The Corsair User Forums  

Go Back   The Corsair User Forums > Corsair Product Discussion > Cooling

Notices

Reply
 
Thread Tools Search this Thread Rate Thread Display Modes
  #1  
Old 01-01-2020, 04:34 PM
TheGameVIX TheGameVIX is offline
Registered User
TheGameVIX's PC Specs
 
Join Date: Dec 2019
Posts: 8
POST ID # = 1029408
TheGameVIX Reputation: 10
Question LL120 RPM's higher than rated

Good evening everyone,

So i seem to have run into a bit of an issue and I have done all i can to troubleshoot the issue.

Setup:
H150I - 3 LL120' all standard RPM's

Exhaust Fans - 3 LL120's All going up too 1700 RPM's.

H150's fans are connected to the three headers from the pump. The three exhaust fan's are connected to the commander pro on fan ports 4,5 and 6.

The issue is that it seems the exhaust fans are going 200rpms above what they are rated for.

From what i can see i think the commander pro is receiving to much voltage based off of the stats below from the ICUE software.

https://imgur.com/a/0PWOTZ7

mind you i did set the case fans to extreme just to simulate them "ramping up" on load.

Im not to sure what could be causing brand new fans that worked 5 days ago at the right rpm's to now be 200 rpms above rated.
Reply With Quote


  #2  
Old 01-01-2020, 04:53 PM
usna92's Avatar
usna92 usna92 is offline
I fight for the users
usna92's PC Specs
 
Join Date: Apr 2019
Location: Seattle, WA
Posts: 92
POST ID # = 1029414
usna92 Reputation: 16
Default

Which LL fans do you have? The black ones or the white ones?
Reply With Quote


  #3  
Old 01-01-2020, 04:55 PM
c-attack c-attack is offline
Registered User
c-attack's PC Specs
 
Join Date: Jun 2014
Posts: 11,557
POST ID # = 1029415
c-attack c-attack Reputation: 116
Default

I think you are looking too closely for an issue here. I have 12 LLs. All of them run 1650-1700. I think the specification is a bit conservative, even without factoring in the normal +-10% variance. They’re all PWM fans, so it shouldn’t matter if you are at 11.99 or 12.15v on line. Even then, don’t take the Commander Pro reading as universal 12v for all connected fans. Not how it works and each line has its own voltage.

As for the differences in your sets, the radiator is a source of physical resistance and will always slow a fan down a bit compared to free air. Dust filters the same, but to a lesser degree. Your speeds match mine factoring in those on my 360mm radiators and are consistent with all the other results I have seen here in the forums. If you had written that all your fans run 1500 rpm, except the the three on the rad at 1350 rpm like the specification, I would be wondering what was going on. Instead, you are just like the rest of us.

Last edited by c-attack; 01-03-2020 at 04:45 PM.
Reply With Quote


  #4  
Old 01-01-2020, 05:08 PM
TheGameVIX TheGameVIX is offline
Registered User
TheGameVIX's PC Specs
 
Join Date: Dec 2019
Posts: 8
POST ID # = 1029416
TheGameVIX Reputation: 10
Default

Quote:
Originally Posted by usna92 View Post
Which LL fans do you have? The black ones or the white ones?
They are the black ones
Reply With Quote


  #5  
Old 01-01-2020, 05:12 PM
TheGameVIX TheGameVIX is offline
Registered User
TheGameVIX's PC Specs
 
Join Date: Dec 2019
Posts: 8
POST ID # = 1029418
TheGameVIX Reputation: 10
Default

Quote:
Originally Posted by c-attack View Post
I think you are looking too closely for an issue here. I have 12 LLs. All of them run 1650-1700. I think the specification is a bit conservative, even without factoring in the normal +-10% variance. They all PWM fans, so it shouldn’t matter if you are at 11.99 or 12.15v on line. Even then, don’t take the Commander Pro reading as universal 12v for all connected fans. Not how it works and each line has its own voltage.

As for the differences in your sets, the radiator is a source of physical resistance and will always slow a fan down a bit compared to free air. Dust filters the same, but to a lesser degree. Your speeds match mine factoring in those on my 360mm radiators and are consistent with all the other results I have seen here in the forums. If you had written that all your fans run 1500 rpm, except the the three on the rad at 1350 rpm like the specification, I would be wondering what was going on. Instead, you are just like the rest of us.
You may be 100 percent correct, the only reason i assumed there was an issue to begin with, was the fact that a few days ago this "issue" was not presenting itself as it is now.

that being the case if this all falls within the normal operating spectrum of the fans themselves. With no harm resulting in them running at a higher than rated rpm's I have no problem. Just trying to insure that any higher than needed voltage issues are resolved as to increase the lifespan of my equipment.

That all being the case, you are more than likely correct i forgot to factor in the resistance of the rad in this case. Just a bit overly cautious with my new setup is all.

Thanks for the input
Reply With Quote


  #6  
Old 01-01-2020, 05:36 PM
c-attack c-attack is offline
Registered User
c-attack's PC Specs
 
Join Date: Jun 2014
Posts: 11,557
POST ID # = 1029419
c-attack c-attack Reputation: 116
Default

If you changed set ups, you might see differences just moving them around the case. Front intake may be slower than rear exhaust because of case design, filters, or just general pressure differences. There may be real reasons for prior results, most likely not related to voltage. If you get one single fan different from all the others, then there may be questions to ask.

You can use the + button in the Performance tab to create a new cooling curve. That will also open up the fixed rpm and fixed % options. That is an easier way to test. Also, you don’t want to use those Commander Pro default fan curves based on cpu temp with no spin up/down delays. Nobody needs their case fans to ramp up when you open a folder. H150i temp (coolant temp) or GPU temp will make a better control variable. Adjust temp scale to match. The coolant temp presets (Quiet, Balanced,Extreme) are in the upper right hand corner to sync with the cooler.

I am actually relieved the LL run a bit over spec. People seem to want 2000 rpm fans, even if they never run them that fast. Less complaining this way.
Reply With Quote


  #7  
Old 01-01-2020, 06:09 PM
c-attack c-attack is offline
Registered User
c-attack's PC Specs
 
Join Date: Jun 2014
Posts: 11,557
POST ID # = 1029426
c-attack c-attack Reputation: 116
Default

Just for peace of mind...

Reply With Quote


  #8  
Old 01-01-2020, 07:24 PM
TheGameVIX TheGameVIX is offline
Registered User
TheGameVIX's PC Specs
 
Join Date: Dec 2019
Posts: 8
POST ID # = 1029436
TheGameVIX Reputation: 10
Default

Quote:
Originally Posted by c-attack View Post
Just for peace of mind...

Oh awesome thank you so much for the assistance this eases my mind quite a bit.

If you don't mind I have attached three pictures of my H150I and CoPro with current settings and fan curve. along with my Ryzen Master current Idle temps.

I know it might be hard to tell from just these pictures but any recommendations you may have such as fan curve or pump preset settings would be very much appreciated.

First time using corsair products so its a lot to take in all at once haha.

H150 + CoPro + Ryzen Idle Temps:
http://imgur.com/a/mfJWlva
Reply With Quote


  #9  
Old 01-01-2020, 07:53 PM
c-attack c-attack is offline
Registered User
c-attack's PC Specs
 
Join Date: Jun 2014
Posts: 11,557
POST ID # = 1029438
c-attack c-attack Reputation: 116
Default

I think you can relax the H150i fan curve a bit. The equation is +1C coolant temp = +1C CPU temp. Coolant temp is the CPU temp with zero voltage or the effective baseline temperature. So if coolant temp is 31C, then minimum possible CPU temp is 31C. Also, if the coolant temp goes up +1C, then the average CPU temp will also go up +1C.

The actual CPU temp is moving all over the place and always warmer than any one likes. Nothing you can do about it. That's how it works. We are all limited by voltage and the rate at which heat with transfer through the CPU from where it is created at the CPU socket pins. The cooler does not cool. It takes the CPU heat and dumps it somewhere else. However, if you don't take out the trash often enough, then that heat is passed back across the cold plate to CPU. That's the +1C coolant = +1C CPU temp thing. So in the grand scheme of things, how much noise is it worth to lower the CPU/coolant temp by 1C? That's what you have to decide.

I used to prescribe some kind of sliding scale for just above normal idle coolant temp to load. You can still do that if you like. Start a gentle slope up about +3C above your normal idle coolant. Ramp up to a tolerable fan speed about +6-9C above the baseline. This is the most you will normally see. Then set a max fan blast at about +15C over. That will give you a clear warning if things are out of bounds. However, I think people can figure this stuff of for themselves. 120mm fans on radiator will move a small amount of air between 750-1000 rpm. That is the quiet zone. Between 1000-1300 is a good work zone where the fan is effective against resistance, but not too noisy. 1300+ may give you another 1-2C on a really heavy load, but probably not on a CPU only loop. I think you will probably top out at about 165W for a 3900X with some overclocking. Still, you can use the max speed to signal if you cross a line. Slow and steady wins the race here. So based on a 29-30C normal idle coolant temp, you might do something like 30C=750 rpm. 33=1000 rpm. 39=1300 rpm 44C=1700 rpm (the blast). You should not go past 40C without strong environmental help (GPU waste heat, super hot room, etc).

For your set-up with the 360mm on the front as intake, the exhaust fans are there to balance the intake from the front and deal with GPU waste heat. The radiator effective cuts the airflow from the fan in half, so even at like speeds, you have plenty of exhaust. You can experiment during GPU load time whether faster slow speeds have any effect on GPU or case ambient temp. Ultimately you might wind up running the top 2 fans to match the front intake and connect the rear exhaust fan to GPU temp. Definitely user choice.
Reply With Quote


  #10  
Old 01-01-2020, 08:03 PM
TheGameVIX TheGameVIX is offline
Registered User
TheGameVIX's PC Specs
 
Join Date: Dec 2019
Posts: 8
POST ID # = 1029439
TheGameVIX Reputation: 10
Default

Quote:
Originally Posted by c-attack View Post
I think you can relax the H150i fan curve a bit. The equation is +1C coolant temp = +1C CPU temp. Coolant temp is the CPU temp with zero voltage or the effective baseline temperature. So if coolant temp is 31C, then minimum possible CPU temp is 31C. Also, if the coolant temp goes up +1C, then the average CPU temp will also go up +1C.

The actual CPU temp is moving all over the place and always warmer than any one likes. Nothing you can do about it. That's how it works. We are all limited by voltage and the rate at which heat with transfer through the CPU from where it is created at the CPU socket pins. The cooler does not cool. It takes the CPU heat and dumps it somewhere else. However, if you don't take out the trash often enough, then that heat is passed back across the cold plate to CPU. That's the +1C coolant = +1C CPU temp thing. So in the grand scheme of things, how much noise is it worth to lower the CPU/coolant temp by 1C? That's what you have to decide.

I used to prescribe some kind of sliding scale for just above normal idle coolant temp to load. You can still do that if you like. Start a gentle slope up about +3C above your normal idle coolant. Ramp up to a tolerable fan speed about +6-9C above the baseline. This is the most you will normally see. Then set a max fan blast at about +15C over. That will give you a clear warning if things are out of bounds. However, I think people can figure this stuff of for themselves. 120mm fans on radiator will move a small amount of air between 750-1000 rpm. That is the quiet zone. Between 1000-1300 is a good work zone where the fan is effective against resistance, but not too noisy. 1300+ may give you another 1-2C on a really heavy load, but probably not on a CPU only loop. I think you will probably top out at about 165W for a 3900X with some overclocking. Still, you can use the max speed to signal if you cross a line. Slow and steady wins the race here. So based on a 29-30C normal idle coolant temp, you might do something like 30C=750 rpm. 33=1000 rpm. 39=1300 rpm 44C=1700 rpm (the blast). You should not go past 40C without strong environmental help (GPU waste heat, super hot room, etc).

For your set-up with the 360mm on the front as intake, the exhaust fans are there to balance the intake from the front and deal with GPU waste heat. The radiator effective cuts the airflow from the fan in half, so even at like speeds, you have plenty of exhaust. You can experiment during GPU load time whether faster slow speeds have any effect on GPU or case ambient temp. Ultimately you might wind up running the top 2 fans to match the front intake and connect the rear exhaust fan to GPU temp. Definitely user choice.

Oh wow, this information is fantastic. I honestly wish i could pay you haha ive been struggling with this stuff for over a week. So for the temp to rpm scale you mentioned would you recommend doing that across all fans or just the h150i fans?

Lastly, do you recommend keeping the pump at Balanced or Extreme. As far as i am aware you can set a custom curve for that.
Reply With Quote


  #11  
Old 01-01-2020, 09:06 PM
c-attack c-attack is offline
Registered User
c-attack's PC Specs
 
Join Date: Jun 2014
Posts: 11,557
POST ID # = 1029443
c-attack c-attack Reputation: 116
Default

The pump speeds are fixed. Quiet=1100, Balanced=2160, Extreme 2850 rpm; all +-30 rpm. Pump speed doesn't matter much on AIO coolers. I suggest most park it on balanced and leave it. People typically report a 0-1C difference between that and Extreme. Go ahead and experiment, but changing it is often more trouble than the 0.5C. There is one exception. The Quiet speed is too low for real work. It is supposed to be an ultra low noise desktop work setting. Perfect for that. No so good for CPU rendering. You will pay a penalty for leaving at 1100 at load.

The suggestion was for the H150i fans. What you do with the others is up to you. In a CPU only environment with no GPU concerns whatsoever, you make the rear fans match the intake air volume from the front three. Now that gets tricky because the radiator alters the flow. No way to accurately measure without an anemometer or other tools, so you estimate. At the speeds you will use on the radiator, those fans move about 50-60% of the free air spec. So at 1300 rpm, 3 free air LL120s would theoretically move about 110 cfm [1300/1500 x (43 cfm x 3 fans)]. On the radiator, that number is more like 55-60 cfm. So theoretically, you only need 60 cfm of air from the exhaust fans (about 750 rpm each) to remove that same volume of air. A little faster, a little slower, it doesn't really matter. You could use other temperatures as guidelines line RAM or VRM temps that might be more affected by the top side fans. Of course the kicker is you likely don't live in a GPU free world and the 300-350W from your Ti will do more to heat the case than anything else. I think there is some merit to using the rear exhaust to try and draw more of the waste heat back, rather than across the center of the motherboard and out the top. Ultimately, that is theoretical and I don't have the same case to experiment with. Try different stuff. Use the GPU and motherboard temps to help make a determination. The CPU and front rad will take care of itself. The GPU is the big fish to keep your eye on.

Last edited by c-attack; 02-09-2020 at 09:42 AM.
Reply With Quote


  #12  
Old 01-01-2020, 11:00 PM
TheGameVIX TheGameVIX is offline
Registered User
TheGameVIX's PC Specs
 
Join Date: Dec 2019
Posts: 8
POST ID # = 1029447
TheGameVIX Reputation: 10
Default

Quote:
Originally Posted by c-attack View Post
The pump speeds are fixed. Quiet=1100, Balanced=2160, Extreme 2850 rpm; all +-30 rpm. Pump speed doesn't matter much on AIO coolers. I suggest most park it on balanced and leave it. People typically report a 0-1C difference between that and Extreme. Go ahead and experiment, but changing it is often more trouble than the 0.5C. There is one exception. The Quiet speed is too low for real work. It is supposed to be an ultra low noise desktop work setting. Perfect for that. No so good for CPU rendering. You will pay a penalty for leaving at 1100 at load.

The suggestion was for the H150i fans. What you do with the others is up to you. In a CPU only environment with no GPU concerns whatsoever, you make the rear fans match the intake air volume from the front three. Now that gets tricky because the radiator alters the flow. No way to accurately measure without an anemometer or other tools, so you estimate. At the speeds you will use on the radiator, those fans move about 50-60% of the free air spec. So at 1300 rpm, 3 free air LL120s would theoretically move about 110 cfm [1300/1500 x (43 cfm x 3 fans)]. On the radiator, that number is more like 55-60 fps. So theoretically, you only need 60 cfm of air from the exhaust fans (about 750 rpm each) to remove that same volume of air. A little faster, a little slower, it doesn't really matter. You could use other temperatures as guidelines line RAM or VRM temps that might be more affected by the top side fans. Of course the kicker is you likely don't live in a GPU free world and the 300-350W from your Ti will do more to heat the case than anything else. I think there is some merit to using the rear exhaust to try and draw more of the waste heat back, rather than across the center of the motherboard and out the top. Ultimately, that is theoretical and I don't have the same case to experiment with. Try different stuff. Use the GPU and motherboard temps to help make a determination. The CPU and front rad will take care of itself. The GPU is the big fish to keep your eye on.

Oh wow i never thought there was this much to the cooling aspect with AIO's. Again i cant thank you enough for all the info.

So i attempted to adjust my intake fans as you mentioned. I am currently leaving my exhaust fans the same as they are running about 150-165 more rpms at anytime than my intake to keep hot air out.

Updated Fan Curve:
https://imgur.com/a/stjAVEg
Reply With Quote


  #13  
Old 01-02-2020, 12:19 PM
c-attack c-attack is offline
Registered User
c-attack's PC Specs
 
Join Date: Jun 2014
Posts: 11,557
POST ID # = 1029506
c-attack c-attack Reputation: 116
Default

Yes, I think that is better. Remember, this stuff is not so sensitive that +-100 rpm has a real meaningful effect on cooling. That might be a fraction of a degree on the H150i at full load and less than that at normal levels. The exhaust fans even more so. +100 rpm will not cool your ambient temp by 1C. When in doubt, always go with your ears. If it sounds too loud, it probably is and your life is not going to be affected by +1-2C more system temps. On the other hand, loud if loud and annoying from start to finish. There is no reason to run that way and too much of the information out there is written by Performance Min/Max-ers. Now I am one of them, but this is by choice and I do not confuse it with necessity.
Reply With Quote


  #14  
Old 02-09-2020, 05:32 AM
BakerPlays BakerPlays is offline
Registered User
BakerPlays's PC Specs
 
Join Date: Sep 2019
Posts: 4
POST ID # = 1034496
BakerPlays Reputation: 10
Default

Quote:
Originally Posted by TheGameVIX View Post
Oh wow i never thought there was this much to the cooling aspect with AIO's. Again i cant thank you enough for all the info.

So i attempted to adjust my intake fans as you mentioned. I am currently leaving my exhaust fans the same as they are running about 150-165 more rpms at anytime than my intake to keep hot air out.

Updated Fan Curve:
https://imgur.com/a/stjAVEg
Would you mind sharing your exhaust fan curves? I am in the process of doing exactly what you're doing, up until now I believe I was creating positive pressure in my NZXT H500 with exhaust fans slower than intake, keen to see your curves and then test my thermals!

Also another +1 for the advice in this post, I've been learning all morning but the lengthy post above helped the most. Thanks!
Reply With Quote


Reply

Tags
fans, ll120, rpm's

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -4. The time now is 01:51 AM.


Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2020, vBulletin Solutions, Inc.