Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Help with Temp Ckt to drive fan

Not open for further replies.


New Member
Hello all. I need a circuit that will provide an output high when the temperature is sensed to be say 160 degree F, but not turn off (go low) till the temperature drops back to say 130 degrees F. I found the LM56 ic and and built it 3 times, still can not get it to work. Rather not use the LM56 anyway, tooooo small when you have 10 thumbs as as I do. As for the switching, I have a circuit that will respond nicely to logic level output of such a circuit.

There are plenty of fan control ckt's, but they all turn on at threshold and off when dropped below threshold. I want it to stay going till a lower temp is reached.

All suggestions will be considered and appreciated:).
I've attached a schematic for a 2 transistor hysteretic controller that has worked for me many times over the years.

Basically it uses the 0.6v Vbe turn-on voltage of a NPN transistor to act as the voltage threshold detector, then a second transistor operates in inverse to the first transistor and provides an adjustable amount of hysteresis based on 1 feedback resistor.

By messing with the main voltage divider and the hyst resistor you can pretty much set whatever hi-low voltage limits you like.

It's very simple and very reliable, one of my home made lab PSUs uses basically this circuit to operate a heatsink fan. It will be similar to comparator accuracy provided it runs from a regulated supply.


  • 33633d1253262274-simple-2-transistor-re-circuit-black_controller.gif
    53 KB · Views: 584
Last edited:
Thanks for the info

Thank you very much. I will build this in a couple days and report the results to you. If this work's as advertised, you are a blessing...
Built it - now ?'s

Hi RB;

Its built, the 100W version, and I am playing with it now. Seems like it will do the trick. I left the FET out for the moment, just trying to find the limits. I originally powered with 12V un-regulated, ooop's on that, I just read your post again, will get 12 regulated on it next.

Does the zener diode need to be of a larger wattage? And just exactly which component is sensing the temperature? I thought its supposed to be the 1st BC337, then thought about it and considered the zener but that makes no sense. Please keep in mind - rookie - here. I'll get back to playing with it now and hopefully find the limits I need.

Oh yea, on both ckt's. the hyst. resistor is listed as 330K ??? I gather you mean that's a good value to begin with and adjust for the target from there?

Thanks for your help.
Got it... mostly...

Hello again. I got it working and can dial in the on temp., tight range with a 10K pot, but will narrow that down a lot on next build. And I confirmed what I thought originally, it is the 1st BC337 that is sensing the temp.

Used a 1 watt zener, just fine. Now the turn off temp I can not seem to get further back down the temp. scale. I got it turning on at 175 F and played with the hyst. resistor (used a pot to cover the 330K range), then went to a 2 meg pot, I gather I can't just add larger resistance there for the hyst.??? Best I can seem to do is about 15 F cooler when it turns off. I would like a larger range. Can you recommend anything?
Hi again, whoops I forgot to say you should lose the zener, that was for the app written on the schematic to sense battery voltage and the zener is used there to make it more sensitive to small changes in the input (battery) voltage.

For temperature sensing, the thermistor should form a voltage divider with a resistor (and/or pot) on the input to the first transistor.

If you had the zener in the top half of the voltage divider with the NTC thermistor the zener was reducing the effectiveness of the thermistor, hence your larger hysteresis and clunky operation.

Try it without the zener, and fiddle a bit with the voltage divider resistor values for the base of the first transistor, you should get it within a couple percent hysteresis with no problems. The one on my PSU just uses a NTC thermistor and a resistor and a pot connected to the first transistor, and it switches the fan on at 45'C and off at 43'C or thereabouts. A 1M or 2M pot should give you lots of adjustment for the hysteresis.

Another thing you can do to make it a bit more sensitive (and switch a bit quicker) is put a 1k or 680 ohm resistor from the base of the second transistor to ground.

The circuits above were drawn from memory for a specific task (battery voltage detect) so you might need to tweak a few parts values, but the basic 2-transistor hysteretic controller design is sound and can be used for just about anything you would use a comparator controller for.
On to next build...

Hi RB. Thanks again. I'll do as you point out. I thought I was doing something wrong, well, I guess I was & didn't know it...

I want to get the turn off temp about 30 F lower than the turn on. Do you think this circuit can be made to do this?

I'll let you again know what I come up with.
Not open for further replies.

Latest threads