It's only the very beginning of school for me, but I've already got my science fair idea approved so I'm starting to work on it now! As the title says, I'm going to be taking sound, and turning it into electricity!

I first thought about this during the time I was ranting about how the clock in the band room keeps stopping because of a dead battery. Then it takes the maintenance people 4 months to change the battery (since it's high on the wall, no one else can reach it without a ladder). I suddenly thought, "hey it's always loud in the band room, why not try and recharge the battery from that?"

With the small amount of research I've done so far, I've learned that I can simply use a speaker to generate electricity. I've also found out that sound energy is abysmally inefficient, especially compared to other solutions like solar power.

Although those results are pretty depressing, I want to try and improve the results. Perhaps by simply using a bigger speaker, a higher quality one, or adding a funnel that'd channel the sound into the speaker. I was wondering, does anyone have any experience with this? Or does anyone have any ideas on how else I could improve the results? It'd be a miracle if I could generate enough power to charge the clock.

While I was researching, I also found out that turning vibrations into energy is slightly more efficient (Yes I know sound is just air pulsing at different speeds). (Be warned, I have no idea what I'm talking about here) Maybe I could have different lengths of metals, each that resonate to a different pitch (like a tuning fork). Since the band is always playing a wide range of notes, perhaps that resonance could be turned into more efficient energy than using a single speaker?

Could you also try multiple speakers? Like 10 small speakers in an array vs one big speaker. Since sound it propagates outward as it moves, you I hypothesize that you'll get more energy from an array of speakers than a single one. Now, does the spacing matter? I don't know how big of speakers you're thinking but is there a difference if they're 1 inch apart or 1 foot apart?

Essentially a microphone I guess, though not all types. The power would indeed be tiny, but it would be a good experiment none-the-less.

Like the original idea. Would a microphone work better, since it is designed for sound input? You could also try piezoelectric materials coupled with the "tuning fork" idea. Pizeolecltric materials might also work if used between that clock and the wall, since the noise would also cuase the vibration of the whole building. However, the strips of metal might produce and audible, and distracting sound if vibrated at the right frequencies.

That's not a result (from some experiment of energy harvesting), that's the actual amount of energy embodied in the acoustic wave- a perfectly efficient harvesting device needs to be very large to get much energy out of it.

There have even been businesses trying to sell this sort of thing, so there are also people who know what they're talking about discussing why it's impractical. Here's Dave Jones, for instance:


You'll note that uBeam is supposed to work with ultrasonic signals which are higher-energy and safer; a literally deafening amount of sound will get you maybe a tenth of a Watt at the sort of range you might care about (a few meters).

On the other hand, clocks require very little energy to run. So you might be able to get away with trying to go extremely low-power, and this application is somewhat more sane than a dedicated acoustic charging system since you don't really care about efficiency as long as you can get enough power out of it.

Wow, thanks for all your replies guys!

I'll probably be turning to either microphones or piezo. You have some great questions I will have to test. I'll probably be using multiple small ones though, packed tight together.


I spent a lot of time discussing this with MateoC yesterday, I'll be most likely using a piezo, unless my test reveal a more efficient method. I was thinking, don't piezos pick up higher frequencies better than lower ones? Since most of the band typically plays mid to low range notes, would something else work better?


Thanks for clearing some of my info up! I'll have to watch the video when I get wifi that doesn't block Youtube.
Mateo and I did some math for a cheap clock I had in my house. It took in 3mAh (roughly) every 1/4th of a second. So that's about 18mAh per day. So I just need to generate at least 18mAh every day to have a self-sufficient clock.

Got to go so I'll cut this post short. Thanks for all your replies and ideas, it really means a lot!

Necro-update! My new science fair partner and I finally started ordering some of the parts we needed. The contact microphones came in today so we borrowed two tuning forks and did a quick test using my multimeter. The results look promising so far! The small microphone (it's slightly larger than a quarter) was able to generate .2 volts however I wasn't able to pick up how many amps it produced (but that could just be due to my inexperience with multimeters) I've got a short video of me recording the voltage.




TLM 1100 Posts!

Neato

Or just use the electricity to power a CE (find a way), then run Nuclear Reactor Sim on it

https://www.cemetech.net/forum/viewtopic.php?t=15191