Ryzen 2600x and H100i Overheating

Greetings all,

 

I'm having an overheating issue with my ryzen 2600x that I seem to be getting nowhere with.

 

I was playing a game when I noticed that I all the sudden was stuttering horribly. I quit out oif the game, thinking that there was just some random glitch, but I noticed even on my desktop, my mouse was stuttering. I rebooted at that point, and on bootup, I was still getting the stutter. That's when I checked my temps and found that my CPU was pegged at 95c. I shut down again and checked out the inside of my tower. For cooling, I use a corsair H100i. I checked for obvious signs of leaking, but found nothing. I let everything cool down and booted back up to check the pump RPM and coolant temp, pump reported ~2200 RPM as normal in corsair link, and the coolant temp was 45c. Temps were now idling around 45-55c. This seemed a bit high, but I thought maybe it was residual from the previous overheating. I ran a CPU benchmark and immediately the temps skyrocketted to 80c, then 85, then 90 and I aborted from there. The coolant temp steadily rose, but also receeded, so it appeared that the pump was operational.

 

 

I proceeded to shut down again and unhooked everything for a cleaning. I cleaned out all the fans and the radiator, reapplied thermal paste, checked all the power connections to make sure the pump was working, although I could tell it was before. Tried again, and got the same behavior. Idle temps 45-55c and applying load skyrocketted the temps in well under 30 seconds. Voltages on the CPU reported as they always have, between 1.3 and 1.41, depending on the load. I am not overclocked, this CPU (for the first time in my life) is running at completely stock settings.

 

 

So now I am a bit at a loss. I don't have a spare heatsink/fan to test with, or spare CPU. It seems to be that there is some kind of flow problem in the AIO cooler perhaps, but I don't know how to verify that without just getting a new one to find out. The cooler is 5 or 6 years old (perhaps older) and it is quite possible that it is just end of life. The cooler is the only thing I can imagine all the sudden going bad that would cause this type of thing.

 

 

Just looking for some advice as to what else I could try before buying a new AIO cooler that I may not even need,

Coolant temperature, like everything else in your case, is affected by internal case air temps and room temps in turn. In your data above, the CPU temp is idling at 45C because the coolant temp is also at that level. We need to figure out why.

 

It is definitely possible for the GPU to heat up the case to the point where it looks like cooler issue. The easiest way to check that is to remove the GPU heat from the equation. After the PC has been off/sleep for a while, power on. H100i Temp (coolant temp) should be just above the room temp and will slowly creep up to about +4-6C above that over the next 10-20 minutes. If you boot up and can watch the coolant temp tick upwards +1C every few seconds, something is wrong. If you boot up and the coolant is already at 35C+, something is wrong.

 

If everything checks out OK at this level, then initiate a mild CPU only stress test. The internal one for CPU-Z (bench tab) is perfect since it is is mild and linear. Depending on your voltage/frequency level, you should see the coolant go up +4-6C in 10 minutes. 10 minutes is long enough. If there is a problem with the cooler's flow, you will know almost immediately as the coolant temp quickly rises 10C+ in 1-2 minutes.

Another option, its less scientific but if you have to remove the cooler, rotate it and listen closely for "water sloshing sound". There is always some of this in an AIO, but if it is excessive you may indeed have had so much liquid evaporation the cooler is now just not able to function well. 5-6 years is pretty good for an AIO as a life service, longer then I have ever kept one but that's very respectable.