Is it possible to hand calculate the phase noise of a VCO?

[iVenky]

Hi,

I am running pss/pnoise sims for a Q-VCO that I am designing. I would like to know if it's possible to hand-calculate the phase noise so that I could reduce by figuring out what exactly is causing the problem. If so, how? I know the basic stuffs that I read on RF Microelectronics by Razavi.

Thanks

 

[dick_freebird]

I would imagine so. Procedurally. As far as accuracy, I'd
be quite suspicious. Being as much PN in real circuits
has to do with sources other than the transistors' own
noise attributes. These set the floor, not the furniture.

 

[Nig]

It is a very tedious process and error prone (you can use ISF method by Hajimiri but i am afraid that it is not easy). Phase noise depends on multiple transistors and their operating point during that instant ( in case of phase noise it is around zero/CM crossing of the output). Phase noise calculation is heavily simulation dependent and if you want to get a feel for it or if you want to improve it you can do the following.

1 - Get the noise contributors for phase noise and try to improve it ( obvious solution of course )
2 - You can run phase noise sampled or PM Jitter and get the noise in Vn at a particular instant and you can use that to improve the noise.