Jump to content
Corsair Community

Possibly Low H100i v2 coolant temperatures during CPU stress test


Recommended Posts

Hi everyone, I’m looking for clarification on whether the coolant temperature on my H100i v2 is in the normal range while conducting a stress test on my 8700K, the details are listed below.


I have no reference point of whether the coolant temps are normal or out of range (hence this post), but to me the coolant temperature seems very low given the CPU temp even with when considering the poor die/IHS heat conductivity. I conducted this specific test twice, once after finishing the OC process and again after having repasted and reseated the cooler.


-Test: Prime95 (v29.3) Small FFT (no AVX), temp readings taken after 2 hours.


-CPU: Intel Core i7-8700K 4.9 GHz @ 1.312 V (1.31 V in Link).


-CPU Max package temp: 89 degrees C.


-H100i v2 max coolant temp: 34.1 C (idle temp 27.6 C).


-Ambient room temp: ~24 C.


-GPU temp during test: 30-34 C.


-H100i v2 pump speed: 2900 RPM (Performance mode).


-SP120L speed: 2700 RPM (Max).


-Case/Fan arrangement: Cooler Master HAF 922 with 2x 200mm fan intake (front and side), top mounted radiator (exhaust, fans pushing) and 1x 120mm rear exhaust.


Changing the pump setting to “Quiet” (1280 RPM) or reducing the fan speed to 50% only sees a coolant temperature increase of 1 C over the listed settings above. A touch test of the radiator and the air coming out of the radiator were both luke warm at best.


Any help is appreciated!

Link to comment
Share on other sites

No, a 6.5C delta is about right. This value is a reflection of watts produced versus the watts the cooler can dissipate. You would only drop another 1C with a 280mm or maybe 1.5C with a 360mm, albeit with much lower fan speeds. I just ran it with my 8700K@1.31v and coolant delta was 5.3C on my H115i Pro (280mm) with LL140 fans fixed at 1200 rpm. Howevever, do note my 8700K is delidded and while this subtantially helps with conductive heat coming from the CPU, it does not make a difference on the coolant temp as this amount of wattage is well below the cooler's capacity. You are limited by the CPU design and voltage (pin side heat), not the cooler's ability to get rid of it.


The CPU temp and coolant temp have an additive relationship (+1C coolant temp = +1C CPU temp and coolant temp = minimum possible CPU temp). However, heat must be conducted into the cooling system and the values in relation to what happens with your CPU temp and CPU voltage is not instantaneous or 1 to 1 in the same way. It's more like the next stage of cooling or an adjacent heat depository. I suppose the better way to think of it is the stove top heating element and the pot of water on it. You turn the heating element to High. It very quickly reaches 400-450F (~225C). The 4 liters of water in the pot is room temperature water to start, then slowly heats up. In the meantime, you can still put your hand in the water without consequence. The bottom of the pan will not be the quite as high as the heating element itself, but it would be folly to touch it. Eventually, the water will reach 100C and boil. The difference here is instead of putting a lid on top the accelerate the process, you connect a radiator and fan system to remove the heat from the water as fast as it can, and thus the water never reaches boil. The coolant temperature is a measure of how much heat is currently in the system and is more a measure of efficiency, with the previously mentioned consequence of increasing the CPU temp by +1C for each measure increase in coolant temperature. The burner or heating element temp is a more akin to the socket temperature on the CPU. Clearly this matters, but the if you want to lower it, you typically must reduce the voltage (turn down the heat) or make the CPU conduct heat faster or more efficiently and that's your delid. On a 8700K, the after effect of this is enormous. Not all CPUs benefit as much, but the Coffee Lake processors most certainly do.

Edited by c-attack
Link to comment
Share on other sites

  • Create New...