Stats

  • Locked Locked
  • Replies 3
  • Subscribers 65
  • Views 16098
  • 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

Best practices for Verilog-a modelling time variant resistance

LDIL

Hello all.

I am simulating a design with time variant resistor, which values can be in [oHM] from thousands to even millions. I now set it as a simple real.

What is the best practice for assigning values/ setups for it, in order to have no convergence issues or runtime issues?

I mean, usually the voltage is no more than 5V, and the currents are less than 1e-2.

Thanks.

Parents

  • Dear LDIL,

    I'm not sure how many values your time variant resistor has - nor the physical mechanism you are trying to model (other than aging perhaps?). If the number of time breakpoints is not too many to code, you could model it however as a piecewise linear source if you combine your piecewise linear source with a controlled source. This would avoid the use of  a verilogA  model if you are concerned about issues relating to convergence with a verilogA model. I can't state this is the "best practice", but it might be something to consider.

    Shawn

  • in reply to ShawnLogan

    Without more information it's really hard to answer this, as Shawn has pointed out. If the resistance has a voltage dependency, you need to ensure that the voltage dependency is stable and also you are modelling reasonably across a very wide range of voltages (not just the voltages you typically see in the circuit). I've seen issues where people (including large foundries) have modelled nonlinear resistors using a quadratic equation and have forgotten that there are two roots - one was physical and the other was non-physical - it ended up where during the route to convergence a large voltage appeared across the resistor (which wouldn't happen in real life), which caused the resistor to become negative, which caused it to get stuck in this non-physical region.

    The other thing you normally want to take care about is ensuring that the resistor value changes smoothly rather than discontinuously.

    Regards,

    Andrew.

Reply
  • in reply to ShawnLogan

    Without more information it's really hard to answer this, as Shawn has pointed out. If the resistance has a voltage dependency, you need to ensure that the voltage dependency is stable and also you are modelling reasonably across a very wide range of voltages (not just the voltages you typically see in the circuit). I've seen issues where people (including large foundries) have modelled nonlinear resistors using a quadratic equation and have forgotten that there are two roots - one was physical and the other was non-physical - it ended up where during the route to convergence a large voltage appeared across the resistor (which wouldn't happen in real life), which caused the resistor to become negative, which caused it to get stuck in this non-physical region.

    The other thing you normally want to take care about is ensuring that the resistor value changes smoothly rather than discontinuously.

    Regards,

    Andrew.

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.