Perhaps my understanding of how magnets work is off, but would it possible to play some waveforms into the positive and negative leads of a DC fan (think motherboard fan) and get a coherent sound? I understand if this were possible, that you'd need amplified sound to be driving into the fan to get a sound due to design constraints, equipment I don't have so I can't test it.

Crazier... If it was possible to play sounds from the fan, would it be possible to attempt this with the fan spinning?


You could try that, but if you get any sound out then it would be by accident. It's equally likely that you'll damage something trying. Standard DC muffin fans want a DC input, not an AC input. They don't work well with anything AC.

There is a type of device called a rotary woofer, which is basically a fan that changes the pitch of the fan blades to move air. The speed of the fan remains constant, but only the blade pitch is changed. It's wacky, for sure.

  • \$\begingroup\$ The rotary woofer is effectively an amplifier - it modulates an air stream using low energy pitch control to modulate the energy input from a high energy air driver. A good enough idea [tm] if fidelity is acceptable for the cost. \$\endgroup\$ – Russell McMahon Apr 4 '12 at 2:37
  • \$\begingroup\$ Learn something new every day! and that's a very fascinating contraption. Thanks for the link David Kessner. \$\endgroup\$ – Kyle Apr 4 '12 at 3:31
  • \$\begingroup\$ Would there be any trouble if we were to drive an AC fan using sound output? \$\endgroup\$ – AndrejaKo Apr 4 '12 at 5:35
  • 2
    \$\begingroup\$ @AndrejaKo it wont work as firstly you would need to amplify the audio signal to some usable voltage to drive an AC motor and secondly the motor's inertia would prevent it from changing speed fast enough to give an accurate representation of the sound. \$\endgroup\$ – Konsalik Apr 4 '12 at 8:51
  • \$\begingroup\$ The rotary woofer also has frequency response down to DC (that is, it can raise the static pressure of the entire room). Of course it depends somewhat on the specific installation. \$\endgroup\$ – mkeith Sep 3 '16 at 5:10

Most fans (I have a brushless DC fan here that does it) starts to whistle when you drive them from PWM. By changing PWM frequency and duty cycle, you change the sound. It would sound more like an 80's computer game than a hi-end stereo though.


I have done this before on Sumo Robot motor driver. The drivers played a sound using the motor coils when they booted up to let you know they were all alive.

The idea is to pulse the direction back and forth at the frequency of the sound. An Arduino Example can be found on this site


Requires the ability to reverse the direction so may not work with the computer fan. May be possible to do while moving with a drive/stop motion.

  • \$\begingroup\$ There's no need to reverse, pulsed unidirectional power will work. \$\endgroup\$ – Chris Stratton Apr 30 '13 at 10:13

Maybe not exactly what the posting person had in mind but this may be of interest: Ham radio operators used to use a DC fan's rotating blades to modulate a He-Neon laser beam passing through (actually between) the rotating blades.

I have done something similar using a CPU cooling fan salvaged from a computer. I mounted the fan in front of a small mirror, (approx. 5" x 7") that reflected sunlight onto a photo cell about 50 feet away. This produced a nice appoximately 600Hz tone in my little Radio Shack "Mini Amplifier" (US$13).

The modulated sunlight in turn passed through a home-made shutter. This is a minature version of those used on WW2 warships. This rig enabling me to send Morse code via sunlight.

I have not tried using voice modulation of the fan motor's current but a highly qualified EE Ham assured me that it will work. Thanks for listening, Ed KF6DXX

  • 1
    \$\begingroup\$ This has rather little to do with the question. \$\endgroup\$ – Olin Lathrop Nov 25 '12 at 0:34
  • \$\begingroup\$ I agree, but this is a facinating tidbit. \$\endgroup\$ – Kyle May 1 '13 at 23:50

It works with a brushless DC computer fan. It is the coil whine that makes the sound. You can get intelligible music out either with the fan spinning or not. They seem to work better with the blades locked. I used a soft handled flat blade screwdriver to lock the blades and transmit sound to a desk. They vibrate when running too so a hard suface is useful. I've done it with half an LM324 or a single LM386. the idea is to feed the fan a DC-biased, amplified signal. For the LM386, look at the datasheet. You don't want the cap on the output though. For opamps, such as LM324 or LM358, try a summing amplifier. I ran a pot from Vcc to GND, unity gain buffered it and tried a few different summing/subtracting topologies. The idea is to add or subtract the signal and baseline and amplify the result. You can drive the gate/base of a transistor if you need more power. A combination of these may be even more effective but I'm not there yet. Older CPU fans seem to work best. i think the anti-lock mechanisms in newer fans may have a part to play. Adding a resistor (10-100 ohm, careful with power!) before the fan can help. I used 11.75 ohms on a 12V 700mA fan running a baseline of 6V, gain 200 (LM386) and got reasonable sound quality, for a fan at least.

  • \$\begingroup\$ Welcome to EE.SE, Matt. Please capitalise and punctuate your posts properly as in standard English. It will make them more readable and give a more favourable impression of the author. \$\endgroup\$ – Transistor Sep 2 '16 at 21:22
  • \$\begingroup\$ Almost there. "LM324" and "DC" but good enough for first post. If you ever need to there's a very good and easy to use schematic tool - button on the editor toolbar. Have fun. \$\endgroup\$ – Transistor Sep 2 '16 at 21:33

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.