Originally Posted by Assyle
today Ive bought h100i, and Im a little bit dissappointed. I had antec 620 cpu cooler before corsair h100i. Antec had 120mm rad, and I thought that 240mm rad on h100i gonna perform at least better, but after installing it I found out that the temps are even higher over antec. I still got a feeling that Im doing something wrong. Could someone say if my temps are too high for h100i or not?
i5 3570k (non overclocked), gtx770, psu corsair tx750. case bitfenix prodigy.
with antec I had
idle 27-33 C, and 67C under a full load. (ambient temp was 23-25C)
and h100i has
idle 29-34 C, and 71C under a full load. (ambient temp was even smaller 20C)
both configs were with corsair SP120 QE, with one exception that antec had 2 sp fans in a push/pull mode, and h100i is on a pull.
I even tried to make a push pull with h100i stock fans, but the situation didnt improoved at all.
My antec rad got hot under a load, but h100i is cold as rock and becomes a little bit warm under a full load.
I didnt overclocked cpu at all, and I think people have even smaller temps with the same cpu, but after overclocking.
Does any one has any suggestions to this situation?
I think the problem can be caused by backplate issue. I saw many people were telling that it doesnt holds good enough. When I placed the h100i for first time, I had temps over 80 under a load, so I uninstalled it and noticed that thermal compound applied only to a half of the cpu. Co apllied some other thermal compound and placed a pump tubes on the other side. And temps were 2-3 degrees smaller than now. Than I wanted to look if thermal compound was applied on the hole cpu so I took the pump off and it was ok. After thatre moved old TC I applied new thermal compound, this time some of cooler master tc. I think its not good enough that why the temps are 2-3 degrees higher, but still I think that temps are too hight even with a chep thermal compound. Could it be because of the backplate issue?
thanx for responses
Originally Posted by Assyle Warning: Spoiler! (Click to show)
I thought by head with blow up, when I was placing it.
I also was thinking about placing them at the front, but fan controller doesnt allows me to do that. Now Im thinking of modding side panel to place fan controller there.
Well Im kind a noob, and after opening it I could find nothing that can help me to find a reason of BSOD. WIndows got me a message after reboot that it will send some report to microsoft. I dont know how to look it through.
Anyway thanx for trying to help me.
Now Ive got another theory why the temperatures can be so high with my non overclocked 3570k.
Could this be because of cheap MOBO?
I ve got msi b75 mini itx mobo, and it was one of the cheapest motherbords that were at that time for 1155 socket.
The first processor I had in this rig was Intel Pentium G840, which was sandy bridge. I was using it with Cooler Master Hyper 212+ EVO. And if I remember idle cpu temp was 47C. And even then I thought it was very high.
After upgrading it to 3570k the temp were even higher. maybe it wasnt a cooler issue, but some motherboard cpu connection issue? as it was told above?
Maybe if I will change the mobo the tamps are gonna be much more better? Or it is impossible theory?
today I applied new thermal compound arctic mx-4 with my h100i and Ive got a feeling that the temos even increased :/here is a screenshot.
For a stock 3570K @ 22c-24c ambients your load temps are on average about 15c-20c to high
You've already tried several remounts and checked the TIM was spread evenly (very good!) ...
So it must be a faulty pump or some kind of restriction for your radiator airflow. A properly functioning motherboard will have NO effect. ...Try a test with the radiator unmounted, outside the case with no restrictions to airflow ... I would also prefer to see the slower rpm/SP120 QE's in the push position if those are the only 2 fans you have circulating air through the rad.
I don't think it is a problem, but we might also find that the slightly severe hose bending may be putting enough side pressure on the block to dislodge it from making good contact.
A 2500K is pretty similar to the 3570K @ stock speeds for a load temps comparison. Below are my stock numbers from several tests I use for overclocking stability for you to compare with
Also note that your H100i should be 3c-4c better than my H100 when all things/settings are equal/stock. My H100 was set on "Low" stock fans @ appx 1350rpm's for the tests below, but I'm also using my CM200mm fan in the pull position for a push/pull setup improving my temps by appx 2c over just the stock fans in push.
2500K stock 3.3GHz (H100L) / A75F [24c] Ambient temps
GSkill RJ-X 2133 CL11 - Man@1866 9-10-9-28 1N 1.6v
tWR-12 / tRFC-130 / tRRD-6 / tWTR-7 / tRTP-7 / tFAW-30
BUS SPEED 100.0-->100.1
CPU-Z - 3404.50 / i.960v L1.208v / 51w
Hyper PI 0.99b - 1M - 11.777s A75F / i32-29-28-30c / L44-43-41-42c
Hyper PI 0.99b - 8M - 2m 11.250s A75F / i32-29-28-30c / L46-45-43-45c
Performance (GFlops) Prob 9977 Mem 768 / 5 runs No Errors 1m 28s
Size LDA Align. Average Maximal A75F / i32-29-28-30c / L48-50-47-47c
9977 9992 4 49.2527 49.2722 RealTemp 60w
XS Bench Real Temp- score 1507 / 9.708s / 3694.89GHz
CPU-Z - 3404.50 / i.960v L1.216v / 78w
P95 Blend (@10min) FFT 448K - A75F / i31-28-28-29c / L49-50-49-48c
P95 Blend (@20min) FFT 8K - A73F / i31-27-28-29c / L50-52-50-50c
IBT v2.54 / Standard - 5 runs 77.84 sec / ram 1024 / 88.6 - 90.2 GFlops
HWMonitor 1.208v / 82.28w A73F / i31-27-28-29c / L48-51-48-49c
All temps were recorded with RealTemp and/or HWMonitor
Sorry didn't have any "screenie's" from base/stock tests
Originally Posted by EniGma1987
I asked pretty much the same question last week and all the charts I saw show the SP120 performance edition being the best, however it is also loud. The AP-15 performs close to the same level at a much lower volume.
... also see [HERE]
for more info/confirmation Edited by TomcatV - 8/10/13 at 5:28pm