1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Harvesting Sound Energy

  1. Apr 6, 2015 #1
    Hello! I'm new to this website, so apologies if I'm a little off with the way I format this question.


    This last year, I've been researching sound energy and methods of focusing, harvesting, and storing this energy.

    Essentially, I'm looking at the possibility of harnessing sound energy by a means of unpowered, and potentially, recycled materials.

    The common understanding is that the amount of electrical energy generated by most sound energy harvesting is ridiculously small, and of little to no use.

    However, the last few months, I've come across a few interesting reports that show using certain piezoelectric materials, a significant amount of voltage can be generated from a sound source.

    For example, a Korean experiment in 2010 showed 50mV AC could be generated from 100Hz at 100dB:


    I know this is a very small amount, but my idea (albeit theoretical) would be to employ this material in an environment like an airport runway (London Heathrow has 1,300 air traffic movements per DAY) where levels reach close to 110dB (which frequencies exactly, are yet to be calculated...)

    This energy would then be stored in battery circuit, specific for piezoelectric harvesting, such as the following here:


    Most importantly, I'd be looking to focus a frequency of 100Hz by placing the piezoelectric material (as mentioned above) inside a Parabolic dish, similar to a parabolic dish microphone:



    Does anyone have any advice on focusing lower frequency sound waves? What kind of material and dimensions would I require for this dish?

    Does anyone have any advice on the project as a whole? Any major mistakes I'm missing or important things I haven't considered properly?

    Any comments at all would be greatly appreciated :)
  2. jcsd
  3. Apr 6, 2015 #2
    There are different ways to do this

    One is to amplify the power of the sound to a point where you could practically use anything

    Or find a suitable conductor (think cup phones) because sound is both an energy and a particle
  4. Apr 6, 2015 #3


    User Avatar
    2017 Award

    Staff: Mentor

    Voltage alone does not help, you also have to know the current to get the power.

    Charging a battery is probably pointless - grid electricity is so cheap even the battery costs will make the whole thing impractical if you want to replace grid power. The applications are places where grid electricity is impractical because you don't want to have hundreds of cables for every sensor, for example.

    The parabolic dish helps for localized sources only. If the sound is coming from everywhere, it does not do much. 100 Hz has a wavelength of ~3 meters, that is "coming from everywhere" independent of the source unless your parabolic dish has a diameter of 10 meters or more.
  5. Apr 6, 2015 #4
    Thanks Christianinsci and mfb,

    In response to your points Christianinsci, I'm looking to design a parabolic dish along these lines:

    Of course I can't build something like this by myself, but in terms of unpowered amplification, this looks like a possibility, especially on a runway. Whether it channels lower frequencies (more power) is what I'm trying to figure out.

    Thanks for the advice mfb, I should definitely have thought of this (the current!). My main objective is to calculate how much power can hypothetically be generated using a number of dishes placed along a runway at Heathrow over the course of a year. By no means do I attempt to power anything large, but rather see if this energy can be used to power anything at all. After all, the noise from the airplanes will exist regardless (and is infact considered noise pollution), the dishes would be unpowered, essentially its free energy. I'm just curious to figure out how much, and how it can possibly be improved.

    Your point about the dish is also really helpful, sorry if this seems like a silly question, but does this mean a diameter of 10 metres or more is a good thing? Where is the formula that explains this? (I'm considering a really large dish size). The absorption coefficient of the dish material I imagine is important too?

    Thanks again guys, this is really helpful
  6. Apr 6, 2015 #5


    User Avatar
    Education Advisor
    Gold Member

    I suggest you compare how much energy is used by large powered speakers and the amount of noise they generate. This is your 100% theoretical recovery value. Is it economically worth it (you could use the speakers to backfeed a circuit) to recover this energy?? I suspect not, but that would provide you with a ROM (rough order of magnitude) calculation to determine the feasibility of your project.
  7. Apr 6, 2015 #6
    My initial thought was that generally the amount of ambient noise in the environment could never produce a worthwhile amount of electrical power.
    Harvesting sound next to major hub airport though .. hmm.
    Well you probably could get enough to charge a car battery over few days.
  8. Apr 6, 2015 #7


    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    The word "harvesting" always gives me a problem in this context. It implies there is enough of a resource available to be of any use. Before you launch out on any project, you need to do two things. Firstly decide how much Energy you want to get, over a given period. You then have to decide whether there is actually this amount of energy available, in this case, in the form of sound. Those are the basic principles behind Energy Engineering and they always apply.
    You need a specific source of high energy (obviously) sound - say a jet aircraft taking off or the noise in a turbine hall or motorway tunnel. You then need to quantify how much sound power is present at a given position.
    But, before getting in too deep, it may be worth while reading this article from MIT, about the notion of Harvesting Sound Energy. It is pretty unenthusiastic about the idea, based on some realistic figures about the actual amount of energy available.
    Microphones 'detect' sound energy and it is at such a low level that the first thing that is needed is amplification by a factor of 100 or more (in volts). What comes out of the amplifier has not been 'harvested' because the power to produce the enhanced signal has come from the Power Supply and not from the original sound source.
  9. Apr 6, 2015 #8
    Thanks for the feedback guys,

    Calcnerd, would you be able to go into more depth about backfeeding a circuit? Or perhaps provide a link on how to do this? I'm not much of an electronics expert, so would need some thorough guidance...

    sophiecentaur, I understand where you're coming from, and yes I've read that article from MIT (infact it was the first thing I came across on the subject haha). I think you're right about the idea of starting with a desirable figure for the power and what you want it to be used for. I was planning on taking the approach of just calculating how much could be generated using the ZnO piezomaterial (I mentioned in my opening post at the focal point) of a parabolic dish (which I'm still figuring out how to design properly) on a runway known to produce levels of 110dB, 300 times a day, 365 days a year. As I say, I don't mean to assume it will be enough to power something hugely significant, but perhaps the staff coffee machine? :D

    Would you have any suggestions on practical research I could carry out in my garage or at home to help contribute to this idea? I was thinking of sending a higher frequency sine wave from my monitors and seeing how well a miniature scale model of the dish (relative to the frequency) would amplify the signal? I'm sure there's probably plenty of better ideas, but I don't have loads of spare time unfortunately...
  10. Apr 6, 2015 #9
    Sorry, this was meant to read:

    "using the ZnO piezomaterial (I mentioned in my opening post) at the focal point of a parabolic dish (which I'm still figuring out how to design properly) on a runway known to produce levels of 110dB, 300 times a day, 365 days a year."

    put the bracket in the wrong place :)
  11. Apr 6, 2015 #10


    User Avatar
    2017 Award

    Staff: Mentor

    Let's take that as a starting point, and use that massively oversized 10m parabolic dish to get 1W of sound energy. Only a small fraction of this will be converted to electric energy, but I'll assume 100% efficiency here. How much money do we make with 1 W? A year has about 30 million seconds, which leads to about 8 kWh per year for 1 Watt. Selling this gives ~0.3 euros (or dollars or pounds, does not matter for the precision of this estimate). Setting up a massive parabolic dish does not even give a single euro per year!
  12. Apr 6, 2015 #11
    Hey mfb,

    Thanks for the response, but I'm a little confused by your answer? Could you explain a little further how you came to the calculations after 'How much money do we make with 1 w ?' Just want to make sure I fully understand.

    What if we were to consider the time at which it takes a jet engine to take off and land, and the distance of the south runway at Heathrow (3 km) meaning that multiple dishes could be set up (lets say hypothetically, 50) and plus, given the calculations I've performed so far for a dish with a 6.8m diameter, a gain of 16dB @ 100Hz

    Also, the amount of piezomaterial used and how it's used could play an important part as well, no?

    I appreciate the constructive criticism :)
  13. Apr 7, 2015 #12


    User Avatar
    2017 Award

    Staff: Mentor

    I made it easier, I assumed you can constantly extract the quoted power, which is an overestimate.
    You can set up multiple dishes but that will cost more, so it does not help much. A smaller dish diameter will lower the extracted power - I also assumed 100% focussing which is certainly not possible with a real dish and receiver. And I assumed 100% conversion efficiency for the receiver, which is probably a massive overestimate.

    Even if the result would be too low by a factor of 100 (and given the very optimistic assumptions that is not likely), such a system makes 30 pounds per year - just getting the permission to build anything will cost thousands of pounds, and the dish alone will cost more than 1000 pounds as well. That is not even including the money for the receiver, for the required place to set them up, for the electricity conversion to grid values, for monitoring the system, repairing the dishes or the receiver, making contracts with someone to sell the produced power, ...
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook