Fan Controller

fan controller

Unhappy with commercial fan controllers Jos van Eijndhoven decided to build his own. The circuit supports three LM60 temperature sensors with pots to adjust the turn on temp. A PIC 16F676 microcontroller reads the temps and controls three groups of fans. A potentiometer is also supplied to control the minimum fan speed. To prevent oscillation the fan speed is reduced slowly in response to temperature drop.

[thanks Alan]

18 thoughts on “Fan Controller

  1. Yes, Chris, but his responds to temperature, whereas it appears that the one you linked does not have the capability to do anything except manual control.

  2. I’ve been thinking of doing something similar to this for a while, but have been working on other projects. This guy did a good job.

    I would want to add an LCD to display current temperature and settings, and maybe a serial connection so it can be controlled via the serial port.

  3. I say, I like this hack. I’ve had similar experience with commercial fan controllers that don’t do EVERYTHING. Either they only work off of temp, or only work manually, whereas this does both, and has little things like decreasing the speed slowly to avoid the stress of changing airflow quickly. Good job.

  4. “How silly to hack like that”

    *Sigh* some people just don’t get it…

    PS: Seriously, can we stop with the “fist post” nonsense? It’s really stupid and annoying.

  5. i think this is a good hack for a beginner, no? well, lets keep our comments on-topic please. no more “this hack sucks” type commenst, they do nothing for the community.

  6. T_T I’m currently making a similiar version to this, and its too bad that I havnt had enough time to finish it before this was posted. Even though its simpler, hopefully mine should turn out looking nicer, and has a pcb. >_>;

  7. I’m not an electronics expert, but shouldn’t it be possible to exchange the tempature sensors for other sensors? I mean, some people wan’t a fan turned on when a certain ammount of carbonoxide is measured (kitchens,boilerrooms).

    – Unomi –

  8. Its a nice hack, but I have a question… Why not use disceet components with a comparator window instead of going to all the necessity of programming? I can rob dicreet components, use the same amount of board space or less, and buy less or cheaper with similar results.

  9. Its a nice hack, but I have a question… Why not use disceet components with a comparator window instead of going to all the necessity of programming? I can rob dicreet components, use the same amount of board space or less, and buy less or cheaper with similar results.

  10. I like this alot. I built a manual one for my small soho server room that uses a radon abatement fan, some 4″ dryer hose, and a board that fits in the window like an Air Conditioner would. It The fan collects heat from the top of the room and pumps it outside in response to a Granger thermostatic control but its a single step — you set the temp and the room is cooled. In winter, its all in need. In summer, an actual AC unit has to be used instead.

    With this, I could use a relay on it to turn on the AC when needed,or the fans for intake to the room and output based on a variety of factors. For example, if the room is too warm and outside is cool, pump warm air out. If the room is warm, the outside is below freezing, pump the warm air into the rest of the house or the garage to make use of that heat. If its too warm outside to cool the room, kick on the AC unit. It works out to a very nice solution.

  11. 11. In my experience, using a microcontroller actually reduces required board space and can be cheaper when compared to discrete components. You also gain an extremely high amount of flexability. YMMV

  12. id say the reason for using an mcu is because it can do numerous independent compariting and pwm (I assume this is the speed control method). if you were going to use opamps & logic or discrete transistors to do this the board size will increase quite a bit which means more that can go wrong when using a prototype board! but thats not to say its not possible by any means.

  13. I’d agree with bronzeg3 and mahoney that using a microcontroller can definately reduce cost. To reduce the oscillation, you could use PID control. The control used at the moment is known as proportional control, where the output is change proportional to the ‘error’ in. PI control changes the output depending on the error and the integral of the error. This has stability problems also, but will always reach the target. PID control has a derivative component, so the output changes relative to the rate of change of input (thus if you’re approaching the target temperature quickly, this will slow you down). It’s a lot more complex than that (google Control Theory, Root Locus, s-plane). Putting digital PID control would be easy though, now you’re using a micro controller already.
    (just an idea :)

Leave a Reply to mahoneyCancel reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.