What is the optimum height of a Floquet-Port in modeling FSS band stop filter?

Using HFSS for*simulating of FSS dual bandstop filter (0.9 and 1.8 GHz), a Floquet-Port needs to be assigned at an appropriate height over the surface. How much is that height supposed to be? and how can I be so sure of the results if any small change in this height dramatically changes the results?

what should be the Deembed distance?

and what is the value of Solution Frequency in the Driven *Solution Setup?

More than one-quarter wavelength at your lowest frequency is a good rule of thumb. All you need is appropriate distance for near-fields on your surface to decay towards the port.

Err on the side of increased distance.

The distance from the port to the first "face" of your surface.

The frequency you want the simulation to converge at.

Thank you PlanarMetamaterials for your reply.

[Err on the side of increased distance.]
can you please explain more?

I have to frequencies : 0.9 GHz and 1.8 Ghz "dual band stop passive FSS filter , the sweep is from 0.5 to 2.2 GHz I put Solution Frequency to 1.8GHz is this is best and close to real measurements.

Mode distance between the floquet port and the surface will allow evanescent modes to decay to a smaller level, increasing the accuracy of the result (which is based only on the fields corresponding to propagating modes).

Usually the highest frequency (2.2 GHz), although it depends on your structure. You want the frequency to be high for accuracy, but not so high that it converges on an undesired mode.