I will make a half bridge again, with 2x2 mosfets in series.

I forgot to mention that at high voltages of the impulses, I saw the current ripple of L2 disappear again, which I first saw, right before I blew up my capacitor.
So it really works, but only at 3kV impulses.

that's why I need 2x 2 mosfets in series.
to get those high voltage positive and negative impulses for parametric excitation.

very exciting this.
I will use 134nF for L2 and 11nF for L3 again as this seemed to work, but at 145kc/s which is very high, I want to stay below 100kc/s So I will try different values also.

dc offset is useless with both polarity impulses, so it wont be used

Also will make a new  L4 0.75mm2 but now equal weight to L3.

First. I will test the principle with only 2 mosfets, as That is already almost build.
if it works,
I can build another pcb, to complete the 2x2 with the 1700V mosfets.

then...
Yeah lots of work. Love it.

I found another parametric excitation video.
a bit longer due to the mathematical analysis.

https://youtu.be/1c_Zqos-FFs

D@mn!
I tested the Idea. but the half bridge wont work that way.
the thing is, L1 always sees 2 body diodes. one of each polarity. So the first impulse always finds a path.
There is no solution to only get the second half wave back into the cap for re use, as the first half cycle also always get stored.
leaving you with no impulse at all.

that really bums me out ... I had hoped for this to work...
 :-[

There still is one solution. And that would be to use AC in stead of a DC supply.
Which means I cant use diodes, that would also mean I can use the mosfet as it has a body diode.
the switch would need to be ... hardware... nope nope nope...

Dear evostars, I’ve been following your work for a short time and I’m very excited about the prospects of your experimentations. Your YouTube videos and notes have been very informative and have encouraged me to join in along this line of research.

While researching parametric excitation in LCR circuits, I came across this paper that I think you’ll find interesting: http://aias.us/documents/miscellaneous/LCR-Resonant.pdf

Some quotes from this text:
It is shown by simulation that energy from spacetime
is possible in certain cases.
Then the voltages at the inductance and capacitance
are about 4000 V, which is technically feasible.
With driving force of 5 V peak signal the current of 10 A fows
through the voltage source, this could be considered
as an input of 50 W compared to 4 kW peak output.
This is a remarkable COP of 200. The proposed design
is adequate for a small generator in households for example.

We have designed and simulated a solid state version
of a mechanical parametric capacitor in a resonant circuit.
The parametric switching can be realized by transistors,
leading to a complete solid state design.
One recommendation is to use Ideal MOSFET diodes
for higher switching frequency, and other designs deserves further study.
It's common knowledge that the higher the frequency
the smaller the electrical components need to be;
however, it is noted here that we lose the ability to handle higher power
with smaller components. Nonetheless, the designs described in this article
are good candidates to construct solid state renewable energy devices that shuttles energy from the time domain.

Quote from lfarrand on December 10th, 2021, 11:20 PM

Dear evostars, I’ve been following your work for a short time and I’m very excited about the prospects of your experimentations. Your YouTube videos and notes have been very informative and have encouraged me to join in along this line of research.

While researching parametric excitation in LCR circuits, I came across this paper that I think you’ll find interesting: http://aias.us/documents/miscellaneous/LCR-Resonant.pdf

Some quotes from this text:
It is shown by simulation that energy from spacetime
is possible in certain cases.
Then the voltages at the inductance and capacitance
are about 4000 V, which is technically feasible.
With driving force of 5 V peak signal the current of 10 A fows
through the voltage source, this could be considered
as an input of 50 W compared to 4 kW peak output.
This is a remarkable COP of 200. The proposed design
is adequate for a small generator in households for example.

We have designed and simulated a solid state version
of a mechanical parametric capacitor in a resonant circuit.
The parametric switching can be realized by transistors,
leading to a complete solid state design.
One recommendation is to use Ideal MOSFET diodes
for higher switching frequency, and other designs deserves further study.
It's common knowledge that the higher the frequency
the smaller the electrical components need to be;
however, it is noted here that we lose the ability to handle higher power
with smaller components. Nonetheless, the designs described in this article
are good candidates to construct solid state renewable energy devices that shuttles energy from the time domain.

Thank you for sharing this excellent article!
It is inline with what I hoped to achieve with the halfbridge circuit.
But I still don't see how I can create the alternating impulses, which would make the capacitance of the L2 L3 coil capacitor, vary Parametrically.

So now I only have the negative impulse. Which already works, which is great. but... I also need that positive impulse.

or... maybe... I could drive L3 at an octave lower then the L2 impulse. maybe?

edit:
part 2 of the paper:
http://aias.us/documents/otherPapers/LCR-Resonant-2f.pdf
and 3rd related article:
http://aias.us/documents/uft/UFT382.pdf

maybe with 2x L1 coils, and 2 mosfets and a single L2
I could do it.
but the impulses would enter L2 from the opposite ends?
mosfets are driven out of phase

edit: L2 never charges, can only discharge, so will not become series resonant.

I tried many variations. my head is breaking over this. So I'll let it go again, and return to the coil cpacitor tread and proceed there

best option of success would probably be using (pulsed) AC instead of DC.

the AC would need to be at the resonant frequency of L2.

AC could be generated by the half bridge.

AC is then fed to the L1 coil by a single mosfet.
mosfet should then open at max current of ac through L1. giving alternating impulses into L2.

L2 could easily be made series resonant with an (pulsed) AC source.

3 mosfets in total. 2 for the AC generating half bridge, one for impulse generation

but... again, will the body diode mess everything up when using AC?

To handle AC you could connect two MOSFETs back to back. First one would handle the positive half oscillation and second one would handle the negative half oscillation.



More details here: TI Designs - Low-Cost AC Solid-State Relay With MOSFETs

Quote from lfarrand on December 13th, 2021, 07:55 PM

To handle AC you could connect two MOSFETs back to back. First one would handle the positive half oscillation and second one would handle the negative half oscillation.



More details here: TI Designs - Low-Cost AC Solid-State Relay With MOSFETs

yes I know.
but the problem is the positive impulses will always pass through the body diodes of the mosfets.
I need both positive and negative voltage impulses (alternating) so a half bridge wont ever work. H bridge also wont work.

bipolar/center-tapped coil help to adapt to this setup perhaps? maybe naive question

Hey Evo. I saw your latest YouTube video. Sounds like you cracked the problem of how to handle the +ve and -ve impulses. Bravo, well done!