The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Simulators >> Circuit Simulators >> pss convergent issue in osc phase model.
https://designers-guide.org/forum/YaBB.pl?num=1281262419

Message started by casual on Aug 8^th, 2010, 3:13am

Title: pss convergent issue in osc phase model.
Post by casual on Aug 8^th, 2010, 3:13am

I simulated the osc phase model but fail to run pss & pnoise.

"Analysis `pss' terminated prematurely due to error."
It is due to convergent issue in pss run.

The input node is connected a voltage source vdc or idc with a ideal res.

Code:

`include "disciplines.vams"
discipline phase
potential Angle;
enddiscipline

`include "constants.vams"
`include "disciplines.vams"

module phasemodel_osc(in, out);

input in;
output out;
phase out;
voltage in;

parameter real gain = 1 from (0:inf); //Transfer gain, Kvco (Hz/V)
parameter real n=0 from [0:inf); //white output phase noise at 1Hz (rad^2/Hz)
parameter real fc=0 from [0:inf); //flicker noise conrer freq (Hz)

analog begin
Theta(out) <+2*`M_PI*gain*idt(V(in));
Theta(out) <+ flicker_noise(n, 2, "wpn") + flicker_noise(n*fc, 3, "fpn");
end

endmodule

Title: Re: pss convergent issue in osc phase model.
Post by casual on Aug 8^th, 2010, 6:06am

I am a novice on RF analysis.

I want to study the phase noise of each components in the whole PLL. I know CPPsim can make it easily.

I just test out the osc alone. Since the input is a voltage, i set a voltage in the input. The reason to use pss is to observe the pnoise, which require pss run.

Title: Re: pss convergent issue in osc phase model.
Post by pancho_hideboo on Aug 8^th, 2010, 6:43am

supermoment wrote on Aug 8^th, 2010, 6:06am:

The reason to use pss is to observe the pnoise, which require pss run.

You can't understand concept of "Phase Domain Model" at all.

supermoment wrote on Aug 8^th, 2010, 3:13am:

The input node is connected a voltage source vdc or idc with a ideal res.

Show me your netlist.

In your netlist, what is large signal ?
"Phase Domain Model" is not autonomous circuit.

Again see http://www.designers-guide.org/Forum/YaBB.pl?num=1280215169/7#7
Also see http://edocs.soco.agilent.com/display/ads2009/Phase+Noise+Simulations+using+Small+Signal+Model+Loop+Components

Here conventional small signal AC(Noise) Analyses are used.

Title: Re: pss convergent issue in osc phase model.
Post by Ken Kundert on Aug 8^th, 2010, 10:24am

You are simulating a phase-domain model of the oscillator alone? Not in a feedback loop? And you are doing it with PSS? That won't work. The phase-domain model of an oscillator is an integrator, which does not have a periodic solution.

You don't need PSS when you are simulating a phase-domain model, a simple noise analysis will do because the signals themselves are formulated in terms of phase, any noise you see on a phase signal with naturally be phase noise.

-Ken

Title: Re: pss convergent issue in osc phase model.
Post by casual on Aug 8^th, 2010, 10:56pm

Thank you.

I eventually have made it by putting everything together. It could be analyzed in noise analysis and pnoise analysis (same result as in noise analysis)
The problem previously i got is that I did not have phase noise source :p

Again, thanks for your patience.

the next thing is to extract transistor level behavior from PSS & Pnoise run.