Can Piezo Energy Harvesting Charge a Battery Without Constant DC?

[gabismm15]

There is this voltage source that generates different types(piezoelectric device, the pressures exerted are not the same everytime), it is not a sine wave properly(A), then passing it to a full bridge rectifier, the outcome is like B, then you use a smoothing capacitor and the outcome is like C, what do you use to generate a steady voltage? I tried with a 5v voltage regulator, but it is not working, and the inputs are usually higher than 5v. It is to charge a battery (3.7 v Li-ion ), do you strictly require constant dc ?

 

[davenn]

what is your source of that horrible looking AC voltage in A ?

what do you use to generate a steady voltage? I tried with a 5v voltage regulator,

A transformer with the appropriate secondary voltage --> bridge rectifier --> smoothing capacitor --> voltage regulator giving the needed voltage out for the battery

It is to charge a battery (3.7 v), do you strictly require constant dc ?

it depends on the battery, which you haven't told us what it is??

different batteries require different charging methods

some require steady current and a voltage that varies with charge level
others are opposite ... a fixed voltage and a varying current
others have a mix of varying current and voltage

cheers
Dave

 

[meBigGuy]

There is no information in your drawing regarding the voltage level and frequencies you are achieving.

If, in diagram C, your voltage is always above 5V when loaded, then a 5V regulator will do the job. If the regulator isn't doing the job (and the input is always above 5.5V or so), then you are doing something wrong and we need to address that.

I hope you understand that properly charging Li-ion is complex, and generally requires a controller IC (or equivalent controls). Also, it becomes even more complex when your supply can come and go (brown out) at random times (since Li-ion charging is state based).

 

[sophiecentaur]

It strikes me that you need to rectify, then integrate to smooth out the short term spikes and, assuming your resulting voltage exceeds the battery volts, use a switch mode circuit to charge the battery. If you don't have enough volts to start with then a transformer (of suitable design) could achieve a suitably high voltage from the integrator.
There is nothing you can do about the brown-out periods; that's what the battery is there for.
You would need to do the sums to make sure that there is enough total energy available in a given time to satisfy the requirements of the battery. (A step-up transformer will not help with any short-fall - it merely matches the source to the load better.)

 

[berkeman]

If you don't have enough volts to start with then a transformer (of suitable design) could achieve a suitably high voltage from the integrator.

I think you mean a DC-DC boost converter. After integration, you would have DC, not AC, right?

 

[sophiecentaur]

No. I meant a transformer (/resonant circuit perhaps) to boost the peaks of the initial pulses. After rectification and integration, the mean value of the 'varying DC voltage' would be greater.
But there would be nothing wrong with using a switch mode combined regulator and DC DC converter, to produce whatever volts you wanted.

 

[berkeman]

Ah, I misread your post then. I thought you were saying to put the transformer after the integration. Makes much more sense as the first component that the AC signal is fed into. Thanks.

 

[sophiecentaur]

The question to be asked, of course, is the actual amount of energy available from this 'energy harvesting'(?). before proceeding with the realisation of this it would be useful to put a resistive load and see the heating effect. A range of resistor values should be used, to find the optimum. That is the acid test - go/no go - assessment.

 

[jim hardy]

Last night i googled on "Piezo Energy Harvesting" and found several articles. Google has got unfriendly and i wasnt able to link to them, though they readily loaded in pdf format.

There's more going on than i had suspected.

OP hasn't said whether he's using one of those newfangled high current piezo arrays or a barbecue grill striker. The latter is, i believe, hopeless as a battery charger. Not enough charge per push.
I'd suggest he try the magnet and coil from one of those "shake" flashlights instead.

But to the question, "do you strictly require constant dc ? ."
No, provided you have enough of it.