OK guys, I've seen a lot of threads about this topic. I'm new here, and I wanna start talking about this.

One of the mains goal of this topic will be the design of the circuit and how to save the energy.

Since the Piezoelectric generates AC voltage, the first thing that we have to do is rectified that voltage. I've seen a lot of circuits with diodes. But I found one with Schottky diodes, apparently due to the forward voltage drop. After this, we have to think how to harvest the energy. So the common way is using a capacitor, but we have to keep in mind that capacitors save voltage, not current. So, it should be a way to keep the voltage with a good current. I've searching and I found a lot of circuits with transistors, some others with DC/DC converters. I'll leave a link here to one of them.

http://www.telefonica.net/web2/carloscossio/Harvest%20Energy%20Using%20A%20Piezoelectric%20Buzzer.pdf

The link above use a max 1675, so go and check it out.

My last goal with this project, is at least reach the way how to charge a battery. If I have a mistake, please, correct me. I'm not an expertise in this field. I know about electronics. But I've never though in something like this.

 

There are ICs you can buy for "energy harvesting", capturing low voltage, low energy signals and accumulating them. They're too expensive for my taste, eg. $20, but if you really want to pursue this project, these would save you a lot of time and trouble. Otherwise you will struggle to get your source energy to survive the many losses.

Any passive rectifier arrangement will lose at least about 0.5v or more directly off the top of your source signal, so I think converting to DC as the first step is hugely wasteful. Do you know what AC voltage you might have coming from your piezo? This will have a major effect on design.

Oh, and you'll need a LOT of elements to produce enough power to charge a battery. Just compare the power you might get from the piezo assuming 100% optimal recovery, to the energy stored in a battery. Whopping difference.

 

You are right... would you like to show me where can I find those IC's???.

Another question that I've been keeping in mind. Is, how should I connect them, in series, or in parallel???. I don't thing that I can't connect them in parallel because I need to squeeze them all at the same time, so, I think the best way is connecting all of them in parallel, I don't know if this configuration is going to give me more current, it's a doubt that I have, if anybody knows the answer, I have my ears open.

 

Here's one, and another related device.

Are you really talking about a floor made of piezo elements, to generate power? Other than being an expensive conversation piece, why? There's no way it will ever pay for itself, IMHO.

 

Sorry... I've made a mistake in the second post, talking about how to connect them. My question is, should I connect them in parallel or in series??? I thing that the best way is connecting them in parallel, but, can I get more current in this way if I squeeze several at one time???

 

And take a look to this video... it's possible, I'm sure it does...
http://www.youtube.com/watch?v=RCOBA3Yfm1k

 

I don't know how you will vibrate the floor strongly enough for a piezo to produce enough voltage to be rectified. Also the current from a piezo is almost nothing.

The article uses a "buzzer" beeper piezo transducer and something whistles to it loudly at its resonant frequency. You cannot do that with a floor.

 

Capacitors store energy in their electric field where:

E = 1/2 C V^2 (energy in joules is half the capacitance times the square of the voltage)

That link you post is interesting. The actual "harvesting" portion is the left most 5 parts: the 4 1N5820's and the .47F super cap. The zener (1.3 watt, a big one) is there to bleed off excess energy from destroying the cap.

The MAX part is a step-up converter to raise the voltage to something you may find more useful.

 

Take a look to the video that I posted before, also. Take a look tho this other one... it's possible, I'm sure it does... http://www.youtube.com/watch?v=YFITXmzB9EI&feature=related

The Clue about all of this, is the circuit. The charging part at least is the most difficult. The Piezo generates a voltage when a force is applied to it.

 

Is there any chance to increase the current with some Transistors??? Is there any why to amplify the current???. Let say that we are charging a Ni-Mh Battery, about 5v and 650mA. The voltage it's not the problem, but, how about the current??? If we amplify the current, problem solved. I think.... Am I right???

 

The new video is in a foreign language.
You do not "squeeze" a piezo to produce a tiny amount of power. Instead you must bend it but not too much or it will break.

 

Is there any chance to increase the current with some Transistors??? Is there any why to amplify the current???. Let say that we are charging a Ni-Mh Battery, about 5v and 650mA. The voltage it's not the problem, but, how about the current??? If we amplify the current, problem solved. I think.... Am I right???

Don't be silly.
Where does the high current come from? From power supply of the amplifier, not from the tiny input signal.

 

I'm not sure about it. There is several ways to use a transistor. To switch a signal. Or to amplify the current. If I have a tiny input signal, there is a way to increase the current due to the base resistor. I think, correct me if I'm wrong.

 

The output power from a transistor comes from its power supply. If you have a power supply then you don't need the piezo transducer.

 

OK. But one piezoelectric disc can power itself 3 LED'S, so at least it's generating more than 60mA.

 

So, the question will be... how do they do it??? It's possible folks!!!!!

 

Blinking 3 LEDs for a couple of milli-seconds each second dimly (a couple of mA) is completely different to charging a 5V battery continuously at hundreds of mA.

Maybe the "floor" video uses hundreds of piezo transducers. Then the LEDs blink only for a moment. (very little amount of power)

 

I see no fundamental problem with the basic concept. If you bang a piezo then yes it makes a voltage. It converts mechanical energy into electrical energy. The issue is generating enough power to make it pay off in a "three rules of thermodynamics" sense. You can't get a free lunch, meaning you can't simply amplify the current to get more current unless that current come from somewhere else.

Each bang on the piezo makes a small pulse of energy, the trick is to accumulate enough of these small pulses to make the system pay for itself.

Say you want to illuminate a stairway. In theory, you could put piezos all over the stair treads so people walking up and down are inputting energy, then you capture that energy to run the lights. As long as lots of people are using the stairs you have a constant input so you can draw a constant output of energy.

The first complication you run into is each piezo element needs be isolated from the others least the energy from banging on one just goes to flexing the others. So it may be more efficient to tie each piezo element thru a single diode to the accumulation cap: you toss away half the energy (you only store on one half cycle), but you don't have the 2nd diode drop loss and you need 1/4th the number of diodes.

 

I see no fundamental problem with the basic concept.

Depends on the concept we're talking about. Can you make an LED pulse for an "ooh" and "ahh"? Sure. Can you meet a break-even economic hurdle. Not likely. So it comes down to what the OP really wants to do. If a floor covered with piezos charging a single AA battery is acceptable, go for it.

Before designing anything, I'd investigate exactly what each piezo element generates when "stepped on". Voltage and current versus time, so you can calculate power per element per step. Then figure you'll be able to capture maybe a fourth of that, losing half at the battery and another half with the rectifiers, regulators, whatever. If you need 100 elements to charge a battery, maybe it would be fun. If it takes a million of them, not so much.

 

There was an "energy harvesting" project where a guy put piezo transducers in his shoes and used an expensive IC to rectify its output and boost its voltage. He charged the little battery in his MP3 player by walking all day long.

Maybe it will charge a cell phone battery in one week.