My son is trying to build a piezo-electric device using piezo-ceramics that generates enough electricity to recharge a small battery. I don't know enough about electricity nor about piezo-electrical devices so I'm reaching out to you guys to get some help.

I understand that piezo devices generate VERY little current (amps) but reasonable amount of voltage. I think it also generates these micro-amps in small bursts so it's not a continuous generation.

Is there a way to accumulate and store the small currents and then once enough has been saved, move that current into a battery? Is this even possible without building a huge machine. Without getting into the details of his idea he needs the whole circuitry to be very small.

Sorry if I sound naïve about this project, but that's exactly what I am - so any advice and resources would be greatly appreciated.

I give my 2 cents (I am not an expert) :-)

You need something which stores pulse current, and converts voltage/amperage to reasonable levels. You can use first a capacitor to store the energy from piezoelectric ceramics. Then you have to connect the capacitor to something more complex (but small and not expensive, much like the circuit present in all battery chargers) to tranfer the energy in the capacitor to the charger, as the capacitor does not give you the desired current, voltage or stability.

Some expert advise would be much appreciated.

@Aurora - thanks for the idea.

I think you are right - a capacitor seems like a way to go. Although I'm trying to learn (maybe my kid can learn faster!) about these widgets and understand what constraints they have. I know that they tend to discharge quickly - especially the lower-end ones, but maybe the discharge rate is low enough that it still works out. I guess it's like a leaky bucket that we're trying to fill - if the flow of water is fast enough the bucket will fill up even with a small leak, but if the leak equals inflow, then nada... So the question what kind of capacitor will work under these conditions...

Hi rpray2007,

I have studied capacitors only in books, thus I can be very unaccurate here :-)

Polypropylene plastic film capacitors and Teflon plastic film capacitors have comparable lower discharge rate (but I have really no idea of costs), most probably low enough to charge a battery (Vacuum capacitors are out of discussion of several reasons).
To have a low loss capacitor both dielettric leakage and ESR must be low, but low ESR can be hazardous as it allows large currents when accidentally shorted.

Believe it or not, I actually am an expert in this field. :)

The piezoelectric device will indeed generate small amount of electrical energy from mechanical forces. And you're correct that the voltage will be fairly high, but the current will be very low. (Another way of saying this is that the material has a very high impedance).

The circuit you'll need is just a little more complicated than the simple capacitor you already mentioned. First let me give you some background info:

The atoms of a piezoelectric material are arranged in a crystalline structure. The atoms are covalently bonded to each other (meaning that they share electrons to form a stable compound). Since some of the atoms give up electrons while others receive those electrons for part of the time, the former end up with a net positive charge while the latter end up with a net negative charge. From a macroscopic point of view, this is all irrelevant under equilibrium conditions. The positive charges exactly balance the negative charges within the material, and both are dispersed evenly throughout. In fact, the same can be said of basically any chemical compound, crystalline or not.

What makes a piezolelectric material special is that when you deform it (such as squeezing it or bending it) along certain directions, the crystalline lattice gets deformed in such a way that the positively charged atoms shift mostly in one direction, while the negatively charged atoms shift in the opposite direction. This creates a small dipole moment (a net separation of charge within the material), which results in a voltage from one side to the other. If you were to put metal plates on either side of the crystal, you could measure this voltage.

We're not talking about a lot of charge here. You can't just squeeze a piezoelectric material with a clamp to create a battery or anything like that! Sure, the initial squeeze will generate a voltage, but as soon as you hook your piezoelectric "battery" to any sort of load (say the bulb of a flashlight, for example), electrons will flow from the negative side of the piezoelectric, through the wires and the light bulb, and into the positive electrode. Your flashlight would light up for a brief instant (okay, not really, since an incandescent bulb needs a little bit of time to heat up before creating light, but you get the point) as those electrons flowed, but after a few microseconds, enough enough electrons would have moved from one side to the other to counteract the initial voltage. You'd be back to equilibrium conditions again, and your flashlight bulb would remain dead forever more. Hence your statement that it generates micro-amps in "small bursts" was pretty much accurate!

Once your piezoelectric battery has rereached equilibrium and died out, if you then released the clamp that was squeezing it, the material would spring back to its equilibrium shape, and the internal dipole moment would go away. Now all those excess electrons that previously had flowed from one electrode to the other would be creating a voltage of their own! (In the opposite direction as before). Assuming the device was still hooked to your flashlight bulb, the current would then flow back in the opposite direction as before to get back to the "unsqueezed" equilibrium, and the bulb would once again light up in the process.

Thus, you can see that if you were to rapidly and repeatedly squeeze and release the piezoelectric device, you'd cause current to flow back and forth through the bulb, and you could actually cause it to light up. In other words, a piezo device can be used as an AC generator (alternating current, as in back and forth). This works great for something like a light bulb which doesn't care what direction the current is flowing. It cannot be used, however, as a DC generator. That means you can't just hook up a piezo transducer to a battery or capacitor and expect it to charge it up. Since the current from the piezo is going back and forth, every time it charged up the battery or the capacitor, it would turn right around and suck the charge back out, over and over again.

Therefore, if you want to use the piezo device to ultimately generate DC current, you need to rectify it. You can build a very simple full-wave rectifier with four diodes. If you were to then use that to charge a capacitor (they call it a "smoothing capacitor" in the circuit diagram at the link I just gave you), the voltage on the cap should slowly increase over the course of time if you repeatedly squeeze and release the piezo device. Using something to vibrate the surface would be good for demonstrating the effect (Power drill? Muscle massager? Blare loud music at it?)

As far as the capacitor goes, the larger the value, the more charge will be able to be stored. However, larger values mean that it'll take more vibration to demonstrate an appreciable increase in voltage. You may need to experiment a little with some different values. I'd try a plain old electrolytic cap to begin with (or whatever else you can find that's cheap; I doubt leakage will matter much here), maybe something in the 1 uF range. Hook a voltmeter to it and watch it charge up. Assuming that works, you should be able to hook a dead rechargeable battery to it and actually charge it up.

@jcaron2 - I have scoured the internet for good information about the piezo-electric effect and they were either too technical (graduate engineering level) or too basic. Your description is really helpful and I finally "get it" and understand what the issues are. Thank you!

I might come back for more information as my son tries out what you are suggesting. An update since my first post - he has added three discs in parallel to maximize current and he is able to measure about 10-15 mA when the discs are compressed - enough for an LED but not enough for a flash-light bulb. But maybe, we have to "vibrate it" as you suggested to get the filament heated up before we can visually see the bulb lighting up.

BTW - I was suggesting hooking up the discs to charge a fuel cell, but he wasn't too keen as that doesn't fit well with his grand scheme. But out of curiosity - do you think a fuel cell would be a good "battery" for this?

Again, thanks for your detailed and very understandable response.

As long as the piezo plates can generate enough voltage to ionize the hydrogen (or whatever happens chemically inside a fuel cell), it should work fine. I assume that potential is just a few volts, so I'm guessing it would work just fine.

In the case of an LED, by the way, an AC piezo stimulus may or may not work like it would with a regular light bulb. In it's simplest form an LED is a diode. Therefore, current only flows in one direction through it. Once you squeeze the piezo plate and charge builds up in that one direction, subsequently "unsqueezing" won't do anything because the current can't get back through the diode to re-equilibrate. At that point, the charged would be locked in place, and further vibrations wouldn't do anything.

Some LEDs actually consist of two diodes stacked in parallel in opposite directions inside one plastic housing so that they can be hooked up in either polarity. Such an LED would work just fine in that case.

Definitely feel free to ask for more detail as you go along! It's not every day I actually get to answer questions about my specialty. (Actually, my real specialty is very, very high frequency piezoelectric devices - up in the GHz range, but close enough!)

Josh

By the way, a good analog for this whole setup would be one of those big fat hand pumps that you use to pump up air mattresses - the kind that they sell at Walmart that pump both on the down-stroke and the up-stroke.

If the pump was a simple piston with no valves, whenever you depressed the plunger, air would flow into the air mattress, but as soon as you lifted the plunger back up, the air would get sucked back out again. Once you add a set of valves to "full-wave rectify" the air flow, however, air gets pushed out regardless of which way you move the plunger. Then you can pump up the mattress as much as you want.

In your son's experiment, the electrons are the air, the piezo plates are the pump, the full-wave rectifier is the set of valves, and the capacitor/battery/fuel cell is the mattress.

Just to follow-up on this. My son has completed the project (for now) but wasn't able to store the energy in a meaningful way.

He did use a rectifier (as you suggested) and also used a capacitor (he tried 47uF) but to enough to light something up. Someone else suggested using a step-up voltage circuit (something called a Schmitt trigger?) to increase the voltage from ~2V to 6-9V but no luck either.

At the same time, in doing research on this, there are people who are able to charge a 40mAh NiMH battery (they used a piezo vibrating plate) within an hour. All they used (at least shown in the paper) is a rectifier, capacitor (about 1mF) and the NiMH battery.

My question is, what are the requirements for trickle-charing a battery of that size and/or is he simply not providing enough current and charge to get this thing charged. The poor kid banged on the discs for over 250 times but didn't make any difference. But directly each step measured up to 18mA? So, what's missing. Maybe the components are simply to too cheap and leaky or is it something more fundamental that we're not understanding?

Thanks again for your help (which he read thoroughly and really appreciated).

I have a couple of ideas. Just brainstorming here. Do you have any idea how much voltage was generated when he banged on the disks? It might be hard to measure with a simple voltmeter because the voltage pulse will be very short (in time), so the voltmeter would probably average it over a longer time. The best way would be with an oscilloscope, but I'm not sure if you can get your hands on one.

At any rate, the reason I ask is that if the voltage is small, it won't be able to overcome the threshold voltage of the diodes in the rectifier. (It takes about 0.7 volts to get through the diode, so if the original signal is less than that, nothing will get through to charge the capacitor.

Increasing the voltage would be the way to fix that particular problem. I can't really see how a Schmitt trigger would accomplish that, but I have to admit it's been a long time since I've even thought about a Schmitt trigger, so I can't rule out that you already have a working solution. A small transformer would be another possible way. But probably the simplest way would be to change the configuration from three plates stacked in parallel (which triples the current, as you said above) to three plates stacked in series (which triples the voltage rather than the current).

Speaking of stacking plates, there's also the possibility that the plates are currently stacked in a way that two of them cancel each other out. Keep in mind that they have a specific polarity.

Finally, there's the possibility that the type of plates you're using aren't sensitive to the type of physical stress being applied. Going back to my original explanation about how piezoelectricity works based on the arrangement of atoms in the crystalline lattice, it's easy to imagine that the effect is very sensitive to the direction and nature of the applied stress. For example, some crystallographic orientations are very sensitive to compressional stress (which I suspect is the type of stress your son is applying to the plates), while others are completely insensitive to compression. In the latter case, the piezoelectric effect is limited to shear stress and strain. (If you picture a "brick" made of Jello, compressional stress would squeeze the brick to make it thinner, while shear stress would "rack" the brick, changing its profile from a rectangle to a parallelogram). For the experiment your son is trying to perform, it would obviously be preferable to have a transducer that was sensitive to compression.

In case you're curious why one type of transducer would be preferable over another, a few examples might be helpful. For instance, if you're designing a piezoelectric audio speaker (or microphone), you want to generate (or detect) compressional waves in the air (a.k.a. sound waves). On the other hand, if you're designing a high-quality crystal oscillator, the generation of or sensitivity to compressional waves through the air is something to be avoided since it would be an unwanted energy loss mechanism as well as an extra source of noise in the system. Hence, in that case you would choose to use a shear-mode transducer.

It's possible that you guys just got unlucky in your choice of transducer. If these are shear-mode plates, they may not be generating much voltage when you bang on them. Ideally, you want something that would be used in a piezoelectric speaker or microphone.

Jcaron: I wanted to follow up and let you know that my son entered the science fair (LA County) and won 3rd place in his division. I'm a proud dad but more importantly, I'm excited because people seem to appreciate his idea and are encouraging him to keep working on it.

At the time of the fair, he still wasn't able to store any energy so that's something he needs to work on. Furthermore, one of the judges asked him whether the discs were potentially canceling each other out sometimes and this to me sounds like a fair question. So he's working on adding a rectifier to each piezo directly and then adding up the voltage. His current circuit is quite simple - piezo to rectifier to 1000uF capacitor to 40mA NiMH button battery. I think he tried it last night and still hasn't been able to store it. He'll keep trying and playing with it now that he's gotten a little acknowledgment for his project!

Again I just wanted to thank you for your help and the detailed answers which I think greatly helped in his understanding of piezo electricity.

Hello there,
I am doing the similar research right now. However, I think I may need more current than lighting a bulb. Therefore, is the method you talked about still applied my reseach except I may need to change a little bit about the capacitance?

As an update to this, the device was able to light a small LED after running about a mile. He charged a capacitor, disconnected it and connected to an LED. The LED lit for about 150ms which isn't much but a nice start.

BTW - his project got 4th place in at the CA State Science Fair in his division.

Thanks jcaron for your help.

Quote:

---

Originally Posted by rpray2007

As an update to this, the device was able to light a small LED after running about a mile. He charged a capacitor, disconnected it and connected to an LED. The LED lit for about 150ms which isn't much but a nice start.

BTW - his project got 4th place in at the CA State Science Fair in his division.

Thanks jcaron for your help.

---

Excellent! I'm glad to hear he got it to work somewhat. If he just had access to an oscilloscope, he could figure out how to improve it. Still, fourth place is nothing to scoff at. Congratulations!

Hello Mr. J Caron,

Good Evening!! Very detailed and informative post about Piezo Transducers. Can you please point me to some articles or research papers which has detailed information about this? I am doing a Science Fair project on this and want some good Research Papers reference about this.

Thanks and have a good night!!
Sincere Regards,
Tanu

Quote:

---

Originally Posted by tanya425

Hello Mr. J Caron,

Good Evening!!!Very detailed and informative post about Piezo Transducers. Can you please point me to some articles or research papers which has detailed information about this? I am doing a Science Fair project on this and want some good Research Papers reference about this.

Thanks and have a good night!!!
Sincere Regards,
Tanu

---

Unfortunately I don't really know of any good research papers to point you to. :( The stuff I work on is very specific, so the papers I read wouldn't be very helpful for gaining g a general understanding of the subject. It's been a long time since I learned the fundamentals, and the textbook I used (B.A. Auld, vols 1 and 2) in grad school is probably WAY too complicated for a science project primer. Unless you're well versed in tensor math and systems of partial differential equations, I'd just stick with what you can find on the web and at your local library.

Sorry I can't be of more help. Good luck! If I come across something, I'll let you know.

http://users.ece.gatech.edu/rincon-mora/publicat/journals/iscas09_pzt_harv.pdf

well rpray2007 , your son's curiosity is quite appreciable and I'm too working on that! http://bindedworld.blogspot.in/search/?q=piezoelectricity
Download the springers book on piezoelectricity from the above link,it helps me a lot!

Quote:

---

Originally Posted by tanya425

Hello Mr. J Caron,

Good Evening!!!Very detailed and informative post about Piezo Transducers. Can you please point me to some articles or research papers which has detailed information about this? I am doing a Science Fair project on this and want some good Research Papers reference about this.

Thanks and have a good night!!!
Sincere Regards,
Tanu

---

Hello miss.tanya425,
Search results for piezoelectricity | Binded world
Download the ebook of springer's from the above link.This may help you,I guess so:)

This is pretty cool. Thank you for posting this question. I'm working on something like this myself. Its just a simple setup using a rodin coil design in conjunction with a stacked capacitor to charge batteries. I was hoping to use silicone dioxide sheets and building up the mechanical pressure through the magnetic flux and hopefully have it self sustained by keeping the capacitor in a closed loop. With auxiliary wires to power just a lightbulb for now.

I've been working with rodin coils, tesla coils and the like for while now. They're amazing things. The ability to nearly double the amount of energy you give them just through the mathematical precision of the wiring. I could post videos for your interest if you like.