Unfortunately the magnetization is wrong.
It would have to be magnetized circumferentially, but it is axially

Why not simply stick two speaker toroid magnets together in attraction with the piezo transducer in between, along with some inert spacers to regulate the correct pressure or airgap, and allow room for the windings. Cutting the ferrite is a bad idea.

There is a simplified formula for calculating piezo power consumption, but the best approach would be to just wire a resistor in series with the transducer and coil and measure accross it. The power consumption while in resonance drops considerably as demonstrated by Tinselkoala from 5 nto 2 volts at the 24khz resonant peak in his video.
 

Hi Guys,
I did a few tests today - to test the hypothesis of the opening post - that mechanically vibrating a magnetic core give a useable output in a coil. Using:

 - A signal generator. 7v Pk-Pk. Sine wave.
 - A 20mm(ish) piezo element
 - An electromagnet - which makes an excellent pickup coil
 - A 2" ferrite magnet
 - A small stack of N42 neo disks

I did a number of tests with different arrangements - but mostly with the piezo in physical contact with either the magnet, coil, or both. (Actually I put it in a plastic bag to prevent direct electrical contact.)

Attached photo shows one arrangement with the ferrite magnet.

I found two main resonance points - no matter the arrangement, at about 100KHz, and 10MHz. The exact resonances did change, but not by a lot.
The voltage output from the coil was always less than the input - in the example above I got 1.76v(P-P) out at 10.3MHz

I suspected that the resonance points may be the natural resonances of the Piezo element itself, but I think its actually the coil...
I tried just the piezo and coil - without the magnets - and again got similar results.

In the end, suspecting capacitive coupling, I replaced the piezo with 2 sheets of aluminium plate as a flat-plate capacitor, and put the coil on top.
I got virtually identical results.

So - in conclusion:
 - Capacitive coupling from the piezo may look like an acoustic excitation - but it isn't.
 - Did the author of the patent make this mistake?
 - If there are acoustic resonances to be found - the piezo's capacitance / electrostatic field would likely prevent it from being observed.
 - A non-electrical means of acoustic excitation would be preferable.

Regards, Tim

hi everyone,

I have just uploaded my first working overunity project into youtube.This is based on Magnetic Resonance Amplifier using Barium Titanate which was purchased from ebay around 2 years ago.
The input current consumption is "0.5uA" via Barium Titanate output current  maximum achieved 500uA from output of another bifilar coil.That's estimated 1000x ou in uWatts.

Magnetic Resonance Amplifier using Barium titanate ver 1.0
http://www.youtube.com/watch?v=MbQJexS4M4k

I don't know why i need to whisper while creating secret video in my room,can't help it.

If anyone know of any signal generator circuit or I/C which uses less than 100uA current and able to produce around 18khz.Please do advise me so that i'm able to demonstrate a close loop system to blink led forever at least.

I have attached a photo to show which of the sample which gave me the best result.

Do google "Magnetic Resonance Amplifier" my experiment is actually based on old research done in the 90s by a group of scientist.

If anyone know of any signal generator circuit or I/C which uses less than 100uA current and able to produce around 18khz.Please do advise me so that i'm able to demonstrate a close loop system to blink led forever at least.

Hi Magpwr,
  you can't use a multi-meter to measure HF... It won't give you accurate results. You have that super-duper Rigol scope there - why not use that?

Once again - another application for Verpies digital HF wattmeter... It'll be good when it's done.

Regards, Tim

Hi Magpwr,
  I think I bought some of the same barium titanate off that guy on ebay...

TBH - I couldn't actually hear much of what you said in the vid... You did whisper. lol.

It's certainly interesting... You should get an increase in voltage over the supply - because it's a series tank circuit. That might account for the LED lighting up.

Can you measure the input power to the sig-gen? That might be interesting - does it increase under load etc...?

You can get These DDS modules for about $5 on ebay - easily(ish) programmed with an arduino (for example).

The datasheet says it uses "155 mW @ 110 MHz (3.3 V)"

Regards, Tim

...
The input current consumption is "0.5uA" via Barium Titanate output current  maximum achieved 500uA from output of another bifilar coil.That's estimated 1000x ou in uWatts.

Magnetic Resonance Amplifier using Barium titanate ver 1.0
http://www.youtube.com/watch?v=MbQJexS4M4k


If anyone know of any signal generator circuit or I/C which uses less than 100uA current and able to produce around 18khz.Please do advise me so that i'm able to demonstrate a close loop system to blink led forever at least.

....

Hi magpwr,

You have interesting results for sure.     

Luc (gotoluc) has built an LC MOSFET oscillator which run at 20 kHz and drew 14 uA from 3V DC  i.e. 42 uW input power, here is his link: http://www.overunity.com/8892/self-running-coil/msg235133/#msg235133    He used an 1 Henry coil and tuned it with Neo magnets, the capacitors for the oscillator were the MOSFET interelectrode capacitors.  Later he managed to reduce power consumption : http://www.overunity.com/8892/self-running-coil/msg235505/#msg235505    I cannot recall schematic but he used two windings on a toroid, one of them was in series with the drain and +power supply input and the other coil was directly between the gate-source pins of the power MOSFET (IRF640), that was all for the oscillator, no more components.

Here is another oscillator which needs no toroidal coils of 1H but uses 10 MOhm (and 22 MOhm) resistors and two MOSFETs in a normal astable multivibrator:  http://discovercircuits.com/DJ-Circuits/astable.htm     of course the 22 MOhm values should be changed to get your oscillating frequency needed.

Off the shelf circuits are also a possibility: http://touchstonesemi.com/wp-content/uploads/2013/03/TS3003PB-r1p0_Final-Mar-13.pdf

Good luck for looping! Make sure to use some voltage or current limiter means in the feedback loop  to prevent a runaway situation.

Gyula
 

I know the multimeter do have it's limitation especially at low current.

The majority of the limitation is not low current but high frequency.  (even as low as 1kHz).  See this and this video.

Your experiment is not measured properly. You cannot multiply AC Volts by AC Amps and expect to obtain average Watts - that calculation works reliably only for DC.

I would appreciate if you measured input voltage and current across a 1Ω current sensing resistor (non-inductive!!!) simultaneously with two channels of your scope.

The output voltage and current can be measured a similar manner ...or with a brightness of an incandescent light bulb with a straight filament (e.g. from auto dome light) and sensed by a PV cell in a dark box (calibrated with DC).

0.5 micro Amps x 4.5volt at input = 2.25uW estimated output power achieved 4.5mW

You cannot do that with non-DC !!!
Finally, you cannot draw any COP conclusions from the fact that the power output does not affect the power input - even if these power levels are measured correctly.  Relative power levels don't mean much - only absolute levels do.

@ all on this thread,

Please check figure 9 on the page http://philica.com/display_article.php?article_id=219

Its quite similar to what is being done on this thread, the piezo acts as a variable capacitor due to parallel vibrating plates across the piezo crystal plate in the middle.

Dr Turtur has given a mathematical explanation for extracting ZPE from this circuit on the same page.

Please comment.

Best,