self_osc

[Fob]

Hi,
I am a little out of brain with this simple buzzing supply. I know that F=Ui/(4wiBsatAe) with saturable cores, but why this start oscillation on my simulator since I have only inductance as parameter? Is there any relation between L in coupled inductors and physicals wiBsatAe? In simulation only base currents affect the frequency.
I am currently working on 50-250mA constant current supply with minUo off 24V and maxUo up to 72V. I think to use self-oscillating with post-regulated current source with MC34063. But if any have idea how to do this with simple flyback I will be very grateful.

[megajocke]

When collector current reaches high enough to desaturate the transistor currently conducting di/dt becomes zero as the transistor will work like a current source and inductor voltage collapses. This cuts off the transistor currently conducting which makes the voltage switch polarity and turn on the other transistor.

It's really this mechanism that is responsible for the oscillation. If the current at which the transistors turn off is set above the saturation current the saturation stabilizes the operating frequency though.

[Fob]

Hi megajocke,
You are right but for the blocking (flyback) generator. This is pure forward and I don’t suppose to have any di/dt except on magnetizing inductance that is Lm = w1^2/R, where reluctance is pure physical – R = le/u*Ac. And the simulation is with coupled inductors with inductance and some relation coefficient of %D 0.999.

[Fob]

I’m sorry for wasting your time, It seems that I found what I’m looking for ( U2 ) . But this bring the next question: how to simulate current fed resonant push pull? I will make some CFLamp converter from 24V to 36W lamp for exercise. It will be suitable for 12V operating with boost and 48V operating with buck preregulator…. I think.