SAMSUNG 860 PRO Suddenly Died

[Pandemicc]

Hello everyone, today my issue is with my very personal SSD, yes it finally happened to me as well

Samsung 860 Pro P/N: MZ7KH512HAJQ MODEL: MZ-76P512 2017.12

Today I was testing a brand new SSD that I bought for a customer to see if it has any defects out of the box. I hot plugged the SSD, as I did hundreds of times before, tested the drive and ejected. It was at that very nano second, my 512gb 860 Pro died. It was my boot drive. PC locked itself, restarted, and now the drive won't even show up on any PC I plugged into. Now I have three set of questions for you guys.

First: Does any of you know that samsung RMA recover data? if that is the case, I will send my drive to them, I will not break the seal

Second: I will break the seal, take photos of the PCB and do some resistance check, if it is not anything related to PCB I will take it to a pro.

Lastly: Do you think that this is just random, a merely unfortunate event? Do you think that what happened here is somewhat related to my PSU? Nothing happened to the SSD I tested and also one of my HDD was also plugged, nothing happened to that too

Thanks in advance people!

[Pandemicc]

By the way, I was always checking SMART every week, like I do with all of my drives, it was always in pristine condition.

[westcoast]

usually wont show up in BIOS. try booting linux and see log, often times they are detected there

[WebClaw]

See if it draws anything; and check the efuse. You can also inspect w/a thermal camera to isolate hot spots.

[Pandemicc]

By the way, I was always checking SMART every week, like I do with all of my drives, it was always in pristine condition.

it won't show up in linux as well

[Pandemicc]

See if it draws anything; and check the efuse. You can also inspect w/a thermal camera to isolate hot spots.

the drive definitely gets hot when plugged but I need more insights before I break the seal

[Pandemicc]

I finally decided to break the seal and took the photos of the PCB. The red circled areas gets really hot really quick when I plug it to SATA power, orange areas are somewhat hot.

Attachments

[Lardman]

It's the e-fuse, bridge vcc in and out for recovery or replace with another IC .

[pepe]

there might be some shorts, coz things get hot nearby...
So better find out why it heats up first.

[Pandemicc]

It's the e-fuse, bridge vcc in and out for recovery or replace with another IC .

thanks a bunch for your reply! Im thinking about removing that ten-legged IC and see if anything still gets hot. Would that be right step to take?

[Lardman]

Start with the IC circled in red on your final picture, remove it (keep it safe) and retest the VCC rail for shorts. It only hits a couple of caps and then disappears into the pmic, I expect the heat you see elsewhere is just residual.

[Pandemicc]

Start with the IC circled in red on your final picture, remove it (keep it safe) and retest the VCC rail for shorts. It only hits a couple of caps and then disappears into the pmic, I expect the heat you see elsewhere is just residual.

I removed the IC and found that fzabkar shared its datasheet: http://www.st.com/web/en/resource/techn ... 095426.pdf

but I am kind of lost about "vcc rail" is it 5v SATA power?

[Lardman]

but I am kind of lost about "vcc rail" is it 5v SATA power?

You can see Vin and Vout in the manual. Check Vin and and Vout with your meter in resistance mode and one probe on ground - if either are 0 that's bad. Once you have done that checked those check the resistance to ground of each of the rails from the pmic, francs probably already posted his test points but each of the inductors will do anything 0 ohm is not good.

[Pandemicc]

but I am kind of lost about "vcc rail" is it 5v SATA power?

You can see Vin and Vout in the manual. Check Vin and and Vout with your meter in resistance mode and one probe on ground - if either are 0 that's bad. Once you have done that checked those check the resistance to ground of each of the rails from the pmic, francs probably already posted his test points but each of the inductors will do anything 0 ohm is not good.

those three beeps to ground on resistance mode. The rest of the legs reads OL.

and I couldn't find anything about franc's test point guide. I appreciate any direction you can give me

Attachments

Untitled.png (27.99 KiB) Viewed 533 times

[Lardman]

If you're beeping you're probably in diode mode, if this is a client device you really need a little more experience before offering to work on them.

You should be checking the PCB not the IC iteself. All 3 Vin (1,2,3) pads are joined so you should have continuity between the 3 of them, if the last one of them beeps when checked against ground all of them should you likely had the probe misaligned on pad 3. You should not have continuity between any Vin and ground (pad5 / screwhole). Likewise you should not have continuity between Vout and ground (10,9,8) and ground (pad 5 / screwhole).

Check and confirm.

[Pandemicc]

If you're beeping you're probably in diode mode, if this is a client device you really need a little more experience before offering to work on them.

You should be checking the PCB not the IC iteself. All 3 Vin (1,2,3) pads are joined so you should have continuity between the 3 of them, if the last one of them beeps when checked against ground all of them should you likely had the probe misaligned on pad 3. You should not have continuity between any Vin and ground (pad5 / screwhole). Likewise you should not have continuity between Vout and ground (10,9,8) and ground (pad 5 / screwhole).

Check and confirm.

This is my own SSD I would have said so if it was otherwise

Thanks for the clarification. Im definitely not in diode. Silly me I was checking the IC itself. All Vout shows resistance but nothing on Vins, yes all Vins are have continuity with each other.

[Lardman]

If you're showing a short on the IC itself it sounds very much like you have found the problem. I am concerned you are getting a direct short to ground on Vin though, there's not exactly a lot between the sata power connector and the fuse to go short, just 1 cap.

Check for a short between ground and the 5v rail at the sata connector itself to be sure. After that there are only 2 resistors and a cap on the back of the pcb before the fuse.

Attachments

[Pandemicc]

If you're showing a short on the IC itself it sounds very much like you have found the problem. I am concerned you are getting a direct short to ground on Vin though, there's not exactly a lot between the sata power connector and the fuse to go short, just 1 cap.

Check for a short between ground and the 5v rail at the sata connector itself to be sure. After that there are only 2 resistors and a cap on the back of the pcb before the fuse.

Attachments

[Lardman]

They're not fuses, just resistors. Your measurements aren't making much sense. If you have no short to ground on either side of them you should not have a short on any Vin pad. You have removed the efuse from the pcb haven't you? Check the pads are clear and not bridged and double check pads 1,2,3 for continuity to ground.

The cap : Find the plastic pin in your red circle, move up to the 4 empty horizontal pads , move left to the tiny vertical cap. Should be ground on 1 side only - the bottom pad is on the same plane as the bottom pads of the 2 resistors so I can't see how it would be anything other.

[Pandemicc]

They're not fuses, just resistors. Your measurements aren't making much sense. If you have no short to ground on either side of them you should not have a short on any Vin pad. You have removed the efuse from the pcb haven't you? Check the pads are clear and not bridged and double check pads 1,2,3 for continuity to ground.

The cap : Find the plastic pin in your red circle, move up to the 4 empty horizontal pads , move left to the tiny vertical cap. Should be ground on 1 side only - the bottom pad is on the same plane as the bottom pads of the 2 resistors so I can't see how it would be anything other.

Attachments