Hello everyone, I’m building my first microwave oscillator and I’m trying to enforce the reflection-coefficient condition—namely, that the product of the two reflection coefficients has a real part greater than 1 and a zero imaginary part. To achieve this, I generated a negative resistance looking into the transistor base. Since the impedance seen at that terminal turned out to be inductive, I closed the loop with a capacitive resonator.


By examining the S₁₁ reflection coefficient at port 1, I observed that |Γ| > 1 and that the trajectory crosses the real axis counterclockwise, thus satisfying the Barkhausen startup condition. Now I’d like to test for oscillation using OSCPROBE, but I can’t get the algorithm to converge. I understand that if OSCPROBE converges it confirms oscillation, but non-convergence doesn’t necessarily imply there’s no oscillation. Since this is my first time, it would be reassuring to have the simulator confirm it.


So my questions are:

1. Does it matter on which node I place OSCPROBE, given that if the circuit is oscillating, every node will see oscillation?
2. Why does the simulator report “at sweep point 1 (FREQ=1 GHz)” and how can I force the analysis to sweep at 5 GHz?

 

Welcome to AAC.

This appears to be coursework for academic credit, is that the case?