Question Intel SSD 320 series fried diode (?) CR3, one contact, should I remove it?

[kornsoud]

Hi! Could you please help me with my SSD? I plugged 6-pin PCIE to 8-pin CPU, I also plugged 24-pin to 24-pin and 8-pin CPU to 8-pin CPU (my motherboard has two 8-pin CPU power slots) and pressed POWER button, I don't remember what happened next. SSD died after that, undetectable in bios. I see one fried component (A2 P0d) inside. Also it looks like the component have only one contact connected, but I'm not sure, it's so small.

Could you please take a look at the photos and give me any advice? ANYTHING WOULD BE HELPFUL!

Spoiler: Photos

Spoiler: Zoomed In

 

[kornsoud]

In your earlier photos you show "0.6 (200)" for the inductor adjacent to 1V8, and "0.2 (200)" for the inductor adjacent to 3V3. Now you are showing different numbers. :-?

Different points, I marked the points where I put the red probe. It's always GND for the black probe.

 

[kornsoud]

In your earlier photos you show "0.6 (200)" for the inductor adjacent to 1V8, and "0.2 (200)" for the inductor adjacent to 3V3. Now you are showing different numbers. :-?

The photos once again:

Spoiler: Photos

 

[kornsoud]

52 ohms is too low for 3V3. We must identify this IC's part markings. Are you sure you can't see them if you observe them under a strong light?

It's just empty as inductors, I don't know why.

 

[fzabkar]

Maybe the heat has rendered the markings unreadable??

As for the HDD, the preamp requires professional data recovery. It's definitely not a DIY proposition. You are so unlucky. In fact, the failure could be an unrelated fluke. :-(

This is what happens to the preamp in catastrophic cases:

https://www.hddoracle.com/download/file.php?id=9531&mode=view

 

[fzabkar]

I still think you need to remove both ICs. Forget the 1V8 and 3V3 test point markings. They are confusing me. Just use the inductors themselves.

Anyway, I need to find a good photo that shows the markings on the 3V3 IC, otherwise we can't move on. I'll come back to you tomorrow.

 

[kornsoud]

Maybe the heat has rendered the markings unreadable??

As for the HDD, the preamp requires professional data recovery. It's definitely not a DIY proposition. You are so unlucky. In fact, the failure could be an unrelated fluke. :-(

This is what happens to the preamp in catastrophic cases:

https://www.hddoracle.com/download/file.php?id=9531&mode=view

Nice photo, thank you! Unfortunately, I live in a small city in a poor country. So far I have seen only dumb repairmans as you see and I don't think I will find a good one. Do you think it's possible to replace the preamp by myself? As far as I understand, in case of HDD, the PCB is okay, right? So now I need to replace the preamp or the whole module, and I need a donor, right? Do I need something else to do that? Do I need expensive gear? I'm really considering doing it by myself.

About the SSD, to sum up it, about the resistances, here is a new all-in-one image with all the measurements I did on that side:

Spoiler: Photo

 

[fzabkar]

Forget the preamp. Believe me, you have almost no chance of pulling this off. We are assuming that there is no platter damage. If there is, then replacing the heads would be pointless, as the new heads will crash as soon as they fly over the bad media.

I'm a bit concerned about your PSU. You said that your repair guy tested the diodes, yet you found that one was shorted. Either he was wrong, or your setup killed the diode after coming back from the workshop. Now your drive is clicking. Who is right and who is wrong? I'm very confused.

 

[kornsoud]

Forget the preamp. Believe me, you have almost no chance of pulling this off. We are assuming that there is no platter damage. If there is, then replacing the heads would be pointless, as the new heads will crash as soon as they fly over the bad media.

I'm a bit concerned about your PSU. You said that your repair guy tested the diodes, yet you found that one was shorted. Either he was wrong, or your setup killed the diode after coming back from the workshop. Now your drive is clicking. Who is right and who is wrong? I'm very confused.

What about the resistor the repairman replaced? Maybe he installed the wrong one. So I'm <Mod Edit> I guess. I don't have another way to replace the preamp. I'll try to find a competent repairman but that's unlikely to happen. Could you please give me more info about the preamp replacement? Do I need a donor like the same exact model? If I need a donor I need to start looking for the donor right now.

 

[kornsoud]

NEWS! I connected the HDD to a power cable (PC was ON), it SPINNED. I restarted and connected a SATA cable. I copied the most valuable text file (35 KB) and the HDD died again. Is there any low power safe mode or something (maybe some dos program) to not put excessive pressure on HDD, to make it work and copy all the files (slow but reliable)?

Or is there another reason for the HDD to die and ressurect on the next day?

Wrong resistor maybe? It now has 10.9 (200) resistance (since removing the diode). Photo:

Spoiler: HDD photo

Could you please help me to test the PCB of HDD with multimeter again to make sure everything is good on PCB side? Photos of PCB:

Spoiler: HDD_PCB

 

[kornsoud]

With isopropanol I was able to see the markings on the U33 element of the SSD with my eyes: It's WEA 045.

U30=QKG 039
U33=WEA 045

Spoiler: U30+U33

I'm sorry for the potato camera, I can't do high quality photos with it

I tried a "heat test". I connected the SSD to the PSU. U33 got burning hot, this is why the markings was unreadable. Is that a good sign?

Maybe the heat has rendered the markings unreadable??

Yes! But now it's readable.

 

[kornsoud]

I'm a bit concerned about your PSU. You said that your repair guy tested the diodes, yet you found that one was shorted. Either he was wrong, or your setup killed the diode after coming back from the workshop. Now your drive is clicking. Who is right and who is wrong? I'm very confused.

I'm 99% sure my new PSU is good. How to check it? I know about a BLACK PROBE to the black wire and a RED PROBE to the other ones. But sata power cable is different, I don't know how to check the voltage and don't blow up the multimeter or the PSU, contacts are so close to each other.

 

[fzabkar]

MIC23153YMT, MICREL, buck regulator, 4MHz, 2A, 2.7V - 5.5Vin, marking WEA, MLF-10:

https://ww1.microchip.com/downloads/en/devicedoc/mic23153.pdf

 

[kornsoud]

I found a donor for my SSD, but FW is different and some numbers are different, will it work for me? Photos:

Spoiler: My SSD FW 0302

Spoiler: Donor SSD FW 0362

 

[kornsoud]

I tried a "heat test". I connected the SSD to the PSU. U33 got burning hot, this is why the markings was unreadable. Is that a good sign?

I made a mistake. The capacitor C241 (?) is getting burning hot, not U33, resistance = 60 (200), photo:

Spoiler: Hot capacitor C241

 

[kornsoud]

I measured resistances of all the capacitors, there were many shorted ones 0.2-0.6 (200), I marked them with a pink dot.

Spoiler: Shorted capacitors

 

[fzabkar]

This thread has gone down a rabbit-hole. :-(

If you send your SSD to a professional data recovery company, they will probably transfer your NANDs (and possibly controller) to a donor SSD. That would be very expensive, probably US$2K or more. However, we still don't know if any of these parts have failed. A failed NAND would probably mean that no data could be recovered.

The alternative is to try to repair your PCB. The hot capacitor (60 ohms) is an unusual failure in this kind of situation. These capacitors are usually rated for a working voltage of 10V or 16V. A 12V overvoltage shouldn't normally kill them.

Those other capacitors which measure 0.2 - 0.6 ohms are probably OK. They are connected to one of more of the voltage rails. This means that a bad DC-DC converter IC, or a bad NAND or flash controller, could be making these capacitors look bad.

I would still remove those two ICs and retest the board for short circuits. You can obtain these ICs from places like Digikey, Mouser, Farnell, RS Components. In general I would avoid Chinese sources as there are too many fakes. The only Chinese source I trust is lcsc.com.

 

[USAFRet]

As was asked long ago, what is the end goal here?

To retrieve what data may have been on it?
Return this particular storage device to actual working condition?
Both?

 

[kornsoud]

The alternative is to try to repair your PCB. The hot capacitor (60 ohms) is an unusual failure in this kind of situation. These capacitors are usually rated for a working voltage of 10V or 16V. A 12V overvoltage shouldn't normally kill them.

Those other capacitors which measure 0.2 - 0.6 ohms are probably OK. They are connected to one of more of the voltage rails. This means that a bad DC-DC converter IC, or a bad NAND or flash controller, could be making these capacitors look bad.

I would still remove those two ICs and retest the board for short circuits.

Should I remove the ICs first and measure resistances or should I put there new ICs from the donor immediately and measure then?

 

[fzabkar]

Remove them and measure the resistances. That will tell us whether the ICs were responsible for the short, or whether the problem is somewhere else. You really need a hot air station to do this work, otherwise you will damage the PCB.

 

[kornsoud]

As was asked long ago, what is the end goal here?
To retrieve what data may have been on it?

To retrieve the data. And repairmen in my country are super dumb so I need to understand something for them to not destroy my data. I need to supervise every step to be safe.

 

[kornsoud]

Remove them and measure the resistances. That will tell us whether the ICs were responsible for the short, or whether the problem is somewhere else. You really need a hot air station to do this work, otherwise you will damage the PCB.

Got it! I will remove U33 and U30 soon and remeasure what I measured before. This is about the SSD.

Also about the HDD I'm still waiting for your professional opinion. TLDR:

1. Removed the diode
2. It worked for 2 hours (36% of data has been copied)
3. Died
4. Next day I connected it, it spinned, I copied a 35KB text file
5. Died again
6. Haven't touched it since then

I'm looking for a safest way to copy all the data left on it.

 

[fzabkar]

The only DIY opportunity for your HDD would be to clone it with HDDSuperClone. It can handle a "difficult" HDD, and it maintains a log file so that it can resume after an interruption.

https://www.reddit.com/r/datarecoverysoftware/wiki/hddsuperclone_guide/

A SMART report can check the health of the drive:

https://www.reddit.com/r/datarecoverysoftware/wiki/index/smart/

 

[kornsoud]

The only DIY opportunity for your HDD would be to clone it with HDDSuperClone.

So cloning puts less pressure on HDD?

 

[fzabkar]

This is a rough block diagram of your SSD:

            DC-DC
           converter IC                  .-------------------.
              __                         |                   |
           -o|  |o-     inductor         |                   |
+5Vin ------o|  |o---------UUU----+------|   NAND flash      |
           -o|  |o- Vout          |      |                   |
           -o|__|o-              ---     |                   |
              |                  ---     | flash controller  |
             ===        capacitor |      |                   |
             GND                  |      |   SDRAM           |
                                 ===     |                   |
                                 GND     '-------------------'
                                                  |
                                                 ===
                                                 GND

 

[fzabkar]

Cloning under Linux puts less pressure on the SSD. HDDSuperClone will only try a "weak" sector once before moving on to the next one. Windows, on the other hand, will thrash a bad sector for several seconds, or even minutes, before giving up. HDDSuperClone is a multi-pass cloner. It will attempt the more difficult sectors during its final pass.

The author of HDDSuperClone has created a Live CD and a GUI to make it easy for Linux novices.