the magnetic induction push pull isn't capable of getting the voltage high enough on the HV coil. which is like an extra coil.
it sparks over around 3kV

but, there is also dielectric induction from the quick change in voltage of the HC coil which to gether with the HV coil acts like capacitor plates.

the high dV/dt induces current flow, which amplifies current on the HV coil, which transforms into high voltage.

I hope to see this soon, when my back emf of the M mosfet is under control

ok, I placed a snubber circuit between drain and V+ of the M mosfet
spikes are gone now. clean signals.

green is the HV coil voltage, blue is the HV coil current. (higher then the current coil which also produces a sine)
orange is the HC coil at the M drain
yellow is the HC coil at the choke.
Voltage dips, but... also dips back.
DC voltage offset is almost instantly restored which should not happen.

somehow, there also is a slicht negative half wave. that isd visible due to the DC offset, causing the body diode to work later.
also not visible, is th epositive half wave, as this is rectified through the snubber.

Snubber is from M drain to mur460 diode, with 100nF ceramic capacitor in series with the diode to the V+ of the dc off set of the current coil. with a 100ohm parallel to the 100nnF cap to drain it.

I also measured current on the output, 200mA pp. but it is near the HV coil, and when I measure the current near the HV coil in mid air, there is loads and loads of current. even more then when probing the HV wire...
so it does appear to work. displacement current is present.

M discharge is timed at minum voltage point, but as can be seen the max current is somewhat shifted, so, I could play with the delay even more.

I should move the load away from the HV coil fields.
And rectify into cap, for DC measurement.  I expect low voltage, due to the few windings of the output coil

this is the current of the HV coil arrow towards coil, measured at the choke side.
Current on the other side of the coil is equal but opposite polarity
notice the phase shift of current with the previous post, of the hv coil at the same side

dc offset is 16.6V  with  0.03  A
PP gets 7.5V    0.56 A
still no diode to prevent negative half to flow back to supply

I playes with the delay,
whgen whatching the HV current, noting changes.

But when watching the HC coil current, there is a change.
The current is highest when no pulse is there.
The current is lowest when the pulse is at the max current point where voltage is minimal.
So it appear the phase is wrong.
not assisting, but dampening.

at least something is happening.

I want to flip the HC coil connections back again, so the currents are 180 degrees out of phase again.

but... still Why is that HV currennt lagged? is that from the push pull coil hysterisis, but then the coil would also delay.
or... is it from the capacity of the coil? that must be it

I am wondering if the earth ground of the probe is properly connected, since it picks up som much current in mid air, near any current source

I reversed the push pull primary connections to change the phase.

but still current acts the same, no change.

I suspect the dc voltage is to low, for the Low voltage of the HV coil.
no were near 30kV just 2kV.

I need to be able to get that voltage higher in the hV coil.

I am working on the HV coil the voltage is way to low from the resonant push pull secondary.
So I decided not to make it resonant. but instead generate a stepped up high voltage square wave with current in phase with voltage. I aim for 600V

This then connects annd powers a variable high voltage inductor, which is made of a 7 toroid core.
I will separate the windings in groups, so that the high voltage is better separated between itss windings.

To get a high Q I again intend to use 0.5mm diameter wire.

I removed a lot of the windings of the secondary PP so its resonant frequnecy will be very high. By driving it well below Fr I should get a good square wave. like a regular transformer does.

and by driving it not resonant the power stays available, meaning voltage and current are in phase.
I also removed 5 or 6 of the PP primary windings, so it can handle more current

I used ty wraps with hotglue to create the separation between the wires

This is the voltage (orange) and current (blue) of the pp secondary while not being loaded.
PSU is 0.02A 16.6V dc

The reactance of the tuning coil will give it a load.
but... the tuning coil will be resonant with the coil capacitor...
oh boy... I hope this will work.

I calculated the needed inductance to be 37.8mH from the 180pF and Fr=61kc/s

I now have way less inductance when the ferrite cores are inserted, so I will need much more windings.

I wound 2 layers per section.
square of the windings is inductance.

so I need around 3x as much windings?
so 6 layers per section. that is doable

orange is the pp output,
green is the HV coil open end.

all connected.

resonant frequency is 67kc/s so I only need a few more windings on the tuning coil. very well done.

The push pull gives a much to high voltage, and no square wave. AND the PP is resonant. it produces a sine wave instead of the square wave I wanted.

So basically the tuning coil is now 1/2 resonant, as I expected it to be. but that is not how I wanted it...

I need to think this over, how will I get that voltage down on one side, of the tuning coil, while the other side goes up?
pp needs more reactance?

tuning coil is series resonant, so low impedance.
this is pretty advanced physics...

I had an already wound large toroid, which I probed with yellow and orange (200v/div and 500v/div)

on top I wound a joule thief circuit with thick 1.5mm wire.

hooked up a 6V battery, with a mur460 between positive and the coils.

used a 25k trimpot which is set to 1.25k ohm to get a 1.6V at the base of a
2sc5200 npn transistor.

as can be seen the secondary gives higher voltage spikes (+1400V) due to the step up.
the orange signal is only around -500V which is due to the base winding giving less voltage.

The plan is to rewind a coil, only wind the common winding with the secondary, and connect a 12V battery,
to get really high voltage impulses that are equal but opposite polarity.
then I can proceed.

A 5k or 2.5k trimpot would probably be enough. but 10k or 25 k also works

I could make the joule thief windings only on part of the coil, and wind the secondary on a much larger section, to get even higher voltages.

the transistor can handle 230V max over collector emittor
and 15A continuously at the collector max
Vbe(on) =1.5V
123=B C E

this is also something for what I have worked out long ago.
but this is not best as a pancake, it is better as a solenoid.

basically better then the joule thief,
and with series mosfets you can get proper high voltages. stepped up by the L2

the beauty is that when the mosfet conducts,
it makes the L1 a and b series connected like a true bifilar coil.
so which increased (series) inductance and charged capacity.

but when opened up, only the mutual inductance stays, the capacity was already charged so basically will remain like a true capacitor.
but the magnetic field which was built by a larger inductance, now has a lower inductance, so it can quicker move the current, thus produce a higher voltage impulse with shorter duration.

 but the capacity will charge up, faster from the  current of the transforming magnetic field.

still the capacity will keep the voltage slower.

hmm is it really going to give a higher voltage, faster, with all that capacity?

isn't it better to just have two uncoupled solenoids, with a mosfet in between.
made from thick wire.
with the L2 made from thin wire wrapped aroumd the two L1 coils in one series.

Interesting this even makes me think of don smith. hmm

I can even space the windings of the L1 to decrease capacity. but the thick pvc coating of speaker wire basically already does that.

I will enjoy building and testing this.
using series mosfets again. it has been a while.

the mosfets can even be higher Rds as they are only used to open fast, and I don't need much current since the impulse will be so fast.

still the secondary L2 on top will present a load.

parts are in, so lets build and test this.

I worked onnthis years back, but now I understand much deeper.

capacitor reactance is extremely low when empty.

I still have much to learn.
The mosfet body diode, shorts the secondhalf wave out. not healthy for the mosfets. if the bifilar series connection is used.
This could be prevented with a diode in series with the drain and or source. blocking the seconda half wave.
but, I can also work with just a positive or negative impulse.

So I placed the mosfets on the low side, to only generate a positive impulse.
That also works.

BUT the secondary has a much larger impedance. So I dont get a proper impulse.
But that will be fixed soon, when I provide a low impedance path to ground

yellow is the drain (+200V)
orange is on of the endings of the secondary (-800V)

I didn't use any diodes for this setup. so the negative half passes through the body diode, to the negative side of the PSU cap.

tried, failed, reversed polarity, still the same.
maybe 470uF 100V elco has to high ESR

I took all the windings of the core, and made a new joule thief.
secondary has not been wound yet.

signals look good. 32 kc/s gives 1.5V at the base, and 200 impulse at the common.
I used a 12V lead acid as PSU
transistor is 2sc5200 (230V max)
base resistor is 9k4 ohm

secondary will only be wound on the winding connected to the collector.
diode from + to coil is mur460

second Goal is to step up the voltage, and create simultaneous positive and negative impulses on the secondary.
I have 0.28mm diameter wire for this.

I have problems with the joure ringer.

If I earth ground the emittor than every thing is fine. 1.5V on base, 200V on collector.

but as soon I remove the ground from the emittor (probe aligator clips)

collector drops to 80V
and base becomes over 15V erratic.

dinner time
lets think this over

earth ground the transistor seems to be the solution

I replaced the 25k trimpot and earth grounded the emittor. works fine  again aroumd 29kc/s

I now have 5pF 6kv caps, 2 im series makes 2.5pF  this gives a reactance of 2.2Mohms
at 29kc/s

the large 470uF is replaced by 2x 10uF in parallel, which have 0.27441 ohms reactance at 29kc/s
with 10 in parallel this would be down to 0.05488 ohms. very very low
not counting ESR but that would also be divided by 10.

I now have a total of 2000V which is still very low since the dV/dt also still isn't that great.

I do see the effects I want. but barely.

thus. I need a higher voltage, and or a quicker change of voltage.
Since I use ferrite and need to use it, I will Aim for higher voltage. and found a nice H bridge High voltage coil on AliExpress

 found this on AliExpress.
 €6,25
DC 7V-15V 12V To 15KV-20KV
High Frequency AC High Voltage Generator Inverter Boost Spark Arc Ignition Coil Module For Igniter
https://a.aliexpress.com/_ExEOQzB

AC which is needed. all flyback or CO2 laser tranaformers are DC and have an internal diode so they wont work.

it is a tiny device, but that is the charm, I don't need power or current. just need high voltage.
altough. this AC might be full wave
then half of the cycle is passed through ground. still not ideal.

sh#t why is it often so hard to find proper parts? Must I build everything myself?

This is the high voltage measured at the 2.5pF cap (5pF x2).
-700V in yellow and + 1300V in orange from both ends of the secondary.
and that is within a 3uF time frame. that is... SLOW
as said, this is due to the ferrite. so the only way to go is to step up the voltage

barely visible, but it is there.
200mV/A setting on my current probe.
blue is the collector positive impulse.
green is the current. very very small.
it takes a lot of seconds to even reach 10V.
it is quicker than without ground. but not quick enough to brake the "barrier".

Maybe my tesla plasma globe has a high voltage transformer that I can use, but I think that one is DC also.

excellent

I have an idea. what about using 2 flyback transformers in parallel. one for positive voltage, the other for negative voltage.
driven from one 97% duty pulse generator to drive both primaries, simultaneously.
secondaries are series connected. so the outputs are  positive and a negative simultaneous dc impulses

a variation of this
High Voltage Power Supply   5KV - 40KV CX-400  ,Electrostatic Ionization   , Electrospinning ,Spraying ,
https://a.aliexpress.com/_EyAqMTF

I ordered 2 TV flyback transformers.
they have built in diodes so give dc out,
but I earlier already found out, you can reverse connect the secondary to ground and get negative voltage out.

so I plan to connect them in series, with the middle to ground, I will get a simultaneously positive and negative out put impulses.

the primary will be a single coil wound around both ferrite cores.

hopefully they both will then give an synchronous output high voltage

https://chat.openai.com/share/36a24e2f-3c86-4df8-8a04-550401b816a9

I had several AI chats about the phase relationship between voltage and current in an electrical resonant system.

And it kept persistent in sayimg both voltage and current reach maximum and zero at the same time.

which is not the case.

I tried to correct it. but the chat was even closed by AI!
hilarious.

I also said that energy would be produced and consumed in such case,
and that real power was created, if so.

again it responded that reactive power was not real power.

Amazing how persistantly wrong AI can be.
I reported and gave feedback (bing copilot chat gpt4 and gpt 3.5)