simulation PSS

Discussion in 'Cadence' started by kallel.firas, Oct 9, 2007.

  1. kallel.firas

    kallel.firas Guest

    Salut,

    Est ce que la simulation PSS permet de réaliser une analyse
    transitoire?

    Si oui, comment réliser cette simulation PSS permettant l'analyse
    transitoire d'un oscillateur ring oscillant à 480Mhz (il a 16 noeuds
    oscillants).

    merci d'avance,

    Firas.
     
    kallel.firas, Oct 9, 2007
    #1
  2. Well, given that this news group is predominantly conversed in English, you'll
    stand a much better chance of getting an answer if you write in English. My
    French is certainly very rusty, hence me replying in English (and I may have
    misunderstood you). There seem to have been a sudden trend in posts in French
    over the last few days!

    Simulating a ring oscillator with shooting PSS should be fairly straightforward.
    On the PSS form, select the "oscillator" checkbox, and give it the nodes at the
    output of the oscillator - (you can leave the negative node blank if it is
    gnd!).

    Then give your approximate frequency in the PSS fundamental field.

    Finally give a tstab time which is long enough for the oscillator to be broadly
    up and running (it doesn't have to be settled).

    First of all, make sure you can simulate it in transient analysis though. If you
    can't simulate it in transient, you're not going to be able to in PSS. PSS is
    primarily going to be useful if you then want to analyse something using a small
    signal analysis afterwards (e.g. pnoise).

    Regards,

    Andrew.
     
    Andrew Beckett, Oct 9, 2007
    #2
  3. kallel.firas

    kallel.firas Guest

    thanks for your help,
    after simulation, I found the following error:

    "Error found by spectre at time = 144.222 ns during periodic steady
    state
    analysis `pss'.
    After the dynamic adjustment of tstab, PSS still failed.

    Analysis `pss' terminated prematurely due to error."

    so please I want to know the origine of this error and its solution if
    it is possible.
    thank you.

    Firas.
     
    kallel.firas, Oct 10, 2007
    #3
  4. Without seeing the full log, it's a little hard to say. Normally PSS adjusts the
    stabilisation period (which is essentially a straight transient) to try to reach
    convergence in difficult cases - but probably it's not detected the period.

    You should turn on the "save initial transient" and look at the transient
    waveforms (you can see the pss_tran results in the results browser) to see what
    happens. Is it oscillating?

    Regards,

    Andrew.
     
    Andrew Beckett, Nov 15, 2007
    #4
Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.