I have a cheap automotive radiator fan ( 12VDC, < 7A ) I want to use as a variable speed exhaust fan for a trailer. Hooking it up to a variable power supply for testing, the speed was easily adjustable by either voltage or current.

When I tried hooking it up to a couple different PWM controllers, the fan only ran at full speed. Not certain, but I think the PWM's are 13kHz. Am I right in assuming something in the 20kHz+ range will work better, or am I missing something here?

 

Am I right in assuming something in the 20kHz+ range will work better, or am I missing something here?

I think that's unlikely. A motor expert will explain why but in the meanwhile I wouldn't waste time on a higher frequency.

 

You need to find out the technology of the fan, is it simple brushed DC or does it have a PWM controller built in?
If brushed DC I would expect it to perform on a PWM controller?
Max.

 

I have used PWM to produce a dc voltage to control speed of a fan.
Here is where I got the idea: http://arduino.cc/en/Tutorial/PWM

So basically, using arduino example linked above (digital values: 0-255; output 0-5V):
0% duty cycle--> 0V
25% duty cycle--> 1.25V
50% duty cycle--> 2.5 V
75% duty cycle--> 3.75 V
100% duty cycle--> 5 V (fan at full speed)

Figure out what type of signal and what amounts your fan needs to control its speed. If it is straight dc, 0-10 V, then we can come up with a circuit to control the speed.

 

The fan is your standard China made plastic 12" radiator fan, 2 wires, no markings. My 10A pwm's work fine on LED arrays but have no effect on fan. I read somewhere that the frequency wanted to be at least 20kHz to make a difference. Maybe i misunderstood. Maybe a diode between the leads?

 

The fan is your standard China made plastic 12" radiator fan, 2 wires, no markings. My 10A pwm's work fine on LED arrays but have no effect on fan. I read somewhere that the frequency wanted to be at least 20kHz to make a difference. Maybe i misunderstood. Maybe a diode between the leads?

I find it odd that it is two wire. I think the fan you got is only On and Off, there is not speed control.

 

The fan is your standard China made plastic 12" radiator fan, 2 wires, no markings.

Yes but it is possible it is a PWM controlled fan internally, or a simple DC brushed as the older models were?
Max.

 

Nothing in the specs about PWM. Just 2 wires, so I'm pretty sure it is a straight fan. Retested my 10A PWM module on some LED's. Works OK, bright to dim.

But no difference on fan speed. The PWM module gets very hot, but there is no effect when I adjust the trimpot. Fan stays at maximum speed. I tried putting a diode across the fan inputs, but no difference.

 

Nothing in the specs about PWM. Just 2 wires, so I'm pretty sure it is a straight fan. Retested my 10A PWM module on some LED's. Works OK, bright to dim.

But no difference on fan speed. The PWM module gets very hot, but there is no effect when I adjust the trimpot. Fan stays at maximum speed. I tried putting a diode across the fan inputs, but no difference.

If you turn the fan on and off as needed, that is PWM. A very very basic form of PWM.

 

The fans in your PC are only two wire also and in all probability have PWM internal circuitry.
Measure the resistance in both directions, if DC brushed they will be very low res.
Also on the voltage range they will generate DC if spun.
Max.

 

Max,

Hooked to a DVM, resistance is 2 Ohms when fan is still, 150 Ohms when spun. Voltage 850 mV when spun.

Shte,

Point well taken.
I want to automate the fan so it operates unattended. Based on temp sensor, louvers open and fan runs at low speed as a function of temperature and battery level. Can be manually increased to any level up to 100% for flushing. Noise is critical tradeoff so I would like a variable duty cycle. I know there are $150-200 automotive fan controllers, but I was hoping for a cheaper solution. And most of them use a thermostat, I would like to drive this with TTL levels from Arduino MCU.

 

IMO, the test is not quite conclusive, I would expect a higher dc generated voltage, if you were able to spin the motor by using a compressed air hose on the blade and reach the approximate rpm as when it runs on 12v.
The DC output should equal ~12vdc.
As a rule a DC brushed motor has no problem of operating on PWM with a 20khz switching rate.
Max.

 

If the motor turns out to be a single-phase BLDC type you could perhaps control speed by a method I developed on the ETO forum:-
http://www.electro-tech-online.com/articles/brushless-dc-motor-speed-controller.551/

 

OK , found out it is a brushed fan

You think I need a diode or capacitor across the fan leads to mave it work with PWM?

I have a 150W step-down converter coming in the mail ( fan is 12VDC, 7A ). I was thinking that I could replace the trimpot with a digital potentiometer IC and adjust the fan speed with the MCU. Am I missing something here?