H150i Elite Capellix Push/Pull

[unseen_uk]

I'm new to using an AIO (air cooled builds up until now) but wanting to keep a 5900x relatively cool on long render sessions. I've had a look in to a push/pull config and the recommendations are to try to keep the fans matched. Ok so stick with ML120 Pro RGB but they don't have the same max RPM so could this cause an issue with the AIO supplied ML120s?

 

Only time I could see this being an issue is if the AIO went to full fan could the slower extra ML120s cause a problem?

 

Again new to all this side of things and just want to make sure I've got everything worked out before pulling the trigger on the extra fans.

[c-attack]

The Commander Core makes this a bit more complex because it uses PWM % as the control factor for the programmable curves vs RPM on the Commander Pro which would allow you to side-step the issue. One workaround on the Commander Core is to run two different fan profiles for each side of the radiator -- one for the 1600 rpm max ML120-RGB and one for the 2400 ML-Elite. You would physically check RPM matching at various PWM % intervals, but that fan is pretty linear and you can do the mathematical equivalent. There is really nothing to be gained by going past 1600 rpm on a 30mm radiator in push pull aside from noise, so capping the ML-Elite set at 1600 (75%) would prevent that.

 

The trickier bit is setting this up. You could burn all 6 PWM ports on the Com Core to handle these 6 fans. That then leaves the rest of the case to fend for itself. Another option would be to run PWM repeaters to offload the radiator fan current to that device while using 1 PWM header for control. However, because of the difference in max RPM you likely need 2 PWM hubs, one for each side. They are relatively inexpensive and 2 of them is cheaper than buying another full fledged fan controller, although it does add some to the bulk of gear to be stashed.

 

Given all that, you may want to assess whether push pull is worth while. Push-pull on a 300W load into 360x30mm radiator will typically get you about 1C at 1300 rpm vs single layer 3x120@1300 rpm. That's not much to go chasing after. However, that is the free air result on lab bench. Actual case installation may bring more resistance into the mix. Restrictive front air opening or heavy filters can increase that 3 v 6 fan differential out by several degrees. I am going to assume you are looking at front mounting the 360mm if this is a 5000D since I am not sure that works in the side or top mount positions. You might also ask in the 5000D thread about actual user experience with the front inlet vs the theoretical analysis I am offering.

 

https://forum.corsair.com/v3/showthread.php?p=1088672#post1088672

[unseen_uk]

The Commander Core makes this a bit more complex because it uses PWM % as the control factor for the programmable curves vs RPM on the Commander Pro which would allow you to side-step the issue. One workaround on the Commander Core is to run two different fan profiles for each side of the radiator -- one for the 1600 rpm max ML120-RGB and one for the 2400 ML-Elite. You would physically check RPM matching at various PWM % intervals, but that fan is pretty linear and you can do the mathematical equivalent. There is really nothing to be gained by going past 1600 rpm on a 30mm radiator in push pull aside from noise, so capping the ML-Elite set at 1600 (75%) would prevent that.

 

Ahh didn't realise it was PWM % rather than RPM. Hmm so it could be balanced but sounds like the effort might outweigh the benefit.

 

 

Given all that, you may want to assess whether push pull is worth while. Push-pull on a 300W load into 360x30mm radiator will typically get you about 1C at 1300 rpm vs single layer 3x120@1300 rpm. That's not much to go chasing after. However, that is the free air result on lab bench. Actual case installation may bring more resistance into the mix. Restrictive front air opening or heavy filters can increase that 3 v 6 fan differential out by several degrees. I am going to assume you are looking at front mounting the 360mm if this is a 5000D since I am not sure that works in the side or top mount positions. You might also ask in the 5000D thread about actual user experience with the front inlet vs the theoretical analysis I am offering.

 

I'm planning to top mount the AIO quite a lot of videos seem to suggest it fits although making sure cabling to the motherboard is done first and then use the front and possibly side as intake with one extra exhaust fan. Still looking at options for those fans. Completely sat on the fence between all 3 for those.

[c-attack]

Ahh didn't realise it was PWM % rather than RPM. Hmm so it could be balanced but sounds like the effort might outweigh the benefit.

 

Commander Pro or Core, you still have to work out power/control for the 6 PWM fans. That will mean at least 1 PWM repeater or using the entire controller for just those 6 push pull fans.

[Corsair HP]

Instead of going push/pull on the AIO I would suggest making sure that your case has the best possible airflow instead. So mount the 3 fans in exhaust positions in the rear/top of the case.

The 5000D has excellent airflow, and even more so when you populate all fan spots, so you should not run into any issues with temperatures. The Ryzen CPU's gets crazy hot per default so don't let that frighten you. I have a 5950X on a crazy custom loop, and it idles at 45-55c and loads at 65-75c on stock settings.