Full Loop Pump Speed. 100% Max or custom?

Associate
Joined
28 Apr 2008
Posts
672
Hi all

Finally have my full water cooling build up and running.

Quick question on Pump/Rez combo speed. Should I be running it at 100% maximum RPMs? Or Should I be setting it lower i.e. 50%

I've read many conflicting posts, some say to run at 100% others say 35%.

Thanks in advance.
 
Soldato
Joined
11 Sep 2013
Posts
12,310
My old D5 I just had at 100% all the time.
My current one goes off an averaged component temperature curve and is quieter as a result... but it's also a fancy AquaComputer one, so I like showing off all the bells and whistles of what it can do.
 
Associate
Joined
10 Dec 2020
Posts
335
Flow rate has a marginal effect on water temperature in relation to fan speed, but it does help. For most people the conflicting issue is noise. Those who have suggested about 35% have probably done so because of noise.
FYI my pumps idle at about 30% and I can't hear them. I use a water temperature sensor to ramp them up to about 50% at 40C when they are a bit louder but so are the fans. My tests show that this reduces the peak temp by only about 1C, but that allowed me to change the fan curves a bit to optimise the noise.
I think my 40C target is arbitrary and I suppose I could let it go higher and reduce the noise from the fans and pumps, but for now I'm happy with the noise level during gaming, which is about the same as my previous air-cooled PC was at idle!

If you keen enough to optimise your pump speed for your taste, you need to run tests to measure noise and temperature.
 
Associate
OP
Joined
28 Apr 2008
Posts
672
Thanks for your input chaps, much appreciated.

I'll try running at 50% and see how I get on. Running at 100% I can hear the pump make what I can only describe as a bee hive low level hum, which I've only picked up on yesterday and I wasn't too sure if I set the pump to a lower speed would it decrease the performance of the loop and if temps would go up.
 
Soldato
Joined
21 Dec 2019
Posts
6,532
Location
Planet Thanet
Yeah as others said
There's very little if any gain over a certain %
And every body has different hearing
Sometimes I prefer a set %
Rather than it altering itself up and down
As sometimes as it goes up through a certain %
It makes an audible noise
Yet as it reaches a higher %
That noise disappears
Either it's resonating at that % or my ears are more
Susceptibile to that frequency
Bottom line is I wouldn't worry about what others do
Adjust it to what works for you
 
Associate
Joined
20 Jun 2013
Posts
1,831
Like Mike and McNumpty I run my 2 D5s at 80% to keep them quiet, loudest thing in the system is the crappy rgb fans for the lighting
 
Associate
Joined
5 Mar 2017
Posts
2,252
Location
Cambridge
No reason to run at 100%, unless necessary.
I always find the balance where the pump runs at the highest RPM yet being quiet. Some D5 were quieter at 60% than at 50%. Best way to play around, in 5% intervals, and when you find the pump quiet, run some tests to check if the loop isn’t struggling for flow.
Depending on your rads and blocks, anyway.
Triple EK (2 pe 360 and 1 xe 360), 3090 ek waterblock and active backplate plus EK monoblock (Asus X570) and a flow meter. A single D5 at 60% was fine. No improvement going faster, maybe 1C.
When tried triple Corsair 360 30mm rads, same blocks, the pump at 100% struggled. A DDC 3.2 was too loud at 80% and just about coping with the loop. Dual D5 at 60% was enough. my point, depends on your loop, but unless very restrictive, a single D5 at 50-60% should be very close to its optimum performance at a dead quiet noise.
 
Associate
Joined
20 Dec 2020
Posts
515
Location
Leeds
I run mine at around 2250 rpm, I ran it once in a gaming session at full tilt 4500 rpm and it made no discernible difference to my water temp so I can`t find any reason to run it at 100%.
 
Associate
Joined
7 Oct 2015
Posts
814
i used to have my D5 at full speed since it was quiet either way.

For my DDC, i have mine set to 40% idling. They produce a whiney sound after 45% which puts me off and settting the speed from 60%+ made no difference, so i range my DDC from 40-60% static (not via pwm) depending on workload
 
Back
Top Bottom