HFSS capacitor simulation

[slickbomb8]

hi all,

has anyone been successful in simulating parallel plate (square) capacitors in HFSS?

I created two sheets with PEC boundary cond. to represent the plates, and between them created a rectangular sheet connecting the two plates. I gave it lumped port excitation. I tried making the lumped port connecting the centers of the plates, and also moved the lumped port away from the center of the plates towards the edges.

I also tried to model the plates as PEC rectangles to give some small thickness to the plates, instead of sheets.

Compared to the simple formula for parallel plate capacitor, I have not been able to extract the right capacitance, from C = 1/(2*pi*im(Z11)). I know that there could be fringing fields at work here, but I made the plate areas really big compared to the separation distance. My extracted capacitance is wildly off what is expected from the basic model.


Can somebody offer some advice?

 

[e_m_c]

Can you show more details about structure? Size?

 

[slickbomb8]

an example would be a two 30mm x 30mm PEC sheets with 0.5mm gap spacing, simulating at 1 GHz. With capacitor plates modelled by 3d box, I'd do 30mm x 30mm x 0.5mm with PEC material. Radiation airbox is greater than wavelength/4 (0.3m) away in all directions from my structure. Lumped port is rectangle with dimensions 0.5mm x 0.5mm, connecting the two plates. Terminal line set from lumped port's end to end.

Perhaps this frequency is not good for the dimensions of my capacitor, but if so I do not know why. In general, is there a rule of thumb of for when HFSS doesn't give accurate results for a specific frequency relative to the size of my structure?