Royer oscillator - phasing dots

Right transistor collector pulls down, right transistor base gets pushed up, the other vice-versa, looks like positive feedback to me?

What an awful schematic diagram. I think you are right, the phasing dots are wrong.

Jeroen Belleman

The base current has to come from the two (two? why?) resistors from the transformer ct. There are better ways.

This is kind of ancient.

I think the dots are correct, but bases are driven directly from the aux winding on the transformer, a low impedance source and they're swinging up to 0.6v and turn the NPN on and down below ground to...whatever! and the emitters are grounded. The max reverse voltage across the base-emitter junction for most NPN power transistors is not high.

that is to say this is an "academic circuit" don't ever build circuits from Wikipedia verbatim, jesus. Even a lot of the stuff from old magazine articles from the 70s and 80s you can find on e.g. seekic.com is generally better at least someone probably tested it.

It has to be positive feedback. The royer works with collapsing magnetic field in saturation

This page is correct:

Cheers

Klaus

Not correct. The resistors are only there for kick starting the royer. One branch will have lower Vbe, so that BJT is turned on first, positive feedback turns it on fully

The base drive winding delivers all the base current needed after startup

Please show a better solution. I am willing to bet that you cannot find a better solution.

Ancient, yes. But still very good. Royer Jensen varianter is better (efficiency, short circuit protection)

Cheers

Klaus

I have worked on Royer converters. There is nothing wrong with the circuit. The real life circuit does not have more components, nothing academic about this for once :-)

Cheers

Klaus

Not in that circuit. The base current can only come from those resistors to V+.

A better design would center-tap the base winding, and only get a little startup current from V+. Even better, use mosfets.

This is my version, back in ancient times when I made CD ignitions for sports cars and motorcycles.

R1 supplies a little startup current, and R2 provides the serious base drive.

How much was that bet?

Most designs has a diode to GND to protect reverse bias of the diodes. But, it works anyway, during turn off of the other BJT, current is drawn out of the base which helps turn on of the other, aided by the resistors as you mention

MOSFETs at high voltage are expensive for the same equivalent conduction voltage

That is copied from Billings book, figure 2.15.1 from 1989. No offense, but I am betting it is not your design. Nothing wrong with copying, getting inspired. I do that all the time, just don't pretend you dreamed it up

Some times one comes up with the same circuits. I am doing Active EMI design right now, and got a circuit I was pleased with. Then a little later I found the same on IEEE

I do not see it is simpler in any way or better (less a diode to base/GND)

Cheers

Klaus

Yes, that's what I was saying

I invented that myself, in about 1970. You are being offensive. And defensive.

If the load is shorted, as can happen in an SCR-based CD ignition, it can be designed to beta limit, stop oscillating, and not fry the transistors. But once oscillation starts back up, there's tons of base drive. It's fairly subtle.

I suppose you don't see it.

Yeah, maybe the wording was a bit harsh. Sorry for that, just wrote it as I saw it

When the Royer Jensen is shorted, it just goes to higher frequency, it auto protects. For the Royer the frequency falls at higher load.

The Royer is sensitive to temperature since it operates at the BH knee. The Jensen fixes that, and is a lot more stable design where the saturation is defined by a non-loaded xformer

Cheers

Klaus

If you just stuff in base current, from resistors to V+, the transistors fry. My circuit allows independent tuning of oscillating and not-oscillating base current.

The usual CD ignition inverter from that time didn't have the diode; it had a 2-resistor divider from V+/gnd to the base winding CT. It was halfway to being a good solution, but had temperature issues.

Surely other people invented the same diode-based circuit; it's pretty obvious.

You're right!

It's also not a Royer, the inductor on the feed makes it a Baxandall. the original post by "The Lightning Stalker" was superior to this revision.

The frequency is dependent on ferrite saturation. So primarily input voltage, temperature, spread of ferrite ur

Cheers

Klaus

That circuit image was posted once, corrected twice, and still won't work.

There is *so* much bad electronics around.

Thanks, Jason et al

Phasing dots are correct in the diagram.

RL