Stats

  • Locked Locked
  • Replies 3
  • Subscribers 125
  • Views 15105
  • 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

Sweeping different design variables at the same time within an (qpss) analysis

Emiel

Hi,

I'm using virtuoso ICADVM18.1-64b.83 and spectre 18.1.0.077.

I have a port (analogLib) generating two tones, they have a power of P1 and P2 as design variables. I made those variables global variables in Meastro. I assigned a value to P1 and I set P2 equal to P1. I have a qpss analysis in which variable P1 is swept. In the simulation, only P1 is swept and P2 is constant, although I set P2 equal to P1 in the global variables. The global variables only seem to be evaluated at the start of the simulation. I want to sweep them simultaneously. I first thought of adding a second sweep for P2 in the qpss analysis, but I assume that that will be treated as an individual sweep and not that they (P1 and P2) are swept at the same time.

I prefer not to set the power parameter of the second tone equal to P1, because I would like to run another simulation in which these powers are not equal.

Creating two testbenches, one with a port with P1 for both power parameters and one testbench with a port with P1 and P2 for either power parameter is also possible. But this is not very neat, because I than have to apply a change in the testbench twice.

What would be a neat way to achieve this?

Regards,

Emiel

edit: another, also not very neat solution is to add a second port (with 0R) in series with the existing port: one port with for instance variable 'P' for both tones for test1 and the other port with P1 and P2 for the tones for test2. And during test1 use P and set P1 and P2 to a very low value and for test2 visa versa.

edit2: there is some odd bug, when entering a power of for instance (<) -10000dBm in the port it in fact outputs a 10dBm tone....

Parents
  • Andrew Beckett

    Are you using ADE Explorer to do this by any chance? If I use ADE L, it works fine, but there's a current issue in ADE Explorer where the netlist ends up with P2 being set to the initial value of P1 rather than the parameters line in the netlist showing P2=P1.

    The workaround is to define your variable as P2=(P1) i.e. with parentheses around the variable name. There's a CCR to get this fixed, but it isn't fixed yet.

    Of course, this is guessing which ADE you're using - it's always useful to know that kind of information.

    I don't see the issue you're seeing with the second edit - however, it doesn't make sense to set a port with such a low power. For harmonic balance and shooting, the signal is still present - a better solution is to parameterise the frequency and then set the frequency to 0 - that way the signal doesn't get generated at all.

    Andrew

Reply
  • Andrew Beckett

    Are you using ADE Explorer to do this by any chance? If I use ADE L, it works fine, but there's a current issue in ADE Explorer where the netlist ends up with P2 being set to the initial value of P1 rather than the parameters line in the netlist showing P2=P1.

    The workaround is to define your variable as P2=(P1) i.e. with parentheses around the variable name. There's a CCR to get this fixed, but it isn't fixed yet.

    Of course, this is guessing which ADE you're using - it's always useful to know that kind of information.

    I don't see the issue you're seeing with the second edit - however, it doesn't make sense to set a port with such a low power. For harmonic balance and shooting, the signal is still present - a better solution is to parameterise the frequency and then set the frequency to 0 - that way the signal doesn't get generated at all.

    Andrew

Children
  • Emiel
    in reply to Andrew Beckett

    I'm using Meastro. The workaround with the parentheses works great, thank you.

    I see, setting the frequency to 0 makes more sense to turn off a signal than giving it very low power.

    About the second edit: indeed these low powers doesn't make sense, it was a (non elegant) way to turn it off. Still I think there is a bug, entering an extremely low power in the power field (-10000dBm) shouldn't result in a tone of 10dBm.

  • Andrew Beckett
    in reply to Emiel

    I didn't see the issue with entering -10000 so I must have done something different from you - I suggest you contact customer support with the precise details of how you produced that bug so that we can fix it (just telling me here won't get it fixed, by the way as I'm doing this in my spare time... it's always best to come from a customer report).

    Regards,

    Andrew.

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.