Storing Cookies (See : http://ec.europa.eu/ipg/basics/legal/cookies/index_en.htm ) help us to bring you our services at overunity.com . If you use this website and our services you declare yourself okay with using cookies .More Infos here:
https://overunity.com/5553/privacy-policy/
If you do not agree with storing cookies, please LEAVE this website now. From the 25th of May 2018, every existing user has to accept the GDPR agreement at first login. If a user is unwilling to accept the GDPR, he should email us and request to erase his account. Many thanks for your understanding

User Menu

Custom Search

Author Topic: A Pendulum should really work  (Read 54380 times)

elgersmad

  • Jr. Member
  • **
  • Posts: 72
Re: A Pendulum should really work
« Reply #15 on: April 01, 2011, 02:25:34 AM »
Quote
@elgersmad: sorry, I have difficulties imagining a circuit from words. May be you could draw your circuit and post it.

I am afraid, the piezo elements are not able to power an OpAmp. The one you mentioned has a
Quiescent Current per Amplifier of 25µA at 6 Volt resulting in 150 µW.

150 µW is a top performing multilayer special production piezo element for "energy harvesting". What we will see from inexpensive run of the mill piezo elements is less (hopefully 10 µW)



Equation Reference

Piezoelectric materials produce voltage and current based upon Force and Time.  The more force that is displaced over a shorter period of time, the more energy is produced.  Hit it hard, get a high voltage, press on it, get a low voltage. If the leads are shorted, all you get it current, if the leads are open all you get is voltage.  For you to say, you know how much energy that there will be, is not even based upon a real answer.  In effect you're just spouting.  You can't tell me how much energy my pendulum will produce when I can predict that.  But, if I change the weight of the pendulum, that changes the peak voltage and current.  There's a list of things that can change and would change the output.  But, you are not doing the math and really are taking about some kind of sensor result from the top of your head.

If you were so right, this device wouldn't even work, and Piezoelectric Transformers are typically 90% efficient.
A Piezoelectric Transformer

90% of 40 watts is 36 watts.  The Piezoelectric Vibration Harvesters are really garbage.  But, what you are saying doesn't even allow a piezoelectric lighter to have enough energy to light a flame.  Check that one out in Joules.

For the amount of stress PZT can handle, it needs a lever and a fulcrum like a pair of pliers to pinch on the piezoelectric ceramic disks.  That fulcrum would be at a 90 degree angle away from the pendulum's axis.  Every time the weight comes down, the pliers bite onto the ceramic disks.  0 to 80% of the peak amount of stress that the element can handle is ideal.  So, when the pendulum swings, the pliers bite starts at zero, goes up to 80% then drops down again.  But, you are wrong.
« Last Edit: April 01, 2011, 02:52:32 AM by elgersmad »

conradelektro

  • Hero Member
  • *****
  • Posts: 1842
Re: A Pendulum should really work
« Reply #16 on: April 01, 2011, 10:44:32 AM »
What I said about the energy (current) coming out of a piezo crystal is what I read in articles about piezo electricity. The intention is to use all available information in order to have reasonable expectations. Yes, I know nothing, but I can read.

Only experiments can tell how much electricity can really be harvested by piezo crystals from a pendulum or a vibrating table (carrying an imbalanced wheel). So, till I have built something, I have nothing more to tell.

A pendulum is nice, because it is slow and quiet, but it will need a lot of room. An imbalanced wheel can be smaller. So there are advantages and drawbacks which have to be used or overcome by real experiments.

Greetings, Conrad

elgersmad

  • Jr. Member
  • **
  • Posts: 72
Re: A Pendulum should really work
« Reply #17 on: April 02, 2011, 02:04:27 AM »

A pendulum is nice, because it is slow and quiet, but it will need a lot of room. An imbalanced wheel can be smaller. So there are advantages and drawbacks which have to be used or overcome by real experiments.

Greetings, Conrad

You are right!  There is a way to use a balanced wheel, and I've figured out.

I was watching this Video

Basically, what you need is a potter's wheel, and for some-one to make a rotor.  The real difference between it and the SEG, is that the rollers are in a race just like a set of roller bearings.

Roller Bearings Picture

The outside race would need to be made from ceramic.  The bearings would be turned from the center by a motor, and not kicked along like the SEG by a solenoid.  They would be heavy rollers but, the whole wheel and assembly would be balanced.  Now, the ceramic race, would have 1cm segments of piezoelectric ceramic PZT embedded inside the wall.  So, as each roller passed over it, the centripetal force would press on the Piezoelectric Ceramic as the roller passed over it, and relieve it as the roller moved away.  The RPM would generate centripetal forces that would increase the apparent weight of the rollers.  So, if everything rolls freely, and there are spokes that lead to the roller bearings mounting as in this picture:

roller bearing Mount, that should have spokes to be turned from the center

It's just that the rollers should be in slots, so that when they go to press on the sides of the ceramic race, they can press on the sides as hard as the centripetal forces will allow.  Ceramic, can be as hard as steel, even harder and retain allot of stress.  But, if you drop it, it breaks.

After that, the piezoelectric ceramic, would need to be machined into place to make sure that the race was smooth and retained a low friction.  For the most part it's a centrifuge.  The larger the radius, the more force is experienced at a given RPM.  Being balanced, it doesn't require allot Horsepower even in a fractional sense with reduced friction being the main point of the whole assembly.  PZT is very hard.  It really only compresses by a few germs Micrometers.  Even normal metals do that much flexing under a microscope.  Manufacturing the device would be the most difficult.  You could couldn't put the electrodes on it until after the race was machined smooth so that the rollers just roll on it, and don't hit any bumps.



gyulasun

  • Hero Member
  • *****
  • Posts: 4117
Re: A Pendulum should really work
« Reply #18 on: April 03, 2011, 08:24:13 PM »
What I said about the energy (current) coming out of a piezo crystal is what I read in articles about piezo electricity. The intention is to use all available information in order to have reasonable expectations. Yes, I know nothing, but I can read.

Only experiments can tell how much electricity can really be harvested by piezo crystals from a pendulum or a vibrating table (carrying an imbalanced wheel). So, till I have built something, I have nothing more to tell.

A pendulum is nice, because it is slow and quiet, but it will need a lot of room. An imbalanced wheel can be smaller. So there are advantages and drawbacks which have to be used or overcome by real experiments.

Greetings, Conrad


Hi Conrad and All,

Here are some useful links on energy harvesting piezo devices, discussing their possible performance efficiency etc.  I just found them in the Kapanadze thread here:
http://www.overunity.com/index.php?topic=7679.msg280281#msg280281 

The last link shown there leads to a patent application, without Figures, here is a direct link to the full text + Figures + photos of the setups:
http://www.wipo.int/pctdb/images4/PCT-PAGES/2010/522010/10151738/10151738.pdf 

Gyula

EDIT: here is another paper: http://www.heathhofmann.me/IEEETransPelvol18no2march2003.pdf 
« Last Edit: April 03, 2011, 09:00:15 PM by gyulasun »

conradelektro

  • Hero Member
  • *****
  • Posts: 1842
Re: A Pendulum should really work
« Reply #19 on: April 04, 2011, 06:23:13 PM »
Hello Gyula!

Thank you, very nice information about piezo elements, specially the step down converter clarifies many questions I had. The setup from the patent would be difficult to reproduce, but it shows what could be done.

I ordered a few parts and will build a simple "shaker" with an imbalanced wheel driven like a pulse motor beating on piezo elements. It has been done before, but I want to see it myself.

I want to use toroids with an enameled copper wire winding, because I built such a pulse motor when the STEON-Craze was starting. It turned nicely with little energy but was not OU of-course. I posted a reed switch circuit and a photo of the toroids further up in this thread.

It will take some time to build, but I will post it here eventually. I just want to see how much one gets out of a stack of standard piezo elements (for buzzers) and how well a simple shaker (based on the pulse motor principle) performs. Does it really take just a little energy to lift an imbalanced wheel over its dead point?

The pendulum idea also has to be really built to learn more. It seems to allow for heavy weights bearing down on piezo elements (although with very little frequency). It also needs some experimenting to come up with a very efficient  "drive" for a pendulum. Everybody seems to know that it takes very little to keep a heavy pendulum going, but I never saw a convincing drive mechanism using and producing this famous "very little energy". It would also be useful for a Milkovic pendulum.

I learn most by building crazy stuff and it is also fun. It needs a childish type like me to enjoy all sorts of Joule Thief lamps in my house day and night. And a turning motor (however stupid it is) gives me pleasure.

For people who like microprocessors, I want to report that the TI LaunchPad serves me very well lately when building little projects. It is very well priced, robust and simple to use (all necessary software is for free on the Internet). Some sneer at the 16 MHz, but it is plenty for many applications. One also has communication with a PC via USB, which allows to display measurements at the PC or to control something from the PC (although you also have to write a program on the PC, e.g. with VisualBasic or VisualC):

http://processors.wiki.ti.com/index.php/MSP430_LaunchPad_(MSP-EXP430G2)?DCMP=launchpad&HQS=Other+OT+launchpadwiki

http://newscenter.ti.com/Blogs/newsroom/archive/2010/06/22/ti-s-new-4-30-launchpad-is-a-complete-development-kit-for-harnessing-the-ultra-low-power-and-16-bit-performance-of-msp430-value-line-mcus.aspx

Greetings, Conrad

Low-Q

  • Hero Member
  • *****
  • Posts: 2847
Re: A Pendulum should really work
« Reply #20 on: April 06, 2011, 11:48:16 AM »
First of all, a pendulum are slow, and the frequency are totally depending on its length and gravity. So, I agree that an unbalanced wheel are the correct approach, because its frequency (or rpm) are not depending on diameter and gravity. It can be made much smaller, with much higher efficiency.

Using piezo electric power source might be useful. However, the output are, as far as I know, depending on the frequency as well as the pressure. A good thing about piezo electric devices is (as probably already mentioned) the requirement of pressure difference rather than mechanical movement to provide energy.
Because an unbalanced wheel have to stay perfectly still during the run in order to not loose energy by its possible viberations. Viberations will slow down / break a spinning unbalanced wheel grately. So therefor, implementing a piezo electric device would probably be the best way because of the minimum viberations, and maximum pressure.

It might therfor be possible to feed back the energy from the piezo electric device into a high efficient motor which powers the unbalanced wheel. Even use the pulses from the piezo to feed directly into an electromagnet which in turn propells a rotating magnet.

The weight which provide the unbalanced wheel must be applied before the magnet approach the electromagnet in order to determine the direction of rotation, and rotation itself.

The magnet can be this weight, and if the coil are not placed at the bottom, but a few degrees later, the coil will be fed with energy in advace of the magnet, so it can pull the magnet towards it and force rotation.

Attached there is a drawing of a simple brushless motor which is powered by pressure on a piezo electric device.

When the magnet is on the bottom, the pressuere on the piezo are greatest. When the magnet is on the top, there is a "negative" pressure so the upper right coil have swapped polarity in order to attract the magnet. Also there should be coils which is repelling the magnet also.

On the other hand, the electricity are probably generated during the pressure difference from no pressure to max pressure, and the electricity reverse from max pressure to negative pressure. So there might be another placement of the coils, but I think you got the idea.

EDIT: The generation will occour when the magnet/weight are between top and bottom. This means that the wheel as drawn will rotate counter clockwise if the piezo are placed either at the bottom, the top, or both top and bottom.

Vidar
« Last Edit: April 06, 2011, 02:33:21 PM by Low-Q »

conradelektro

  • Hero Member
  • *****
  • Posts: 1842
Re: A Pendulum should really work
« Reply #21 on: April 06, 2011, 07:29:39 PM »
I did some tests with a piezo element (for a buzzer, sits on a metal plate):

See the attached drawing, photo an scope shots.

The piezo element needs "hits", pressure alone did not do the trick. So, if one builds an imbalanced wheel, the setup must produce "hits", like with a little hammer.

The stroke of the "hit" does not have to be very high, but the "hammer" must hit the element, leave and hit again, leave and hit again, ...

I could charge up a 100µF electrolytic capacitor (replacing the LED) within seconds to 3 Volt (about 5 hits per second), done by hand with a wooden stick (like rapidly making dots with a pencil).

The piezo element was sandwiched by two acrylic disks to protect it. With or without acrylic disks, the performance was about the same.

The low power red LED only gives fairly dim light pulses. I plan to use up to twenty such elements (all in parallel, each one having four diodes, feeding into an electrolytic cap) in "bouncer" with an imbalanced wheel.

Greetings, Conrad

Low-Q

  • Hero Member
  • *****
  • Posts: 2847
Re: A Pendulum should really work
« Reply #22 on: April 07, 2011, 08:17:13 PM »
I did some tests with a piezo element (for a buzzer, sits on a metal plate):

See the attached drawing, photo an scope shots.

The piezo element needs "hits", pressure alone did not do the trick. So, if one builds an imbalanced wheel, the setup must produce "hits", like with a little hammer.

The stroke of the "hit" does not have to be very high, but the "hammer" must hit the element, leave and hit again, leave and hit again, ...

I could charge up a 100µF electrolytic capacitor (replacing the LED) within seconds to 3 Volt (about 5 hits per second), done by hand with a wooden stick (like rapidly making dots with a pencil).

The piezo element was sandwiched by two acrylic disks to protect it. With or without acrylic disks, the performance was about the same.

The low power red LED only gives fairly dim light pulses. I plan to use up to twenty such elements (all in parallel, each one having four diodes, feeding into an electrolytic cap) in "bouncer" with an imbalanced wheel.

Greetings, Conrad
Could you load the piezo with a 10 - 20 ohm resistor, and measure the output voltage over the resistor?

DreamThinkBuild

  • Hero Member
  • *****
  • Posts: 574
Re: A Pendulum should really work
« Reply #23 on: April 07, 2011, 09:20:29 PM »
Hi Conrad,

I've attached pager motors to piezo's they generate about 119khz sine wave around ~3-5volts ac out not a lot of current(uamps) from my current piezo speakers. The pager motors use very little power 1.5v+@30ma+(lowest).

http://www.goldmine-elec-products.com/prodinfo.asp?number=G17959
http://www.solarbotics.com/products/vpm2/

Low-Q

  • Hero Member
  • *****
  • Posts: 2847
Re: A Pendulum should really work
« Reply #24 on: April 07, 2011, 10:25:06 PM »
Does anyone here know if  there is a direct relationship between the force and deformation (Force times distance) and the energy output of a piezo?

I have imagined a pendulum with a narrow magnet on it. The magnet pass by another stator magnet which is fixed to a piezo device. Each time the magnet pass by, there will be generated a mechanical pulse into the piezo device (Therfor a narrow (and maybe long magnet). The electric current should then power a small electromagnet which pulls the pendulum in correct phase to sustain the pendulum.

Vidar

lumen

  • Hero Member
  • *****
  • Posts: 1388
Re: A Pendulum should really work
« Reply #25 on: April 08, 2011, 01:46:16 AM »
Vidar,

Interesting concept!
I have a stack of these piezoelectric disks I bought on eBay some time ago for a similar device. I have been distracted on other projects so it hasn't been built but it was going to be a rotary device with disks around it and each disk would have a full wave bridge soldered to it to convert to DC. The DC was going to charge a capacitor (did some tests on this already) and power a small motor to keep the magnet rotating. The disks deflect very little for a good output and should have almost no drag on the rotating magnet.

Your idea gave me some insight to some other ways to do this also. Thanks!

Low-Q

  • Hero Member
  • *****
  • Posts: 2847
Re: A Pendulum should really work
« Reply #26 on: April 08, 2011, 09:40:31 AM »
Vidar,

Interesting concept!
I have a stack of these piezoelectric disks I bought on eBay some time ago for a similar device. I have been distracted on other projects so it hasn't been built but it was going to be a rotary device with disks around it and each disk would have a full wave bridge soldered to it to convert to DC. The DC was going to charge a capacitor (did some tests on this already) and power a small motor to keep the magnet rotating. The disks deflect very little for a good output and should have almost no drag on the rotating magnet.

Your idea gave me some insight to some other ways to do this also. Thanks!
I also have an idea of a rotor with lots of narrow and long magnets, and a stator with the same amount of magnets fixed to a piezo each. In theory, the rotor will not require energy to rotate as long as the stator magnets doesn't deflect. However, ther will be cogging, but that is not energy loss by itself. But if the stator magnets deflect due to the compression in a piezo, accordingly to the energy output from a piezo, it will not be possible to make a sustaining motor.

So I have asked myself this:
If we apply 1N of force into a piezo, and the piezo compress 1um, the input energy is 1uJ. The question is then: Will the piezo provide at least the same energy, or will it provide more? Will the piezo provide energy according to the force times distance? Or will the piezo provide energy mainly due to the force applied rather than the force (1N) times distance (1um)?

The problem, as you sure have figured out, is the deflection, and how this will affect the efficiency. Because if the stator magnet doesn't have the same distance to the rotor magnet when the rotor magnet has passed, it will take more energy to enter the stator magnet, than it gives back after the rotor magnet has passed, we have energy loss.

Hmmmm....I wonder...

Vidar

conradelektro

  • Hero Member
  • *****
  • Posts: 1842
Re: A Pendulum should really work
« Reply #27 on: April 08, 2011, 11:27:10 AM »
It is obvious that the amperage coming out of a piezo element is very low (because it is an electrostatic effect). Whatever we do, we will not make the piezo effect more powerful in any significant way.

Therefore one has to come up with a "drive" that uses very low power (in the range of µA).

May be I am wrong, but I am aiming at a heavy imbalanced wheel or heavy pendulum which by itself (because its weight overwhelms air drag and friction in the axis bearings) comes almost back to its initial position. The heavy imbalanced wheel would almost complete a revolution and the heavy pendulum would almost complete the same amplitude at each swing. And because it is heavy it will bend the piezo element also at low speed.

And with the microamperes from the piezo elements one could hope to overcome the "almost" if the contraption is carefully built.

The imbalance in the pager motor (which is in itself a nice gadget) is not heavy enough to bend the piezo element when turned slowly. So, one needs Milliamperes to impart speed on the imbalanced wheel of the pager motor which only then gives the necessary "weight" to bend the piezo.

The question can be reduced to: is it possible to overcome the losses (air drag, friction in the bearing) in a heavy imbalanced wheel or a heavy pendulum with Micro-Watts? By "heavy" I am thinking about 1 kilo.

Greetings, Conrad

« Last Edit: April 08, 2011, 02:14:08 PM by conradelektro »

Low-Q

  • Hero Member
  • *****
  • Posts: 2847
Re: A Pendulum should really work
« Reply #28 on: April 08, 2011, 02:32:13 PM »
I have now ordered 50 pcs 43mm piezo's on ebay. It should do for trial and error. 35 dollars for all 50pcs. It's worth a try.

Regarding heavy imbalanced wheels. They suffer from the very same air drag and friction as any other wheels with same shape. However, if the weight is the main reason why there is applied more force to the piezo, this must be the only, and important, reason why a heavy imbalanced wheel will work better.

I choose a lighter, but probably at least as much efficient design using neodym magnets (Which at most repel and attract with several kg for each small magnet). The short pulses from the passing magnet represent a higher frequency, which in turn will increase the current from an electrostatic unit, and hopefully make it more efficient.

Well, I do not know the outcome yet, but therefor I want to test this out. I also have extracted a bunch of small long neos from brushless RC-motors lately. So I am prepared for some tests :)

Vidar

Low-Q

  • Hero Member
  • *****
  • Posts: 2847
Re: A Pendulum should really work
« Reply #29 on: April 08, 2011, 03:17:38 PM »
A nice link for those who want to do some research on piezo units:
http://www.americanpiezo.com/knowledge-center.html

Vidar