Desk Fan

It all began when I decided I wanted some sort of fan on my desk. I have this 4-1/2 inch muffin fan I removed from an old server a long time ago that moves serious amounts of air and can be slowed down to be very quiet and mellow. So I built a manually controlled desk fan with it. But I wanted it to be temperature controlled and as automatic as possible. It also didn’t have nearly enough lights on it.
I decided this would be a great opportunity to revisit my electronics obsession from years past. So I dug out my books and got to work.

Features:

  • Temperature driven speed control from off to max.
  • Fan shut-off below minimum speed to prevent stalling.
  • Hysteresis to eliminate erratic fan cycling at set temperature.
  • Set temperature is user-adjustable between 72 and 82 degrees F.
  • Sensitivity is user-adjustable between +10 and +3 degrees F.
  • Digital temperature display.
  • Eight step multicolor fan speed indicator.
  • “Max/Sens” switch to force max speed, bypassing temperature sensor.

I completed the desk fan on Aug 17, 2008. I’m really happy with the final product (2nd image on the right) and immediately put it to work on my desk at home. It’s going to permanently live on my office desk, however since the air conditioning there is so bad.

One thing I did in this project that I had not planned on was to fabricate a new chassis out of raw materials. It proved to be a lot of fun to do this. But I went into the project expecting to buy a project box and mount everything to it. I’m glad I made my own chassis because it looks much nicer than I would have gotten otherwise.

The circuits utilize op amps for the temperature control, transistors to drive the fan and comparators to run the LED speed display. The temperature display has its own tightly calibrated sensor and is completely separated from the fan control circuit. This was not intentional however. I had planned to drive everything from the same sensor, but the pre-made display had a circuit design that prevented me from doing it that way. It worked out better anyway, because the display would change based on the fan speed when it was attached to the same sensor. So the two sensors are mounted in contact with eachother in order to keep them as close to the same temperature as possible.

This project opened up a whole world of ideas. I’m currently considering revising my bike light with circuits that monitor the battery voltage and do things based on that.