Jump to content
Corsair Community

Very high temps on aio (Corsair h100i v2)


Recommended Posts

It all started when I ordered a 5 pack of cheap Chinese rgb fans to put in my pc, when swapping out all of the fans I had the radiator hanging down from the pump, I then put everything back in and immediately on boot I go on nzxt cam I see my cpu is going mental, hit 100C and turned off. I've tried everything, vertically mounting it horizontally mounting it having the rad upside down having the rad the right side up. I even took off the aio and let the air go to the top of the rad then put it back in. When I didn't have the aio installed I went to my old wraith prism. On windows start-up it was at about 50c and gaming never went above 75C. When i tried taking the prism off the cpu stuck to it and i bent some pins in the process, fixed it put it all back together with new thermal paste and clamped the aio tight. Upon bootup I went straight to BIOS where i had just cleared the cmos so I put xmp on and all of my fans to max, one thing to note is that all my fans are connected to a Chinese fan hub which is powered by something (4holes connected) I used the remote I got with these fans to turn up the speeds to max but idk. Boot into windows temps go straight to about 65-75C literally just booted in. I hop onto csgo and monitered the temp, it was at 30%usage and it went from 70 to 85 where I decided enough was enough. My aio should be sat at a maximum of 60c while gaming and while idle from 20-40c. Idk what the problem is tbh.

My cpu is the ryzen 7 2700x and my aio is the corsair h100i v2.

Link to comment
Share on other sites

You need to look at the coolant temperature (H100i Temp) in CUE.  This is the measure of the cooler's efficiency and a good indicator of overall health.  Most people see an idle coolant temp of +4-7C above room temperature.  However, if you have some type of blockage or other flow issue the H100i Temp will go above 40C immediately at boot and then slowly and steadily continue to climb.  This can usually be observed on the BIOS level as well.  If you boot into the BIOS and see the CPU temp skipping upwards 45...50..55..60...80..85.. and up, you don't have moving liquid regardless of what the pump RPM reads.  That is usually the end of the line for an AIO.

Link to comment
Share on other sites

7 minutes ago, c-attack said:

You need to look at the coolant temperature (H100i Temp) in CUE.  This is the measure of the cooler's efficiency and a good indicator of overall health.  Most people see an idle coolant temp of +4-7C above room temperature.  However, if you have some type of blockage or other flow issue the H100i Temp will go above 40C immediately at boot and then slowly and steadily continue to climb.  This can usually be observed on the BIOS level as well.  If you boot into the BIOS and see the CPU temp skipping upwards 45...50..55..60...80..85.. and up, you don't have moving liquid regardless of what the pump RPM reads.  That is usually the end of the line for an AIO.

So nothing I can do about it? Is this rare or not. I don’t plan on forking out another 150 for an aio!

Link to comment
Share on other sites

55 minutes ago, E36jacob said:

So nothing I can do about it? Is this rare or not. I don’t plan on forking out another 150 for an aio!

Not really. Whether or not you'll need to fork out for another AIO depends on how old the AIO is - if you are still under warranty, Corsair should replace it.

Link to comment
Share on other sites

All-In-One liquid coolers are no maintenance products.  You don't have to care for it in the same way as a custom water cooling set-up.  However, that means eventually the coolant levels will evaporate since you can't replace or the cooling block will get gunked up.  The warranty period is 5 years, so that gives you some idea of the expected lifetime.  

 

It might be a good idea to verify the suggested temperatures before calling it a failure.  There are other possibilities if the coolant temp is within expected levels.  

Link to comment
Share on other sites

Ok I’m in iCue at the moment pump on extreme (3150 rpm) and the temp says 37C. My nzxt can is spiking from 45-65 when opening apps not even in game. What does this mean?

Link to comment
Share on other sites

Why are you running CAM? Do you have any NZXT devices as well?

And the spikes when you open apps are normal - it's a burst of CPU activity. This is pretty common with modern chips.

What's more concerning is that your coolant temp is 37C at idle. Have you been playing any games lately or stressing the system in other ways? How is your radiator's airflow configured? (e.g. placement in the case and airflow direction)

Link to comment
Share on other sites

So there are two temps involved when liquid cooling, fluid or component being cooled, and it gets super hard to tell which you might be referring to in several of the posts above. So, the CPU temp is 37C or the fluid temp? And the spiking from 45-65 is that also the CPU? Fluid should not get up to 65, that is bad. A CPU up to 65 should be no concern but fluid should not be more than about 15 C more than the room's air temp. Please always refer to fluid or CPU when mentioning a temp.

These "cheap Chinese" fans you installed, are they blowing into the radiator or pulling through? Cheap fans make so-so blow through and really crappy at pull through. If you have them in pull through I suggest reversing them. If you still have the original fans maybe there is room to put them on the other side so the radiators get air coming and going?

Link to comment
Share on other sites

49 minutes ago, E36jacob said:

Ok I’m in iCue at the moment pump on extreme (3150 rpm) and the temp says 37C. My nzxt can is spiking from 45-65 when opening apps not even in game. What does this mean?

As mentioned above, these are two different values.  The coolant/liquid temp (H100i v2 Temp) is the baseline or lowest possible CPU temp with zero voltage.  The radiator and fans are adjusting the baseline and this is a slow moving variable.  The CPU temp is directly related to the amount of voltage applied in any on instant and is added/removed at in fractions of a second.  This will often make the CPU temp appear to jump around when the monitoring program polls every 1-2 seconds.  It would be very normal to see CPU temps of 45-65C with a coolant temp baseline of 37C.  

 

37C is a bit warm for baseline temp, but this is also directly affected by room/case temperature like everything else.  A 37C power on temp in the middle of Winter in a 16C room is cause for concern.  A 37C power on temp in a 29C room on a hot late Summer afternoon is just about right.  The key to distinguishing "hot environment" from problem is what happens next.  An AIO with any kind of blockage or flow problem will not be able to maintain a steady coolant temp, even at idle.  So if you power on in your 27C room and the coolant starts at 29C and moves up to 35C and stops, this is normal.  If you power on and sit on the desktop and the coolant slowly but continuously climbs over time and never comes back down, there is a developing problem. 

 

The trick is to test this at idle or by using a CPU only load, something mild like the stress test in CPU-Z.  It's linear so you can see the +1C coolant to +1C CPU relationship during the load.  You only need to run for 5 min and then stop.  Warning signs are a large increase in coolant temp during the 5 min (more than +10C) and then especially what happens after you stop the load.  The coolant temp should start dropping 1C every 20-30 seconds.  If it just hangs there and can't release the heat, there may be an issue.  That is also the problem with using gaming as the test data.  The GPU can heat up the local area and bring the coolant temp up along with it.  Then you quit the load, but the coolant temp still reads 40C because everything around it is also 40C.  This makes it look like a problem with the cooler, but its just the normal hot air and case ambient temp is the real baseline coolant temp.  

Link to comment
Share on other sites

Right just read through that all sorry all that I didn’t clarify. The 37 was in CL4 and was the coolant temp (I didn’t see what it was in game, sorry) and the 65-75+ was NZXT CAM software. I don’t own anything nzxt however I just use the software for monitoring. I don’t really stress the cpu out at all especially these past weeks and months as I’ve only been playing csgo which barely takes 40% load on it and has been running great. No overclock on it just cpb like everyone has. Now my graphics card does get hot as it is an rtx 2080 but with a blower style fan (I know dreadful looking to sort this soon) I have it undervolted but at idle the temps for the 2080 where no more than 54° while the cpu was going from 45 on startup to 70 something idle just light monitoring with a few apps open. Now my configuration. I have the Corsair 275r carbide and I have the aio vertically mounted with the coolant lines at the bottom of the rad. I have 3 exhausts, one at the back and two up top all 120mm. Only intake is the pull through configuration on the rad. The top of the rad is higher than the pump and cpu. I have also tried this with the coolant line at the top of the rad no difference. My fan configuration is pull through, I can feel it when it’s on if I put my hand next to it but yeah okay, I will try swapping the fans over one last time before I try and shake the pump and rad as it could be blocked up. I’m not sure about the warranty because I did buy this second hand and asking the guy I bought it from says he doesn’t have the receipt. So yes thanks all again hopefully I can sort this out with the help of you guys! Cheers, Jacob.

Link to comment
Share on other sites

15 hours ago, c-attack said:

As mentioned above, these are two different values.  The coolant/liquid temp (H100i v2 Temp) is the baseline or lowest possible CPU temp with zero voltage.  The radiator and fans are adjusting the baseline and this is a slow moving variable.  The CPU temp is directly related to the amount of voltage applied in any on instant and is added/removed at in fractions of a second.  This will often make the CPU temp appear to jump around when the monitoring program polls every 1-2 seconds.  It would be very normal to see CPU temps of 45-65C with a coolant temp baseline of 37C.  

 

37C is a bit warm for baseline temp, but this is also directly affected by room/case temperature like everything else.  A 37C power on temp in the middle of Winter in a 16C room is cause for concern.  A 37C power on temp in a 29C room on a hot late Summer afternoon is just about right.  The key to distinguishing "hot environment" from problem is what happens next.  An AIO with any kind of blockage or flow problem will not be able to maintain a steady coolant temp, even at idle.  So if you power on in your 27C room and the coolant starts at 29C and moves up to 35C and stops, this is normal.  If you power on and sit on the desktop and the coolant slowly but continuously climbs over time and never comes back down, there is a developing problem. 

 

The trick is to test this at idle or by using a CPU only load, something mild like the stress test in CPU-Z.  It's linear so you can see the +1C coolant to +1C CPU relationship during the load.  You only need to run for 5 min and then stop.  Warning signs are a large increase in coolant temp during the 5 min (more than +10C) and then especially what happens after you stop the load.  The coolant temp should start dropping 1C every 20-30 seconds.  If it just hangs there and can't release the heat, there may be an issue.  That is also the problem with using gaming as the test data.  The GPU can heat up the local area and bring the coolant temp up along with it.  Then you quit the load, but the coolant temp still reads 40C because everything around it is also 40C.  This makes it look like a problem with the cooler, but its just the normal hot air and case ambient temp is the real baseline coolant temp.  

Right I’ve just ran the cpuz stress test, at the start I had the pc running at idle for 5 mins, coolant was at 32-33c and cpu was 45-60C. I ran the test and in the first maybe 20-30 seconds the temps shot right up. The coolant temps also rose to about 36. I stopped it when it got to 85C and was thermal throttling the cpu to 3700mhz. I turned it off and the temps dropped on the cpu quite a lot quite fast and it was down to 60° in about 20-30 seconds. The h100i took a bit longer maybe a minute to get back down to 33-34C. What do I do, this is still with the Chinese fans.

Link to comment
Share on other sites

No the fans affect the coolant temperature and not the cpu directly. They matter more over a long duration and have no impact on the immediate load temp. +4C coolant is as expected and the difference between good and terrible fans is just a few degrees. 
 

The notable value is 85C cpu with 36C coolant. That is a +50C difference and typically what someone running a high overclock will see. So either you have a lot of voltage being put through the cpu or it’s possible the contact between cpu and cold plate is not optimal. 

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

9 hours ago, E36jacob said:

the 65-75+ was NZXT CAM software. I don’t own anything nzxt however I just use the software for monitoring

So I'm assuming that's your CPU temperature? And why are you using CAM if you don't have any NZXT products? There are plenty of other things that will do the same monitoring (including iCUE) and won't present issues with potential conflicts. Both NZXT CAM and iCUE, by the way, use the same library to monitor system sensors (CPUID).

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

×
×
  • Create New...