[SOLVED] CPU overheating after changing GPU

Nov 15, 2021
12
0
10
I recently installed a 3060 Ti which caused my CPU temperatures to skyrocket at 100C on some games and 50-60C when idle. I am using a stock cooler to cool my CPU.

Specs:
B460M Gaming Motherboard
Intel i5 10600
32Gb Ram
Rtx 3060 ti
700W EVGA PSU
 
Last edited:
Solution
The CPU cooler has been in use for about a year but I do recall accidentally touching some of the thermal paste during installation which smudged a little bit of it off which could potentially be the problem
Yep, this can cause it. A smudge can leave a pocket of area that neglects the transfer heat to the CPU heatsink.
Clean the old paste with 70 or 90% rubbing alcohol, let it dry and then re-apply new thermal paste.
Avoid touching your CPU with your fingers too as fingers are usually dirty and can have oils that interfere with the job of the thermal paste to cool.
Hello 7scooby1. Intel 'K' SKU CPUs do not come with a stock cooler. Question: Before you purchased the RTX 3060 Ti, did you use the integrated graphics or another discreet GPU. If GPU, what was it?

The Intel stock cooler is insufficient for a 'K' SKU CPU. There is no doubt that you'll need to upgrade it. What's your budget, and do you prefer liquid AIO or a fan cooler? I'm sure that others will be happy to chime in with recommendations.
 

Kona45primo

Honorable
Jan 16, 2021
524
143
9,890
I'm guessing a better CPU cooler might be a good idea. Chances are the new GPU can put out more frames, which means your CPU is tasked with sending more info, which makes it consume more energy and get hotter.
 
So you actually have an Intel core i5-10600 (non 'K') and it's hitting 100C during gaming.

FYI: 100C is the Junction Temp for that processor.

"Junction Temperature is the maximum temperature allowed at the processor die."

https://ark.intel.com/content/www/u...10600-processor-12m-cache-up-to-4-80-ghz.html

100C seems very unlikely for a non-overclockable SKU, assuming that the stock cooler is installed correctly with it's pre-applied thermal paste. Which utility is reporting that temperature to you? And what made you look up the temperature in the first place? I just want to be certain that you actually need an after-market cooler, because even though I have no experience with your specific CPU, a stock cooler should be fine for it.

Addendum: Kona45primo poses a good question. Matter-of-fact, would you mind taking a photo of the inside of your build, uploading it to Imgur and sharing the link here? Something about this scenario isn't quite adding up to me. Also, could you please find out and list the exact mfg. and model of your motherboard?
 
CPU Overheating after changing GPU

So you actually have an Intel core i5-10600 (non 'K') and it's hitting 100C during gaming.

FYI: 100C is the Junction Temp for that processor.

"Junction Temperature is the maximum temperature allowed at the processor die."

It might actually be that you are complaining about GPU temps, because the recent hardware change you've done is change to a new GPU (RTX 3060 Ti). I'd try MSI Afterburner and reducing its power limit.

By default, most GPUs fresh out of the box run at 100% power limit.
The latest RTX 30 series draw much more power, for instance my 3080 Ti I set it to as low as 45% power limit in MSI after burner.
+ an undervolt tweak as well.

Its no surprise that the latest RTX 30 series GPUs have high voltage transient spikes, and I wouldn't be surprised as well if it were the same case in Nvidia's upcoming love lace RTX 40 series too

Decrease the power limit by 5% and record in game temps and FPS
Keep doing this until you can no longer have an "acceptable FPS" by your standards, and then move it up a notch by 5%
That's where you determine what power limit to use.
If your game is complaining too little FPS, increase the power limit by 5% again.

I have experienced no crashes or errors or any FPS lags with 45% power limit, at least on my RTX 3080 Ti
 
Can you download and run HWinfo64
Use sensors mode
Post a screenshot of the CPU temps while gaming
Post a screenshot of the GPU temps while gaming

I do agree with rcald2000, 100C for your CPU is highly unlikely. Your CPU would then be throttling down to reduce temps at that rate.
 
I don't know the answer yet. But I believe the CPU would initially throttle, but at 100C I would expect the system to reboot to protect itself, which isn't happening.

observations:
RTX 3060 Ti model consumes 200 watts, which would explain the excess heat leaving the case.

model: MSI Ventus GeForce RTX 3060 Ti VENTUS 2X 8G OCV1 LHR

Gigabyte B460M GAMING HD doesn't appear to have heatsinks on VRM; not sure if that's relevant to the high CPU temps or not.

https://www.gigabyte.com/Motherboard/B460M-GAMING-HD-rev-10#kf

CPU cooler recommendation: I haven't personally installed this cooler, but I've heard good things about it. Your temps seem so odd to me. It's almost as if it has no pre-applied thermal paste on it.

Scythe Mugen 5 Rev.B CPU Air Cooler, 120mm Single Tower, Intel LGA1151, AMD AM4/Ryzen
$49.99

[Black Edition] Scythe Mugen 5 Rev.B
$59.99
https://www.amazon.com/Scythe-Mugen-Cooler-120mm-Single/dp/B08X4HX858


View: https://www.youtube.com/watch?v=GXbZyo9rEqA
 
Yes, there's something extremely off here by a mile...

It may be that the CPU cooler is the culprit and would advice changing to a CPU cooler from noctua for your model. Intel CPU stock coolers are usually durable and I don't suspect them to fail, but if the bearing of the fan fails then it is a possibility. How many years has it been in use?

and changing power didn't stop it.
Are you sure you did try to set the power limit lower in MSI After burner?
Would be interesting to see if you get the same temperature results in Hwinfo64 if you have a power limit setting of say 45% in MSI After burner while gaming (post screenshot). I expect GPU temps to lower because of the set power limit, but would be interesting to see the resulting CPU temps as well.

Some old suggestion: as well

Your new GPU is consuming more power than your old one. This will make the inside of your case hotter. Try running with the sides of your case removed. If this solves the overheating, then add more fans to your case.
 
Nov 15, 2021
12
0
10
The CPU cooler has been in use for about a year but I do recall accidentally touching some of the thermal paste during installation which smudged a little bit of it off which could potentially be the problem. So I might try a new CPU cooler and see what it is like. Thanks everyone for your help.
 
The CPU cooler has been in use for about a year but I do recall accidentally touching some of the thermal paste during installation which smudged a little bit of it off which could potentially be the problem
Yep, this can cause it. A smudge can leave a pocket of area that neglects the transfer heat to the CPU heatsink.
Clean the old paste with 70 or 90% rubbing alcohol, let it dry and then re-apply new thermal paste.
Avoid touching your CPU with your fingers too as fingers are usually dirty and can have oils that interfere with the job of the thermal paste to cool.
 
Solution
I just asked a regular from a pc tech discord that I'm apart of. He told me he would build several rigs with that specific chip, and would hit 90+ with a stock cooler, in a poorly ventilated cases. I guess it is indeed possible. My bad for not immediately knowing that answer.
 

Karadjgne

Titan
Ambassador
P=VxI. If power used was just 65w on the cpu, at @ 1.0v, that's 65A at 12v. That's enough to power over 6500 leds at 5v and the average well lit pc has less than 100 leds.

For their size, a cpu deals with massive amounts of power, and it's usually a fraction of what a gpu deals with, closer to 1/2 or 1/3rd. Incorrectly or insufficiently cooled, any cpu/gpu can easily reach 100°C or more.