High Frequency DC High Voltage Arc Ignition Generator Inverter Boost Step-Up ??

[cpuxxx]

i didnt understand flyback - feedback relationship here clearly. Normally a driver switches here transistor's gate but in here what switches transistor's gate ? Gate is closing and opening so fast as polarized by coil or something like that ? What switches here transistor's gate ?

[strawberry]

I have this plug and play ZVS flyback multifunktion driver up to 240W rated output (ATX compatible): eBay auction: #303152318527 and flyback transformer as well eBay auction: #302204237459

Your circuit is similar to '' joule thief circuit ''

[floobydust]

Here is a schematic of the "15kV generator" kit with more details.
I think it's better mostly a saturated-core oscillator, as I understand how it operates:
Initially, the transistor gets base-current through the feedback winding and turns on.
Flux in the core increases and the feedback winding generates (in-phase) EMF to assist and add base-current.
Current ramps up until the flux no longer can build up- either the core saturates OR the transistor plus winding resistance plus battery reach their maximum current. The flux density is no longer increasing.
Then the feedback winding generates no EMF, and the transistor turns off slightly.
Any drop in flux density causes the feedback winding to generate opposition (counter EMF) to base-current which turns off the transistor, until the flux falls to nothing and the cycle repeats.

I did not measure the core's saturation point yet on these cores, I did not see an air gap.
I find these kits have crappy counterfeit transistors which tend to run pretty hot.

[David Hess]

That configuration is called a blocking oscillator and it is essentially one half of a Royer oscillator which operates in the same way but with two transistors in a push-pull configuration.

Drive to the base of the transistor is controlled by the *change* in flux in the core of the inductor.  When the core saturates and the flux stops increasing, the base drive is removed and the transistor turns off.  The decreasing flux in the core then pulls the base negative holding the transistor off until the flux through the core drops to zero and stops decreasing.

Tektronix made great use of blocking and Royer oscillators for the high voltage inverters in their oscilloscopes through to the 1980s.

[floobydust]

I'm not sure what the name of this type of oscillator is because some use an RC delay and not core saturation. So I use different names. Or maybe it doesn't matter.
People seemed to come up with an oscillator, patent it and use their name for it:
1937 blocking oscillator patent is about the tuning capacitance.
1951 transistor blocking oscillator patent
1957 Triggered transistorized blocking oscillator with saturable transformer

Royer has an extra inductor and I only know of it being push-pull, near sinusoidal not flyback.

[David Hess]

Royer has an extra inductor and I only know of it being push-pull, near sinusoidal not flyback.

Royer's design only relied on saturation and produced a square wave output; he used no resonant circuit.  It is the common push-pull saturable transformer oscillator while the "Joule Thief" is the singled ended variation which works the same way and used to be known as a blocking oscillator.  Both use a single transformer core.

The resonate extension of Royer's circuit has various names and includes a capacitor and may rely on constant current drive which may be provided by an inductor but other variations are possible.

[floobydust]

Ah got it - two different types of Royer oscillators (converters).

"The sine-wave oscillators in [Jim] Williams' CCFL circuits use feedback from the tank voltage to switch the transistors, and they do not saturate the transformer core. This type of sine-wave oscillator is more properly called a "current-fed push-pull parallel-resonant inverter."

Jim Williams: "Royer's circuit is not an LC resonant type. The transformer is the sole energy storage element and the output is a square wave."

From letter exchange "A Royer by any other name" between Bryce Hesterman and Jim Williams. Published on November 21, 1996

[David Hess]

I was going to cite that if necessary.  The history of development after Royer is muddled but it is clear that Royer's design relied only on saturated switching and not resonance.