Home > Community > Forums > Custom IC Design > spectre stb analysis to show poles and zeros

Email

* Required Fields

Recipients email * (separate multiple addresses with commas)

Your name *

Your email *

Message *

Contact Us

* Required Fields
First Name *

Last Name *

Email *

Company / Institution *

Comments: *

 spectre stb analysis to show poles and zeros 

Last post Tue, May 27 2014 10:13 AM by potsticker. 7 replies.
Started by apple419 30 Jan 2014 10:24 AM. Topic has 7 replies and 5121 views
Page 1 of 1 (8 items)
Sort Posts:
  • Thu, Jan 30 2014 10:24 AM

    • apple419
    • Top 500 Contributor
    • Joined on Fri, Apr 19 2013
    • Norristown, PA
    • Posts 35
    • Points 550
    spectre stb analysis to show poles and zeros Reply

    My system: IC6.1.5

      In stb analysis, based on the bode plots I am able to approximately see where the poles and zeros are. But sometimes I would like to know the exact locations of poles and zeros. Is there a way to print out all the poles and zeros?

      I know pz analysis can show poles and zeros, but in a feedback system I need to break the loop so that what pz analysis shows are the real open loop gain poles and zeros. So how to break the loop in pz analysis? I know stb has this built in...

     

    thanks for helping. 

    • Post Points: 35
  • Thu, Jan 30 2014 2:39 PM

    Re: spectre stb analysis to show poles and zeros Reply

    They're not really the same thing. The pz analysis is computing the poles and zeros of the entire circuit, not just a specific loop gain. The stb analysis computes the loop gain (it does not "break" the loop) - they are different algorithms that work in different ways - so there's not really a way to do this (right now).

    There have been some requests to implement this, but no plans to do so right now (not sure about the feasibility).

    Regards,

    Andrew.

    • Post Points: 20
  • Fri, Jan 31 2014 1:13 AM

    • Frank Wiedmann
    • Top 75 Contributor
    • Joined on Wed, Jul 16 2008
    • Muenchen, Bavaria
    • Posts 96
    • Points 1,560
    Re: spectre stb analysis to show poles and zeros Reply

    You probably want to fit poles and zeros to the loop gain computed by the stb analysis. One way of doing this (I have not tried it) might be to export the complex loop gain data and use something like zpkfit (see http://www.mathworks.com/matlabcentral/fileexchange/27821-zpkfit).

    • Post Points: 20
  • Fri, Jan 31 2014 8:14 AM

    • apple419
    • Top 500 Contributor
    • Joined on Fri, Apr 19 2013
    • Norristown, PA
    • Posts 35
    • Points 550
    Re: spectre stb analysis to show poles and zeros Reply

    Thanks Andrew, thanks Frank,

       Yes stb analysis calculates the open loop response without really "breaking" the loop. pz analysis just calculates poles and zeros from A to B based on your input. So in my understanding pz analysis is not really useful for stability analysis, because for stability we mostly care open loop response...

        I will try zpkfit in matlab.

    Regards,

    Apple 

    • Post Points: 20
  • Fri, Jan 31 2014 9:01 AM

    • Frank Wiedmann
    • Top 75 Contributor
    • Joined on Wed, Jul 16 2008
    • Muenchen, Bavaria
    • Posts 96
    • Points 1,560
    Re: spectre stb analysis to show poles and zeros Reply

    Your circuit is unstable if the pz analysis gives you any poles in the right half plane (with a positive real part).

    • Post Points: 5
  • Fri, May 16 2014 11:01 AM

    Re: spectre stb analysis to show poles and zeros Reply

     hi Andrew,

     does cadence calculator now has a pole/zero fitting function?

    so that it can form function like zpkfit? thanks.

     Frank

    • Post Points: 20
  • Sat, May 24 2014 12:32 AM

    Re: spectre stb analysis to show poles and zeros Reply
    No.
    • Post Points: 20
  • Tue, May 27 2014 10:13 AM

    Re: spectre stb analysis to show poles and zeros Reply

     thanks.

    • Post Points: 5
Page 1 of 1 (8 items)
Sort Posts:
Started by apple419 at 30 Jan 2014 10:24 AM. Topic has 7 replies.