搜 索

Hi Petrus,

Here you can find a HFSS simulation file of a Square Patch AMC surface, also called EBG or PMC. This example concers the AMC type based on a FSS surface backed by a ground plane.
The patch side dimension is lx, and the height over the ground plane is lh. No dielectric substrate is considered. The infinite periodic conditions are given by the pairs of PEC and PMC boundary conditions.
This design resonates around 5.6 GHz; if you change the patch size, the height and the gap, you can retune this design to your desired frequency.

Bye,

Pere Hi Pere,

I've looking to your simulation file of an AMC, but i don't know how do you calculate the height of the air box over the ground plane. Could you help me with that, please?

thank you,

Wycleef. Hello Wycleef,

The height of the air box over the patch is simply calculated as the total height of the airbox (there's no rule to define this height, just several times the dielectric substrate/object height/thickness, usually more than 5 minimum), minus the height/thickness of the dielectric substrate. This lhh - lh distance is the one used in the waveport definition as the deembedding distance (== where to set the reference surface to calculate S-parameters, mostly for the phase). In AMC/EBG desigbs, the reference surface is just the surface where the metallic patches are placed.

Pere

Hello,

I resimulated your design, and I found some comments:

1) The setup is well defined, but in Analysis-->Setup, you have to increase the maximum number passes up to 30 (it will stop earlier, but sometimes above 10 passes), and the minimum number of passes up to 5, and minimum converged passes up to more than 1 (1 is ok after 15 iterations, but it is not enough in 4-5 iterations).

2) You have to check the Deembed in the Waveport definition, in order to activate the "air_height-via_height" deembeding and properly retrieve the phase of S11 (reflection coefficient).

3) You have to redefine your mushroom. If you want to operate at f0, the width of the patch/mushroom should be about lambda0/2 (where lambda0 = c / f0). And in the case you use a dielectric substrate instead of air, you should add a sqrt(er) (where er is the permittivity of the dielectric). Then an approximate value for the width is lambda0/2/sqrt(er). In that case, you need to increase the width of your design in order to operate around 2.45 (about 32 mm width), because right now, it operates at a higher frequency.

4) I don't know exactly why, but the result is not really nice (wrong results below 2.15 GHz); in addition, magnitude of S11 at the resonance is about -3 dB, too much. It needs some further work, but you can try to use PEC layers instead(thickness = 0) for the ground plane and the patch. This will reduce the sticky points. Moreover, you can play with w and g values, in order to obtain a resonance around 2.45 GHz.

See you soon,

Pere

Thanks

Dear Pere,
thank you for you sample project.

I am an hfss and AMC newer.
Could anyone help me to realize the differences between the AMC sample without thickness and implementation ?
Thank you very much.