Maker Pro
Maker Pro

5V Computer Fan - easy way to reduce speed?

ukspawn

Feb 5, 2022
88
Joined
Feb 5, 2022
Messages
88
I have a 5v 120mm computer fan that i use to extract heat from my HIFI Amplifier. Id like to slow the speed down and i know you can buy the adapters and stuff but i thought if i can do it myself simple then ill do it.

Will i need a resistor of some kind inline on the +5v wire? or can i use a thermistor and plonk that somewhere that gets warm?

What speed it runs isn't exactly important - 800-1000 rpm should reduce the noise enough.

can anyone tell me what part i need? (less spendy the better)

Screenshot 2022-04-21 at 10-35-52 NF-F12 5V.png

Thank you!
 

Harald Kapp

Moderator
Moderator
Nov 17, 2011
13,057
Joined
Nov 17, 2011
Messages
13,057
A 3 pin fan usually has power (+), gnd (-) and a tacho output, see the wiring diagram. I think you probably only use power and gnd, the tacho output is unused.
The fan operates from DC 0 V ... 12 V. Fan speed can be controlled by varying the voltage on power between these two limits.
There are sevral ways to do this.
  1. PWM as suggested by @bertus . You can buy off the shelf modules for that purpose (example). You'll have to try and see (or better: listen) if the switching of the PWM interferes with the amplifier's electronics. You wouldn't want to hear artefacts from the fan controller through your speakers. Albeit at 25 kHz switching frequency this wouldn't be audible for the average person.
    This is the most efficient way in terms of power disipation.
  2. Use a linear power supply that can be controlled between 0 V and 12 V. A linear supply creates much, much less noise on the power supply, so interference with the amplifier is not t be expected. But then again, the fan creates noise by itself, so a PWM solution may be suitable. A linear power source is less efficient than a PWM controller, but efficiency may ot be the premium design aspect in this application.
  3. Use a series resistor in the power line between power (your 5 V) and the power input of the fan. The resistor will be comparatively big as it needs to dissipate some power. If you want different fan speeds, you will need multiple resistors and a selector switch. You can not use a potentiometer - t least not a standard one - as potentiometers are not built for the currents required by the fan.
    This is imho the least elegant solution.
 

kellys_eye

Jun 25, 2010
5,470
Joined
Jun 25, 2010
Messages
5,470
A thermal switch (mechanical) coupled with a 'low noise' fan run at its full speed (the noise difference will be negligible) should do.

That said, why does your amplifier NEED fan cooling? If it was properly designed and is running in the environment the designers created it for then its own passive cooling system should be more than capable.

Is there and issue we are unaware of?
 

ukspawn

Feb 5, 2022
88
Joined
Feb 5, 2022
Messages
88
its not so loud and i only use it on a hot day and i do listen above 95dB(c) SPL so no its really not much of an issue.

I just wanted to slow it down with the simplest solution. i think an inline resistor would be fine. the look or elegance of it isnt important, neither is selectable speeds. i simply want to slow it down a bit...

I guess no one know the value or is there a way to calculate what RPM i will get using what resistor value?
if the full 5v 0.5a give max RPM what value resistor gives 70% of that speed etc...

Or just buy some cheap ens' or take them from some broken electronics and try them...

And thank you all for the replies
 

kellys_eye

Jun 25, 2010
5,470
Joined
Jun 25, 2010
Messages
5,470
There is no practical way to determine rpm using resistors to limit the speed. Suck it and see is the way to go here. Even the airflow path resistance will mean a change in resistor value (i.e. you might slow it down 'on the desk' under testing but when the fan is placed in a path that offers air resistance the actual speed may well be different).

Best way would be to use a PWM speed controller so you can set the speed using a small control rather than going through various resistors.

For all the trouble involved, running at full speed would make little difference to the end result.
 

hevans1944

Hop - AC8NS
Jun 21, 2012
4,792
Joined
Jun 21, 2012
Messages
4,792
Reducing the effective voltage to the fan motor, whether by PWM or using a variable output power supply, may not work if the fan is driven by a brushless motor that operates at 5V and draws only 0.15A or 150 milliamperes. This type of motor requires electronics to switch and route its supply current. At 0.75 watts (maximum) input power there is not a lot of lee-way room for this to occur.

OTOH, if the fan is driven by a permanent-magnet brushed motor, then any of the above responses should work. I would try the fixed series resistor, in-line with voltage used to power the motor, first because it is cheap and simple. PWM modulation of the 5V supply may work for a brushless motor.

For this application, a 33Ω resistor would drop the entire 5V supply across the resistor at 0.15A, leaving nothing for the motor. So, something with less resistance is appropriate. Get a handful of 4.7Ω, 1/4 watt (larger is okay), metal-film or carbon-film or wire-wound resistors and experiment with one or more in series until you get the speed and/or noise-level to where you want it to be. If the resistors get too hot, try series+parallel arrangements until you get the effect you want without excessive heat. Or use higher-wattage resistors. Remember. the motor without resistors only draws 0.75 watts, so any resistance you place in series with the motor power wires will, perforce, dissipate less than 0.75 watts. Be careful NOT to connect any of the resistors directly across the 5V supply. The motor MUST be in the circuit to limit the current in the series resistor(s).
 

ukspawn

Feb 5, 2022
88
Joined
Feb 5, 2022
Messages
88
wow. a very detailed and apreciated reply.

Thank you Hevans
 
Top