Why does Piezoelectric output voltage dropped when load is added?

[grasscut]

Hi,
im doing a piezoelectric energy harvesting project with a standard harvestings circuit consisting of a full bridge rectifier, smoothing capacitor and a resistive load. I managed to get around 28V after rectification without connecting to a smoothing capacitor and load, but when i added a LED parallel to the smoothing capacitor, my voltage instantly falls to around 2.4V and remains the same. Can someone tell me why will my supply voltage from the piezoelectric dropped so much when a LED load is added?
any kind of help would be really appreciated! thanks

 

[anorlunda]

You should be thinking about power, not voltage. My guess is that the piezo system is having difficulty supplying enough power for the LED.

You have done well to get that far. Does the LED light up?

 

[grasscut]

thanks for replying! Yes the LED is able to light up with no problem, i can even add a few LED and all of them will light up. When i conduct the experiment, my piezoelectric is at its resonance. Thus, the supply should be quite consistent. As you mentioned for the power, how should i approach this problem?

 

[tech99]

Measure the voltage and use a variable resistor as the load. When the voltage falls to half the open circuit figure, the generator is giving its maximum power, and that is the optimum resistance for a load. Although a Piezo crystal is a capacitive source at low frequencies, if it is at resonance then it will be resistive.

 

[anorlunda]

thanks for replying! Yes the LED is able to light up with no problem, i can even add a few LED and all of them will light up. When i conduct the experiment, my piezoelectric is at its resonance. Thus, the supply should be quite consistent. As you mentioned for the power, how should i approach this problem?

What problem? You haven't said what you are trying to accomplish.

If you want more LEDs, more power. If you want to supply a your whole house, about 50,000 times more power. You can make more power with multiple replicas of what you already built.

 

[dlgoff]

... when i added a LED parallel to the smoothing capacitor, my voltage instantly falls to around 2.4V ...

 

[Jeff Rosenbury]

Go with tech99's measurement. It should give you the internal resistance of the source. This is a useful value for moving beyond simple LED circuits. With it you can figure the output power.

If your goal is only to light LEDs, put them in series to the half voltage point. There could be a problem with too much current (at least theoretically) and LED's typically have a current limiting device to keep them from burning. Often this will be a resistor where losses aren't relevant, but more active choices are used when power is tight (hand held devices, etc.).

 

[grasscut]

Thank you all for the reply.
So actually the current problem that I am facing is that i couldn't get the impedance matching done to determine the maximum power output. My piezoelectric has a capacitance of 54nf, which around 190Kohm at resonance frequency of 15 Hz. However, when i tried to use various resistors value to find the optimal resistor to measure the maximum output power, there isn't a significant change in power when i match with a resistor load of 190Kohm. Thus, I am wondering how i can find the maximum output of my piezoelectric to determine its performance. I will try out what tech99 mentioned and give it a try. for measuring the open circuit, should i measure with DC or AC? Can i consider the rectified voltage of 28V as the open circuit voltage?

 

[tech99]

If the piezo is oscillating at its natural frequency, then the capacitance is not relevant, because it will become resistive. So try lower resistance loads until the voltage is halved.
If you measure and use the generator under conditions with the rectifier in circuit, that is OK.
If you use LED as a load, it is not so easy; remember they do not obey Ohm's law, as described previously. You need to put LEDs in series to give the required load voltage. Why not use an incandescent lamp for test purposes?

 

[grasscut]

Lets say i have a rectified voltage of 28V , so i should be looking at around 14V for the voltage across the resistance load and that will be my optimal resistance load? am i right to say that? Resistance load that i used for testing a normal resistor.
I just used the LED to test whether the energy harvesting system is working. eventually, i will use a rechargeable battery as to resistance load to charge up a NI MH AA battery.

 

[Jeff Rosenbury]

Thank you all for the reply.
So actually the current problem that I am facing is that i couldn't get the impedance matching done to determine the maximum power output. My piezoelectric has a capacitance of 54nf, which around 190Kohm at resonance frequency of 15 Hz. However, when i tried to use various resistors value to find the optimal resistor to measure the maximum output power, there isn't a significant change in power when i match with a resistor load of 190Kohm. Thus, I am wondering how i can find the maximum output of my piezoelectric to determine its performance. I will try out what tech99 mentioned and give it a try. for measuring the open circuit, should i measure with DC or AC? Can i consider the rectified voltage of 28V as the open circuit voltage?

Normally I would use an oscilloscope. It will give a better picture of what is happening.

Remember with high resistance loads the test equipment loading might be significant. (Know your test equipment. )

Try measuring both the raw signal from the crystal and the rectified output. Then you can compare them to help figure out what's going on. Remember that in theory, theory and practice are the same, but in practice they are not.

 

[tech99]

Lets say i have a rectified voltage of 28V , so i should be looking at around 14V for the voltage across the resistance load and that will be my optimal resistance load? am i right to say that? Resistance load that i used for testing a normal resistor.
I just used the LED to test whether the energy harvesting system is working. eventually, i will use a rechargeable battery as to resistance load to charge up a NI MH AA battery.

Yes, correct. But when you use an LED or a battery instead of a resistor, it is an abnormal sort of load, and you will need to make its rated voltage suitable for the piezo generator. A milliammeter (or multimeter on mA range) placed in series with the load would be a useful way of monitoring the current being supplied. It is possible that when using LEDs, you will not require a series resistor, as the generator probably cannot supply enough current to cause damage.

 

[grasscut]

Yes, correct. But when you use an LED or a battery instead of a resistor, it is an abnormal sort of load, and you will need to make its rated voltage suitable for the piezo generator. A milliammeter (or multimeter on mA range) placed in series with the load would be a useful way of monitoring the current being supplied. It is possible that when using LEDs, you will not require a series resistor, as the generator probably cannot supply enough current to cause damage.

I'll try out the method for impedance matching to find the maximum output power of the piezoelectric. hopefully it works! do you mind telling what is this topic, so i can read up on it.
Regarding making the rated voltage suitable, how can i achieve that?
Speaking of the current, i tried using a multimeter in mA in series with a LED or a resistor, i was unable to get any reading even though the LED lights up. Is it because of the current is too low that the multimeter can't read? Can i determine the current by using ohms law with the resistance load value and voltage across?
thank you for your kind assistance . truly appreciate it:)