Edgar J. Kaiser
Certified Consultant
Please login with a confirmed email address before reporting spam
Posted:
5 years ago
24.05.2020, 07:20 GMT-4
Anna,
an often used method is to circularize in the circular waveguide by a dielectric plate inclined by 45° to the incoming E-field to introduce a lambda/4 phase shift between the two components of the incoming mode. This may also work in the horn itself, but haven't seen it anywhere so far to be honest.
Cheers
Edgar
-------------------
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
Anna,
an often used method is to circularize in the circular waveguide by a dielectric plate inclined by 45° to the incoming E-field to introduce a lambda/4 phase shift between the two components of the incoming mode. This may also work in the horn itself, but haven't seen it anywhere so far to be honest.
Cheers
Edgar
Anna Tz.
Electronic and Electrical Engineering
Please login with a confirmed email address before reporting spam
Posted:
5 years ago
24.05.2020, 16:27 GMT-4
Updated:
5 years ago
24.05.2020, 16:28 GMT-4
Thank you Edgar for your reply,
Could you explain more about this technique? The waveguide is practically a cylinder and the TE11 mode port is at its bottom. Should the plate be inclined relatively to the axis of this cylinder? I don't quite understand the concept...
I was thinking that maybe there is a way I could combine two linear polarizations to produce the circular one during post processing of the model. Do you have any ideas about that?
Thank you Edgar for your reply,
Could you explain more about this technique? The waveguide is practically a cylinder and the TE11 mode port is at its bottom. Should the plate be inclined relatively to the axis of this cylinder? I don't quite understand the concept...
I was thinking that maybe there is a way I could combine two linear polarizations to produce the circular one during post processing of the model. Do you have any ideas about that?
Edgar J. Kaiser
Certified Consultant
Please login with a confirmed email address before reporting spam
Posted:
5 years ago
24.05.2020, 17:07 GMT-4
Anna,
ok, you are only talking about modeling. I was assuming you are interested in a technical realization too. In the model you can simply use two ports with perpendicular E-field settings and a phase difference of 90°. It should be possible to put those 2 ports on the same boundary, but I personally never tried that.
It should also be possible to combine two runs with perpendicular linear polarization and the phase shift in post processing. You would need to combine solutions then or use with() or withsol() operations for that. This is in the documentation and also in some blog posts.
Cheers,
Edgar
-------------------
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
Anna,
ok, you are only talking about modeling. I was assuming you are interested in a technical realization too. In the model you can simply use two ports with perpendicular E-field settings and a phase difference of 90°. It should be possible to put those 2 ports on the same boundary, but I personally never tried that.
It should also be possible to combine two runs with perpendicular linear polarization and the phase shift in post processing. You would need to combine solutions then or use with() or withsol() operations for that. This is in the documentation and also in some blog posts.
Cheers,
Edgar
Robert Koslover
Certified Consultant
Please login with a confirmed email address before reporting spam
Posted:
5 years ago
24.05.2020, 20:17 GMT-4
Updated:
5 years ago
24.05.2020, 20:24 GMT-4
Edgar is correct (as he usually is) and I have done it. You can assign two different TE11 mode ports to the same physical boundary at the end of the cylinder. Make sure that you: (1) orient the modes to be perpendicular to one another - which you do with specifying two different pairs of reference-direction points, and (2) set the excitations to differ in phase by pi/2 radians. Plus pi/2 will give one of the handednesses of polarization and -pi/2 will give the other-handedness polarization. You don't need to use with() or withsol() just to run these circular polarization cases, by the way, and you don't need to combine separate runs.
-------------------
Scientific Applications & Research Associates (SARA) Inc.
www.comsol.com/partners-consultants/certified-consultants/sara
Edgar is correct (as he usually is) and I have done it. You can assign two different TE11 mode ports to the same physical boundary at the end of the cylinder. Make sure that you: (1) orient the modes to be perpendicular to one another - which you do with specifying two different pairs of reference-direction points, and (2) set the excitations to differ in phase by pi/2 radians. Plus pi/2 will give one of the handednesses of polarization and -pi/2 will give the other-handedness polarization. You don't need to use with() or withsol() just to run these circular polarization cases, by the way, and you don't need to combine separate runs.
Anna Tz.
Electronic and Electrical Engineering
Please login with a confirmed email address before reporting spam
Posted:
5 years ago
25.05.2020, 04:21 GMT-4
Many thanks to both of you,
I wasn't aware I can assign two ports on the same boundary. Indeed I followed your advice and it seems to have solved my problem. Again, thank you very much!
Many thanks to both of you,
I wasn't aware I can assign two ports on the same boundary. Indeed I followed your advice and it seems to have solved my problem. Again, thank you very much!