Question Partial System Shutdown (Randomly) after adding 4K Monitor ?

[Farboguy]

I recently upgraded one of my two displays to a 4k monitor. I had no issues with my system prior, but probably about a week after the upgrade I had sporadic issues where both monitors suddenly turn grey for several seconds and then the entire system almost completely shuts down. The monitors are black at this point and the peripherals have no power. The only exception is that the gpu lights are still on as though it is normally powered and the tower power light remains on. Otherwise, it appears to be fully shutdown and I have to use the power button to turn the pc off and on again or use the reset button to turn the computer back on.

This happens anywhere between once a week to once a day, but I believe has only happened when running games (during several different games). After this first started I did a clean install for my video card drivers with the app the community recommends, but it still occurs randomly. I've also ran memtest86 to check my RAM (no errors), and I've run furmark to try and tax my system to see if it's maybe a heat issue or some other stress thing, but I couldn't reproduce it. I couldn't get the gpu to go above about 82 C with it. I even left a youtube video playing on the second display while running the furmark stress test since I often have something playing while gaming, but the problem seems hard to reproduce and very sporadic. Here is a google drive folder link that has my system specs and the event viewer files that may be helpful. Since there was no BSOD, I can't find other dump files and the event viewer notifications seem to all have timestamps from after I restart the pc, rather than whatever caused it to crash. Thanks for any help you can provide!

 

[Colif]

What Make/Model power supply do you have?

sounds like a power problem

 

[Farboguy]

What Make/Model power supply do you have?

sounds like a power problem

I have a CORSAIR CX Series CX650 650W ATX12V 80 PLUS BRONZE
https://www.newegg.com/corsair-cx-series-cx650-650w/p/N82E16817139200?Item=N82E16817139200

 

[Colif]

R5 2600
ROG Strix B450-F Gaming BIOS version 2008
PSU CX Series CX650 650W ATX12V 80 PLUS BRONZE

what GPU?

have you tried with just the old monitor to see if it continues. It shouldn't if only change is monitor.

the event viewer notifications seem to all have timestamps from after I restart the pc, rather than whatever caused it to crash

this indicates windows doesn't know why it was shutdown. Its not doing it, this is hardware as windows is just reacting to what hardware is doing.

what monitor is it? is it a high refresh rate 4k screen or just 60? it might be expecting GPU to a faster signal and the balance of power in system has changed.

On my last PC I had a GTX 960 and then was given a 4k screen as a gift and the 960 couldn't supply 4k 60hz on Display port so I upgraded GPU to stop my monitor from displaying the no signal warning

The monitors are black at this point and the peripherals have no power.

this is what makes me think its power related as if PC is low on resources it will shut off GPU and USB just to make sure CPU still gets what it needs.

you might want to check PSU is giving you the power it should be.
not many ways to test PSU
the paper clip method - https://forums.tomshardware.com/threads/what-is-the-paperclip-method-of-testing-a-psu.1336402/

or multimeter, this is best way as can test under load but not everyone has one lying around.

or in the BIOS to check the +3.3V, +5V, and +12V. - https://www.lifewire.com/power-supply-voltage-tolerances-2624583

another way is log your sensors using something like HWINFO but its not always accurate
downlaod hwinfo = https://www.hwinfo.com/download/
when you start it, click sensors only and click run
in next menu, click the button next to the clock in bottom toolbar, it says "logging start..." if you hover mouse over it.
this open file explorer and lets you create a log file.
you will need to make a new one each day until crash occurs.
it creates a csv file that can be read by excel or google docs

 

[Farboguy]

The GPU is an ASUS AREZ Strix Radeon RX Vega 56:
https://www.newegg.com/asus-radeon-...gaming/p/N82E16814126180?Item=N82E16814126180

and if I'm understanding the monitor specs correctly, its a 60hz:
https://www.amazon.com/dp/B074PK4R2H/?ref=r4m_udw_co_c_dd

I haven't yet tried running it with the old monitor because I couldn't reproduce the problem and it sometimes took up to a week to occur depending upon how much I was using it.

 

[Colif]

oh, i hate problems that only happen sometimes. I put up with one of those for 4 months. No bsod
is other screen 1080p? also 60? I just trying to get info and then i go figure out why. If I can...

i saw vega in the print out, not sure why I asked again. I remember thinking, I am sure an R5 2600 doesn't have integrated, and didn't think anymore of it..

have you looked in reliability history? admittedly it should be the same events as event viewer but might offer clues like display driver turning off. It will see the restarts as hardware errors, that is normal.

 

[SkyNetRising]

The GPU is an ASUS AREZ Strix Radeon RX Vega 56:

Asus recommends minimum 750W PSU for this card.
Chances are - your 650W psu is not enough, when you put extra high 4k workload on it.

 

[Farboguy]

oh, i hate problems that only happen sometimes. I put up with one of those for 4 months. No bsod
is other screen 1080p? also 60? I just trying to get info and then i go figure out why. If I can...

i saw vega in the print out, not sure why I asked again. I remember thinking, I am sure an R5 2600 doesn't have integrated, and didn't think anymore of it..

have you looked in reliability history? admittedly it should be the same events as event viewer but might offer clues like display driver turning off. It will see the restarts as hardware errors, that is normal.

I just checked on the reliability history (I did not know that was a thing!) but I'm not seeing anything different from the event viewer. My second monitor that's also hooked up is an older Dell that's 1080p. The Radeon software tells me that the Hz for the new monitor is 59.9966 and the second monitor is 60.0197.

 

[Farboguy]

Asus recommends minimum 750W PSU for this card.
Chances are - your 650W psu is not enough, when you put extra high 4k workload on it.

Do you think if it's not quite enough wattage, that it would be such a rare occurrence? I tried stress testing the gpu with furmark (which could bring it down to less than 5 fps) and had a youtube video running on the second monitor but I could not recreate the issue after an hour. It makes total sense that a power issue would act this way, but is there any way to tell before I buy a new power supply?

If I buy a new PSU, I'd like to make sure that it's capable of doing what I'm planning in the future. I have an identical PC I built for my partner. My plan is eventually to replace the gpu in one of the PCs with a new card and put the two identical cards in one PC running in Crossfire mode. What power supply would you recommend for running two of these cards sometime in the future? Maybe 1,000W?

 

[Farboguy]

another way is log your sensors using something like HWINFO but its not always accurate
downlaod hwinfo = https://www.hwinfo.com/download/
when you start it, click sensors only and click run
in next menu, click the button next to the clock in bottom toolbar, it says "logging start..." if you hover mouse over it.

I'm trying this application now, thanks for the suggestion! I've always been terrible at wrapping my head around voltage stuff, so if the error happens again I'll upload it here for advice.

 

[Farboguy]

another way is log your sensors using something like HWINFO but its not always accurate

I had the shutdown happen just a few minutes ago (About 9:30 pm, Eastern US time) while the HWINFO was logging. I don't really know how to interpret the file, but I linked it below. Thanks for any help you can provide!
https://drive.google.com/file/d/1dRE1y1L9tCTf8GcrIlFbFZMEiAp8Pnvs/view?usp=sharing