Hi,
I'm trying to run a phase noise simulation on a non-50% duty cycle clock and I'm trying to figure out what tool would be the right one to use. I tried using pnoise sources in SpectreRF to do these simulations but it seems like the phase noise result from pnoise sources varies with duty cycle. My understanding is that the phase noise should be fairly independent of the duty cycle. So, this result does not make sense. May be this makes sense because pnoise sources does time averaging.

I then tried using "modulated" option in pnoise. The "modulated" option gives the AM and PM noise components and, it seems like, the AM noise portion changes with the duty cycle. The PM portion of the noise stays the same though. The part that's puzzling is that, on a non-50% duty cycle clock, I can see AM noise even after an ideal limiter. I'm not sure why there would be AM noise when the highs and the lows of the clock are perfectly ideal (after a limiter). So, there's something I'm missing and wanted to see if anyone could help me understand this better.

Thanks,
Nath.