Overheating CPU with H80i GT Cooler

Hi guys,

My cooler seems to be defective from day one but, I thought this is how usually this CPU cooler works as it had a minor issue of keeping the CPU temps in the normal range and it took me 3 years to figure this out that this is a faulty one. I mostly do 3D Rendering so when my CPU load reaches 100% load, it reaches 100 deg C straight away and I thought this was normal because the system was rendering heavily and then thermal throttling starts kicking in but it gets back to 50 deg C right after rendering process is finished and when it has no load or 0% load on the CPU. My mobo temp was around 40ish degrees all the time and it never went up to 50 deg C. I did check all the mounts on the CPU and Cooler and I'm pretty sure that's it's not because of the mounts being loose or less in contact with the CPU since the mounts were super tight like almost feels bending the Mobo as I tight the thumbscrews of the cooler. I hope everyone knows that there's a flaw with the mounting bracket when you fix it on Asus z97 Pro. The mounting screws go back and forth if you fix it without the washers which are not provided by Corsairs so I had to use few washers from the local market which turns out to be really good in holding up the mounts very tightly. I have assembled a lot of Machines in the past so I'm very well thorough with the technical aspect of all these hardware plus I like assembling expensive hardware. :)

However, yesterday I thought to clean up the system since it was all dusty. I removed GFX card, RAM and then Cooler. I took out the fans from the radiator and cleaned up all the carbons and dust deposits. I also cleaned up the thermal paste with rubbing alcohol and applied the Arctic MX-4 Thermal paste on it which I believe is the ultimate thermal compound available in the market. I did everything with gentle care and fixed the Cooler back on to the CPU mount.

After turning on, I get CPU overheat warning instantly in the BIOS and saw that CPU was at 85 Deg C already, I closed everything immediately. I had to remove the cooler to see what's the problem is. I did a thorough check of backplate mount and mounting screws and everything seemed OK to me. I even checked the formation of the paste applied on the Cooler and CPU and it was really good, evenly spaced in a square form on both the end of the CPU and Cooler. I removed those paste with Alcohol and reapplied using the 5 point method on the CPU. I booted the System and saw it was at 70 Deg C in the BIOS and then gradually went upto 90 Deg C in the BIOS itself, still no luck. I again removed the Cooler and applied the paste again after cleaning up with Alcohol and this time I shook the radiator a little hard before mounting. After booting up, the temps were fluctuating from 55 to 65 and 70-degree celsius. I gave a hard tap a couple of times on the radiator thinking the coolant will pump more. But, the temps were now in between 65 and 70 degrees so I thought to load and initiate the windows to see how much the corsair link shows and if there's any fix to it.

I booted up the system and saw the CPU temps were fluctuating from 70 to 90 a lot on idle. Like for instance, if there's a CPU load of about 20% then temps reach till 100 deg C up to the point where thermal throttle starts kicking in. I shook the radiator a little hard this time again after removing it and it worked a little for some time. The temps were at 50 and 65 but reached only 90-95 deg C even if there's any CPU load of about 10%. I noticed that one pipe is really hot and the other is cold. I tapped the other pipe which was cold but no heat sensation felt and felt no heat on that side of the radiator as well where the cold pipe is connected. After feeling the radiator in my hand, I noticed that one side of the radiator is hot and the other was dead cool. I don't know what this means but I suspect that it is faulty for sure. I started up the system next morning (7 hours later ) and the CPU temps are now at 90 Deg C. It seems like coolant is not cooling up or not functioning properly as it should. The current temps are 98 Deg C as we speak. I have pics and videos but not sure if I'm allowed to post it here?

In short, after cleaning up and fixing the cooler 3rd time on my CPU, The idle temps were stable at 65 Deg C but on the next day, it's idling at 100 degrees C constant which is what the temps are now of the CPUs. If I reboot now then BIOS will stop me in proceeding with the boot right after it throws a "CPU over heating error". Please help!

Shall I RMA this device or are there any fix to it? Let me know what you guys think. Thanks!

My Specs;

OS: Windows 10 Pro 64-bit

CPU: Intel Core i7 4790K @ 4.00GHz

RAM: 16.0GB Dual-Channel DDR3 @ 1199MHz (11-13-13-31)

Motherboard: ASUS Z97-PRO(Wi-Fi ac) (SOCKET 1150)

Graphics: 2047MB NVIDIA GeForce GTX 660 Ti (EVGA)

Monitor: DELL S2415H (1920x1080@60Hz)

Contact Corsair through the support ticket system in the menu header above to start the RMA. You have a blockage of some type and it is not uncommon for that specific model series. There is a lot of info above, but the key points for your support ticket are:

1) While in the BIOS you see the cpu temp gradual but continuously rise. That is rapid coolant heating resulting from the low flow state. When you have a cpu contact issue the temp goes from room temp to 90-100 the millisecond voltage is applied. You never see it “rise”. Smooth and gradual increases like that are coolant temperature changes. If you make it to the desktop, the H80i Temp in Link will be somewhere in the 50-70C range. That is way out of bounds.

2) Hot and cold pipes. Dead give away and this should never happen. The most the radiator can drop coolant in one pass is 1-2C. That is below touch perception and they should always feel the same.

3) Coolant temp is approximately equal to radiator exhaust temp. As the cpu and coolant heat up, you will likely feel cool air coming off the radiator instead of what should be matched to the H80i Temp. A 62C coolant temp and cold air exhaust is another clear indicator of a cooler issue.

4) The “hitting”. Yup, this actually is a troubleshooting method. Sometimes you get lucky and the blockage dislodges, but the stuff has nowhere to go. It will come around again. Either way, contact/mounting issues don’t have temp changes when you fiddle with the hoses. This is a cooler issue.

Try not to use the pc much until you get another cooling solution in place, either a new water system or a cheap stand-in air cooler. You are in a vulnerable state at this point.

Thank you c-attack for replying to my post, much appreciated. My cooler is h80igt and you were referring to h100i but yeah its more of the same thing I guess.

Now it's even worse like when I switch on my system I see the temp in bios at 70 and rises to 80 C like in 30 secs which means I can never power on my system until I get a replacement or a new one. I've contacted their support already and will proceed with RMA.

Know any latest cooler that supports z97 pro lga 1150?

Thanks again c-attack for helping me out with your inputs as I was going nuts because of this issue...cheers!

Yes, apparently I had H100's on the brain. All of the Corsair coolers will be compatible for Intel 115x right out of the box, including the Pro and Platinum line.

This is always a tough spot if you cannot afford to be down very long on your PC. My general advice is get a cheap air cooler that becomes your permanent "emergency back-up" for this kind of stuff or go shopping for a new model and make the RMA the back-up. I used to keep a small air tower for this, but as RAM got taller, heatsinks got bigger, putting that thing in became a 4 hour job. In contrast, I can swap a AIO cooler in and out in 5 minutes. Now I have an AIO backup. The problem is those are a perishable item after a certain number of years. You might give some thought to picking up something new, especially if you have the option to go up a size (240/280mm or more), but this is very much case and use specific.