Jump to content
Corsair Community

Normal delta temp between H80iGT and CPU


Rillenn

Recommended Posts

I'm new to all liquid coolers so I have tested a lot of different settings and stuff for good performance and sound level. And to be honest I'm a little confused so I will share my findings and I'm just asking for second opinion if this looks all fine or if you think something looks off.

 

Statics: ambient temp ~20°C (68°F), All test start at CPU temp~38°C (100°F) and water temp~32°C (89.6°F) Stress test 100% CPU 30min. H80iGT push-pull config pump in performance, i7 4790K @ 4,4GHz, 3st CHA fans @ 1200rpm.

 

Test 1

Fan=100%

Water= 34,6°C (94°F)

Cpu temp=70°C (158°F)

 

Test 2

Fan=60%

Water=36,7°C (98°F)

Cpu temp=70°C (158°F)

 

Test 3

Fan=40%

Water=38,8°C (102°F)

Cpu temp=74°C (165°F)

 

So in my testing CPU temp changes very little with fan changes, even so that I couldn't see any changes between 100% and 60% and very small but noticeable changes with 40%. For me this looks a bit warm, not that i expect magic from the small radiator but it feels like my CPU would run a little bit cooler. It runs about 8-10°C cooler stock.

 

So to end this weird rant, roughly what delta temps between water and CPU is normal with this cooler and do this test result indicate something wrong?

Link to comment
Share on other sites

It's very difficult to give a standard delta for water and CPU temperature. There are numerous factors and it really is somewhat unique to each case, hardware, and voltage. You have done the only thing you can do, which is to run tests at controlled settings.

 

Water cooling works a little differently than air cooling. CPU heat will be transmitted through the lid and contact plate where it is absorbed by the water. The water is the transport mechanism to the radiator fins, where the heat is then hopefully released. The fans only job is to speed the release of heat from the water, not cool the CPU directly. The water is primary on that job. The amount of water in your system and the flow rate will determine how much heat you can remove from the CPU in a given amount of time. As you have just discovered, high fans speeds do not really help much with your results in a controlled environment. There are some circumstances, particularly with increased GPU waste heat, where this may not be as clear cut.

 

So, to answer your question, I don't see an issue with the fan speed scaling. The water temp goes up 2C (less fan speed), the CPU temp goes up 2C. The water temperature is your effective CPU baseline temperature + voltage = final CPU temp. Since the voltage is more or less constant during the 100% load, the change is based on the water temp.

 

The delta I am more interested in is the that between your 20C room temperature and your 32C test start water temp. That is a little bit high over ambient and may account for slightly higher than expected temperatures. How is your case set-up with fans? where is the H80i GT located?

Link to comment
Share on other sites

The delta I am more interested in is the that between your 20C room temperature and your 32C test start water temp. That is a little bit high over ambient and may account for slightly higher than expected temperatures. How is your case set-up with fans? where is the H80i GT located?

 

Thanks for explaining, well the Water temp is not idle temp I didn't have time to let it sit and cool down between test so i just picked a temp well below load temp and started the next test when it was around that to get fair testing. And the CPU temp is what it landing on under light load such as browsers and monitoring programs.

 

I have 0 issues with it now and the CPU is performing good and with 60% I get silent enough and cool enough during my normal loads. Just wanted to run my numbers by someone with more experience because I had massive issues with my first install and had to refit it.

Link to comment
Share on other sites

OK, then. A 6C delta between water and CPU at idle is fine. I don't think you have any install issues. If the block wasn't making good contact the stress test would be untenable and the flow and fans all check out. Your CPU temps still seems slightly on the warm side. If you are at the stock 4.4 (turbo/auto) settings, you likely can decrease your temps by tightening up your vcore and other settings. Most people can run 4.4 well below the auto Vcore, but I'll leave that to someone running a 4790k. One thing to be aware of is if you have mounted the H80i in the rear exhaust as most people do, you will likely see higher than expected "game level" CPU temps. In those situations, the GPU waste heat increases your water temperature and boosts your baseline CPU temp by a few degrees. When I was running an H80i in that configuration, my Prime 95 numbers were a good 10C lower than actual heavy CPU gaming temps. That doesn't mean the rear exhaust position is ill-advised. Just something to be aware of.
Link to comment
Share on other sites

It's not stock it currently running @ 4,4 base and turbos up to 4,6 but the highest I have seen is 4,55 and yes to be honest it probably isn't the best OC and I could probably lower the Core V with only positive result.

 

I have noticed that the water temp rises a little when loading the GPU (thinking about adding another exhaust fan) but it's really nothing that have any meaning in real life situations because the CPU (atm) never runs anything close to 100% in my load situations.

Link to comment
Share on other sites

OK, sounds like a more traditional Intel overclock with one core max at 4.6. When all 4 cores are loaded it will cap at 4.4. Not a problem and not a large factor.

 

One thing to consider if the H80i is rear exhaust mounted is a fixed speed when gaming. Pick something you can stand. This would likely be higher than needed for pure CPU cooling, however the other purpose of the H80i in that situation is to exhaust air. This should enable you to get more air out plus more air through the radiator. You also won't get the fan cycling with GPU load and the constant speed is easier to block from your mind. I found this to be a more useful set-up when running a H80i as the main exhaust.

Link to comment
Share on other sites

OK, sounds like a more traditional Intel overclock with one core max at 4.6. When all 4 cores are loaded it will cap at 4.4. Not a problem and not a large factor.

 

Almost, 2 cores run max at 4.6 well I'm kinda new to all this but I figured that I might as well try and get a little extra from the cores I'm using, not sure if this is right thinking but. I play mostly a really heavy single core load. And the second core is for if I play a game (I don't know if i ever do but for safety) that loads 2.

 

I could probably just raise all of them to 4.6 and be perfectly happy with it, but I don't think it would like stress testing/bench marking at all and i don't like the idea of that.

Link to comment
Share on other sites

No, that makes sense for your usage. 4 cores at 4.6 requires more voltage than 2 cores, and that brings more heat. You are fine as you are.

 

Wow, I love how basic logic get you a long way when you got minimal knowledge. Well everything runs great now Thanks for the help!

Link to comment
Share on other sites

Stock voltage for a 4790k is 1.209v at 4.4GHz(stock max turbo). My chip can run 4.4 @ 1.111v, 4.5 @ 1.121v, 4.6 @ 1.162v and 4.7 @ 1.190v on all 4 cores. Gaming temps are 43-54C, with occasional spikes between 55-65C(1-3 seconds) @ 94-100% load at 4.5GHz. Idle temps are 21-29C. This is very general, but I hope it helps.

https://drive.google.com/file/d/0B007JgCLgXQLU0YwTlZ1cW9Rcms/view?usp=sharing

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...