PC fans from mains (with AC-DC psu)

qbazdz

qbazdz

qbazdz

Associate
Joined
27 Oct 2009
Posts
1,314
Location
West Sussex
I have a small project on the go, where Id like to get a few fans up and running with PWM controller and powered from mains.
It does need to be pretty efficient to make sense and it will be used for several hours in a day so a phone charger is out of the question.
Anyone has had similar needs and can recommend some kit?

Thanks!
 
I was thinking regular 12V PC fans.
The AC infinity stuff looks great, shame in black only.
Do you know if they are bespoke designed for 5V operation?

I forgot to add, that I have an extension lead with USB C available so was hoping I can just get away with powering this up through a C cable, just don't know how to force 12V output into the PWM regulator (these are pretty dumb, they just work of whatever voltage is supplied).
 
You might want to look at some model plane / car kit. Model speed controllers work in those sort of voltages and you could control the speed with a servo tester. Or maybe a servo tester would be able to drive a 4 pin fan directly. I'm thinking supply 12v to the fan and 6 to the servo tester via a UBEC and tap the signal off the servo tester into the fan signal pin.

Suggest looking at Hobbyking or similar.

hobbyking.com

Turnigy Servo Tester

Turnigy Servo Tester
hobbyking.com hobbyking.com
 
Last edited:
Might help sharing what you want to achieved how far you're willing to DIY it, all sorts of fans out there at trade paces etc
 
Last edited:
Ok - lets step back a bit then.
We have a new radiator cover that looks neat, but I am surprised in how much heat it is actually stopping from radiating out. I dont think it helps that the TRV is also within, which means it cuts out even when set to max. In the interim, I can remove the front and it makes a huge difference when we keep it more open.
I am planning to get a few fans arranged under the actual radiator, pulling the air via the bottom vents and push it by the TRV and through the radiator itself, so we can get some of that heat to actually circulate a bit more.
I do have a socket nearby, but was hoping to run off something efficient and low power, rather than a proper power supply. as they are stupidly inefficient. I want to power it on manually for now, later with a timer or arduino with some more temp sensing etc. I want PWM for them to just dial the best possible inaudible revs while keeping the air moving.

I trialled it already with some other fans and it seems to do the job, but it was a pure mockup with a big PSU, jump starting it and reducing voltage rather than pure pwm signal control.
 
qbazdz said:
It does need to be pretty efficient to make sense and it will be used for several hours in a day so a phone charger is out of the question.
Click to expand...
That is actually exactly what I would use (and have used for this sort of thing before). You can get chargers that will happily supply 65W for hours at a time (I presume you won't be using >65W of fans) with good efficiently. Get one that supports USB PD @ 12V, and a 12V USB PD trigger to force it to output 12V.
 
I am thinking the GaN style charger, like you said with the 12V PD. not the likes of switching adapters which to me always felt rubbish.
I did find an AdaFruit 555 board, which has the right fan PWM frequency (25kHz) @thescouselander (servo stuff seems to run at too low freq) but I am not sure if I read their design doc correctly. The adjustment seems to change the frequency at which the pulses are generated, not the pulse on time that is meant to drive the actual PWM duty cycle. Maybe I got that wrong, but think it only does fixed cycle time, but you can change freq.
 
Last edited:
Worth noting that PC PWM control is a negative polarity signal i.e. you need to invert the logic if you're providing a control signal. This is so that if a 4 pin fan is fitted to a 3 pin socket, its default behaviour is to run at full speed rather than minimum. If 0V is seen at pin 4 it will still work.

Personally I think a simple switching power supply is the simplest AND the most efficient. 5V or 12V, your choice really. I wouldn't bother with PWM, I'd just get a simple DC voltage controller and adjust to where it was quiet enough.
 
qbazdz said:
I have a small project on the go, where Id like to get a few fans up and running with PWM controller and powered from mains.
It does need to be pretty efficient to make sense and it will be used for several hours in a day so a phone charger is out of the question.
Anyone has had similar needs and can recommend some kit?

Thanks!
Click to expand...

I have a number of fans on the mains, cooling a cupboard full of routers and switches and things. I have a PWM controller, but instead of that I just use a switchable power supply (What I mean by "switchable" is that I can set the voltage with a switch that's on the front of the PSU). It's just a little thing from Amazon that voltage controls the fans instead of using PWM.

Interestingly, Noctua make some products for this sort of thing. Although they are a bit expensive.
 
Last edited:
So I've got a small PWM controller with a knob on order and will mock this up with the switchable psu for now and test further. I can't believe the switchable PSUs are more efficient then the GaN chargers though? mine is an oldie, but is like a small heater.. it reminds me the days of using 7805 chips which were super hot and needed a heatsink to stop it from melting..
 
qbazdz said:
I can't believe the switchable PSUs are more efficient then the GaN chargers though?
Click to expand...
I think there's a terminology problem here. Modern phone chargers, and indeed almost all small power supplies including everything made with GaN transistors, are "switch-mode" or "switching". Decent quality ones can be very efficient. This is in contrast with older linear power supplies, which are generally not very efficient.

That's not the same as the way SpellowHouse used "switchable" to mean a PSU with a physical switch on it to change voltage. Such a power supply could be either linear or switch-mode. If it's old it's probably linear, with a heavy transformer inside, and not efficient.
 
Mr Evil said:
I think there's a terminology problem here. Modern phone chargers, and indeed almost all small power supplies including everything made with GaN transistors, are "switch-mode" or "switching". Decent quality ones can be very efficient. This is in contrast with older linear power supplies, which are generally not very efficient.

That's not the same as the way SpellowHouse used "switchable" to mean a PSU with a physical switch on it to change voltage. Such a power supply could be either linear or switch-mode. If it's old it's probably linear, with a heavy transformer inside, and not efficient.
Click to expand...

You are right.
As it happens, it is switched mode and switchable! Ha!
 
You must log in or register to reply here.
Back
Top Bottom