The Designer's Guide Community
Forum
Welcome, Guest. Please Login or Register. Please follow the Forum guidelines.
Mar 29th, 2024, 12:57am
Pages: 1
Send Topic Print
The mechanism behind the estimated oscillating frequency/line width (Read 1709 times)
BinaryMonkey
New Member
*
Offline



Posts: 2
Sz
The mechanism behind the estimated oscillating frequency/line width
Sep 26th, 2017, 5:53am
 
I am using spectre 14.1.0.138, MMSIM14.1.0,Cadence IC6.1.6-64b

I am curious about how did Spectre implement the "estimated oscillating frequency" and "estimated linewidth". Is it using FFT? Or is it using auto-correlation function?  How do I know about the metrics for confidence of this estimation? Can I assume that I reach steady state if seeing these things in log? Is this information  enough to for us to infer a more reasonable Tstab to save time in simulation?

In my output log , it says
"Estimated line width of the oscillator is 699.02 mHz."
"Estimated oscillating frequency from Tstab Tran is  = 26.0096 MHz"
"Tstab: runs at least 100 timesteps per cycle,   MaxStep=3.84474e-10"

PS: I will be very grateful if the answers also come with a reference to papers/books/app notes explaining the mechanism rigorously.


Back to top
 
 
View Profile   IP Logged
Andrew Beckett
Senior Fellow
******
Offline

Life, don't talk to
me about Life...

Posts: 1742
Bracknell, UK
Re: The mechanism behind the estimated oscillating frequency/line width
Reply #1 - Dec 21st, 2017, 6:27am
 
There are various ways that the oscillation frequency is estimated. The first is done during DC if the "Calculate initial conditions (ic) automatically" is enabled (corresponds to the option to hb/pss called oscic=lin). This uses some kind of small signal analysis to estimate the frequency and conditions (I'm not sure there's any more detail than that needed).

Then the estimate from the Tstab is based on measuring the difference between the edges of the last period in the transient. The message about MaxStep is telling you how the maximum timestep was controlled.

The line width is related to determining the frequency offset in the pnoise/hbnoise calculation below which the linear assumption breaks down. Again, I don't think there's any  documentation on how it is computed (I'm not sure it tells you anything about how well the circuit has converged as it's related to the noise computation).

It's often a good idea to use the options for "Detect Steady State" and "auto" tstab time to ensure that the tstab runs no longer than it has to. You may be running too old an IC version though to see that (you didn't mention a subversion of IC616 that you're using though).

Regards,

Andrew.
Back to top
 
 
View Profile WWW   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.