Can You Piezo A Peugeot?

Car manufacturers have a problem when it comes to climate change. Among the variety of sources for extra atmospheric CO2 their products are perhaps those most in the public eye, and consequently their marketing departments are resorting to ever more desperate measures to sanctify them with a green aura. Among these are the French marque Peugeot, whose new electric version of their 208 model features in a slick video alongside a futuristic energy-harvesting billboard.

This is no ordinary billboard, nor is it a conventional wind turbine or solar array, instead it harvests ambient noise in one of the busiest parts of Paris, and turns it into electricity to charge the car with an array of piezoelectric energy capture units. This caught our eye here at Hackaday, because it seemed rather too good to be true. Is it a marketing stunt, or could you make a piezo billboard as a practical green energy device? Let’s take a closer look.

What Is Piezoelectricity Anyway?

As most readers will be aware, the pieoelectric effect is a property of some materials by which electric charge accumulates at their surface when they are deformed. In turn a piezoelectric material will also deform when an external electrical charge is applied to it.

There are a variety of piezoelectric substances commonly available including crystaline materials and polymers, and most people will encounter them in domestic applications such as speakers, buzzers, and gas lighters light like that clicky button that ignites your grill. The spark generated by those gas lighters demonstrates that the voltage created by piezoelectric materials can be surprisingly large, but they have by nature an extremely high impedance so they are not generally capable of delivering significant current. Yes, in theory they could be used to generate usable power but it does not appear to a casual observer that they would be an obvious choice. Thus the Peugeot billboard deserves some scrutiny.

That's a lot of greeting cards
That’s a lot of greeting cards

We might expect that the billboard would make use of an advanced material, perhaps one of the piezoelectric polymers in a sheet with a large surface area. It’s a surprise then in the video to see that it is instead composed of a large array of the commonly available piezoelectric disk sounders that can be found in a multitude of cheap toys and musical greeting cards. They are held in an array by a lasercut sheet and each one has a small printed circuit board behind it to hold a power connector, feeding its output to another PCB that presumably contains some kind of regulator.

The use of these components really raises an eyebrow. While they will undoubtedly generate some electricity from ambient sound it is definitely not their intended use. There are several reasons for this, and they lie in both the nature of the sound they are facing and their differences from piezoelectric devices intended for energy harvesting.

Carefully-Tuned Energy Harvesting it is Not

Piezoelectric energy harvesters do exist and they can produce usable quantities of power, but to do so they must be distorted slightly, or bent. They are best suited to harvesting energy from constant vibration at a particular frequency, and will normally be chosen and mounted such that they are working close to their resonant frequency of the system they are in.

The cheap piezo sounder disks will also generate electricity if you bend them and also have a resonant frequency, but here they are not being mounted in a way that causes them to do that and are being fed what amounts to a broad spectrum of frequencies rather than a carefully tuned vibration source. Thus it’s unlikely that they are operating at anywhere close to their peak efficiency, which would not have been as high as that of the purpose-built device anyway.

The system also has a mains power connection.
The system also has a mains power connection.

The clinching moment of doubt comes in a fleeting instant of the Peugeot video itself. The whole project is extremely light on technical details, but we do have a glimpse of a block diagram that shows a connection to the city electrical grid. At this point the story becomes clear, this is not a sound-harvesting charge point but a normal mains charge point which can take some input from the sound harvesting panels. Thus it’s safe to declare it a publicity stunt rather than a serious attempt at power harvesting.

This is a shame though, because it remains an interesting project into which its creators have evidently put a significant quantity of work.  We doubt that Peugeot will be publishing a full technical write-up of its design, construction, and efficiency, but we think that it wouldn’t do any harm to their image in our community at least (much the opposite, we want to see the data!). Electric cars are still a niche product that has more traction among people with a technical interest, so any of the technology surrounding them will bring plenty of readers.

If there’s a final point to be gleaned from this, it’s that there is a significant quantity of energy contained in ambient sound, so how might it be harvested? Our gut feeling from youthful experiments with high-impedance moving-iron earpieces would take us in the direction of something with a much wider audio bandwidth, but we’re not sure whether an array of surplus 1960s telephone parts would be any more efficient or appropriate than one of cheap greeting-card speakers. We’re sure you’ll have your own views, please feel free to make them known in the comments.

55 thoughts on “Can You Piezo A Peugeot?

      1. I am always skeptical of any form of “large gang of high impedence energy harvesters” scheme (smart roads, this thing, and others) because the job of aggregating and balancing those sources so that the lowest current element in a series group or lowest voltage element in a parallel group pull what weight they can without dragging the others down is non-trivial even if you can safely assume all elements are well matched and their inputs are too (e.g. single wafer solar panels evenly lit) but the more heterogeneous the elements (or their inputs) get the more extra support circuitry is needed to keep everything in line, and even with that solved something the size of a billboard will have copper losses.

        This is why serious power generation is much more efficient at large scales and yeah… this is also just silly.

        1. And that’s why some poor fool invented “Binning”. If you’re willing to crank them out in the hundreds or thousands per hour, chances are you’ll also have an automated testing rig. I’m pretty sure photovoltaic panels are binned for the same exact reason.

    1. I think this is correct.

      Figure a flux of 1pW is 0dB (1pW/m^2, by definition), typical outdoor sound levels are in the ballpark of 100dB (loud, irritating, but not quite to the point of damaging over the short term, and convenient), and the area of the billboards is 10m^2 each (again, convenient). Multiply, and get of the order of 1W, not correcting for efficiency. This is consistent with the Wikipedia page (https://en.wikipedia.org/wiki/Sound_power#Mathematical_definition) that 85dB is a flux of about 0.3mW/m^2

      I don’t know what the cost of the billboards was, but the payoff time is likely to be greater than the life of the boards.

      1. This! Sound is already measured in terms of work / area. This sort of calculation would be a good problem for a high-school AP or first-year college physics class. If you hadn’t done it, I was gonna.

        And Jenny’s spot-on with the resonant frequency comments. The actual piezo power harvesters I’ve seen are flexible with weights on the end to tune them to the peak expected energy input.

        Piezo discs in the 1.25″ range come in at 1000 Hz, and go up as they get smaller. 4000 Hz is common for those little buzzers.

        So unless there’s a large fraction of the noise at/around these frequencies, there’s just not going to be that much energy transferred. 10% of total available? Probably a good upper bound. 100 mW? You could do better with a fist-sized solar cell.

        If any of this sounds like bitterness on the part of someone who’s tried to harvest energy from piezos, there’s a reason. :) I learned a lot about diode reverse-current leakage by trying to rectify piezos. You need to pick them very carefully, or they’ll eat up 1/2 of your (tiny amount of) juice. And Schottkys have higher leakage.

    2. You don’t even have to do the math. You can tell at a glance simply by looking at the thin conductors they’re using that there isn’t a lot of energy involved.

      Where there *is* a lot of energy is in the construction and installation of these billboards. Had that same energy been used instead to install some well-tuned passive diffusors, it probably would have gone a lot further.

  1. “If there’s a final point to be gleaned from this, it’s that there is a significant quantity of energy contained in ambient sound, so how might it be harvested?”

    The problem is that there isn’t a significant quantity of energy in ambient sound, even in the noisiest cities in the world. The reference point for sound pressure level (i.e. 0 dB SPL) is 1e-12 W/m^2, meaning that at 90 dB SPL, which will cause hearing damage from continuous exposure, there’s only 1 millwatt available per square meter of collection surface. If you could find a city that were that loud continuously (you can’t) and harvest it perfectly (you can’t), you’d get almost as much energy as exposing the same area of photovoltaic cells to sunlight for half a second per day. It would take tens or hundreds of thousands of years to recover the energy cost of constructing the harvesting device.

    It is, as usual for corporate greenwashing, complete bullshit.

  2. Maybe the energy harvesting could be maximized by using a cochlear shaped funnel lined with piezos tuned to particular frequencies. In addition, audio guides (i.e. horns) could help channel sound waves to the cochlear funnel.

  3. So many questions. Replacing ICEs with EVs claims to reduce city noise by 30% when you are trying to put more electric vehicles on the road. So the stated goal is to deplete the power source needed to expand sales? They would theoretically need to artificially generate more noise to fuel the expanding fleet of EVs. Dystopian “noise gangs” would fight for control over the power grid.

  4. First, if you want these to work you need to install them on runways. That’s where all the sound is.

    Second, if you want to harvest energy from passing motorists a large array of small wind turbines will take advantage of the air displaced as the vehicles pass.

    What we are missing here is common sense.

      1. My first take was: These people are lying about this, and so they probably lie about other things too. “Don’t trust Peugeot” isn’t exactly the message they want to get across.

  5. I did this project last year for the Intel science fair. Its cool to see it being fully realized but I sure wish I was working on that project. Its not cost effective currently, but at least someone’s putting in the steps to make it feasible.

  6. There’s just not much energy there in the first place, and piezoelectrics aren’t an efficient way of harvesting it. A recent student project tried harvesting energy from cars driving over a speedbump fitted with piezoelectrics and an energy-harvesting circuit. They *almost* got enough power to light a LED.

  7. Screw the solar roadway – let’s have the sound harvesting runway.

    Just need a better name for it and some time to write up a bit of a brief and fleece some government department and other green washed individuals :)

  8. Someone should create a non profit organisation with the goal of exposing this kind of bs. This is too much, using piezo to harvest vibration (not sound) energy IS a thing, but only for things like powering highly energy efficient MCUs. Nobody serious in the energy harvesting world would ever think of powering a car with it.
    I find this kind of ads extremely harmful for the public opinion, as it constantly deviates the debates into some conspirationist BS (you know, “cheap solutions exists, why aren’t THEY implementing it? That’s suspicions!”) instead of exposing some real (but less glittering) potential solutions.
    Couldn’t this kind of video be considered as false advertising? They are basically claiming their apparus can be a significant part of a car charge. No sane judge would consider less than 0.00001% being significant… (The number is roughly made up using a 3mW for 10 m2 powering a 100kW car)

    1. +1 – So much marketing/advertisements misleading the general public who has no knowledge of how a particular tech actually works under the hood. AI/ML, this, ‘autopilot’, etc…

  9. Nowhere in the video does it actually say the board charges the car, in fact the designer says in the video that one day such technology might be used to charge a car. The infographic for the billboard during the video even shows grid power charging the car and the piezo “helping”. It is a stunt and gimmick, nothing more. They may get enough energy to illuminate the small sign at ground level telling you what its all about. And what it is about is selling their new electric car.

  10. Now if they had installed piezoelectric material within speed bumps, that actually may have harvested a few orders of magnitude more power (coming at the expensive of reducing the efficiency of the vehicles traversing the piezo-bump, but one could argue that the inefficiency would be there for a normal speed-bump anyhow). The legal issues in an advertisement company being able to modify government owned infrastructure in the middle of a major city for the purpose of selling products, It just would never be allowed to happen in any democratic country (not without a hell of lot of brown paper envelopes) especially on any high traffic route that typically would never need a speed-bump (The only exceptions that I can think of would be places like near airports, schools, hospitals, embassies, stadiums and places like that). And also the intermittent nature of traffic is such that it still would not harvest enough energy to pay for itself or charge an EV in any reasonable period of time.

  11. At night you can harvest the energy from all these city lights and put it back into the grid. People would set up more city lights for the sake of getting energy back at some point.
    Not to forget moonlight and space radiation can be captured, too.

  12. About 10 years ago, I was walking at night in Dubai (I was there on appointment to take pictures of the new years eve fireworks, not for vacation), in a very quiet street, ant then, I heard the sound of an engine, but there was no car, and it was a tiny motor, then walking further I saw the source of the noise : an engine generator, that was used to light up a gigantic billboard (at least 8 meters wide)…
    And the advertising was for… energy saving lightbulbs <I;OD
    Maybe it's the same great advertisement agency.
    For further improvement, they could build an enormous engine that makes a lot of noise in front of the energy harvesting billboard, and then, well, I don't know I'm not that kind of people, I have no imagination for swindles…

  13. right so how does it make sense to use a given area for piezoelectric energy harvesting instead of sticking solar panels on it?
    no need for sun tracking, it’s going to blow the piezos out of the water as is.

    here’s my sincere downvote for peugeot’s unnecessary, inefficient, stupid pr gag.
    because that’s all it is.

    no need for maths of any kind here.

  14. Heuuu Any idea why the piezos are connected to a non populated ADC PCB?!
    Not only the all thing is clearly a scam, but I suspect them to not even try to harvest a sigle picowatt !

  15. Wasn’t it Volvo that painted some sort of catalyst material on their radiators, and claimed that driving their (fossil fuel powered) cars “reduced CO2” because of it? Needless to say, the “green radiators” didn’t convert anything near the amount that that came from the tailpipe!

    Piezoelectric transducers produce a fair voltage, but very little current. Can you get enough power out of one to make more than the power consumed to carry its weight around? No.

  16. The total amount of sound power available over the surface of one of those billboards is a single digit number of milliwatts. (0.3mw per square meter at 85dB) Both together might be able to power a micro LED at 100% capture efficiency.

    “could you make a piezo billboard as a practical green energy device?” – the total energy ever harvested by the billboard if it was in place for many years might account for the energy used to produce one of the piezo elements in it. The whole thing is a ridiculous idea.

Leave a Reply to Bunsen Cancel reply

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