The Designer's Guide Community
Forum
Welcome, Guest. Please Login or Register. Please follow the Forum guidelines.
Jul 16th, 2024, 8:26am
Pages: 1
Send Topic Print
Transient simulation of PLL in Phase-Domain possible? (Read 3463 times)
ussmueller
New Member
*
Offline



Posts: 3

Transient simulation of PLL in Phase-Domain possible?
May 14th, 2010, 2:47am
 
Dear all,

I have written phase domain modells for my PLL according to the "Predicting the Phase Noise and Jitter of PLL-Based Frequency Synthesizers" paper by Ken. I have created a veriloga and symbol cell view for each of the modells. I have then assembled them together in a schematic and simulated it with Spectre. I can simulate the ac behaviour and the phase noise of the PLL.

I was wondering whether a transient simulation of the lock-in behaviour of the PLL is possible. I guess I need to change the reference oscillator with another source. Can anyone please clarify if the transient simulation is possible and if yes, how it is possible.

Thanks,
Thomas
Back to top
 
 
View Profile   IP Logged
Marq Kole
Senior Member
****
Offline

Hmmm. That's
weird...

Posts: 122
Eindhoven, The Netherlands
Re: Transient simulation of PLL in Phase-Domain possible?
Reply #1 - May 14th, 2010, 4:47am
 
Hi Thomas,

The general approach of phase-domain modeling can be applied, but not directly using the models from the paper. If you read through it there are a number of assumptions and modeling choices that make it operate correctly only in lock. However, if you rewrite the VCO, divider and PFD/CP models such that they take both phase and (angular) frequency as signals you should be able to get a phase-domain model that shows locking behavior at very high simulation speed.

Cheers,
Marq
Back to top
 
 
View Profile   IP Logged
Eugene
Senior Member
****
Offline



Posts: 262

Re: Transient simulation of PLL in Phase-Domain possible?
Reply #2 - Sep 10th, 2010, 7:42pm
 
There is a way to write phase domain models of the PFD and VCO that let you quickly simulate acquisition transients. The Cadence SpectreRF user's guide has a section called "Introduction to the pllLib" that explains the model pretty well. You can also find a few charts on it in the tutorial given at the 2001 BMAS (see link below). The trick is to move the VCO integrator into the PFD and then make that integrator resettable. I've had pretty good success with this method. I've also used it with time domain noise sources and CP nonlinearities to quantify sigma delta noise folding.

http://www.bmas-conf.org/2001/index.html
Back to top
 
 
View Profile   IP Logged
Pages: 1
Send Topic Print
Copyright 2002-2024 Designer’s Guide Consulting, Inc. Designer’s Guide® is a registered trademark of Designer’s Guide Consulting, Inc. All rights reserved. Send comments or questions to editor@designers-guide.org. Consider submitting a paper or model.