Simulation of lumped elements at f > 40 GHz
Altair EmployeeHi,
The question is:
In general how does FEKO simulate lumped elements? Particularly for very high frequencies, where wavelengths are of the same order as the size of the edge port and adjacent faces.
In detail:
I'd like to know how to properly set a simulation to calculate structures with discrete elements (e.g. capacitor) when real dimensions (capacitor in package) of the elements are near-wavelength (1/6 lambda in this example).
I know that an edge port could be used to simulate lumped circuits but I wonder how does the size of the edge port and adjacent faces affect simulation results.
The edge port could be also loaded by using S-characteristics of some element, e.g. diode (package with real length of 1/6 lambda), but the S-parameters that are given correspond to a reference plane, which is defined on both ends of the element package. The problem is that in FEKO the diode would be represented by the edge port, that has length of zero.
Is there any way to properly simulate lumped elements in packages in so high frequencies? Maybe there is another way of simulating lumped elements except the edge port?
EDIT: By structures I mean Microstrip, Slotline, Stripline planar circuits.