Hi,

I'm new in this forum, so please excuse me if this issue has already been posted. I'm simulating an LC oscillator running at 3.4GHz with a tank Q=10. What I observed is that by making extremely small changes in the circuitry, e.g. varying the supply by 5%, the PN varies by 30dBc/Hz, but not in a monothonic way, so for example it will be best at 2.5V, worst at 2.4V and in between at 2.3V. Similar things happen by varying other parameters. Other simulations are OK, e.g. center frequency, amplitude, etc. It's as if PN simulations would have some kind of "sweet spot". I wonder if anybody observed this behaviour. Thanks.

Daniel