Help needed to amplify piezoelectric power 4

[zapciname]

Hello everyone, Im doing a project on piezoelectric power harvesting. Right now, my piezo disk prodce 30V AC. After putting through a bridge rectifier and 1000uF capacitor for DC conversion, the output is 20mV DC and 30microA which is insufficient to even use a BJT. Is there anyway I can amplify the current?

Best regards,
Noob piezo

 

[Skogsgurra]

To amplify the current would mean that you use an amplifier. There are no amplifiers without losses. So, unless you are an over-unity believer, don't.

I think that you need to understand the characteristics of a piezoelectric power source. It outputs a charge and, as long as you don't load it, that charge can produce a rather high voltage. It is when you try to do any useful work that voltage collapses.

"But I don't load it" you say. Yes you do, charging an electrolytic capacitor is work. And do remember that electrolytic capacitors have leakage currents. But I agree that you should have more than 20 mV if you have waited for a while. Check if the capacitor is connected with the right polarity. If it isn't, the leakage gets really bad.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[zapciname]

Hey Skogsgurra,

Thanks for your feedback, yes I understand the circumstances of all the power losses through internal resistance the piezo faced when it is put with extra electric components. As such, Im just doing a study on manipulating this low power output to be as optimal as possible . With that, I do not have a problem with voltage as I can easily add a high resistance (Tested and proven with V=IR) to increase the voltage. However, the current remains the same ranging from 10microA - 30mircoA. As such, Im looking for methods to boost the current. Perhaps, a buck converter or other methods? If so can guide me through the math of it?

Thanks!!

 

[ijl]

See thread797-339131

 

[waross]

Im doing a project on piezoelectric power harvesting.

There are a lot of useful applications for piezio-electric devices. Unfortunately power harvesting is not one of them. How many Watts or milliWatts of amplifier losses are you willing to spend to harvest microWatts from your piezio device?

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Skogsgurra]

Qoute(OP): "Im looking for methods to boost the current. Perhaps, a buck converter"

I told you that all converters (or amplifiers) have losses. And then, there's the law about conservation (or loss) of energy. Bill (Waross) summarizes that beautifully in his comment: "How many Watts or milliWatts of amplifier losses are you willing to spend to harvest microWatts"


Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[OperaHouse]

Possible problems. If capacitor is new or hasn't been used fr a while it will have considerable leakage for a while until it forms. Try applying power for a while at expected voltage with a power supply. Try a 1uF film capacitor that has low leakage. Second, your diode may be slow. Almost any diode is fast on turn on. Turn off is slower for generic diodes, that shorts out the signal for a while. Piezoelectric signals are generally high frequency pulses (short duty cycles) and have very low power when averaged by a capacitor.

 

[itsmoked]

Go with a controller designed for it?

[URL unfurl="true"]http://www.linear.com/product/LTC3588-1[/url]

Keith Cress
kcress -

http://www.flaminsystems.com

 

[jraef]

How do you go from 30VAC to 20mVDC by just going through a bridge rectifier? Did you leave off an "m" in the first value perhaps?


"You measure the size of the accomplishment by the obstacles you had to overcome to reach your goals" -- Booker T. Washington

 

[Skogsgurra]

Smoked&Jeff

It is a piece of piezoelectric material (ceramic or foil) and all it does is to produce a charge change when force is applied. The charge is miniscule and such energy harvesting techniques (popular in IoT applications) cannot be amplified or enhanced with external devices that consume energy because there is no other energy available. And the charge produced is very often in the nanocoloumb range. High voltage when left unloaded and no voltage at all if you try to load it more than a few uA.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[OperaHouse]

I worked on a similar low energy harvesting project that sampled energy from magneto points of a field irrigation pump. This provided a few uA (about 20 if I remember) to a 90V LED array pointed towards the field managers hut. On a dark night this was bright enough to see and prevented the half mile treck on a narrow snake infested levee to the pump several times a night to verify it was still operating. It is rare these low energy harvesting projects ever find an application. Could you explain your application to see if the end result is practical.

 

[itsmoked]

Gunnar, have you looked at the data sheet? The device I suggest is entirely capable of harvesting piezo energy. That's its sole function. They work great! Look at the data sheet. They show that chip running a wireless transmitter via a uP! Of course if the OP wants to do inappropriate energy demanding things, heat water, power his bike, of course that's pointless. But one can easily run a lot of micros these days on nano amps.

And, of course amplifying the result is ridiculous since it implies another power source to drive the amplification. LOL

Keith Cress
kcress -

http://www.flaminsystems.com

 

[jraef]

My only point was in exposing either an error in units, or a possibly gross mismanagement of the circuit to lose over 29V of potential just in rectification... I understand that you are all ASSuming he meant 30mV, not 30V, but that's not what he said in his post. If he rectified 30VAC and only ended up with 20mVDC, he is doing something wrong.


"You measure the size of the accomplishment by the obstacles you had to overcome to reach your goals" -- Booker T. Washington

 

[Skogsgurra]

Jeff&Smoked

Yes, I read that data sheet. But the power this guy seems to be needing is way more than a piezo device of reasonable dimensions is able to output if you don't put in some kind of vibrator. The use of a 1000 uF electrolytic capacitor points to that. And I don't think that many of us assumed 30 mV. That wouldn't even open a schottky diode, let alone a Si diode. So 30 V is what the guy is talking about, no doubt.



Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[zapciname]

Hello Everyone,

Thanks for all for your feedback and I appreciated it, although there's been a couple of disagreement in this discussion, I still thank you guys for caring enough to
help me. Just to answer some questions, Yes I talk to my all my professors before in the university in regards to piezoelectric behavior for generating electricity.

Firstly, for understanding purposes, piezoelectric transducers produces electricity in a spike manner when under stress. However, although voltage is high, the current produced
from it is low and to make matters worse, it's heavily subjected to power loss! So I hope that answers the question why is dropped from 30V AC to 20mV DC (with bridge and capacitor). If you read my
earlier post, Low Current (22uA) produced is my problem. I can easily increase voltage by adding a Super hig resistance but current will be constant. Thus, Im looking on how to INCREASE CURRENT.
For your information, 22uA is insufficient to be used on a BJT (min 0.5mA). As such to clear, Im looking on other ways to increase current.

So far, my professor as me to experiment with buck converters since Power input = power output and the voltage and current can be manipulated. So....If you guys got any other methods to give me a shout out
Oh and to be clear, my objective is find out how to improve the low current/ power output from piezoelectric and IF the power harvested can be used in any applicable form.

Best regards.
zap!

P.S THANK YOU ALL ONCE AGAIN

 

[zapciname]

oh and itsmoked,

THANK YOU i will buy that to try instantly, if it worked i will publish my results here!
I believe that since all you guys have passion in engineering i will be more than happy to share my findings!

 

[waross]

Nice solution, Keith. lps

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Skogsgurra]

zap

I think that it is a good idea to first find out how much energy/timne aka power you have available from your piezoelectric device. You haven't told us anything about the device's data or what amplitude, force and frequency your vibrations have.

Then find out how much voltage and current your application needs. And if the load is intermittent or not, duty cycle and so on. If available energy is better than twice what you need, then there's a possibility that it will "just" work with the LT chip. Why twice? because impedance match (optimum match) drops half the energy, I would make it four times for better sleep at nights.

Before you have compared these fundamental data and needs, there's no reason at all to try anything practical.

A comment on "manipulating this low power output to be as optimal as possible": Remember that the internal impedance of a piezo device is purely capacitive. At low frequencies that means that it is mainly a constant current generator and that is why your voltage increases when you increase load resistance. If it is possible for you to find a driving force with a much higher frequency, then the internal impedance goes down and the available current increases.



Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[zapciname]

Hi Gunnar,

Wow you sure list out factors that are needed in this. You do have a process flow mind indeed. I have attach the datasheet for the piezo.
I use finger tap force for experiments and can you tell me the methodology on how to gather force data and frequency in the piezo base on tapping force?

Thank you so much!

 

[HamburgerHelper]

I have always wondered how the pendulum self charging watches compare efficiency wise to what you are trying to do. My gut tells me that if piezoelectric crystals were a good solution, self charging wrist watches wouldn't be so unusual.

If you are going to build a buck converter to get more current, you might want to look into using an ideal diode bridge. It uses a series of transistors, I believe, as switches that are controlled smartly to rectify AC more efficiently than diodes. I came across them when on some boards built for pinball machines. I hate them only for the reason that they are board mounted so they are harder to solder and replace when they fail compare to a standard diode bridge.