Partnered Output Coils - Free Energy

[TinselKoala]

Yes, I've been wanting to point out the effect of leakage inductance and imperfect coupling but I keep forgetting to do it. MileHigh makes me think of it when he has spoken about perfect cancellation with the bucking setup. Even with perfectly matched coils I don't think you'll get perfect cancellation because of leakage and imperfect coupling.

The flyback cores we are using also have a little plastic spacer between the halves inside the coil bobbin part, perhaps half a millimeter gap in there. Mine originally had two stacked spacers and when I removed one of them the performance changed quite a bit.

[TinselKoala]

TK,

It's a MUR460.

GL.

Yes, this diagram I could read. I was talking about the first diagram, where it looks like MUR8320, and the closest I could find was MURS320, a surface mount ultrafast rectifier 25 ns recovery time.

[TinselKoala]

And here is the EMJ circuit, not quite as high voltage spike. The Zener in the emitter is to simulate the high C-E saturation voltage, since I don't have a model for the 2SD1555.

This is very rough, with proper models, a much better simulation can be had.

Have you got a model for 2sd2539?

[Vortex1]

Yes, I've been wanting to point out the effect of leakage inductance and imperfect coupling but I keep forgetting to do it. MileHigh makes me think of it when he has spoken about perfect cancellation with the bucking setup. Even with perfectly matched coils I don't think you'll get perfect cancellation because of leakage and imperfect coupling.

The flyback cores we are using also have a little plastic spacer between the halves inside the coil bobbin part, perhaps half a millimeter gap in there. Mine originally had two stacked spacers and when I removed one of them the performance changed quite a bit.

Yes, varying that tiny gap will have a very large effect on inductance and on the ability of the core to "reset" the remnant flux.

EMJ will never get anything near perfect cancellation spread out the way the partnered coils are unless the partnered windings are co-axial. But then he wouldn't get that big spike!.

Have you got a model for 2sd2539?

In my old library there are no deflection transistors.

[TinselKoala]

@Vortex1: I don't think he's getting such a big spike anyway, since he's underdriving the transistor. Hard to tell from his scopeshot though. In my testbed when I use similar drive levels to his (from comparing scope trace appearance) the main spike only goes to 500 volts or so, when the 10R (actually 11R for me) load is connected as in the EMJ-Meyer diagram.

In my old library there are no deflection transistors.

And in the new one, maybe not too! Who uses HD transistors any more, since CRTs are fossils like dinosaurs from another century!