Which EM Simulator works best for resonant circuits?

Hello and thanks in advance,

I'm trying to simulate a open-loop resonator filter, which is a highly resonant circuit.
Using MWO and CST for simulation works really bad. Only Ensemble gives a good response in comparation with measurments, but I have no idea about how to use design variable in Ensemble, I mean, setting a 'gap' variable for example so it is a parameter I can easily modify so the model is also modified. Drawing in ensemble is a pain in the a$$ and I need quick changes. Using Autocad or another software for dxf edition and import it in Ensemble is not viable

By the way, I've been using Ensemble becouse the circuit is planar, obviously

So, my question is in fact double,

Which EM Simulator works best fot this circuits(and it's is easy to parametrize)?
Is it posible to use design variables (parametrize) in Ensemble?



Thanks a lot!!!

Hi

I think HFSS will do the job

Regards hilper

Thanks a lot

I will try HFSS and tell you

It 's a pitty HFSS is full 3D, it takes a lot of time to simulate.
Have you any idea on Sonnet?

Hi,

You can use Ansoft Designer...it is an updated version of Ensemble. You can parameterize your dimensions I believe. Also, you can link up your HFSS project to Designer and vice versa.

I have used Designer and i slower than ensemble so I preffer not to use it
Thanks alot anyway
Any another idea?

Download Sonnet Lite, that may have all the features you need.

If not then Jim Rautio is sometimes here and he can let you know how applicable Sonnet would be. My feeling is that this is your best bet over HFSS/CST etc.

Thanks a lot for your information

I have donwloaded it, but I cannot import dxf. I will try a full version asap.

So, there is no way to use variable in Ensemble, there is?

Hi Buqpa -- First thanks to Doug for the nice comment above. I am on travel right now (at EuMC in Manchester, if anyone is there, stop by and say hi). Internet access is difficult sometimes.

It should be possible to get a reasonable answer from most any of the simulators. I don't know why there are some problems. I would guess that CST has a problem because it is a resonator. A high Q resonator impulse response lasts a long time and requires a lot of analysis time. But they have ways to handle that kind of situation well. You might find it in their documentation, or contact their support. If you are in North America, you can contact us too.

I have never seen any benchmarks on Ensemble and know very little about how it works, so can not comment there.

The problem sounds ideal for the free SonnetLite (I work for Sonnet). You can even use our fast sweep (ABS, Adaptive Band Synthesis) which is default. Circuit resonances have no effect on the number of frequencies ABS needs, in complete contrast to all other fast sweeps that I know about. It is really amazing to watch Sonnet interpolate data from 4 or 5 frequencies into an almost exact frequency response of a 20 resonator 5% bandwidth filter.

If you can get your circuit into AWR, you can then send it directly to SonnetLite. Also, the in the next release, SonnetLite will interface to ADS for free. So if you can get it into ADS, then you can get it into SonnetLite.

If you have any problems, post them here, or better yet, post them on the Sonnet User's Forum, www.sonnetsoftware.com. It is seeing quite a bit of activity lately.

Hi
I have some problem like this.
see this: **broken link removed**
It may help you. [/url]

Added after 1 minutes:

Hi
I have some problem like this.
see this: **broken link removed**
It may help you.

Agilent ADS-Momentum is helpful in solving this problem.

Buqpa shalom,

It is not true that Designer is slower than ensemble. They tuned it to be more accurate, but you can decrease the Edge Length Factor (advanced mesh settings) and get faster results.

Another feature that you would like is the 3D via features for the coax excitations - that will increase the feed accuracy of the simulation.

Regards,
Itai

For high resonant microwave structure, you should use Frequency-Domain method (FEM, MoM) to analysize it.

For 2.5D microwave structure, it is more effective to use MoM method.

Therefore, you should use Sonnet em or Momentum or IE3D to analysize the microstrip open loop resonator because these simulators have more effecitve meshing and more efficient and accurate than other 2.5D simulators.

Hi Buqpa,

just my comments:

I'm actually wondering why AWR and CST did not give you good results. This seems to be a pretty simple structure and I'm convinced that ALL EM Simulator can handle this task. Which solver module did you use in CST? For highly resonance structures the Frequency Domains Solver might be then best bit. However, even the time domain solver should be able to handle it.

Maybe you can post your design as DXF File together with some design information and we all have a look. I guess there are quite some ARW and CST users here in this forum which might be able to help.

F.

u can try Zeland IE3D software..its pretty simple to use and accurate