SpectreRF can not distinguish phase noise and amplitude noise. Therefore the simulated VCO phase noise is actually phase noise and amplitude noise. However, the help from cadence says that, since VCO has amplitude limit mechanism, close to carrier noise is mostly phase noise.

The question is, for voltage biased VCO, I believe its true. But for current biased VCO, there is no such obvious amplitude limitation. Then, can we still say that most of the noise is phase noise? If not, how can we fairly compare the noise performance of two kinds of VCOs?

The same questions for the VCO buffers and frequency dividers, apprently the simulated phase noise is actually again AM and PM noises. The more the wave form  saturated, the more phase noise than am noise. But there is not way to tell how much of them is phase noise.

Please help me if you have some hints, thanks a lot.

Thanks Ken!

Yes, SpectreRF can distiguish amplitude noise and phase noise, what I am trying to find out is how can they be reported seperately. I just could not find anything about this from cadence documents.

Could you please let me know how can SpectreRF report the AM noise and PN noise seprately, thanks!

The reason I try to find out the exact amount of phase noise and amplitude noise, is that I think different VCOs could have different AM noise suppression mechanisms or different level of suppression, so it will be interesting to know them seperately not combined.