High-voltage ZVS

I have seen the video you are referring to and I built a circuit similar to yours. I had trouble with it and then found a video by zenodilodon called "bare and basic zvs circuits of all shapes and sizes" and it was informational for me... I was lucky when I found an old home projection theater TV and took the flyback and the rest of the circuitry out of it. It is a huge transformer by comparison to the others I found, and almost fried my power supply from feedback. I was wondering how to prevent this when I read your statement regarding the zener diodes. If I run it with batteries it runs great though. You indicated that the zener diodes took the place of the feedback that normally would happen. This Piqued my interest. When it started kicking my little power supply around, I was experimenting with smoke and how it acts upon it when my power supply begin to glitch. Every time I arced the smoke in a bottle from end-to-end the power supply began to make sounds that you don't like to hear LOL. Without smoke it works fine, but ionizing smoke in a bottle will smoke the power supply. Do you know what the smoke is doing to make the difference? Obviously I'm new to this stuff but I find it to be fascinating.

I'd made a ZVS driver using a hand-wound transformer. I got a ferrite E-Core base from a PS and wound 8 turns for the primary and more than a thousand turns for the secondary side. It worked great; the cool thing was that unlike the flyback, the current made the arc thick, red and orange. The total input power consumed was nearly 120W. After a few days of its usage, I ended up frying my transformer. It was mainly composed of several layers of duct tape, which had melted and changed colour. The primary winding got completely fried while the secondary winding was pretty much intact, with the molten tape stuck to it. Well, the problem was that I'd used the type of transformer found in tasers, which usually consumes about 50W (I guess).

@ptufts44 when the circuit starts up it consumes lots of power as it’s constantly charging the oscillation and it hasn’t yet gotten to sufficient voltage to let the energy flow back into the supply. Once the voltage is high enough the energy consumption drops sharply since the energy spent charging the oscillations flows back into the power supply (perhaps try using a diode to protect from the reverse current if the PSU isn’t designed for reverse current and a capacitor to absorb that reverse current so as not to lose the energy). When the oscillation energy is allowed to discharge, e.g. through smoke or an arc it suddenly consumes far more power as the oscillation energy that would normally feed back into the PSU gets used up ionizing the smoke or generating the arc. As such, it continues drawing a lot of power (or even more if the oscillation voltage drops considerably) but this time without returning any of it. Also, the arc is red/orange because unlike a normal electrostatic arc that discharges very little energy and simply ionizes the air making it blue, the ZVS heats it as it dumps all its power into it and this gets it to the temperature of a flame essentially turning it into a sort of artificial flame. This is the principle behind which carbon arc lamps make light. Also, if you want a truly terrifyingly deadly flame made this way (with a different type of circuit) here’s a video by styropyro: https://youtu.be/UNisqZOAaAs

Actually I found a way to make the excess power on one side feed into the other side thereby almost entirely removing the demand on the PSU when the oscillator doesn’t have a load. I’ve modified this circuit accordingly.