Would this work ?

[diachi]

Say I wanted to run a fan in a laser, but I wanted it set up so that It would vary it's speed according to the heat of the laser diode/module .

Could I use an LM317 driver and replace the pot with a heat dependent resistor and attach the fan to the outputs of the driver ?


Like in this image.

 

[diachi]

And ignore the " Values shown for Blue-ray " lol

-Adam

EDIT : would I need to put an inverter in the circuit somewhere ? so that the fan doesn't slow down when It gets hotter.

 

[diachi]

bump

 

[billg519]

The thermistor in this circuit would have to be able to carry the current drawn by the fan, without this current in itself causing the thermistor to heat up. You would need a thermistor that decreases in ohms as it heats up. This is not a problem as NTC and PTC thermistors are available. I'm worried here about the current needed by the fan causing the thermistor to heat up. Fan would become locked on at full speed. Maybe put a transistor's collector and emitter where the thermistor is currently in your circuit, put a thermistor to this transistor's base. Something simple like this should work. Do you have a prototyping board? Could be a fun project.
Hope this little bit helps.

Bill.

 

[diachi]

Thanks that helps a bit, got a prototype board yet, I just wanted to know if it was possible or not first.

I had another idea which was a thermal sensing circuit that switches over a relay when a certain temperature is attained, when the relay switches over it turns on the fan, this idea is not as good however because the speed wouldn't be varied, it would just be on or off.

-Adam

 

[billg519]

In "Experiments & Modifications", kernelpanic has a thread going about a new low dropout current regulator based on a mosfet and opamp. You could add a thermistor to this circuit and contol fan speed without the thermistor carrying too much current. Just an idea ..

 

[diachi]

Hmm sounds interesting .