Asus Prime-P X570 Thermal Shutdown

h4wks

h4wks

h4wks

Associate
Joined
6 Apr 2011
Posts
119
I've recently built a new PC, spec's below, and I'm having problems with the system shutting down after around 1 hour of gaming.

MB: Asus X570 Prime-P
CPU: Ryzen 3700x - cooled with Corsair H115i 280mm AIO
GPU: MSI RTX 3080 Gaming Trio
HDD: 2x NVME m.2 (PCIE 3), 1x Sata SSD, 1x HDD
Case: NZXT H510
PSU: Aerocool Project 7 850W

GPU temp is maxing out at 80C, CPU package at 62C, Ram modules at 60C. I've only got one reading in the BIOS for motherboard temp, there's no Chipset of VRM temp available on the Prime-P. My suspicion is that the chipset is overheating and triggering thermal shutdown. My GPU is nearly completely blocking the PCH fan, which is spinning at 3800rpm under load. The system is outputting some serious heat, the top of the case is really quite warm to the touch!

Is there any way I can actually monitor the chipset temps to see if this is what is triggering the shutdowns? iCue and HWinfo have 12 temp readings for the motherboard but a lot of these look like duplicates and are almost all just labelled with numbers. One of readings does rise from 53C to 77C when shutdown is triggered after 90mins gaming. It was as high as 81C after just 40 mins of gaming though and this didn't trigger a shutdown.
 
It's not a huge case especially with the size of the 3080! The AIO being front mounted is also not helping get any cool air in to the case I guess. I think I'll switch in a Noctua S12 for the CPU cooling and try and pump some good cool air in the front with some 140mm fans and see if this helps.
 
Try leaving the side of the case off see if that helps if it does it is heat related also could it be your gpu I think others have been on about there pc shutting down due to new gpu's but could be wrong have a look at graphic section on forums there was definitely people with 3900 cards
 
Last edited:
mickyflinn said:
Why dont you reverse airflow your case and have the 2 rear fans as intakes and the aio pushing out the front of the case.
Click to expand...

I could try this. Only issue is the top and rear fans are not dust filtered and the front of the case is, so would probably lead to some pretty substantial build up.
 
alex h said:
Try leaving the side of the case off see if that helps if it does it is heat related also could it be your gpu I think others have been on about there pc shutting down due to new gpu's but could be wrong have a look at graphic section on forums there was definitely people with 3900 cards
Click to expand...

Maybe I should try this before switching out the AIO. Hopefully not the GPU :( I'm not getting any blue screens or crashes the PC just completely shuts down, I think the lights are still on on the MB and the power button is definitely still illuminated but all fans are off so it's definitely shutdown. I can't restart it using the power button and actually have to switch it off at the wall or PSU for 30sec to get it to power on again.
 
mickyflinn said:
There have been issues with the 3080 and capacitors, have you looked into this.

Nvidia drivers helped.
Click to expand...

I did read about this yes. I've got the MSI Gaming Trio which has 5 of the better capacitors I believe. I think the number capacitors on the underside of the chip might have been disproven as being the reason for the problems now though?. I'm not completely up to date on the saga I must admit. I've definitely got the most recent Nvidia game ready driver installed.
 
h4wks said:
I did read about this yes. I've got the MSI Gaming Trio which has 5 of the better capacitors I believe. I think the number capacitors on the underside of the chip might have been disproven as being the reason for the problems now though?. I'm not completely up to date on the saga I must admit. I've definitely got the most recent Nvidia game ready driver installed.
Click to expand...

It was due to the quality of the capacitors so nvidia drivers dialled back the clock speeds.
 
h4wks said:
I've actually got the sensor logs from iCue here https://docs.google.com/spreadsheets/d/1NtTEDVM49k3vyngzTE5aFraj5qUARCKeJpszuRxI1HI/edit?usp=sharing

This is starting up from cold and playing Warzone for 90mins or so before the system shut itself down.
Click to expand...
MB Temp #12 is likely chipset temperature with its very high correlation to PCH (chipset) fan speed.

And with one of the hottest/most heat energy ever producing graphics card there along with CPU heat dumped into case you need to increase airflow through case.


Also check Reliability History and Even Viewer for any errors close to shutdown.
 
Thank you, I'll check these logs and see if there's anything untoward. Will also change up the fan configurations over the next day or so and probably remove the AIO/Rad and report back.
 
Removing the side panel as suggested earlier is a good first step.

If it still fails, set up a desk fan blowing into the open case.

If it's still failing after that it probably isn't an airflow/temperature problem.
 
To update:

This evening I've switched out the AIO for a Noctua N12S and have 2x 140mm case fans at the front and 120mm rear exhaust and 140mm top exhaust.

Test 1 - with the above config but side panel still installed. Lasted about 60sec running Heaven benchmark before shutting down! https://docs.google.com/spreadsheets/d/1nYHe0UJw7ayXQ89OTIchYVrrQhz5hgG4r9d1P3afX6c/edit?usp=sharing

Test 2 - The above config but ramped up the fan RPM, side panel still installed. Got about 20min on Heaven benchmark and then system shut down. https://docs.google.com/spreadsheets/d/1KyP8q3pz4-BOcZKqMebbphc1k5L2wCks6rihMB62t70/edit?usp=sharing

Test 3 - Same config and increased fan RPM but with side panel removed - lasted 19min on Heaven benchmark. https://docs.google.com/spreadsheets/d/1_POUHmTRp5fhGU7g8KZi5FXWS5_-EfXW2LnoZwDitHQ/edit?usp=sharing

With test 2 the motherboard, GPU temps plus Temp #12 (might not be the chipset temp?) are still as high as previous. Worrying thing is in test 1 and 3 I'm now getting shutdowns even without the high temp's. Starting to suspect GPU! Currently 15 mins in to a benchmark with my old 2070 super and still running. Problem is as the card is substantially smaller and cooler it doesn't necessarily rule out other components being the issue - PSU, motherboard?
 
Hah, you made me doubt myself but I just checked and yes all blowing in the correct direction! 40 mins in to Heaven and the 2070s is still going strong at only at 66C! Next step I guess I could try the 2070 with all case fans off and see if I hit a thermal shutdown? Or the 3080 with all case fans and GPU fans set at 100%?

Saying that I'm still not convinced it's the chipset overheating since I've managed to repeat the shutdown after only a few mins of benchmarking earlier this evening.
 
Could be your psu mate how old is it I don't know much about your psu to be honest but I was once told never skip on the quality of your power supply it's the most important part of your pc maybe it isn't stable enough to run the 3080 gpu
 
alex h said:
Could be your psu mate how old is it I don't know much about your psu to be honest but I was once told never skip on the quality of your power supply it's the most important part of your pc maybe it isn't stable enough to run the 3080 gpu
Click to expand...

The Aerocool Project 7 is a solid psu, and 850W is easily enough for his setup.
 
You must log in or register to reply here.
Back
Top Bottom