Stats

  • Locked Locked
  • Replies 11
  • Subscribers 65
  • Views 10017
  • Members are here 0
This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

injection locked oscillator phase noise simulation using pss+pnoise

calp

Hi all,

I am designing an injection locked vco.  My simulation scenario for the phase noise is as follows:

1. Run the vco without the injection signal for 2 microseconds ; let it setttle to the free running frequency; 

2. Start the injection signal and observe the phase noise...

 Well, I need a periodic vsine source to generate the injection signal in the schematic; and this is the very source of an error in my pss simulation:

" 'V0' is a periodic input signal, which is inconsistent with autonomous circuits. "

 this is what happens if I set cadence for phase noise simulation using the instructions given in the cadence Spectre manual.

I know... the instructions are for a 'normal' vco; but I was hoping that it would work for my injection locking case as well.

 

Question, as obvious it is by now :), how do we simulate the injection locked oscillator phase noise ?

Is there a way to conceal the periodic voltage source in the circuit so as not to piss the pss off ?

 

thanks,

calp.

 

Parents
  • Tawna

    Here is the article:

    When simulating an injection locked oscillator and looking at the startup waveform, you are looking for the point where the timing of the oscillator output is stable with reference to the injection source. Note that the amplitude must be at steady-state as well.

    When you set up your PSS analysis you:

    - Do NOT select the oscillator button.

    You are applying a signal (vsource) to your ring oscillator which causes the oscillator to oscillate at the injection frequency. Thus it is a driven circuit - as it is driven by the injection source. This is why you don't push the oscillator button on the PSS choosing analysis form. SpectreRF considers this a driven circuit and will error out if you select the oscillator button on the PSS choosing analyses form.

    - Set tstab to 100 periods.

    You want the circuit to be stabilized, i.e. you need to get it close enough to the final oscillation frequency so that PSS can converge. This will depend on the time constant of the circuit. A high Q oscillator will take a long time to settle. A ring oscillator is lower Q by definition, so it should not be as problematic.

    - Check the saveinit button.

    The waveforms for the initial transient before steady state are saved and you can view them with the results browser.

    - You may even want to try to run the circuit using transient analysis.

    Look at the oscillator output compared to the injection source. Look for variations in the timing of the oscillator output compared to the locking source. Hopefully, the circuit will settle fairly quickly.

    Now, once you can get PSS to successfully converge, you can look at things like pnoise jitter

    Best regards,

    Tawna

Reply
  • Tawna

    Here is the article:

    When simulating an injection locked oscillator and looking at the startup waveform, you are looking for the point where the timing of the oscillator output is stable with reference to the injection source. Note that the amplitude must be at steady-state as well.

    When you set up your PSS analysis you:

    - Do NOT select the oscillator button.

    You are applying a signal (vsource) to your ring oscillator which causes the oscillator to oscillate at the injection frequency. Thus it is a driven circuit - as it is driven by the injection source. This is why you don't push the oscillator button on the PSS choosing analysis form. SpectreRF considers this a driven circuit and will error out if you select the oscillator button on the PSS choosing analyses form.

    - Set tstab to 100 periods.

    You want the circuit to be stabilized, i.e. you need to get it close enough to the final oscillation frequency so that PSS can converge. This will depend on the time constant of the circuit. A high Q oscillator will take a long time to settle. A ring oscillator is lower Q by definition, so it should not be as problematic.

    - Check the saveinit button.

    The waveforms for the initial transient before steady state are saved and you can view them with the results browser.

    - You may even want to try to run the circuit using transient analysis.

    Look at the oscillator output compared to the injection source. Look for variations in the timing of the oscillator output compared to the locking source. Hopefully, the circuit will settle fairly quickly.

    Now, once you can get PSS to successfully converge, you can look at things like pnoise jitter

    Best regards,

    Tawna

Children
No Data

Community Guidelines

The Cadence Design Communities support Cadence users and technologists interacting to exchange ideas, news, technical information, and best practices to solve problems and get the most from Cadence technology. The community is open to everyone, and to provide the most value, we require participants to follow our Community Guidelines that facilitate a quality exchange of ideas and information. By accessing, contributing, using or downloading any materials from the site, you agree to be bound by the full Community Guidelines.