My GPU is heating too much and my PC keeps crashing

dino.dedicc

Prominent
Dec 19, 2017
8
0
510
Hello!
I bought my pc maybe 2 years ago, and after a year and a half I've encountered dozens of pc crashes, during the gameplay, when my gpu starts making loud noise when heating (from the fans spinning).
My GPU is AMD Radeon R7 250 1GB DDR5 128-bit
PSU is Rhino Light Core RX420-H12 v1.3 (I guess it's 420W)

Haven't got any performace issues, even if it is an old-ish GPU.
My first thought is that, it may be a problem with the PSU, but I am posting this so I can hear it from someone more experienced than me. This keeps me worried a lot.

Thank you.
 
Solution


90c looked OK, until I reread and saw you have a AMD 250. There is no way that card should be getting that hot. That's not a full power monster 280x or a 780. The XFX Radeon R7 on this benchmark I am reading is around 65-70C under max load. Unless your GPU is backed with dust it shouldnt be that hot. The thermal shut off for that card is about 96C and it likely throttles around 90c.

So your options are remove the headsink/fan on the card and replace...

lindowsosx

Prominent
Dec 17, 2017
13
0
520
Did you overclock it? If not

I would suggest replacing the thermal paste on the GPU. Its probably getting a little crusty.


While you're doing that you should also just give it a deep clean.
 
I would try stressing my CPU and my GPU separately, to see if one or both cause the crashes. You could run something like Prime95 to stress the CPU, and Furmark or similar for the GPU.

Yes, a higher load on the power supply can result in system instability if the power supply is inadequate. When the computer isn't doing much, the low power demands will make it seem like the power supply is functioning well. But even cheap power supplies can work if they don't have to supply much power. Also, 'peak power' does not mean the same as usable power. Quality power supplies will supply more usable power, it's one reason they cost more.
 

dino.dedicc

Prominent
Dec 19, 2017
8
0
510


I have already been cleaning the pc several times, I think that's not the case, it might be the thermal paste tho, haven't checked but I'm to scared to do anything with the GPU, the only thing i did is to remove it from the case and clean the dust of it, that's all.

And no, I haven't overclocked it at any point of using it.

 

dino.dedicc

Prominent
Dec 19, 2017
8
0
510


It gets around 86-7c to about 90c. That's when the fan starts working harder and faster and I can hear it. After that my PC crashes, I can hear that speaker "hitting" sound and my screen goes black as a night.
 

dino.dedicc

Prominent
Dec 19, 2017
8
0
510


Tried both programs now, the CPU isn't the issue definetly, but the GPU stress didn't go as far as it goes in certain games. The stress didn't heat the card over 80c so it just stayed normal with fans spinning faster and that's all.
 

jr9

Estimable


90c looked OK, until I reread and saw you have a AMD 250. There is no way that card should be getting that hot. That's not a full power monster 280x or a 780. The XFX Radeon R7 on this benchmark I am reading is around 65-70C under max load. Unless your GPU is backed with dust it shouldnt be that hot. The thermal shut off for that card is about 96C and it likely throttles around 90c.

So your options are remove the headsink/fan on the card and replace the thermal paste using a guide somewhere, or replace the card if you can afford to. The 250 is pretty weak anyways. The paste issue I see a lot on the 200 series cards along with fan failures. You have to use special non conductive thermal paste like ARCTIC MX-4.

 
Solution