The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Simulators >> RF Simulators >> Consistency of QPSS analysis
https://designers-guide.org/forum/YaBB.pl?num=1097167035

Message started by Rehan Zakai on Oct 7^th, 2004, 9:37am

Title: Consistency of QPSS analysis
Post by Rehan Zakai on Oct 7^th, 2004, 9:37am

Does anyone know why a swept qpss analysis will behave differently than individual qpss simulations performed with the same design variable manually set?

When simulating to estimate IIP3/IIP2, I am sweeping the RF input power variable within the QPSS analysis, I am getting numerous warnings (imax exceeded etc.) and finally zero jacobian error. When I run QPSS simulations individually with RF input power variable manually set to same values simulation runs just fine.

It seems to me that SpectreRF treats the subsequent simulations in a swept QPSS analysis differently than a stand alone simulation. Is there some time saving algorithm kicking in?

Thanks in advance for your comments
Rehan Zakai

Title: Re: Consistency of QPSS analysis
Post by Eugene on Oct 7^th, 2004, 10:36am

I am not 100% sure, but as I recall, SpectreRF uses the solution from the last sweep value and as initial conditions for the next sweep value. The idea was that if you had trouble getting convergence at some parameter value, you could "sneak up" on it using a parameter sweep. It seems odd that you see the convergence warnings during the sweep and not on the individual run.