Hello, I am reading a paper about a SIW-fed antenna. At the input port, there are two cuboids, as shown in the attached file.
Can anyone tell me what these two cuboids are? Are they PEC?
The paper I am reading is:
Y. Li and K. Luk, "A Multibeam End-Fire Magnetoelectric Dipole Antenna Array for Millimeter-Wave Applications," in IEEE Transactions on Antennas and Propagation, vol. 64, no. 7, pp. 2894-2904, July 2016.
I guess they are there so that the input port is well-defined, with a constant geometry for a few mesh cells, which wouldn't happen if round posts were used. I think that if there were no conducting edges to the aperture then the input mode would not correspond to the SIW mode and there would be a mismatch at the input. I expect they are PEC
A field calculator interprets a port declaration as follows, Chuangkai Wang. It assumes that there is a waveguide connected to that port, with a cross-section that is the (flat) side of your model, where that port is located. For example, if that side is an electric wall with a rectangular aperture, your field caculators connects a standard hollow rectangular metallic waveguide perpendicular to the said side, and whose cross-section is that very aperture. In your case, if there were no rectangular posts where the ports connect, the field simulator involved would mistake the ports for connections to parallel-plate waveguides, as Malcolm implied.
Than you for your kind replies, Malcolm White Denis Jaisson. Sorry for my poor English, I didn't describe my question clearly.
What confuses me is why not remove the rectangular post and add a port directly at the round post, as shown in the attached file.
I try to simulate this structure in HFSS and get an error "Non-manifold edges found". I think maybe this is the reason why two rectangular posts are added.
I explained in my first answer. I think CST requires constant geometry for the first few mesh cells of a port. Also, the mode that would be calculated at the port is not the same as the propagating mode because the port is significantly wider at this point.
Round posts would not be a problem for HFSS, but, in the course of your simulation exercise, you might change some dimensions in such a way, that the position of the reference plane with respect to the posts might change. That be the case, the characteristic impedance of the mode assumed by HFSS would change accordingly, possibly without you being aware of it...
Beware that the said mode is only an approximation to the Floquet modes that propagate through your periodic structure. You will find that most application engineers of the vendors of field simulators understand only little about that, actually. I suggest that you look into Collin's "Foundations for microwave engineering" for a didactical introduction to those modes. Collin addressed the subject in his "Field theory of guided waves" too, but his latter treatment is more complex than in the former book.
- Badges
- Science topic
- Similar topics
- Engineering
- Telecommunications Engineering
- Antennas and Propagation