Jump to content

H100i v2 unusual temps

I recently purchased an h100i v2 aio from Corsair, to replace my old ugly tower cooler.
Now the idle temps are about 10 degrees higher, and as soon as I start Prime95 stress test, the CPU (7600k @ 5ghz) hits 100 degrees and the system shuts down. That never happened with my tower cooler.  Pump and fans are working, I already checked the mounting and replaced the thermal paste, but no changes to the mentioned points. Only when gaming it behaves as expected, with themps about 10 degrees lower than with the old cooler.
Is this behaviour normal when water cooling? Its my first water cooling attempt ever.

Thanks!

Link to comment
Share on other sites

Link to post
Share on other sites

1 minute ago, DerBurner said:

"Now the idle temps are about 10 degrees higher.."

"Only when gaming it behaves as expected, with themps about 10 degrees lower than with the old cooler..."

So which is it?

Link to comment
Share on other sites

Link to post
Share on other sites

Just now, MVPernula said:

So which is it?

Sorry, forgot to mention.
Before: Idle about 30
             Gaming 70-80
Now: Idle about 40
         Gaming 60-70

Link to comment
Share on other sites

Link to post
Share on other sites

4 minutes ago, DerBurner said:

Sorry, forgot to mention.
Before: Idle about 30
             Gaming 70-80
Now: Idle about 40
         Gaming 60-70

Have you plugged the fans into CPU FAN instead of AIO/PUMP? It allows you to better control the speeds from the bios.

 

I think it has to do with the auto-monitoring of the fans speed, which try to be as silent as possible when their power isn't needed (and <50°C is not really needed for the cpu)

Link to comment
Share on other sites

Link to post
Share on other sites

Just now, DerBurner said:

Sorry, forgot to mention.
Before: Idle about 30
             Gaming 70-80
Now: Idle about 40
         Gaming 60-70

Watercoolers can lie a bit to you more that towers.

If you have a PC under some sort of load for a long period of time you'll find your average temp slowly increasing, that's becuase aluminum is easier to cool than water.

But for it to reach 100c should be a thing..

Link to comment
Share on other sites

Link to post
Share on other sites

4 minutes ago, ErrantNyles said:

Have you plugged the fans into CPU FAN instead of AIO/PUMP? It allows you to better control the speeds from the bios.

 

I think it has to do with the auto-monitoring of the fans speed, which try to be as silent as possible when their power isn't needed (and <50°C is not really needed for the cpu)

I have them plugged into the pump, cause I only have 1 cpu fan header and no fan splitter available. But even if I manually set them to 100% (in Corsair LINK), temps remain as bad, its almost as if the cooler is to slow to keep up with the temps. As is mentioned, the pc shuts down when starting a stress test, because the CPU reaches 100 degrees immediatly, even with fans at 100%.

Link to comment
Share on other sites

Link to post
Share on other sites

@DerBurner You say you rechecked the thermal paste did you check it had been spread out before? This shows that there was contact between the cold plate and the cpu. If the paste doesn't appear flattened when you take the cooler off then you have a contact issue. Are you sure the pump is connected to a header on the board? Run with your case open can you hear the pump? If the contact is fine and your fans spin up then it sounds like either the pump isn't doing anything. Which could either be not connected correctly or a dead pump.

Link to comment
Share on other sites

Link to post
Share on other sites

2 minutes ago, tom_w141 said:

@DerBurner You say you rechecked the thermal paste did you check it had been spread out before? This shows that there was contact between the cold plate and the cpu. If the paste doesn't appear flattened when you take the cooler off then you have a contact issue. Are you sure the pump is connected to a header on the board? Run with your case open can you hear the pump? If the contact is fine and your fans spin up then it sounds like either the pump isn't doing anything. Which could either be not connected correctly or a dead pump.

It was spread out, so no contact issue. Yes, the pump is connected, I can hear it and the software shows its working. When I´m gaming for an hour or so, one tube is hotter than the other, so its working. That´s why I`m asking for help, cause there is no apparent reason (at least for me) for the aio to be worse than the tower.

Link to comment
Share on other sites

Link to post
Share on other sites

54 minutes ago, ErrantNyles said:

Have you plugged the fans into CPU FAN instead of AIO/PUMP? It allows you to better control the speeds from the bios.

Nah, OP is doing it correct by plugging the Fans into the fan headers coming off the pump. If your fans are plugged into the motherboard then the fans on the H100i V2 will ramp up to respond to this temperature increase on the CPU cores. However, the water temperatures are unlikely to change by more than a degree during this period, and since it is the water temperature that ultimately dictates how much cooling you get (rather than the fan speed, which is cooling the water) and so you will not benefit from the increased noise from the fan.

 

The cooling performance of the H100i V2 will only start to degrade when the water temperatures creep up, and so that is really when you want the fans to creep up. This is the design of the product and this is why I think it is actually much more effective to have the fans plugged into the pump itself and should not be perceived as simply extra headers to only be used if you don't have any motherboard ones, since there is no way to monitor the coolant temperature through the motherboard.

 

25 minutes ago, DerBurner said:

It was spread out, so no contact issue. Yes, the pump is connected, I can hear it and the software shows its working. When I´m gaming for an hour or so, one tube is hotter than the other, so its working. That´s why I`m asking for help, cause there is no apparent reason (at least for me) for the aio to be worse than the tower.

The more important thing that recent purchases of H100i V2 may not be aware of is that it expects a constant supply of power rather than a variable PWM/DC. Is the header that you've connected the pump to running at full speed (or constant 100%) in DC mode? The pump is not designed to be voltage controlled, and will behave weirdly if you try to control the pump speed through the BIOS.

Link to comment
Share on other sites

Link to post
Share on other sites

1 hour ago, DerBurner said:

Now the idle temps are about 10 degrees higher, and as soon as I start Prime95 stress test, the CPU (7600k @ 5ghz) hits 100 degrees and the system shuts down

Which version of Prime95? are you burning the CPU with AVX?

Also, what was your previous air cooler, and when did you last take your air cooler readings (ya know, summer time heat?)

If you had a top of the range air cooler like DRP3 or NHD15, then the performance cooling wise is actually comparable.

Link to comment
Share on other sites

Link to post
Share on other sites

1 hour ago, DerBurner said:

Sorry, forgot to mention.
Before: Idle about 30
             Gaming 70-80
Now: Idle about 40
         Gaming 60-70

Also please report your "H100i v2" temperature found on Corsair Link at idle. This is the coolant temperature and so will let us know if your idle temps are sensible. I predict they are around 35 or so based on these numbers. Lastly, what are your idle temps?

Link to comment
Share on other sites

Link to post
Share on other sites

Out of curiosity, have you ever heard a grinding sound coming from your fans or maybe the pump? I had this issue where the pwm fan controller for the H100i was actually undervolting because there wasn't enough voltage coming from the 1 fan header since that's all it was connected to.

 

I solved this issue by connecting it to the two PWM headers I had on the motherboard, but in your case if you could somehow do pwm to molex or pwm to sata, you'd lose the quietness and some connectors, but you might actually get better thermals in the process.

"The only thing that matters right now is that you're here, and you're safe."

Link to comment
Share on other sites

Link to post
Share on other sites

7 minutes ago, Hiitchy said:

-snip-

Never had this problem. which motherboard were you using? Asetek pumps and 2 fans are very low power consumption so should be no problem with normal motherboards

Link to comment
Share on other sites

Link to post
Share on other sites

42 minutes ago, For Science! said:

Nah, OP is doing it correct by plugging the Fans into the fan headers coming off the pump. If your fans are plugged into the motherboard then the fans on the H100i V2 will ramp up to respond to this temperature increase on the CPU cores. However, the water temperatures are unlikely to change by more than a degree during this period, and since it is the water temperature that ultimately dictates how much cooling you get (rather than the fan speed, which is cooling the water) and so you will not benefit from the increased noise from the fan.

 

The cooling performance of the H100i V2 will only start to degrade when the water temperatures creep up, and so that is really when you want the fans to creep up. This is the design of the product and this is why I think it is actually much more effective to have the fans plugged into the pump itself and should not be perceived as simply extra headers to only be used if you don't have any motherboard ones, since there is no way to monitor the coolant temperature through the motherboard.

 

The more important thing that recent purchases of H100i V2 may not be aware of is that it expects a constant supply of power rather than a variable PWM/DC. Is the header that you've connected the pump to running at full speed (or constant 100%) in DC mode? The pump is not designed to be voltage controlled, and will behave weirdly if you try to control the pump speed through the BIOS.

I actually had the CPU fan header still set at a quiet profile, just switched it to max power, no difference.

 

39 minutes ago, For Science! said:

Which version of Prime95? are you burning the CPU with AVX?

Also, what was your previous air cooler, and when did you last take your air cooler readings (ya know, summer time heat?)

If you had a top of the range air cooler like DRP3 or NHD15, then the performance cooling wise is actually comparable.

It´s version 28.10, yeah I´m using AVX.
Previous cooler was the Alpenföhn Brocken 2, and I bought and switched coolers 3 weeks ago, so the temps are rather up-to-date.

 

36 minutes ago, For Science! said:

Also please report your "H100i v2" temperature found on Corsair Link at idle. This is the coolant temperature and so will let us know if your idle temps are sensible. I predict they are around 35 or so based on these numbers. Lastly, what are your idle temps?

Water temperature is about 32 degrees at idle. CPU idle temps are in the 40s.

 

 

Link to comment
Share on other sites

Link to post
Share on other sites

Link to comment
Share on other sites

Link to post
Share on other sites

3 hours ago, For Science! said:

Sorry, I meant ambient temps :P

Those are around 23 degrees :)

Link to comment
Share on other sites

Link to post
Share on other sites

I just noticed another weird behaviour that could help with troubleshooting: When I start a benchmark (not that maximum heat one that instantly shuts down the system due to high temps, Prime95 has several options, here I´m using the one with max power consumption, not the one that produces max heat ) and let it run for some time, temps are settling at around 70-75, which isnt that bad. But at some point all cores suddenly spike up to over 90 degrees, eventually hitting 100 and shutting down the system.  When I intervene and stop and immediately restart the test, the temps go back to 70-75 and stay there, only to spike again for no apparent reason after some time has passed. I didnt start any other programm or task when that happened, I just sat and watched the temps.

Link to comment
Share on other sites

Link to post
Share on other sites

9 hours ago, DerBurner said:

=snip=

So apart from your spikes in prime95 everything seems normal to me. Maybe consider using something slightly easier on your CPU like Realbench so at least you can do diagnostics without your PC killings itself as a ball of fire....Sorry I can't be any more help

Link to comment
Share on other sites

Link to post
Share on other sites

12 hours ago, DerBurner said:

I just noticed another weird behaviour that could help with troubleshooting: When I start a benchmark (not that maximum heat one that instantly shuts down the system due to high temps, Prime95 has several options, here I´m using the one with max power consumption, not the one that produces max heat ) and let it run for some time, temps are settling at around 70-75, which isnt that bad. But at some point all cores suddenly spike up to over 90 degrees, eventually hitting 100 and shutting down the system.  When I intervene and stop and immediately restart the test, the temps go back to 70-75 and stay there, only to spike again for no apparent reason after some time has passed. I didnt start any other programm or task when that happened, I just sat and watched the temps.

fun fact, the reason why gaming is normal is because you are not using 100% cpu usage while gaming. Stress tests on the other hand do 100% for a consistent period of time but you should test at 100% for 24 hours minimum to make sure that you do not have the issue you are currently running into. How is the configuration of your loop? is the reservoir higher then the pump? Is the liquid flowing fast enough? You don't need it maxed but having a middle ground speed is best in my opinion to make sure that liquid isn't sitting for too long. 

Link to comment
Share on other sites

Link to post
Share on other sites

18 hours ago, For Science! said:

So apart from your spikes in prime95 everything seems normal to me. Maybe consider using something slightly easier on your CPU like Realbench so at least you can do diagnostics without your PC killings itself as a ball of fire....Sorry I can't be any more help

No worries, thank you very much for your help :)

 

15 hours ago, TheSinisterGamer said:

fun fact, the reason why gaming is normal is because you are not using 100% cpu usage while gaming. Stress tests on the other hand do 100% for a consistent period of time but you should test at 100% for 24 hours minimum to make sure that you do not have the issue you are currently running into. How is the configuration of your loop? is the reservoir higher then the pump? Is the liquid flowing fast enough? You don't need it maxed but having a middle ground speed is best in my opinion to make sure that liquid isn't sitting for too long. 

Of course I know that gaming does not stress the system in the same way as a stress test does. My whole problem is that I (currently) am unable to run any stress test for more than half an hour due to very strange temperature behaviours that (in my mind) shouldn't occure with a water cooled CPU, given a proper installation and a working cooler, which I'm sure it is. By the way, I'm using an AIO, so no dedicated reservoir, I have the radiator mounted in the front of my NZXT s340 elite case.

Link to comment
Share on other sites

Link to post
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

×