lets first see what happens, when I remove the neutral from the HV coil, so It wont act up. (output coil still earth grounded)

I removed both earth ground from the output coil
AND  neutral from the hv coil.
still the same, so that diode is conducting.... tss
 :wtf:

this is with the diode parallel to the choke removed.
only the HV coil PSU is now powering. HC PSU=0 0
current is induced in the choke, and the orange spike shows the discharge, when the HC coil is switched to neutral.

put a diode on the neutral of the psu

lower duty cycle to 8%

I changed the choke to a smaller 3mH  to allow more curretn for the HC to become resonant by itself from the psu.

I had a small accident whereby the HC psu shorted out on the buck converter of the delay IC.
It seems not to have caused damage.

But I now checked the HC mosfet gate signals, and they are wrong. I first dialed in a 25% duty cycle.
But now I see, the frequency has halved. giving 1 pulse per 2 cycles
 :-[
If this was always the case?

I connected the choke again to the center series connection of the bifi coil, and both mosfets are connected again on on each side of the parallel tuning capacitor of the hc coil.

I hope I didn't smoke anything.
I need to see why the signal is so slow

definitly something not good with the HC coil signal.
the left is 1:2   on the right it is 1:1 as it should be. no clue why
 :'(

I am going to replace the delay IC its output is unstable
HC123

it was not the IC, I had the delay turned up to high, and that made it somehow go into 2x slower frequency.
well ok... lets continue

ok back to basics.
I removed the mosfets and choke from the HC coil.
Powered only the HV coil, and checked it it still is resonant.

freqeuncy is a little bit higher without the choke and mosfets connected, at 15.88kc/s
this is the same as it was before.
The parallel capacitor measures 5,4uF at 10kc/s so that is also still good.

phase is also good.

Now I need to get that HC coil resonant without the HV coil. at the same frequency.

To be clear, the curretn probe arrow always point towards the coil away from the capacitor.
I can measure the same current and the same phase on both sides of the capacitor.
So a true parallel resonance, whereby the coil endings, simultaniously charge the capacitor plates, and vice versa

I don't really understand that the current on both ends of the coil has the same polarity.

I am used to it having opposite polarity.
Like the ac high voltage coil.
 each end of that coil has opposite polarity voltage (and current)

So is this due to the dielectric induction of the high voltage coil?
which makes a displacement current that is longitudinal in flow?
and thus the capacitive coupled bifilar HC coil has this current flowing to both ends simultaneously?

then it does make sense to put the dc voltage and choke in the series connection of the bifi hc coil.

But what about the mosfets?

What about the voltages on each side of the capacitor plate? lets check that. because if the currents are equal in polarity then what is going on with the voltage on the plates of the parallel capacitor?

I placed two HV probes one on each and of the parallel capacitor.
orange and yellow, show the capacitive coupled voltage of the HC coil. both are in phase. as expected.
But this does not show the much lower voltage of the 5,4uF capacitor.
for that I would need a differential probe.
https://www.eleshop.nl/catalog/product/view/id/6239/s/pintek-dp-15k/category/301/
is nice, but then the low voltage reading is still very not visible. hmmm...

cyan is the math of substracting the voltages (arange yellow) of both sides of the cap.
10V/div shows around 20V on the capacitor, which is in the range where I would expect it to be.
Also the phase angle in relation to the current (green) appears to be fairly correct.
it is very crude but it gives a nice idea of how to proceed.

Note that the polarity of the math is set by which voltage is subtracted. Vc1(yellow)-Vc3(orange)=Vs in this case.
yellow is probed on the HV coil open ending.
orange is probed on the hv coil connection to the HV AC transformer.

my best guess is this is working with around 10 to 15 volt between the capacitor plates at this setting (1.51A + 2.0V HV psu)

I now connected the center series connection of the bifilar HC coil, to the 3mH choke.
This slightly detuned it to F=14.8kc/s
only the HV psu is on, which now uses 3.0A and 3.7V dc

again the current on both sides of the capacitor is equal in polarity.

So It would make sense, to connect both mosfets, and let one side be connected to neutral, constantly alternating this.
lets do that.

Phase should be right, so that the mosfet switches when the voltage is zero.
if needed the mosfets can be reversed connected to the ends of the coil/parallel capacitor

I checked the duty cycle of the HC coil gates, and it is 47%, so 3% dead time which is good.

now I checked the phase between the hv coil gate (orange) and the HC coil gate (blue)
this is also good, or in the neighborhood where it should be -90 degrees or +90 (polarity is not clear yet)

I reverse connected the mosfets to the ends of the HC coil, at the capacitor plates.
I probed both the capacitor plates (blue and orange).
When powered, a clear spike is visible, which is produced by the choke, when the mosfets disconnect it from neutral, allow it to discharge.
When the next mosfet turns on again, the spike is discharged, which makes the system ring.
My mosfets can handle 900V, and both have 200V spikes now.
This was below 9.3V and 0,07A on the HC psu.

The current (green) is low, but present. the large current spike are from the discharge of the voltages.
current also doesnt show a nice sine wave. around 180mA peak to peak.

If I go higher in voltage, my audio starts to make crackling noise, probable from the discharge of voltage.

I lowered the HC psu to 0.06A 7.5V hc psu,
to get lower voltage spikes (as I expect them to rise).
the voltage of the spikes now was around 170V without the HV psu.

then I increased the hv psu to 0.65A 0.8V
Now it can be seen tha the phase seems to be good, the spikes where the mosfet switches, is at the zero voltage point.

when the voltage is rising from negative to positive, the spikes have a larger magnitude around 190V
while when the voltage is declining from positive to negative, the voltage spikes are just 150V.
This indicates the choke, has somehow got less magnetic charge in its field, to produce the spike.

the current in green is not decreased from the lower hc psu, but increased to 200mA pp from the dielectric induction of the HV coil.

I put the current probe in the 200mV/A setting (earlier it was in the 20mV/A setting) to get a better picture
first is only powering the HC coil. 0.06A 7.5V
second is with also powering the HV coil. with 1,4A 1,7V dc

this not clear enough. The larger spike does appear the create a larger current. but it is to vague

I played with lower dury cycle settings, but that didn't make any sense.

when tuning back to 45 % duty the music stopped because my wifi router didn't like the fast changes in voltage....  needed a reset

hmm reset my router and laptop. still no streaming of audio or video...

rest the audio mixer, and now it works again... seems the mixer was the source of problems.

I reduced the duty cycle to 41%, which allows the spikes to fully form and return to zero. This also solved the unequal magnitude of the spikes.

HC psu=0.02A @ 17,0V
spikes are 140V

HV psu turned off

as soon when I tune on the HV psu (1A 11,3V) the spikes become unbalanced again.
But now, the spikes can return to zero again, giving less noise on the current probe (green)

voltage rises and falls. making it impossible to create high currents, without over voltage protection

the unbalance again, is from the choke discharge current. the high voltage displacement current, is in and out of phase with the magnetic current of the discharge(?)

so.
I do need to deal with those spikes.
is a reverse connected mosfets the solution. I think not.
the choke gets charged any way, and will discharged during the dead time of the mosfets.

So to conclude today.
the 3mH choke works (277 ohms at 14,7kc/s),
barely any resonant current in the HC coil. not enough to be amplified by the displacement current (if it is connected in the right way)

Also the spikes. they are a real problem. I cant get to high enough levels with these spikes Ill kill my mosfets

So a smaller choke allowing more current. less windings.
still the spikes will be there.

I tried bypassing the choke, but even then the spikes are present altough less high in voltage, and faster.
naturally I can only use very low voltages, with high currents.
And I need those voltages to get it to work.
So no, not a different choke. or no choke.
 I need fast diodes, to kill those positive spikes.
but... I already have had many troubles there.

BUT also success. due to the higher current, more amps flowed, and these where amplified by the high voltage displacement currents.
So...  back to that path

If the current is the same polarity on both ends of the coil,
and there also is a voltage over the parallel capacitor, is then only the current the same polarity on both ends of the coil,
while only the voltage has the opposite polarity ?

that would mean on one end the voltage is leading
while the other end the current is leading?
?

I can see the dielectric displacement current flowing longitudinal perpendicular into the coil (and reverse), whereby it then on the coil starts flowing towards both ends simultaneously

ending up at the capacitor plates creating opposite polarity voltages on the plates due to ?
what? the coil spin direction? but wouldn't that also mean a opposing current?

tsss
black magic.
turn to white.
enlighten me please.

lets take the open ended tesla coil which I used a year ago, to measure the current, and see if it is opposing on both ends.