The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Simulators >> RF Simulators >> PSTB and PNOISE analysis in spectre
https://designers-guide.org/forum/YaBB.pl?num=1328543268

Message started by Sumit Kumar on Feb 6^th, 2012, 7:47am

Title: PSTB and PNOISE analysis in spectre
Post by Sumit Kumar on Feb 6^th, 2012, 7:47am

Hello all
PSTB simulation in cadence gives loop gain. This loop gain can be used to calculate noise transfer function in loop. Can we rely on PSTB analysis results to calculate noise transfer function? I am having some problem mainly when my circuit consists of analog lib switches and ideal current source. TSTB results are coming as expected and hence there is no problem with periodic steady state operating points (I guess ).

Title: Re: PSTB and PNOISE analysis in spectre
Post by Frank Wiedmann on Feb 8^th, 2012, 1:07am

You could use the separatenoise=yes option of the pnoise analysis, which separates noise into sources and transfer functions. Transfer functions can also be simulated with the pxf and pac analyses.

Title: Re: PSTB and PNOISE analysis in spectre
Post by sandman on Feb 8^th, 2012, 11:06am

In addition to what Frank suggests, the noise separation option in pnoise allows one to also identify noise internal to the device itself (if it is supported by the design kit). For example, the flicker and thermal noise contributions from a transistor can be individually identified at the output as a transfer function versus any swept value.

As an afterthought, if you are using switches (which are inherently non-linear devices), you could consider using qpss which might significantly reduce simulation time, if you haven't done so already.

cheers,