Stats

  • Locked Locked
  • Replies 2
  • Subscribers 126
  • Views 14513
  • 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

Parametric analysis (in ADE) differs from single analysis, at equal parameter

Clidre

Hello,

I'm facing a problem with spectre. I made a system with ideal components: it's an ideal current source that is directed to a capacitor for some time and then redirected to a low impedance node for another time slot. I use ideal switches to rearrange the circuit. The capacitor is precharged with a voltage, depending on a digital code. In my system there are ideal components from analogLib and digital ports in verilogA. A dc analysis doesn't converge. I'm interested in the overall functioning over time and I managed to simulate it with a tran analysis + skipdc. 

I did a parametric analysis, changing the digital code and I noticed that, for a certain code, the waveforms differ from what I expected. By running the same simulation just once (I set the problematic code in the Design Variables list in ADE), the waveforms are sensibly different from what I get in the parametric analysis at the same code.

I cannot understand why, any suggestions?

Thanks a lot!

  • Clidre

    By initializing the verilogA flip-flop output with @(initial_state)  begin x=0; end  , I managed to do the dc simulation and a transient analysis with skipdc=no. Now both the results seem to be consistent: for each parameter, if I run a parametric analysis or a single simulation, I get the same result.

    Any insight would be helpful. Thanks!

  • Andrew Beckett
    This is hardly surprising. If you skip the dc, the initial starting point may be uncertain and unrealistic and the transient is attempting to recover from that. When you do a parametric sweep, you are typically starting from the last solution and that may be different from if you start from scratch. Any circuit with multiple operating points (such as a flip-flop) can be sensitive to such uncertainty - the simulator doesn't know which operating point you want, and can find others - just depends on which the numerical algorithms hit first.

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.