in the fixed pin of J6.

Prior to this accident, there was a voltage in the 4 pins of J6 but only during the first minute. But now, there is no voltage at all in the pins of J6 (not even during the first minute).

In order to backup the ROM embedded in the Marvell MCU:

Is it really a problem that one of the pins of J6 is broken?

Thank you for your replies.

Attachments

: I have tried to backup the ROM embedded in the Marvell MCU with the WDR Demo v2.1.

Give another try with WDMarvel Demo, menu "ROM -> Read".
Looking at the LED and the empty drive ID, it's not sure that you have selected the right port in WDR.
Hoping that the PCB is not fall in a state that block you from reading the ROM...

Reply to Diybit:

Yesterday, I made several attempts and it is not recognized by the WDR Demo.

Therefore the PCB is failing.

Just in case the PCB gets totally damaged, I have not made more intents.

Reply to Fzabkar:

Now, the PCB is clearly failing (still more than previously).

I have taken new measurements to the old PCB. I have measured the voltages after plugging the PCB into the SATA power connector of the PSU.

These are the current problems:

There is no voltage of 3.3 V (neither in transistor Q4 nor in transistor Q2).

Transistor Q4 gives a voltage of 0 V in its two pins (on the left and right) and a reading of 5 V in the middle pin.

Transistor Q2 gives a reading of 5 V in its two upper pins (on the left and right) and a reading of 0 V in its remaining 4 pins. (A few days ago, these mentioned 4 pins were giving a reading of 3.20 V. Now, all of them are 0 V).

There is also no voltage in pins #1 to #3 of the SATA power connector of the PCB. (Although this was already happening before now).

Pins #1 to #3: 0 V (it should be 3.3 V as far as I know).

The voltages from the other pins of the SATA power connector on the PCB are right:

Pins #4 to #6: 0 V (ground)

Pins #7 to #9: 5 V

Pins #10 to #12: 0 V (ground)

Pins #13 to #15: 12 V

I get 5 V in catode of diode D3 and 12 V in catode of diode D4.

Also I have noticed the following differences in comparison with the measurements I took a few days ago:

Transistor Q3 gives a reading of 5 V in its upper two pins (on the left and right). But it gives a reading of 0 V in the remaining 4 pins. (A few days ago, I had gotten a reading of 1.35 V at those 4 pins. But now, all of them read 0 V).

The voltage before and after resistor R45 is 0 V (a few days ago I had gotten readings from 1.20 to 1.30 V). But now, I get 0 V.

[Note: I have also measured the resistance of resistor R45 (selecting the scale of 200 ohms) and I got a reading of 0.7 ohms. (By the way, it is necessary to take into account that the multimeter resistance is 0.5 ohms. Therefore the real resistance of R45 is 0.2 ohms)].

Also I get no voltage before and after inductor L2. (A few days ago, I got readings of 3.20 V).

[Note: By the way, I have mesured the resistance of inductor L2 and I got 0.5 ohms and since the the multimeter resistance is also 0.5 ohms. Therefore, I suspect that inductor L2 is shorted].

I also get a voltage of 0 V (both in the anode and catode of diode D2). [Note: A few days ago, I got readings of 3.20 V in catode and anode of diode D2].

When measuring the resistance of diode D2, I get the following results (selecting the scale of 200 k-ohms):

61.8 k-ohms (i.e. 61,800 ohms): red probe to its catode and red probe to its anode.

135.8 ohms in the reverse direction (black probe to its catode and red probe to its anode).

Regarding diode D1, both catode and anode give a reading of 0 V. (On the contrary, a few days ago I got readings of 1.38 V in its catode).

When measuring the resistance of diode D1, I get the following results (selecting the scale of 200 ohms):

56.2 ohms (red probe to its catode and black probe to its anode)

81.1 ohms in the reverse direction (black probe to its catode and red probe to its anode)

Regarding the contacts of the preamp at J1, NOW I get the following voltages:

Point #1: 0 V --------- Point #2: 0 V

Point #3: 0.03 V ------ Point #4: 0.03 V

Point #5: 0.03 V ------ Point #6: 5.05 V

Point #7: 0.01 V ------ Point #8: 0 V

Point #9: 0 V --------- Point #10: 0 V

Point #11: 0.01 V ----- Point #12: 0 V

Point #13: 0.01 V ----- Point #14: 0 V

Point #15: 0 V -------- Point #16: 0 V

Point #17: 11.21 V ---- Point #18: 0 V

Point #19: 0.43 V ----- Point #20: 4.94 V

This can be compared with my previous readings of the preamp connector J1 that were as follows:

Point #1: 0 V --------- Point #2: 0.05 V

Point #3: 3.28 V ------ Point #4: 3.26 V

Point #5: 3.01 V ------ Point #6: 5.09 V

Point #7: 3.01 V ------ Point #8: 2.40 V

Point #9: 0 V --------- Point #10: 0 V

Point #11: 0.10 V ----- Point #12: 0 V

Point #13: 0.10 V ----- Point #14: 0 V

Point #15: 0 V -------- Point #16: 0 V

Point #17: 0 V -------- Point #18: 0 V

Point #19: 0 V -------- Point #20: 3.36 V

As clearly noticed, pins #3, 4, 5, 7, 8, 17 and 20 currently give very diferent voltages than in previous days.

I have also measured the resistors around the MCU (just in case): I have measured exactly resistors R48, R52 and R56. And all of them gives me a reading very close to 0 ohms. Therefore, they seem to be OK.

Again, I have measured the voltages of the 4 pin legs of J6 (i.e, the motor connectors of the PCB).

Because, as stated in my last message but one, I accidentally broke a pin leg of J6.

The base of such a pin is intact but its raising part is broken,

[Note: I am enclosing a picture where the broken pin leg is circled in red. See picture #11].

I got the following results when measuring the pin legs of J6:

There is no voltage at all in any of the pin legs (not even during the first minute just after powering up the PCB).

Currently, 0 V at all the 4 pin legs (all the time).

On the contrary, 10 days ago, I usually got the following voltages:

A) At Bottom:

I get a reading of 6.25 V on the left pin.

And I get a reading of 4.16 V on the right pin.

B) At top:

I get a reading of 7.40 V on the left pin.

And I get a reading of 5.92 V on the right pin.

Finally, a few more questions:

Is it possible that my trying to backup the ROM embedded in the Marvell MCU with the WDR Demo has caused the PCB to fail still more than before?

How can I check (for sure) whether the ROM is still intact?

But if the WDR Demo did not manage to recognize my PCB, then it would be very strange for the ROM to be damaged by this program.

Is this so?

Fzabkar, thanks in advance for your replies to my questions.

Attachments

: The base of the pin is intact but the raising part is broken. I have circled the damaged pin in red.

There appears to be no activity in the SMOOTH motor controller. This IC controls the Vio (Q4) and Vcore (Q3) regulators. If these voltages are absent, then we will not be able to talk to the Marvell MCU (88i6745). I think your DIY adventure is at an end. :-(

It is necessary to take into account that resistor R47 is damaged. I have not replaced it yet because I do not find a store in my area where these components are sold. I would have to purchase it in an online store.

Of course, if necessary, I will do so.

If resistor R47 is shorted, can this cause the above problems (i.e., the absence of such voltages and that the Smooth Controller is inactive)?

Again, the fact that a pin leg of the motor contacts (J6) is broken. Does this have an influence on the activity of the Smooth Controller?

Finally, I managed to retrieve all the data from the damaged HDD. All the files were intact.

Above all, I want to thank Fzabkar for his guidance and efforts.

He rebuilt a ROM file from another HDD which had exactly the same firmware as mine. That is: 005C003G (or 01.01B01).

The process was exactly as follows:

I purchased an HDD very similar to my damaged HDD: it has the same model of PCB (2060-701444-004 REV A). And also the donor HDD has the same capacity (500 GB).

In addition, the firmware of the donor HDD is very close to that of the patient drive.

Donor HDD firmware: 005C003H (or 01.01B02).

Later, I placed the donor PCB into the patient HDD. And I started the HDD within Windows.

Windows told me that the HDD had a size of 0 GB. And therefore I could not access it from Windows,

But I used the WDMarvel Demo and I managed to retrieve about 70% of all the modules from the Service Area.

All this data was sent to Fzabkar. And on the basis of a ROM file from an HDD with exactly the same firmware as the patient drive (i.e., 005C003G or 01.01B01), Fzabkar made some modifications to that ROM file by using the data taken from the SA modules of the patient drive.

Finally, he sent me that rebuilt ROM file.

Later I placed the donor PCB into the donor HDD and I wrote the rebuilt ROM file into the PCB by using the WDR Demo.

After doing that, I placed the donor PCB (already flashed) into the patient HDD. And I started Windows and the HDD was totally recognized by my computer.

Since then, I can already access all of my data with no problems. Fortunately, all the files are intact. This HDD was including all the personal data for the last seven years.

And finally I copied all the files to a new HDD.

In short, the only problem was the PCB. The supply of -5V was missing. The origin of the problem was the Smooth controller which was working badly.

After this experience, I will back up the most important files from time to time. This is the best remedy to all this.

The beginners in data recovery are prone to think that by replacing the PCB HDD, you will solve the problem. But even if this is the case and one of the TVS diodes is burnt. There is also the risk of the preamp being damaged at the same time as the PCB got damaged. This is not usually taken into account by beginners. And in this last case, one could not solve the problem by oneself and he/she would have to go to a specialist in data recovery.

In addition, the newer HDD PCBs are encrypted. And in order to retrieve the SA modules and/or the ROM from a damaged HDD you must previously unlock the drive. And therefore, nowadays, if you want to retrieve the data from a recent HDD that is damaged, it might cost you a lot of money.

In short, back up your most important files from time to time. This is the best remedy.

Once more, I want to thank Fzabkar for his instructions and help.

Many, many thanks to him.

Without his help, I would not have been able to retrieve the files from my damaged drive.

I am also grateful to the HDDGuru and HDD Oracle forums. Because when you have a damaged HDD, both forums are a very valuable source in data recovery to the newbies.

Finally, I want to thank Diybit for his interest and good intentions. He took some measurements to some HDD PCBs just in case that might be useful to me.

Wait a minute... Wait a minute...

You had only 2 problems on your original PCB correct?:

-R47 needed to be replaced (easy, with appropriate soldering skills)
-SMOOTH controller needed to be replaced (easy with appropriate soldering skills)

Why didn't you try to replace these?
That way, you could have preserved the use of your base firmware/SA data without extraction/modding etc...

It seems to me you did an incredible amount of analysis and research here on the issue! Yet in the end, someone else just fixed it for you? (No offense to the experts involved).

Problem with a HDD PCB

Re: Problem with a HDD PCB

Re: Problem with a HDD PCB

Re: Problem with a HDD PCB

Re: Problem with a HDD PCB

Re: Problem with a HDD PCB

Re: Problem with a HDD PCB

Re: Problem with a HDD PCB

Re: Problem with a HDD PCB

Re: Problem with a HDD PCB

Re: Problem with a HDD PCB

Re: Problem with a HDD PCB

Re: Problem with a HDD PCB

Re: Problem with a HDD PCB

Re: Problem with a HDD PCB

Re: Problem with a HDD PCB

Re: Problem with a HDD PCB

Re: Problem with a HDD PCB

Re: Problem with a HDD PCB

Re: Problem with a HDD PCB

Re: Problem with a HDD PCB