the use of mechanical wash relay ! but here it works and how it is possible such a short very fast pulse?
when tensions 120V. It provides the inverter able to weld the relay contacts?
for long period of time it will not work

time 1:35   https://www.youtube.com/watch?v=4ZqHtsc537Y

  Tomtech:
  Interesting, that last Ruslan video.  He is showing how the induction circuit lights the bulbs by itself, without the Kacher. 
Each additional bulb added is dimming the brightness of all the previously lit bulbs, by quite a bit.
  He almost showed what I've been waiting for. To show how much more output the Kacher crt adds when connected to a full load 2000 - 4000 watts worth of bulbs. But he didn't, this time.  Bummer!
  Without the HV circuit connected his bulbs are lighting similar to how mine do, nothing special, considering his input voltage is twice as high as mine.

  I've modified my Kacher circuit to run on 24 volts now. And, I also have a new yoke wound up for 24 volts, as well. It runs similar to my previous yoke core, on 12v, but soon I'll connect it up to my Mazilli, which hopefully will be running on 24v also.
This might also help to keep the Mazilli's fets running cooler. We'll see about that...
  About 10 more days before my Techtronics scope gets delivered. So, I'm doing what I can until then.
 

https://www.youtube.com/watch?v=4ZqHtsc537Y

The main purpose of that video was to show how frequency shifts from different loads(from 27.154kHz without load to  29.12kHz with 3 light bulbs load) and how PLL keep resonant condition with spikes to control Tesla coil driver. Also the circuit had Tesla coil disconnected and was drawing 6A from 24V battery.

Also the relay switch there you just seen is used there most likely to power circuit after capacitor is charged and the button on top is pressed...

The main purpose of that video was to show how frequency shifts from different loads(from 27.154kHz without load to  29.12kHz with 3 light bulbs load) and how PLL keep resonant condition with spikes to control Tesla coil driver.

This part is still a little disconcerting for me.  If the frequency changes, then it's no longer in the 1 to 100 ratio.  Seems to me this would kill the effect we are after.  Have you found any more information that would explain the derivation of this ratio?  How it was found and what its actual purpose is?

 
NickZ
I wonder if your system will get better performance? or is the result of larger diameter ring volume is controlled or has a very big impact here
 I found a uniform ring without connecting space of two halves
but is less fear that the winding 28 winding 4mm2 will be quite a challenge

induction heater would not look like we have a 810 watt load on the source 6Amp relatively need to take into account the voltage drop as can be seen in the light bulbs

This part is still a little disconcerting for me.  If the frequency changes, then it's no longer in the 1 to 100 ratio.  Seems to me this would kill the effect we are after.  Have you found any more information that would explain the derivation of this ratio?  How it was found and what its actual purpose is?

The ratios 1:50 1:60 and 1:100 still blend on resonant harmonics. And the purpose of using high harmonics is to bring frequency into kHz range for induction heater circuit.
With changing loads I assume you will need to tune system for maximum load first so when smaller load is applied the output power will be lower due a bit out of tune.  Just the moment when the Tesla coil driver is fired is still on same position in wave no matter how frequency shifts.

Hi Guys,  I have got my current sense transformers today, so I am gonna wire it with my microcontroler.  Dog- what resistor value you use to shunt it?

Can somebody explain me please how that Push-pul transformer actually works? Ruslan said it is the part where magic happens as you phase there your current and voltage.  I am thinking a bit different way now. Ruslan many times point to how fast capacitor charge from Tesla. The problem is we can't drain capacitor and at the same time to charge it. Is possible that push pull transformer works here something like the switch or blocker- like magnetic amplifier?  DC voltage controls the AC current flow.  When Tesla fire- it charge the cap and at the same time  DC part blocks the output from that cap,   etc?...

Hi Guys,  I have got my current sense transformers today, so I am gonna wire it with my microcontroler.  Dog- what resistor value you use to shunt it?

I started with 100 ohm and if by some magic the current goes way up (self runs), I'll add another resistor in parallel to bring the reading back on scale.  For the moment I just need a clean 0-5 volt reading for the input of the micro-controller.  One hundred ohms work very well for the oscilloscope which I needed to get the tuning capacitors set correctly.

I have noticed something counter to what T-1000 is saying by way of experimenting...

When you have maximum amplitude sine wave current flowing through the induction heater coil, you can place a steel screw driver into the tube.  The screw driver heats up no problem.  When you do this, the amplitude coming out of the current sense transformer drops by about 30-50 %.  No matter how much retuning you do (frequency changes and capacitor changes), you can only get this amplitude back up by about 10%.  So clearly the resonant energy gets transformed into heat.  Whether this holds true when you increase the load on the grenade coil, I can't say for sure yet.  It does have me wondering about the effectiveness of a PLL.  At the moment it appears sticking to a locked frequency regardless of load would make more sense.  You can adjust frequency all you want and still not get anywhere close to the original tuned amplitude.  In an actual running system, you can only change frequency dynamically, you can't change the capacitors.  So thinking ahead a little, leads me to believe you do not want to tune this device under no-load conditions.  Better to tune it with the load you expect to run at, that way the capacitor values are optimal and the frequency can be held constant.  Ideally, you want the system to become better tuned the more load you put on it, not the opposite.

As LC circuits go, this is a pretty high Q device which means the margin for error is very narrow.  You need to be in this narrow band or the resonance is gone.