Jump to content
Corsair Community

GTX 780 reaching 100c after PC wasn't used (H80i V2)


gigiga

Recommended Posts

Hello all, below are my specs:

 

CPU: i7-4770k

MB: MAXIMUS VI GENE

GPU: NVIDIA GTX 780 Ti

GPU COOLING: H80i V2

CPU COOLING: H100i V2

 

My PC was working just fine for many years, successfully running heavy graphics.

 

I went abroad and the PC was off for about 3 weeks. After turning it on for the first time, after a few seconds of running Windows you can hear the fans go to full speed, and the computer ends up crashing.

 

I repeated this process several times and was able to run my Corsair LINK program to see what's going on. It turns out that the GPU temps are idling at 70-75C, and quickly reaching 100C which is shutting down the PC.

 

Nothing physically has changed with the PC since a few weeks ago when it worked. The H80i V2 GPU cooler starts off at room temp + 6C or so, but reaches close to 45C when the GPU temp goes up to 100C.

 

From what I can tell, the pump and fans are working fine. I've also re-applied thermal paste on the GPU, because it's been a while since the last time, but that didn't help it all.

 

I also checked for malware, bitcoin miners etc. but nothing came up. The GPU temp is reaching 100C even though task manager says under 1% utilization on the GPU.

 

I've also set the CHA FAN in the BIOS to always max (and the H80i is connected to the MB's CHA FAN port), to ensure the H80i V2 is getting the full 12V.

 

What am I missing here? what could have caused this odd behavior?

 

Thanks!

IMG_8695.thumb.JPG.94c020793dc303b2ab09df1134eca1d5.JPG

Link to comment
Share on other sites

The GPU temp is reaching 100C even though task manager says under 1% utilization on the GPU.

 

That part is telling since the coolant is only reaching 45C. At the idle state with stepped down voltage, the GPU diode temp should only be a few degrees over the coolant. However, double check the GPU voltage with Afterburner, Precision (or whatever) to see if it really is at the idle state. Link/iCUE open may be enough to cause a voltage increase. Even then, the GPU temp should basically tick a degree here or there like the coolant. This reads like a contact problem, even though you have already checked and reapplied TIM.

 

The other way to look at this is the coolant really should not hit 45C unless you are at sustained near maximal load. If you are getting there on the desktop, something is wrong with the cooler (unless the case temp is 40C+). A continually rising coolant temp at the desktop suggests a flow issue and unfortunately this is a common complaint on the GT/v2 series. This could be air bubbles or something else more permanent. Air bubbles make a lot of noise, typically described as clickity or static like. A physical blockage does not, except when near complete flow block you might hear some high pitch squeal. At 45C, the exhaust temp of the the back should also be 45C. That is clearly warm. If it feels cool, heat is not getting there. You should contact Corsair Tech Support through the ticket system to set up an RMA or seek another replacement.

 

Short term, there isn't much you can do. You can try tilting the case backwards, tapping the hoses, or with the smaller 120mm radiator even take it off the back and shake a little to see if you can dislodge the blockage. Even if successful, what's in there will come around again and this is a temporary and unpredictable fix. You still need a replacement if the flow is getting blocked.

Link to comment
Share on other sites

×
×
  • Create New...