Question GPU overheats for no reason

Apr 26, 2022
24
0
10
0
Hello guys, I have a problem.

Yesterday, I decided to clean my GPU of dust and I've used a brush. Everything was fine until I turned my PC on again. It entered system and gpu temp was 35C. I wanted to play TheForest, but when I opened the game, fans went crazy and my PC cut off. I tried replacing Thermal Paste again to see if it works but this time something new happened. I pushed power button and Card fans immediately went crazy around 4000 RPM which is not normal, and GPU cut off but PC was still on.
I appreciate everything you do for me!

Looking forward to your replies!



My config:

Intel Core I5-4690k 3.5GhZ
Sapphire radeon r9 280x Dual-x OC 3GB
8GB RAM
PSU: Strike-x 800W
Mainboard : B85m-g
 
Last edited:

Ralston18

Titan
Moderator
Update your post to include full system hardware specs and OS information.

Include PSU: make, model, wattage, age, condition (original, new, refurbished, used)? History of heavy use for gaming, video editing, or even bit-mining?

What kind of brush did you use?

Power down, unplug, open the case.

Confirm by sight and feel that all connectors, cards, RAM, and jumpers are fully and firmly in place.

Use a bright flashlight to methodically inspect everywhere and everything for signs of damage.

All-in-all, it may be that cleaning with a physical brush caused something to be disconnected or unseated.

Or worse, broke something.

Then again, there is the possibility that the PSU coincidentally has failed in some manner.
 
Apr 26, 2022
24
0
10
0
Everything is fine, I will try with another PSU tomorrow and tell you what happened.

Nothing seems to be damaged, only thing I did is cleaned an old paste and got a new one.
I used painting brush but it did really well.

PSU: Strike-x 800W
Mainboard : B85m-g i think
 
Apr 26, 2022
24
0
10
0
Update your post to include full system hardware specs and OS information.

Include PSU: make, model, wattage, age, condition (original, new, refurbished, used)? History of heavy use for gaming, video editing, or even bit-mining?

What kind of brush did you use?

Power down, unplug, open the case.

Confirm by sight and feel that all connectors, cards, RAM, and jumpers are fully and firmly in place.

Use a bright flashlight to methodically inspect everywhere and everything for signs of damage.

All-in-all, it may be that cleaning with a physical brush caused something to be disconnected or unseated.

Or worse, broke something.

Then again, there is the possibility that the PSU coincidentally has failed in some manner.
I've found the problem. There were some black sticky things and I don't know what they were for, I just removed it and GPU works.

Before I removed them, I've noticed a big gap between heatsink and GPU.

I don't know what those black sticky things on Sapphire R9 280X DUAL-X 3GB OC are. If anyone knows, please let me know.

They look like this:
https://snipboard.io/uBVvSx.jpg


Best Regards.
 

ttower2020

Reputable
Those appear to be the thermal pads that cool the GPU memory. They are important to have installed, in order to keep the memory from overheating. Its hard to tell from your message if you have them removed or installed, but they should absolutely be installed on the card.
 
Apr 26, 2022
24
0
10
0
Those appear to be the thermal pads that cool the GPU memory. They are important to have installed, in order to keep the memory from overheating. Its hard to tell from your message if you have them removed or installed, but they should absolutely be installed on the card.
But when I install them, firstly they don't stick good and secondly, my GPU is overheating..
 

ttower2020

Reputable
But when I install them, firstly they don't stick good and secondly, my GPU is overheating..
If the Thermal Pads are not correct, and the wrong thickness, they can prevent the cooler plate from touching the GPU the way it should. If the pads are too thick, this would cause the GPU to overheat, because the cooler is not making good contact. If the thermal pads are missing, or too thin, the GPU could be fine, but the memory temps will be very high. I am not an expert on GPU internals, such as thermal pads, so I cant confidently recommend anything specific to do. It also does not explain why the GPU was fine until you cleaned it, unless cleaning it happened to cause other damage, and the thermal pads are just being a distraction from the real problem right now. It is also possible (maybe?) that the thermal pads are only good for one use, and should be replaced. I do not think this is normal, but the pads are supposed to stick to the memory for installation.
 
Apr 26, 2022
24
0
10
0
If the Thermal Pads are not correct, and the wrong thickness, they can prevent the cooler plate from touching the GPU the way it should. If the pads are too thick, this would cause the GPU to overheat, because the cooler is not making good contact. If the thermal pads are missing, or too thin, the GPU could be fine, but the memory temps will be very high. I am not an expert on GPU internals, such as thermal pads, so I cant confidently recommend anything specific to do. It also does not explain why the GPU was fine until you cleaned it, unless cleaning it happened to cause other damage, and the thermal pads are just being a distraction from the real problem right now. It is also possible (maybe?) that the thermal pads are only good for one use, and should be replaced. I do not think this is normal, but the pads are supposed to stick to the memory for installation.
After I removed them, my GPU temp is rly fine, but how can I check memory temps?
 
Apr 26, 2022
24
0
10
0
They are definitely thermal pads for the memory and are important. Did you clean the cold plate? Where is the thermal paste/pad for the GPU?
Yes I did clean everything, and I forgot to mention that thermal pads don ´t stick well with GPU so I cannot stick them properly.
 

ASK THE COMMUNITY