I'm trying to determine the spacing between resonators for a planar bandpass combline filter, such that the even and odd mode impedances match those that I've calculated as part of the filter design. If someone can spot any errors in how I've gone about this, please feel free to let me know. I've attached a screenshot containing my history list, model, and the error messages I've received.
I've set the simulation (time-domain solver with hexahedral meshing) up as follows:
I had hoped to eventually run an optimiser (time-domain solver with hexahedral meshing) on both the even and odd mode port definitions to determine the optimal spacing value, but can't get past this error.
Does anyone have advice on how I can get the even and odd mode impedances without running into the aforementioned error? There's not much literature or examples on performing parameter extraction on a planar combline filter, so I'm simply following a procedure that makes some sense to me.
On a separate but related note: If I only run the solver (without making use of the port mode calculator), what result should I be looking at to see whether my specification has been met? I also have coupling values M_{ij}, could I perhaps go about extracting the spacing value "s" in a different way? I can't find relevant literature that works when I follow their steps.
Thank you for your time.
I think I would use a waveguide port but not use multipin, just ask for several modes. It is an option in the solver setup. Don't use shorting or connecting pins. I think CST will calculate the mode impedance and pattern for each mode before the calculation and the impedance will be available on-screen when you display each mode pattern under the 2D/3D results. You can get CST to just calculate the modes and stop, it is an option in the solver setup.
Thank you, Malcolm White . I'll try that. Is it the wave impedance that I need to match to my calculated even and odd mode impedances, or the line-impedance?
I'll remove the vias and see how that influences my results.
I think it is the line impedance. I think the wave impedance is the free-space impedance of the dielectric, but I'm not sure.
- Badges
- Science topic
- Similar topics
- Computer Communications (Networks)
- Simulators