Circular Polarization in CST MWS (again!)

A search of the threads in this forum shows that, in far-field analysis at least RHCP and LHCP can be shown from antenna structures.

This is not actually very useful if one is needing to use a microwave port to excite a circular polarized wave into a device model, such as a polarizer, or square waveguide. The plane polarization angle at a port can be set, but this is not the same thing as having it rotate at the excitation frequency. The kit being modeled is vacuum cut out of a universe of perfect conductor. There is no place to have far-fields in it. Using a far-field source with a antenna seems a hard way to get something up a waveguide.

The one example referred to, an Ortho-Mode Transducer, is not the same thing as a polarizer. The analysis is done in two separate components, offering first a 0° plane polarized wave, then one does it all again for a 90° plane polarized wave to make it exit the other port.

Is there some way to define a port to be circularly polarized, or maybe use 2 ports at the same place, one delayed in phase by 90° ?

Maybe I have to ask some help from colleagues who can do HFSS :(

OK - we only partly got there!

Instead of placing a microwave port, select "Edit Plane Wave". It has to go on the end of a boundary box, and the condition has to be "open".
The dialogue box then allows rotation LHCP, RHCP, elliptical, axis direction, etc.
This is the kind of thing one would use to excite (say) a feed or patch antenna.

Unfortunately, a PEC perfectly conducting environment is not allowed with plane wave excitation, so this method is not useful to put a circular polarized wave up a port of anything shaped of vacuum cut out of PEC.

Making the entire waveguide shapes out of PEC in a vacuum environment, complete with exterior walls and all that mesh when all the action is supposed to remain inside the waveguides just seems wrong, so we continue to search for the trick.

I did not have opportunity to test example (borrowed PC) but this does seem sensible.

So far - we appear to have struck out on this one.
Whether it be HFSS, or CST, or any other, if there is a general approach on how to get a circular polarized wave going inside a waveguide (vacuum) cut out of a universe of PEC conductor, using ports..whatever, it would be nice to discover.

Hii,
I am doing my final yr grad, for my proj, m required to design a circular patch in cst, m totally new to this software,
from online info,
i have designed a cicular patch at 10.45GHz,
substrate - FR4(lossy) material
substrate height - 1mm
radius of the patch- 3.56mm
ground nd patch height - 0.1 each, PEC material,
probe radius- 2mm
feed radius- 0.3mm


when i started the solver, an error occurred showing wave guide port is too short,
can someone help me with this?

At the present, I do not have any access to the PC that ran CST, so regrettably, I cannot help right now.

In general, you would be driving the feed point of the patch antenna via connection to a transmission line of some kind. It might be microstrip, or coplanar waveguide, or it might be the short cylinder of coaxial that typically connects through from behind the ground plane. Here is where you can make it longer.

Certainly the position of the feed point, and most other dimensions affects the driving point impedance, and also the radiation pattern. Where a nice combination has been found that unfortunately presents an inconvenient impedance, then one trick is to make the approaching transmission line with a quarter-wave matching section. This also gives you length.

If any of these approaching structures are so short that the solver cannot converge, you get a problem. You can make the port transmission line temporarily artificially longer, and then set the position of the port measurement plane in the ports dialogue to the place you intend to put the connection when you actually make it. It gets over the "too short to solve" problem.

thanx, i appreciate ur suggestion
m gonna try it