CPU overheating with corsair h100i v2 cooler

Status
Not open for further replies.

Oats2010

Commendable
Jan 4, 2017
9
0
1,510
Last night I woke up after drifting off to find my cooler's fans were at max. I shut my computer off and went back to sleep without really thinking much of it.
This morning I turned it back on to find my cpu overheating (up to 100 degrees celcius). The cooler's "corsair" light was solid red, I'm guessing because of high temps. So I dusted my computer out good, reapplied new thernal compound, and made sure it was seated good. Nothing seems to be working. When I start it up, after a couple mins the temps are back up in the 80s/90s.
I've been online all day trying to find out the problem but I'm kind of at a loss now. If anyone could help, that would be great.
 
Solution
Oats2010,

On behalf of Tom's Moderator Team, welcome aboard!

What you've described is a pump failure, which is a very common problem. It's inevitable that all pumps will fail, which is the reality of liquid cooling. It's not a question of "if", it's a question of "when". Depending on the pump manufacturer and quality, pumps only have a finite number of operational hours until failure is eminent. It's called Mean Time Before Failure (MTBF), which is rated in thousands of hours, and is typically overestimated.

There's 8,760 hours in a year. If you leave your rig running 24/7/365, then you've put needless hours on your pump, fans (dust) and hard drive(s), all of which have moving parts with motors and bearings that will eventually...

Oats2010

Commendable
Jan 4, 2017
9
0
1,510
Well I wiped all the old stuff off then added a small line 1mm thick like the Arctic silver instructions on the website said, but I think I got a bit much because it was still running hot. Then I removed some to where it seems to have a thin layer across the entire surface but not thin enough to see through to the lettering on the cpu.
 

Tumeden

Honorable
Oct 15, 2016
449
0
11,160


You should completely clean off the CPU before reapplying new thermal paste, and be sure to not actually touch the top of the CPU as the oils on your fingers will cause issues. Personally I use like 5-10 qtips to gently remove all of the old thermal paste, it usually does the trick. One pea size dab in the very center of the CPU before placing the heatsync back on is plenty.. any more then that and it's too much.
 

Oats2010

Commendable
Jan 4, 2017
9
0
1,510
Ok I'll try to wipe it all off and reapply a pea size dab on there then I'll post back to let you know if it worked. (Not sure if this is how to reply, I'm new to the site and on mobile). It should cover the entire cpu without spilling over the edges, right?
 

Tumeden

Honorable
Oct 15, 2016
449
0
11,160


Once the heatsync is properly seated, then it will spread it evenly over your CPU, you should never add too much as it causes a thermal barrier and causes your CPU to overheat. Just a small pea size in the center.
 

Oats2010

Commendable
Jan 4, 2017
9
0
1,510
I thought it might be the pump also and don't know how to check, other than feeling the 2 tubes and when I do that, one feels warm and one feels cool so I would assume it's pumping. I really don't know though other than feeling the tubes
 

CompuTronix

Intel Master
Moderator
Oats2010,

On behalf of Tom's Moderator Team, welcome aboard!

What you've described is a pump failure, which is a very common problem. It's inevitable that all pumps will fail, which is the reality of liquid cooling. It's not a question of "if", it's a question of "when". Depending on the pump manufacturer and quality, pumps only have a finite number of operational hours until failure is eminent. It's called Mean Time Before Failure (MTBF), which is rated in thousands of hours, and is typically overestimated.

There's 8,760 hours in a year. If you leave your rig running 24/7/365, then you've put needless hours on your pump, fans (dust) and hard drive(s), all of which have moving parts with motors and bearings that will eventually wear out. If a bearing is beginning to seize, it can cause a pump head or the motor to intermittently run too slow or just stop. Also, a worn sloppy bearing can cause a seal to leak, which we all dread.

Sorry, but even though you still have some thermal differential between your hoses, it sounds like you'll need to RMA / replace your cooler.

CT :sol:
 
Solution

Oats2010

Commendable
Jan 4, 2017
9
0
1,510
Thanks for the info computronix. I've had it for about 8 months (I think). But I do leave it running all the time so I'll quit doing that from now on. I didn't think about all the moving parts. Do you think I should go with a regular fan cooler? If so, which do you recommend?
 

Oats2010

Commendable
Jan 4, 2017
9
0
1,510
I live in the Midwest and have AC, so it stays pretty cool but can get a little warm in the summer. I'm running an i7 4790k, and no I haven't over clocked yet. I got into computers about a year and a half ago so I'm not too familiar with overclocking.
 

Oats2010

Commendable
Jan 4, 2017
9
0
1,510
About the same as the h100i so around $120 I'd say. Also I installed the stock cooler and temps went way down finally. But they randomly spike up to 60 but go right back down to 30s and 40s so could it be too much thermal compound? Or maybe it needs to settle?
 

gladiusz600

Prominent
Jan 9, 2018
1
0
510
Ok so I have had the same prob But Corsair Link tells me that the pump is running and the fan is running. So be for you go pulling you're Pc apart see what Corsair Link is telling you. Also AI Suite is telling me the same thing all is running fine.

So there has been an update from Mirosoft that has screwed with peoples CPU's https://doublepulsar.com/important-information-about-microsoft-meltdown-cpu-security-fixes-antivirus-vendors-and-you-a852ba0292ec

So I had a fixed yesterday CPU was Idel @ 35-36C nps under load 67c Max ,Now as of this morning with another "Fix" from MS it is overheating again. I have NO Solution as of yet.
 
Status
Not open for further replies.