Small-signal analysis of transformer BJT oscillator

Question

In an attempt to improve my circuit analysis skills, I've been trying to understand how this photo flash charger circuit works.

I am mainly focusing on the oscillator part on the left, which I've redrawn as follows

^{simulate this circuit – Schematic created using CircuitLab}

To analyze the circuit's stability, I "broke the loop" at the base of the BJT to derive an expression for the loop gain. Unfortunately, my expression for the loop gain does not at all mirror the simulated loop gain obtained by simulating the circuit in LTSpice. From what I can tell, there are three possible sources of error:

1. My small signal equivalent circuit is wrong
2. Due to the oscillator's large signal swing, the linearized small signal model is not representative of the actual circuit behavior
3. I did the algebra wrong.

This is the small signal equivalent circuit I used for the analysis. Does anyone see anything wrong with the way I drew it?

^{simulate this circuit}

Answer 1

The redrawn oscillator does not work. The polarity of one of the coils is reversed compared to the original. The operation of the oscillator is significantly affected by the capacitance of the secondary winding with a large number of turns. The diode load does not change the essence, it would only greatly prolong the simulation time. Charging the large capacitor to 300V is a long time. The circuit is strongly nonlinear, the linear model cannot be used here. The transformer is the standard flyback, with approx. 10uH primary inductance. This determines the (no load) frequency of the oscillator.