rf parallel plate capacitor simulation

Hi everybody! i'm new to Hfss. I need to simulate a parallel plate capacitor with air as dielectric. I've designed it according to the hfss guide i've found on the web, but i would like to calculate the S and Y parameters for this structure (i think i'm gonna post the hfss file as soon as i'm home). How do i have to excite the capacitor and where? Is it possible to get a plot of the impedance value vs frequency? What boundaries do i have to use (i just have assigned the radiation boundary to the air, which sorround the whole structure)?
Looking forward to read your answers.
Thank you very much

Luca

https://www.edaboard.com/viewtopic.php?p=982577#982577

Hi electroluke,
I got your PM. I guess you are Italian. Me too. Why don't you send me your design? I can try to have a look at it.

I.

Look in the post processor section and you will see s and y parameter outputs. You can also output impedance (vs. frequency if you include a sweep).

A radiation boundary is reasonable but I would keep it away from the cap elements.

You have a number of choices for the feed. When in doubt I try to make the feed look like the actual structure and use a wave port. The user's guide has a discussion towards the back that might be helpful.

Here i post the Hfss file.
The main fact is that i don't know what a wave port is, or better i don't know how to use it. I simply have this Mems capacitor (interdigitated) and i'd like to get information about its Impedance vs Frequency and Q factor (isn't it im(Y)/re(Y) ?) vs frequency. I'm used to SPICE where i just apply an AC voltage and get the frequency response.
My question is how can i do the same thing in HFSS?

Per Ivan:
Ti ringrazio molto per la disponibilità e ti ho postato il file hfss e se puoi darci un'occhiata ti sarò eternamente riconoscente :D. Si tratta di un condensatore a denti di cui vorrei ricavare l'andamento dell'impedenza e del fattore di qualità Q (che dovrebbe essere il rapporto tra im(Y)/re(Y)?) rispetto **** frequenza (fino a qualche GHz).
Per qualunque domanda puoi mandarmi direttamente un e'mail!
Grazie e fammi sapere se hai ricevuto il file!
>Ciao

Luca

Luca,
I had a lot of work to do but I looked at your design this morning. I'm gonna sebd you a pm. Let's look at it.
Ciao

Added after 1 hours 21 minutes:

Luca, as you said in your last pm, the use of that ground plane is just for convenience but it does not represent the real situation. I think you'd better use lumped ports in this case: remove the ground plane (this one affects anyway the circuit behavior) and wave ports, draw a rectangular sheet, one for each capacitor terminal line, and create a lumped port for each rectangle, defining the correct current flow direction. Then draw a 2 very very thin rectangles (one for port) touching the side of the lumped port that doesn't touch the circuit terminals. Apply pec to these thin rectangles, which will behave as reference for ports. Enlarge the substrate in order to let the ports to be far enough from the borders (lumped ports are internal ports). Should be working.
Ciao

Luca, yesterday my connection was very bad. Here,'s something you may find useful... There's a cpw MEMS switch, it uses lumped ports as I adviced at first. How did u get on with cpw feeding port? Moreover, have a look at this:

http://www.cmc.ca/news/events/mems_w...pacitor_WT.pdf

break a leg!

Ivan

It is possible to get a plot of impedence vs fequency.As far as ur design is concerned,you should use a lumped port instead of a wave port for the purpose of excitation.It is better that you use mica instead of air as a dielectric.If u want to view the fundamentl antenna parmeters you should include a far field setup in ur design by making a infinite sphere too