Jos Hideky Posted February 5 Share Posted February 5 (edited) Hi im using 13900KF with H150i i have also CPU temp issue, at 100c when stress test with Cinebench R23 the load curve is attached with the temp, watt tdp and clock speed. The picture is at the end of 10 min test, when the temp, clock and power fall down. Is my installation of the AIO or the AIO itself have problem when the CPU temp at constant 100c and the coolant temp is only at 38~41 at the whole time when test going ? Its like there is no heat transfer from the CPU to the AIO as its show the coolant temp is not much differ (only raise about 3-4c while the CPU raise from 40c to 100c) or this is normal considering that after the test finish the temp also almost instantaneously cool down ? Any answer from other experienced user will be much appreciated thank you Edited February 5 by Jos Hideky Link to comment Share on other sites More sharing options...
LeDoyen Posted February 5 Share Posted February 5 it's normal. the power limits are disabled in the bios. your AIO can't cope with that. you'd need custom loop cooling to do it and still, it would be quite hot. you should look up in the mobo manual how to set power limits back to what Intel normally sets, or how to disable their version of "multicore enhancement", that handy option that let a cpu boost as high as it wants. 1 Link to comment Share on other sites More sharing options...
PAlvarez05 Posted February 5 Share Posted February 5 1 hour ago, Jos Hideky said: Hi im using 13900KF with H150i i have also CPU temp issue, at 100c when stress test with Cinebench R23 the load curve is attached with the temp, watt tdp and clock speed. The picture is at the end of 10 min test, when the temp, clock and power fall down. Is my installation of the AIO or the AIO itself have problem when the CPU temp at constant 100c and the coolant temp is only at 38~41 at the whole time when test going ? Its like there is no heat transfer from the CPU to the AIO as its show the coolant temp is not much differ (only raise about 3-4c while the CPU raise from 40c to 100c) or this is normal considering that after the test finish the temp also almost instantaneously cool down ? Any answer from other experienced user will be much appreciated thank you How many watts are you putting out? 1 Link to comment Share on other sites More sharing options...
Jos Hideky Posted February 5 Share Posted February 5 Just now, PAlvarez05 said: How many watts are you putting out? Uhm you can see in the pic, its almost constant 311w, except when the cinebench reload the pic from the beginning, you can see the dip in the cpu power and also in the temp Link to comment Share on other sites More sharing options...
PAlvarez05 Posted February 5 Share Posted February 5 (edited) 3 minutes ago, Jos Hideky said: Uhm you can see in the pic, its almost constant 311w, except when the cinebench reload the pic from the beginning, you can see the dip in the cpu power and also in the temp Sorry, I was looking at it from my phone. You are maxing out the cooling capabilities of the cold plate, is unable to dissipate anymore heat. I have the bigger h170i with six fans and i max out around 380-390 watts, so i assume you'd max out around 320-330. FYI I also have a 13900k, Edited February 5 by PAlvarez05 1 Link to comment Share on other sites More sharing options...
Jos Hideky Posted February 5 Share Posted February 5 6 minutes ago, LeDoyen said: it's normal. the power limits are disabled in the bios. your AIO can't cope with that. you'd need custom loop cooling to do it and still, it would be quite hot. you should look up in the mobo manual how to set power limits back to what Intel normally sets, or how to disable their version of "multicore enhancement", that handy option that let a cpu boost as high as it wants. But why the coolant temp is only at 38~41c ? From my understanding coolant temp can be much higher right ? Is like there is no heat transfer from the CPU to the coolant temp .. Link to comment Share on other sites More sharing options...
Jos Hideky Posted February 5 Share Posted February 5 1 minute ago, PAlvarez05 said: Sorry, I was looking at it from my phone. You are maxing out the cooling capabilities of the cold plate, is unable to dissipate anymore heat. I have the bigger h170i with six fans and i max out around 380-390 watts, so i assume you'd max out around 320-330. Yeah, how bout your coolant temp ? is it also around 40c no matter what happen ? Link to comment Share on other sites More sharing options...
PAlvarez05 Posted February 5 Share Posted February 5 Just now, Jos Hideky said: But why the coolant temp is only at 38~41c ? From my understanding coolant temp can be much higher right ? Is like there is no heat transfer from the CPU to the coolant temp .. Yeah it can go higher, but the cold plate isn't able to efficiently transfer the temp into the coolant anymore, hence why it stays the same You are throwing more heat out than the cold plate/pump can handle 1 Link to comment Share on other sites More sharing options...
Jos Hideky Posted February 5 Share Posted February 5 (edited) 3 minutes ago, PAlvarez05 said: Yeah it can go higher, but the cold plate isn't able to efficiently transfer the temp into the coolant anymore, hence why it stays the same You are throwing more heat out than the cold plate/pump can handle Even at idle, the coolant temp is sitting around 36c and CPU at 40 Full load, coolant temp 40c and CPU 100 Is it useless to try to reinstall the thermal paste ? Do you ever try to full load your CPU ? Stress test ? Can you print screen the load curve and put it here ? i want to compare with the one that installed correctly .. Edited February 5 by Jos Hideky Link to comment Share on other sites More sharing options...
PAlvarez05 Posted February 5 Share Posted February 5 Just now, Jos Hideky said: Even at idle, the coolant temp is sitting around 36c and CPU at 40 Full load, coolant temp 40c and CPU 100 Is it useless to try to reinstall the thermal paste ? Hmm... what voltage are you running the cpu at? 1 Link to comment Share on other sites More sharing options...
LeDoyen Posted February 5 Share Posted February 5 The CPU temperature is much more related to Vcore than to the cooler attached. If Vcore is excessive, the silicon will be a lot hotter than the IHS, that's just how CPUs are. the transistor density is enormous compared to GPUs that have a much larger die. They are that much harder to cool down. A temperature delta of 40 - 50° is not uncommon. On GPUs, the die can get 10 - 15 hotter than water, usually not more. Here you have close to 70° of difference which points to either.. a contact issue or excessive Vcore. Seeing how much power the CPU draws, it's clearly Vcore. It wouldn't hold 5.7Ghz with a contact issue, but throttle heavily half way down or more and not register anywhere close to 300W. It's just too much power for the AIO, you have to reduce power limits where it stays maybe close to throttling but doesn't get there, or just restore Intel limits (Boost to 253W, then step back to 125 after boost expiration). 2 Link to comment Share on other sites More sharing options...
PAlvarez05 Posted February 5 Share Posted February 5 Just now, LeDoyen said: The CPU temperature is much more related to Vcore than to the cooler attached. If Vcore is excessive, the silicon will be a lot hotter than the IHS, that's just how CPUs are. the transistor density is enormous compared to GPUs that have a much larger die. They are that much harder to cool down. A temperature delta of 40 - 50° is not uncommon. On GPUs, the die can get 10 - 15 hotter than water, usually not more. Here you have close to 70° of difference which points to either.. a contact issue or excessive Vcore. Seeing how much power the CPU draws, it's clearly Vcore. It wouldn't hold 5.7Ghz with a contact issue, but throttle heavily half way down or more and not register anywhere close to 300W. It's just too much power for the AIO, you have to reduce power limits where it stays maybe close to throttling but doesn't get there, or just restore Intel limits (Boost to 253W, then step back to 125 after boost expiration). Yeah i just asked him about the vcore, sounds more like a vcore issue, than a cooler issue Link to comment Share on other sites More sharing options...
LeDoyen Posted February 5 Share Posted February 5 With the motherboard power limits disabled, they usually boost vcore quite generously to keep stable even on the worst silicon, so it will be excessive ^^' Link to comment Share on other sites More sharing options...
PAlvarez05 Posted February 5 Share Posted February 5 Just now, LeDoyen said: With the motherboard power limits disabled, they usually boost vcore quite generously to keep stable even on the worst silicon, so it will be excessive ^^' He can keep max power limits and undervolt the vcore to get a better stable OC. It just takes time Link to comment Share on other sites More sharing options...
Jos Hideky Posted February 5 Share Posted February 5 6 minutes ago, PAlvarez05 said: Hmm... what voltage are you running the cpu at? Uhm what is the default volt ? I really new at this, i never change anything for the CPU settings Link to comment Share on other sites More sharing options...
PAlvarez05 Posted February 5 Share Posted February 5 Just now, Jos Hideky said: Uhm what is the default volt ? I really new at this, i never change anything for the CPU settings But you are overclocking though? You have an all core frequency of 5.8, it shouldn't boost that high FYI Cinebench r23 is not a good stress tester VCore default varies by cpu, is all about the silicon lottery 1 Link to comment Share on other sites More sharing options...
Jos Hideky Posted February 5 Share Posted February 5 7 minutes ago, LeDoyen said: The CPU temperature is much more related to Vcore than to the cooler attached. If Vcore is excessive, the silicon will be a lot hotter than the IHS, that's just how CPUs are. the transistor density is enormous compared to GPUs that have a much larger die. They are that much harder to cool down. A temperature delta of 40 - 50° is not uncommon. On GPUs, the die can get 10 - 15 hotter than water, usually not more. Here you have close to 70° of difference which points to either.. a contact issue or excessive Vcore. Seeing how much power the CPU draws, it's clearly Vcore. It wouldn't hold 5.7Ghz with a contact issue, but throttle heavily half way down or more and not register anywhere close to 300W. It's just too much power for the AIO, you have to reduce power limits where it stays maybe close to throttling but doesn't get there, or just restore Intel limits (Boost to 253W, then step back to 125 after boost expiration). Ok so no amount of cooler going to effect the CPU temp ? Because from what i can see, the AIO is not that hot with coolant temp 40~41c right ? So maybe reapply the thermal paste ? Link to comment Share on other sites More sharing options...
Jos Hideky Posted February 5 Share Posted February 5 (edited) 1 minute ago, PAlvarez05 said: But you are overclocking though? You have an all core frequency of 5.8, it shouldn't boost that high FYI Cinebench r23 is not a good stress tester VCore default varies by cpu, is all about the silicon lottery Oic, thnx for the info .. Maybe the one that installing the PC overclocked it ? Me myself not touching anything in the MSI afterburner, installing it just for monitoring purposes. what is your coolant temp when underload ? Edited February 5 by Jos Hideky Link to comment Share on other sites More sharing options...
LeDoyen Posted February 5 Share Posted February 5 Not overclocking ^^ it's what happens when power limits are disabled by motherboard manufacturers out of the box. People get a system that heats up because the CPU doesn't behave as it was designed to. It needs either manual tweaking (longer, needs some savvy on overclocking), or set power limits (easier) and let the CPU reduce its own vcore as it wants. If you are not too savvy when it comes to cpu tweaking and testing, it may be easier to just reduce the power limits in bios, and the CPU will just adapt to it and reduce the Vcore on its own. Link to comment Share on other sites More sharing options...
PAlvarez05 Posted February 5 Share Posted February 5 Just now, Jos Hideky said: No, no overclocking, im not touching anything in the MSI afterburner, installing in for monitoring purposes. what is your coolant temp when underload ? MSI afterburner is for gpu overclocking, not cpu overclocking, Coolant temp depends on ambient temp, so you can’t compare mines to yours my Copland idles at 30 and under load I’ve seen 45ish Link to comment Share on other sites More sharing options...
Jos Hideky Posted February 5 Share Posted February 5 Just now, PAlvarez05 said: MSI afterburner is for gpu overclocking, not cpu overclocking, Coolant temp depends on ambient temp, so you can’t compare mines to yours my Copland idles at 30 and under load I’ve seen 45ish Ah ok2 haha sorry really not savvy with this ... Owh so your highest coolant temp is also at 45ish ? And is your 13900k is at 100c too ? Or can you drive your CPU at 100c with coolant temp 45c ? What i mean is can you recreate my condition so that i know its not contact issue ? Link to comment Share on other sites More sharing options...
Jos Hideky Posted February 5 Share Posted February 5 5 minutes ago, LeDoyen said: Not overclocking ^^ it's what happens when power limits are disabled by motherboard manufacturers out of the box. People get a system that heats up because the CPU doesn't behave as it was designed to. It needs either manual tweaking (longer, needs some savvy on overclocking), or set power limits (easier) and let the CPU reduce its own vcore as it wants. If you are not too savvy when it comes to cpu tweaking and testing, it may be easier to just reduce the power limits in bios, and the CPU will just adapt to it and reduce the Vcore on its own. Owh ok2 thnx for the info man. So you think its not the contact to AIO problem ? Link to comment Share on other sites More sharing options...
PAlvarez05 Posted February 5 Share Posted February 5 Just now, Jos Hideky said: Ah ok2 haha sorry really not savvy with this ... Owh so your highest coolant temp is also at 45ish ? And is your 13900k is at 100c too ? Or can you drive your CPU at 100c with coolant temp 45c ? What i mean is can you recreate my condition so that i know its not contact issue ? Is not a contact issue, LeDoyen explained if it were a contact issue you would thermal throttle is a “your cooler can’t handle the cpu “ issue Link to comment Share on other sites More sharing options...
Jos Hideky Posted February 5 Share Posted February 5 Just now, PAlvarez05 said: Is not a contact issue, LeDoyen explained if it were a contact issue you would thermal throttle is a “your cooler can’t handle the cpu “ issue stable at 100c is not thermal throttle ? if cooler cant handle the CPU, i think we will see high coolant temp right ? Link to comment Share on other sites More sharing options...
PAlvarez05 Posted February 5 Share Posted February 5 The 13900k is a beast is not meant for a 360 AIO… Link to comment Share on other sites More sharing options...
Recommended Posts
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 accountSign in
Already have an account? Sign in here.
Sign In Now