Question Random PC "Death"

[cuddlyShambler]

Hi everyone

I ran into a pretty bad problem with my PC. I cannot even get it to POST anymore.
A few days ago i was just using it normally and all of the sudden, the PC cuts its own power and since then i cannot get to start again.

The system is bit old yes, but i took good care of it over the years, cleaned it on a regular basis,
never overclocked it and always paid attention that CPU and GPU temps don't exceed temps above 70°C.
It ran rock solid for over 8 years now. Also in all the years i'm using debian-based linux-distros the OS never crashed on me once, so you can probably understand as to why this is bit of a headscratcher to me.

The only sign of live i get from the PC right now is a, for cosmetic purposes only, red LED on the motherboard and that's about it.

Further down, in SECTION 3, i will describe my actual question, so feel free to skip the other ones if you don't need further context.


[SECTION 1, System-Specs]

CPU: i7 6700k
CPU-Cooling: Air-cooling (scythe mugen)
GPU: Geforce GTX 980ti
Motherboard: ASUS Maximus Ranger VIII
RAM: Corsair Vengance LPX 32 Gigs
PSU: Coolermaster V1200

[SECTION 2, Inspections/Troubleshoot]

After the crash i inspected the inside of the case thouroughly.

I couldn't find any:

- leaking caps
- leaking batteries
- corrosion
- moisture
- dislodged cables
- dirt/insects

I took a look a the powersupply (without dissambling it) and couldn't find anything
out of the ordinary. I saw the usual white goop around the heatsinks and copper-coils.
i poked it with some non conductive material to check if it was solid. It was indeed
solid, so i figure it is safe to assume that it is the usual epoxy/glue you find in all
psu. At the time of the crash, there was no suspicious smell of burnt electronics in
the air. I know this doesn't mean much, since for example, caps can fail without
smelling horribly burnt, but at the very least the psu didn't fail catastrophically.

Next i had suspected the power-on circuit/button (on the outside of the case) somehow became bad. I removed its
cables from the motherboard. Then i tried to turn it on, with the start button, which is
soldered right onto the motherboard. The motherboard has a onboard POST-code indicator, so
my hope was to get the motherboard at least in a state, where it shows me a code. It didn't.

Going from there i can report the following: The LEDs inside the motherboard
turn off for a few seconds, then nothing happens at all, then they come back on and nothing
else happens. The fans don't spin up and there is no onboard POST-Code at all.

[SECTION 3, Going forward]

My next hunch would be that maybe the cpu-fan-connector has corroded and/or the cpu-fan has gone bad.
I heard that some motherboards refuse to turn on if they don't get a response from
the cpu fan. Not sure about this, but i think it's a good place to start(?).

Now to test this theory, i need to clear a few questions/problems first.

If just the fan-motor has gone bad i don't see any problem of just removing
and replacing it, for testing purposes, with another fan. If the system boots i
will either get a proper replacement fan, if not available, i'll replace the
whole heatsink, since it's about time to replace the thermal grease anyway.

But if corrosion on the connector is the root cause, then i'm unsure on how
to go about this. The connector is in a really awkward spot. Yes i could
remove and reattach it multiple times, in the hopes of loosening some of
oxides and thus make the connector conduct properly again, but since this connector
is in such a awkward spot, there is no chance that i can assure that none of
the oxides get onto surrounding components.

It is therefore my concern, that i might "sprinkle" potentially conductive material upon
spots on the motherboard, where i really should not. Maybe i am overly cautious
here, but i lack the knowledge here to make a properly informed decision.
So if you could share your thoughts about this matter it would help a lot.

I also would love to hear you suggestions on how i should go further in the
process of troubleshooting this machine, because without any postcodes, i find
it really hard to get to the bottom of this. Maybe there is something else
i can do, to get to the bottom of this problem.

Thanks in advance

 

[Ralston18]

How old is that PSU? History of heavy use for gaming or video editing?

Disk drive(s): make, model, capacity, how full?

= = = =

This motherboard?

https://dlcdnets.asus.com/pub/ASUS/...I_RANGER_UM_WEB.pdf?model=MAXIMUS VIII RANGER

Verify that I found the correct User Manual.

Read through the User Manual first and make notes about anything that seems to relate to the problem(s) the system is exhibiting.

No Q_Code LEDs per Page 1-26?

Check the on-board buttons per Page 1-22.

= = = =

Try clearing CMOS and reinstalling a new CMOS battery per the User Manual's instructions.

Is it possible for you to obtain another known working PSU to swap in for testing purposes? Remember to use only the cables that come with the test PSU.

Is all important data backed up?

 

[cuddlyShambler]

How old is that PSU? History of heavy use for gaming or video editing?

Disk drive(s): make, model, capacity, how full?

= = = =

This motherboard?

https://dlcdnets.asus.com/pub/ASUS/mb/LGA1151/MAXIMUS_VIII_RANGER/E10485_MAXIMUS_VIII_RANGER_UM_WEB.pdf?model=MAXIMUS VIII RANGER

Verify that I found the correct User Manual.

Read through the User Manual first and make notes about anything that seems to relate to the problem(s) the system is exhibiting.

No Q_Code LEDs per Page 1-26?

Check the on-board buttons per Page 1-22.

= = = =

Try clearing CMOS and reinstalling a new CMOS battery per the User Manual's instructions.

Is it possible for you to obtain another known working PSU to swap in for testing purposes? Remember to use only the cables that come with the test PSU.

Is all important data backed up?

The PSU is also 8 years old. It has a 10 year warranty, hopefully i don't have to make use of it 😉

I didn't stress the PC too much, i am more of a oldschool gamer, the most intense game that was ever installed on it, was the witcher 3. I also didn't use the PC for intense workstation type loads. I have a separate machine built for that purpose.

As for the disk drives i have 1 Samsung SSD 860 EVO 1 Terrabyte (2 Years old), filled up to about 50% and 1 Samsong 860 Pro (About 6 years old), filled up to about 30%. The EVO runs linux, the PRO runs Windows 10. There are also two backub HDD (each 2TB) for each operating systems, which i already removed, to prevent accidental damage to them, while working on the PC. So my data is safe at least .

Yes you got the right motherboard and yes no Q_code LED as per Page 1-26.
CMOS Battery is fresh, i replaced it this year. I also checked the manufacturing date of the Battery, it was produced this year, so aside from really bad luck, it should be good.

About obtaining another PSU: I should be able to test a replacement PSU, but it could take a week until it arrives at my home. I brought the PC to the house of a friend, because he has a workbench which should make working on the PC a bit easier, than in my small apartment. I will first try to reset the CMOS then, if that leads nowhere I think i can start working on replacing the PSU, in the beginning of the next week. My friend is also an electrician, so he should have equipment handy, which would allow me to check the voltage rails of the current PSU. Maybe then some obvious fault will pop up.

Any opinions about my concerns about possible corrosion?

 

[Ralston18]

8 years is a long time for PSU.

You do not necessarily need to stress a PSU per se.

Like many other products the PSU likely has a built in EOL (End of Life) and will begin to falter and fail as that time nears or goes past.

And a PSU can be tested to some extent.

FYI:

https://www.lifewire.com/how-to-manually-test-a-power-supply-with-a-multimeter-2626158

Not a full test per se because the PSU is not under load. However any voltages out of tolerance would make the PSU a likely culprit.

 

[cuddlyShambler]

8 years is a long time for PSU.

You do not necessarily need to stress a PSU per se.

Like many other products the PSU likely has a built in EOL (End of Life) and will begin to falter and fail as that time nears or goes past.

And a PSU can be tested to some extent.

FYI:

https://www.lifewire.com/how-to-manually-test-a-power-supply-with-a-multimeter-2626158

Not a full test per se because the PSU is not under load. However any voltages out of tolerance would make the PSU a likely culprit.

Ok so i was able to do some more troubleshooting this week. First i reset the CMOS, which did nothing. Then we began to test the PSU and i appeared to be completely dead. Well apart from that one time, where it was putting out completely wrong voltages. Like 8 volts instead of 12 volts and so on. A few days later the replacement PSU arrived, so we swapped it out and lo and behold the PC started normally again.

In the past two days i ran a few software based tests and was using the PC normally and so far it appears that the PC is working normally, so luckily no other parts were damaged during the initial fail.

I looked up my receipt and saw that the PSU died right after the warranty ran out so... Well it's a bit of shame really, because usually i see PSU's live much longer than this, but then again i'm happy nothing else has gone bad.

So thank you for your help with my problem, i wish you well

 

[Ralston18]

👍