Jump to content
Corsair Community

H100i v2 Temperature question


sean_nl
 Share

Recommended Posts

Hello,

 

I have a question about my Cpu temperature.

Idle temps are fine, its ~30c

But under load the temp stays around 70 ~ 75c

I have a Intel 6700k running @ 4.4ghz with 1.3v

Are these temps normal, or is there anything else going on?

Also the radiator/hoses don't get warm.

I already reapplied thermalpaste, and i've set fan/pomp speed to max, but temps stay the same.

 

Any thoughts/comments?

Thanks in advance

Edited by sean_nl
Link to comment
Share on other sites

There are a large number or reasons for CPU core temp variations and there is no specific delta over idle temperature for a 100% load. It is very CPU and voltage specific. it is also program specific. An Intel XTU 100% run is not the same as a Prime95 or OCCT 100% load.

 

If you want to assess cooler functionality, you need to look at the water temperature (H100i v 2 Temp - last box on Link main page). It should be slow to change up and down. If you are idling at 30C, then your H100i v2 temp will also be about 29-30C. Load up your stress program and run for 5-10 minutes. The water may climb 6-10C depending on the load and program. When you stop the program, it should drop 3-5C almost immediately and then over the next 5 minutes or so, slowly drop back to 30C. Other things like case set-up and design can affect retained heat levels, but these are reasonable approximations.

 

If you water temperature increases drastically or will not come down after a load (of any level), that usually suggests a problem. However, in this case, my hunch is your CPU core temp are the result of the program in use and nothing more.

Link to comment
Share on other sites

Thank you for the reply,

 

I ran a 15minute Realbench test, and the h100i temp did go up 6c in total, and after the test it did go down slowly.

 

I guess its normal then to get those temps.

I just installed the cooler last weekend, and didn't really test it in games yet, i hope the temps are lower now than with my 212 evo.

I thought it was a bit high, because my 212 evo had the same temps in realbench.

 

Again, thank you for the comment

Link to comment
Share on other sites

I don't think you are out of line. Real Bench is funny test and hard to make sure predictions. It is warmer than the easier AIDA/XTU types, but also introduces GPU load and can affect water temp in small cases or when in the rear exhaust. It also can pull quite a bit of power. Total power draw during the final phase is almost 550w on my set up. That's a good 200 watts more than most other stress tests for me.

 

Peak values can sometimes misrepresent the current state. When you slam 1.40v (or whatever) into the CPU, it instantly heats up. The cooling is on the other side and the heat must still pass through. Another way to measure total cooling performance is to look at the average CPU core temps over the course of a fixed internval -- like a 15 min Real Bench run. The amount of heat retained on a spike is going to be similar with similar cold plate materials (air tower vs water). However, the water system has a far greater capacity to hold heat and can handle more for longer periods of time. That should be apparent in the averages over time.

Link to comment
Share on other sites

 Share

×
×
  • Create New...