Temperature controlled fan circuit.

Community Projects
1

My first post here, guess it’s sorta an introduction, kinda…

This is really my first attempt at designing, modeling, and implementing a functioning low voltage circuit. I say low, to me anything under 120v is low voltage…

The problem… My A/V receiver runs hot and I want to cool off it’s innards.

Solution…ish… So I wanted to run two separate computer case fans running on two separate temperature sensors, essentially a HIGH and LOW temp fans, I decided to make it fancy and add different color LED’s so I would know when what is running. I found myself in a conundrum of sorts as all of my knowledge is on mechanical jobs; I can TIG weld you a trailer, rebuild your car motor, and frame/plump/wire/drywall an extension to your house. Well all of that gets me nowhere in this current project.

So I wandered through some electrical engineering articles, brushed up on the components and their uses, watched a few basics on youtube and eventually found myself at [Everycircuit and what a neat little site that is. Amazon has since furnished my tinkering needs as far as components go, that and Fry’s Electronics.

So if you’ve made it this far I thank you and commend you, trying not to be too long winded but also what to convey that I don’t want the easy answer. I want to learn how to do this properly.

Here is an image of what I have so far… In the circuit I’ve replaced the switches with 10K ohm thermistors and the lights with 12v 80mm case fans

Help%20please%20Fans%20w%20LEDs%201_zpsd5as39gc.png1024×870 211 KB
http://i1083.photobucket.com/albums/j39 … as39gc.png

And every thing works, kinda… With a 10k trim pot I get 4v to the fans, enough to turn them on but not enough to move any real quantity of air. So I try a 1K trimp pot, now I’m getting 1.5v to the fans… weird… so I put a 500k trim pot and I get 5.5v to the fan… Now we’re talking. I don’t get it though.

What am I witnessing here?? Because it’s going AGAINST what I thought should happen. AND it’s not what is predicted by the computer application which says I should be seeing 10.2v at the “fans”, HOWEVER the schematic doesn’t account for the thermistors…

Do I need lower resistance thermistors? On top of all this I’m still trying to figure out how to determine which transistors are best for this application.

Feed a man for a day, or teach him… I’d rather be taught. Thank you for your patience.

Nick](http://www.everycircuit.com)

2

Well, there’s the kinda easy way and the not so kinda easy way. It looks like you’re going with the not so kinda…

First a few pointers. Schematic drawing convention has power supply lines at the top, ground at the bottom. If there’s a negative voltage bus, that goes below ground. Inputs (sensors, controls) on the left, outputs (fans, transistors) closer to the right. Signal flow from left to right, top to bottom. Little things like that make it much easier to read your diagrams.

OK, now back to the matter at hand (been listening to vintage Snoop Dogg recently…)

This circuit: http://www.cedarlakeinstruments.com/Sha … ontrol.png is a differential fan controller that switches the fan on when the battery sensor is hotter than the ambient sensor and it’s an example of a somewhat “harder” way to do it. It was designed for the parts the person I was working with had on hand. Don’t remember if I actually built it or not!

The “easy” way is with a semiconductor temperature switch that can control the fan directly. Problem is most of them these days are surface mount and don’t lend themselves to easy breadboarding and I can’t think of any parts off the top of my head that would do it in a single package anyway.

So let’s talk concepts you can Google and then come back for more detailed help.

A comparator (e.g., LM339) is a useful device to take a sensor input and a setpoint and switch the output on or off as the sensor goes above or below the setpoint. A comparator can generally drive a FET (sf has logic level FETs: https://www.sparkfun.com/products/10213) easily and the FET can switch fairly large motors. Comparator output by itself won’t switch much. It’s certainly possible to do this all with just a few transistors, but honestly it’s not worth the effort. Comparators are [cheap.

You can use a thermistor since you have them on hand, or a linear semiconductor temperature sensor that gives a predefined output, e.g., 10 millivolts per degree F and is easy to set. For one-off’s I prefer semiconductor sensors since they are just so easy to use (argh! tried to link to a blog post I did on sensors but realized I haven’t published it yet).

So, take the output of an [LM34 sensor (10mv/deg F), feed it into the + input of an LM339, give the - input of the '339 0.90V, and tie the output of the 339 through a 1k resistor to +12V or whatever regulated voltage is available (you want it regulated so you can use a simple & cheap voltage divider or potentiometer to generate the setpoint. Regulated voltage is coarse, but Good Enough) and watch the output. When the temperature is below 90F, the output will be off, when it’s above 90F it will be on. Now tie that output to the gate of your FET, ground the FET’s Source pin, and connect your fan between +12 and the Drain pin et voila: fan goes on and off with temperature.

I’d draw a diagram but CircuitLab is apparently no longer free. grrr.](http://www.bgmicro.com/TRNLM34DZ.aspx)](http://www.bgmicro.com/ICSLM339.aspx)

3

I really like the possibility of using comparator’s, this may be a more feasible solution. My curiosity is however still piqued with my results. I’d like to understand the results I get from this circuit even if it won’t be the final solution to my problem.

I’m ordering from BG the comparators, they have what appears to be a similar MOSFET, IRFZ34 which i’m going to order as well only for the sake of reduced shipping. At work now, will redesign the circuit with a comparator as well as the correct schematic etiquette. :wink:

Nick

4

Ok, so I finally received the components from BGMicro but to be honest I’m having a hard time figuring out how to supply .9V to the comparator. Every time I try and solve for resistance I get .45 Ohms and frankly I don’t have any resistors that low. So, dumb it down for the new guy, what the heck am I doing wrong here? Maybe I’m over thinking this…?

5

Simplest way would be to use a 10k multiturn pot across power & ground (preferably regulated power) and adjust it for 0.9V, although it will be difficult to adjust. A better way would be to make a voltage divider first.

What resistor values do you have available?

6

Basically majority of the most popular 1/4W, it was a 640 piece kit with 10 of every value. I’ve got a plethora of trimmer pots from 500 Ohms to 500k Ohms.

7

Try this: I can’t draw a schematic, but use an 820 ohm and 10k resistor voltage divider with a 1k pot between them apply 12VDC to the 10k and ground at the 820 ohm. Take the output at the pot’s wiper. With a regulated 12V supply that will let you adjust the voltage between 12 * (820/11820) = 0.83V and 12* (1820/11820) = 1.84 easily

8

I’ll try that tonight. Thank you sir.

9

I made a cooling fan for my receiver using old tech… A 12v power brick, two computer fans, and a mechanical thermostat. It works great. I got the thermostat from Home Depot in the attic fan section. And its adjustable.

10

OK, so an example circuit is shown here:

http://www.cedarlakeinstruments.com/blo … ivider.jpg

The entire blog post is at [this location. Thanks for giving me the idea for a new post! It was fun to write, even if hurried and Lefty and Poncho gave me dirty looks…

[edited to fix image link]](http://www.cedarlakeinstruments.com/blog/archives/337)

11

http://i1083.photobucket.com/albums/j39 … locdoq.jpg

Back in business.

Tried using my thermistor(s) last night, no dice. Going to have to do more math it looks like. Room temp (80F) they are 8k ohms and at body temp (95-98ish) they are 6.8k ohms. So on the current circuit they are not impeding at all, as soon as I plug in the fan gets powered up.