Based on the "h bend waveguide 3d" model in version 4.3b I try to simulate a dielectic loaded circular waveguide geometry. The waveguide consists of three cylinders with identical inner diameter and wall thickness in series. The dielectric is not present in the attached minimal (non-)working model. The excitation frequency range is 25.3 GHz to 25.5 GHz.
To calculate the transmitted power from one end to the other I introduce a circular excitation port 1 (with arb. power of 100W) and a listener port 2. As long as I only use two stacked cylinders (Cylinder 1 and 2 or 2 and 3) in the model the excitation on port 1 works perfectly well and I obtain the expected field distribution for the excited TE11 mode.
As soon as I add a third cylinder without any gap and the same material choices as the first two cylinders and assign the excitation port to the new end face, the calculated value of the electric field is on the order of E-13 V/m instead of 5E4 V/m.
Searching the forum for "circular port" I couldn't find any answers related to my problem.
I'd appreciate any helpful comments on what is going wrong when adding the third cylinder.
To calculate the transmitted power from one end to the other I introduce a circular excitation port 1 (with arb. power of 100W) and a listener port 2. As long as I only use two stacked cylinders (Cylinder 1 and 2 or 2 and 3) in the model the excitation on port 1 works perfectly well and I obtain the expected field distribution for the excited TE11 mode.
As soon as I add a third cylinder without any gap and the same material choices as the first two cylinders and assign the excitation port to the new end face, the calculated value of the electric field is on the order of E-13 V/m instead of 5E4 V/m.
Searching the forum for "circular port" I couldn't find any answers related to my problem.
I'd appreciate any helpful comments on what is going wrong when adding the third cylinder.