• 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.

PWM fan speed control by various thermistor

Status
Not open for further replies.

jocanon

Member
I want to be able to control the speed of a simple 4 wire PWM computer type fan based on the temperature of various components. I bought this item which will arrive eventually on a slow boat from China:

http://www.ebay.com/itm/221878598666?_trksid=p2057872.m2749.l2649&ssPageName=STRK:MEBIDX:IT

It appears this little gadget will do what I need, except I want to make one minor tweak, that is, instead of controlling the temperature based off of the change in only 1 location, I want to expand it so the fan will rev up if the temperature heats up in one of several (probably about 4) different locations. I have a soldering iron and some soldering skills and I know enough about circuits to know (or at least think I know) that this is pretty easy to do, I just do not know enough to actually know how to do it (i.e. what components I will need).

So in short, I would like to create a schematic and BOM for parts that in the end will allow me to put a varying amount (at least up to 4) separate thermistors in 4 separate locations that all link back to one central location where if any one of the separate thermistors heat gets above a certain level then it will cause the speed to increase on the PWM speed controller (instead of just the one thermistor that comes standard with the eBay device I purchased).

Is this as easy as just connecting the desired quantity of thermistors in parallel and then plugging them all into the speed controller? Would this mean that the thermistor with the least resistance would set the speed (in other words the one connected to the hottest device at the time would be the one that would set the speed)?

edit:
now that I think about it, the conductivity of the thermistors would probably be additive. For instance, if I had 2 thermistors and both were open half way while being connected in parallel to the speed controller, then the fan would go at full speed, versus if I just had one that was open half way and the fan would go half speed. Am I thinking of this correctly? If so, is there a way to make a circcuit where they would not be additive, but just the hottest thermistor would set the speed and override all the others as if only the hottest one was the one connected at any given time?
 
Last edited:

MikeMl

Well-Known Member
Most Helpful Member
Do you have any idea of the resistance vs temperature curve of the thermistors? At what temperature do you want to the fan to come on?

This is likely to get quite complicated; like opamps, comparitors, and some logic. Are you up to that? How about an Arduino?
 

crutschow

Well-Known Member
Most Helpful Member
So if two locations become warm (say at the same temperature) do you want the PWM to have a larger duty-cycle so that the fan runs faster than if only one location is warm?
Or do you want to run at a speed proportional to just the hottest location.

For the first case you could use a summing op amp circuit.
For the second case you could use a maximum value circuit.

Below is the LTspice simulation of a maximum value circuit, since that sounds like what you want.
It's output is equal to the value of the highest input.
It's shown for two inputs but can be readily expanded to 4 or more.
If you need the minimum instead of the maximum just reverse the diodes and add a pull-up resistor from Out to V+.

upload_2017-5-1_14-51-18.png
 

Attachments

Last edited:

jocanon

Member
Thank crutschow, I think that is exactly what I am looking for. I just want the speed to run proportional to the hottest location. As a side note, I am not too concerned with exact values at this point. My plan is just to build the circuit then experiment with a range of thermistors and ultimately place them either nearer or further away from the heat source as needed. It doesn't have to be too exact, I just want to be sure that once the devices get warm then the fans kick on and move some air. The application I am using this in is debatable whether or not fans are even required at all, but I figure better safe than sorry, so as long as they come on and move some air then that's all I need, no need to get very exact with speeds of the fan.
 

jocanon

Member
I might even have some, or all, of those parts laying around from a previous project. I'll check when I get home and let you know if it works.
 

jocanon

Member
...Below is the LTspice simulation of a maximum value circuit, since that sounds like what you want...
View attachment 105833
I have the LM324AN and LM358AN. Do you know what the "AN" at the end of the name indicates? Will these work? Also, I am not sure I have the 1N4148, but I do have some other diodes. What are the key specs on the diode that I need to be sure it will work on this circuit?
 

crutschow

Well-Known Member
Most Helpful Member
I have the LM324AN and LM358AN. Do you know what the "AN" at the end of the name indicates? Will these work?
I think "A" has slightly improved specs such as input offset voltage and bias current.
An "N" is the DIP package.
It will work fine.
Also, I am not sure I have the 1N4148, but I do have some other diodes. What are the key specs on the diode that I need to be sure it will work on this circuit?
Just about any small silicon junction diode will work.
Its specs are non-critical.
 
Status
Not open for further replies.

EE World Online Articles

Loading
Top