I've been unsuccessful over the last month in looking for a circuit that fits exactly what I'm looking for and was hoping one of you might be able to help me out.

I'm working on a project for my aquarium where I want to take a 0v-10v analog signal from my neptune apex controller and use it to digitally control 4 x 12v PC fans.

From my research there are two ways to control fan speeds, one is to reduce the voltage using either a potentiometer, resistors, or rheostat. The second is to use PWM to control the frequency sent to the fans.

My preference would be to build a circuit that uses PWM. Please let me know if anyone has already built a circuit or could help me design one that uses a 0-10v analog control signal, and a 12v 2A power source to do what I'm trying to accomplish.

I'm pretty good with a soldering iron, and have built several circuits in the past, I'm just not great at designing the circuits from scratch.

Thanks!

 

Couple comments ...

(1) You described two ways to control fan speed, PWM and potentiometer. I would characterize them as PWM and voltage. The second method is to change the voltage to the fan and this can be done better ways than with a potentiometer.

(2) Some DC fans can not be PWM'd very well because they contain internal electronics (motor control circuit) that does not respond well when turned on and off quickly with PWM. I had a couple of these recently here at work, and found that when PWM'd they ran rough and made abnormal sounds. The only way to control their speed was to control the DC voltage directly (no PWM).

(3) Referring back to #1 above, if you want to control fan speed by changing the voltage, a potentiometer wastes a lot of power and dissipates a lot of heat. Another way is to use a DC-DC converter (buck regulator) with an adjustable output. In my recent design I used a DAC to generate a voltage which I then connectied through a resistor to the feedback network on the buck regulator. This allowed my micro-controller to set the DAC voltage which changed the ouput of the buck regulator and thus change the speed of the fan.

 

Harder that it would first seem...

Most modern DC fans use an internal motor controller chip that will not take kindly to PWM input.

Varying the input voltage will only control the speed over a narrow range.

The basic problem is that the controller chip inside the fan needs a steady supply at some minimum voltage to operate.

How about using a small RC servo motor to move some louvers in front of a normal fan? This would give you complete control over the airflow.

 

If it would help, here's the fans I'm working with:

http://www.newegg.com/Product/Product.aspx?Item=N82E16835226042
SilenX EFX-12-15 120mm Effizio Quiet Case Fan
120mm
1400 RPM 74 CFM
15 dBA
Operating voltage: 6-12V
Current draw: 0.11A
Power consumption: 1.32W

 

Here are some links that might be helpful

http://www.techpowerup.com/forums/showthread.php?t=107135

http://www.bit-tech.net/modding/2001/12/03/pwm_fan_controller/1

http://www.devhardware.com/forums/pc-cooling-8/howto-homebrew-fan-speed-controller-283464.html

http://www.bearblain.com/fan_speed_control.htm

http://lifehacker.com/5866009/contr...rmance-when-you-need-it-silence-when-you-dont

There is lots of stuff regarding fan control.
I used a site (I can't find it now) that was about computer fan control using a 555, and I changed a few things to move it over to a simple 3V DC motor.

 

Forget it.. Not gonna happen (NOT simple at all).
That fan like others have said includes its own IC/hall sensors built into the fan. You can ONLY vary the input voltage (from like 7V to 12V) to control its speed. Your controller cannot control these. Not to mention that the fans that ARE controllable use PWM inputs NOT an analog voltage input for speed control.
Trust me..forget it.

If 12V is too much just use a lower voltage power supply (like 9 or 10V) and live with that set speed.

 

I appreciate you all posting comments, and thanks for clarifying the PWM with these fans. If PWM is not an option I won't be heart broken, but I'd still like to create a way to digitally control the voltage then.

For example, I'm going to use the fans running at low speed the majority of the time to keep airflow in my aquarium hood to keep the PH up (lets say 6v or 7v), but in the event of an overheating problem I want the ability to programatically (using my aquarium controller) to ramp the fans up to max speed (12v) and cool down the water.

So with knowledge from the given posts, I'd like to take a 0v-10v analog signal coming from my controller, and use it to control how much voltage is being supplied to the fans using a secondary power supply which runs at 12v, and upto 2A.

 

Some fans can tolerate PWM if it has the right frequency.
For others you need to use linear regulation.

The interesting point is most 12V fans turn on at 6V or so, but once they spin, you can go much lower than that. If you use a controller, you can easily program that.

Depending how you sample the RPM, or temp. etc., it can also make sense to add hysteresis. For instance you can do this in hardware, or you only program a small number of preset speed levels, let say 5 to 8 different speed levels.

Some new PICs include a 16 or 32 level D/A output, maybe could be used to control a MOSFET linear regulator. Have not tried that, just an idea.

 

It's interesting that you point this out, because I've noticed exactly what your talking about. If I plug the fans into one of the variable voltage wall warts to test them, I'm able to run them on the lowest setting at 3v as you've described. I can easily program this and hysteresis using my controller, I'd just need a way to design the circuit to vary the voltage based on the 0-10v signal wire.

 

It's interesting that you point this out, because I've noticed exactly what your talking about. If I plug the fans into one of the variable voltage wall warts to test them, I'm able to run them on the lowest setting at 3v as you've described. I can easily program this and hysteresis using my controller, I'd just need a way to design the circuit to vary the voltage based on the 0-10v signal wire.

One idea I developed for that was to use a capacitor of some uF on a MOSFET gate, together with a suitable discharge resistor.

The gate is then pulsed with PWM. A voltage will develope, and this voltage can be varied by the PWM ratio.

I have never tried to implement that. A digital MOSFET such as FDV303 or 2n7002 are suitable for this, directly using uC output.

There will be some jitter but maybe it can be designed/programmed so it is still useable. I don't know the dimensions for the cap., maybe it would rather be some 100nF.

 

See attached docs for two ways to control a voltage to drive a high current device like a fan, with a low current DC voltage.

The first is a Power supply like I described earlier. It could be a linear or a switching power supply. A linear will get hot and require heatsinking.

The second is an op-amp with a transistor driver. This is a linear circuit so it will get hot - the transistor needs heatsinking. Also the op-amp may not be able to drive enough base current into the transistor depending on your load (fan) current. You probably need to use a darlington transistor.

Oops! I noticed that my two files are labeled wrong. Sorry!

 

The thermal time constant of your aquarium is probably measured in hours.

You could employ a very long period PWM cycle of one minute for example, 16% cooling power = 10 seconds on, 50 off. The fan would have no trouble dealing with the start and stop. Then you don't even need to bother with controlling the speed.

Convert the 0-10 V signal into a Reaaaaaaly Sloooow PWM and use it to switch the motor on/off