• Skip to main content
  • Skip to search
  • Skip to footer
Cadence Home
  • This search text may be transcribed, used, stored, or accessed by our third-party service providers per our Cookie Policy and Privacy Policy.

  1. Community Forums
  2. RF Design
  3. Problem when using RFDE momentum blocks in Cadence (Drop...

Stats

  • Locked Locked
  • Replies 9
  • Subscribers 63
  • Views 5954
  • 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

Problem when using RFDE momentum blocks in Cadence (Drop across Resistance w/o any current)

dipal261
dipal261 over 8 years ago

Hi,

I have a problem with the use of RFDE mom symbol view in Cadence schematic. The image is attached below for reference. The momentum block is basically a MOM capacitor for coupling ac input . In theory, there should not be any drop in the 3KOhms resistance connected to the gate of the transistor in one end and the capacitor on the other end. This entire block works fine if I simply remove the MOM model with an ideal capacitor from analoglib. In that case, I see no drop across the 3K Ohms resistance. But when using the MOM model, it simply destroys any gate biasing that I try to put. What is the problem here? and how can I get a work around it?

This is a dc analysis and shows the current and voltage at each node.

Thank you.

  

Regards

Dipal

  • Cancel
  • Andrew Beckett
    Andrew Beckett over 8 years ago

    Dipal,

    Given that Momentum is a tool from Keysight (not Cadence) and I don't know which version of Virtuoso, Momentum, or Spectre (assuming that it's spectre you're using as the simulator - you didn't say that either), it's quite hard to give a definitive answer.

    So, assuming that the simulator is spectre, can you post what the lines in the input.scs look like for I741 (the highlighted component above)? Also can you mention the versions of the tools you're using?

    Thanks,

    Andrew.

    • Cancel
    • Vote Up 0 Vote Down
    • Cancel
  • dipal261
    dipal261 over 8 years ago

    Hi Andrew,

    This is a fresh simulation where I removed everything else. I am using ADS 2016.01, MMSIM15.10.627 and IC6.17.704

    Here is the spectre simulator data from the input.scs file for the below schematic

    PORT0 (net70 0) port r=50 num=1 type=sine freq=100G dbm=0

    R3 (net70 net03) resistor r=1u

    R2 (net06 net42) resistor r=1u

    R1 (net42 net010) resistor r=3K

    R0 (Amp_Vdd Drain_device) resistor r=100

    V4 (net010 0) vsource dc=550.00m type=dc

    V0 (Amp_Vdd 0) vsource dc=Vdd type=dc

    I405 ( net03 0 0 0 net06 0 0 0) nport interp=linear \

           file="workdir/mommdl/text.txt" \

           datafmt=citi

    and the mommdl/text.text file is an S parameter file generated for the momentum block I405 (schematic below)

    Please let me know if you need any other data.

    Thank you.

    Regards

    Dipal

    • Cancel
    • Vote Up 0 Vote Down
    • Cancel
  • Andrew Beckett
    Andrew Beckett over 8 years ago

    Dipal,

    Some questions:

    1. Does the MoM component give you a choice of interpolation method? It would be good if you could pick "bbspice" as this is generally a better choice for passive components. Otherwise you could try using the nport component from analogLib and picking bbspice as the interpolation method, and pointing at the s-parameter file
    2. How low in frequency does your s-parameter file go down to? Does it have a DC point in the file?
    3. What frequency grid did you pick when you ran your Momentum simulation? We would generally suggest a linear sweep and ensure that there are sufficient points.
    4. If you add "-cc" (to clear the cache) on the spectre command line (which you can access via the userCmdLineOptions field on Setup->Environment in ADE), and re-run, does spectre complain about passivity or causality when processing the s-parameter file? Usually this is only done the first time  you use a particular s-parameter file, so the clear cache option forces the checks to be run on subsequent simulations (usually you'd not have this set for best performance though).

    Regards,

    Andrew.

    • Cancel
    • Vote Up 0 Vote Down
    • Cancel
  • dipal261
    dipal261 over 8 years ago

    Hi Andrew,

    1. The MoM component does not have an option for interpolation method. I just get a warning when running the DC analysis. It says "Risky extrapolation to DC of data given in S-parameter file". I tried the nport from the analogLib with the "bbspice" as the interpolation method and also selecting the option "Extrapolate to DC". The problem still persist and I still see a drop across the resistance connected to the gate.

    2. As it takes a long time to simulate, the lowest point is at 80 GHz. Like you suggested, there is no DC point in the file. Is there anything that I could add into the S Parameter file?

    3. I did the simulation with linear sweep and a step size of 1GHz.

    4. It does not throw any complaint about passivity or causality.

    Please let me know if you need any other data.

    Regards

    Dipal

    • Cancel
    • Vote Up 0 Vote Down
    • Cancel
  • Andrew Beckett
    Andrew Beckett over 8 years ago

    Dipal,

    Well, I think it's a little optimistic for any model to be built which will simulate at DC and in the time domain if the lowest simulation point you ran in Momentum was at 80GHz! How the spectre is supposed to know that the capacitor is open circuit at DC (after all, it doesn't know it's a capacitor - all it has is the s-parameters). I nearly fell off my chair when you said the minimum frequency is 80GHz!

    A 1GHz step is pretty big too.

    I'm not sure we can help much more without seeing the data, but to be honest I think the main problem is that the s-parameter data is missing too much low frequency and even high frequency data, plus is too sparse.

    Andrew.

    • Cancel
    • Vote Up 0 Vote Down
    • Cancel
  • dipal261
    dipal261 over 8 years ago
    Hi Andrew,
    I just finished a simulation and it seems indeed to be a problem with the step size! I reduced the lowest frequency as well. You are right, the S parameter file was having no data at lower frequency range which was causing this silly problem!
    Thank you so much!!
    Regards
    Dipal
    • Cancel
    • Vote Up 0 Vote Down
    • Cancel
  • Tawna
    Tawna over 8 years ago

    Hi Dipal,

    You might want to take a look  at the IMS MicroApp “7 Habits of Highly Successful S-Parameters” is on our Cadence website.

    On Cadence Online Support,  the in-depth AppNote is here:  20466646 .

    Best regards,

    Tawna

    • Cancel
    • Vote Up 0 Vote Down
    • Cancel
  • dipal261
    dipal261 over 8 years ago

    Hi Tawna,

    Thank you for the document. It is really informative.

    Regards

    Dipal

    • Cancel
    • Vote Up 0 Vote Down
    • Cancel
  • Tawna
    Tawna over 8 years ago
    You are very welcome, Dipal!
    • Cancel
    • Vote Up 0 Vote Down
    • Cancel

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.

© 2025 Cadence Design Systems, Inc. All Rights Reserved.

  • Terms of Use
  • Privacy
  • Cookie Policy
  • US Trademarks
  • Do Not Sell or Share My Personal Information