Menu

Discussion Forum

New Discussion
Filter

Defining port in 2D (not axisymmetric) with microwaves launched from open-ended circular waveguide

Posted Mar 27, 2025, 4:28 p.m. EDT Electromagnetics, RF & Microwave Engineering, Modeling Tools & Definitions Version 6.2 6 Replies

Natalia Nigay Beamed Microwave Propulsion
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Hi COMSOL folks,

Here is my problem: I am trying to focus microwaves launched from an open-ended circular waveguide at 2.45 GHz to the inside of a small diverging nozzle (essentially a cone with metal walls). I am working inside of a vacuum chamber, so I am limited in space, therefore I am trying to implement a custom off-axis parabolic (OAP) mirror, assuming a mostly collimated incident beam which will be reflected and focused to a point, at which location I'd place the nozzle.

In the image attached, I have the OAP, section of circular waveguide, and the conical nozzle modeled in the domain. I would like to put this setup into a 2D simulation, but I am having a hard time knowing how to correctly define the port as there is no circular port option. I don't have the strongest EM background, but I believe I would need the e-field distribution for TE11 mode with a circular waveguide.

I would really appreciate any advice, comments, or guidance on how to approach this simulation or improve the setup as a whole. Thanks in advance!

-------------------
Natalia

Attachments:

Reply
6 Replies Last Post Mar 28, 2025, 5:03 p.m. EDT
Robert Koslover Certified Consultant
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 4 days ago Mar 27, 2025, 9:14 p.m. EDT
Updated: 4 days ago Mar 27, 2025, 11:09 p.m. EDT

Your overall geometry is not axisymmetric and it is not well suited to an ordinary 2D model either. However, you could use a 3D RF model with a symmetry plane. Alternatively, if you wish to pursue a simpler 2D axi-symmetric model of the horn alone, you can represent the azimuthal dependence of the TE11 mode analytically. This hybrid approach is illustrated in examples provided by Comsol in the Application Library. See https://www.comsol.com/model/fast-numerical-modeling-of-a-conical-horn-lens-antenna-18695 and also https://www.comsol.com/model/corrugated-circular-horn-antenna-15677 .
FYI, your "nozzle" is referred to in RF literature as a "conical horn."

-------------------
Scientific Applications & Research Associates (SARA) Inc.
www.comsol.com/partners-consultants/certified-consultants/sara
Your overall *geometry* is *not* axisymmetric and it is not well suited to an ordinary 2D model either. However, you could use a 3D RF model with a symmetry plane. Alternatively, if you wish to pursue a simpler 2D axi-symmetric model of the horn alone, you can represent the azimuthal dependence of the TE11 mode *analytically*. This hybrid approach is illustrated in examples provided by Comsol in the Application Library. See https://www.comsol.com/model/fast-numerical-modeling-of-a-conical-horn-lens-antenna-18695 and also https://www.comsol.com/model/corrugated-circular-horn-antenna-15677 . FYI, your "nozzle" is referred to in RF literature as a "conical horn."
Edgar J. Kaiser Certified Consultant
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 3 days ago Mar 28, 2025, 7:11 a.m. EDT

Hi Natalia,

2.45 GHz radiation has a wavelength of 12 cm. Any focus spot won't get smaller than this and the little cone seems to be much smaller. Also the wave won't propagate into that cone, because its dimensions are smaller than the cutoff wavelength.

For any useful hints we would need to know what the purpose of the setup is?

Cheers Edgar

-------------------
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
Hi Natalia, 2.45 GHz radiation has a wavelength of 12 cm. Any focus spot won't get smaller than this and the little cone seems to be much smaller. Also the wave won't propagate into that cone, because its dimensions are smaller than the cutoff wavelength. For any useful hints we would need to know what the purpose of the setup is? Cheers Edgar
Natalia Nigay Beamed Microwave Propulsion
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 3 days ago Mar 28, 2025, 3:44 p.m. EDT

Hi Natalia,

2.45 GHz radiation has a wavelength of 12 cm. Any focus spot won't get smaller than this and the little cone seems to be much smaller. Also the wave won't propagate into that cone, because its dimensions are smaller than the cutoff wavelength.

For any useful hints we would need to know what the purpose of the setup is?

Cheers Edgar

Ah, I did not know that a focus spot does not get smaller than the wavelength. This is for a propulsion application; beamed microwave energy propulsion. The idea is to use beamed microwaves to generate plasma inside of the diverging portion of a nozzle (the cone, simplified for the simulation). It is indeed quite small, because we are working with water vapor and are limited in our flow rates due to power requirements.

-------------------
Natalia
>Hi Natalia, > >2.45 GHz radiation has a wavelength of 12 cm. Any focus spot won't get smaller than this and the little cone seems to be much smaller. Also the wave won't propagate into that cone, because its dimensions are smaller than the cutoff wavelength. > >For any useful hints we would need to know what the purpose of the setup is? > >Cheers >Edgar Ah, I did not know that a focus spot does not get smaller than the wavelength. This is for a propulsion application; beamed microwave energy propulsion. The idea is to use beamed microwaves to generate plasma inside of the diverging portion of a nozzle (the cone, simplified for the simulation). It is indeed quite small, because we are working with water vapor and are limited in our flow rates due to power requirements.
Natalia Nigay Beamed Microwave Propulsion
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 3 days ago Mar 28, 2025, 3:50 p.m. EDT

Your overall geometry is not axisymmetric and it is not well suited to an ordinary 2D model either. However, you could use a 3D RF model with a symmetry plane. Alternatively, if you wish to pursue a simpler 2D axi-symmetric model of the horn alone, you can represent the azimuthal dependence of the TE11 mode analytically. This hybrid approach is illustrated in examples provided by Comsol in the Application Library. See https://www.comsol.com/model/fast-numerical-modeling-of-a-conical-horn-lens-antenna-18695 and also https://www.comsol.com/model/corrugated-circular-horn-antenna-15677 .
FYI, your "nozzle" is referred to in RF literature as a "conical horn."

I have it set up in 3D, I was just hoping for computational simplicity to be able to do this in 2D as well; what is the reason it's not well suited to a 2D model? I will take a look at that example.. However, I suppose I should have clarified in my post that the conical structure will in fact be a nozzle for an electric propulsion thruster, but does in a way function like a receiving horn antenna. The crux of my problem lies in the fact that the target area is small relative to the wavelength, and unfortunately we do not have power sources at higher frequencies or the capability to increase the nozzle size, therefore I am trying to come up with a creative solution.

-------------------
Natalia
>Your overall *geometry* is *not* axisymmetric and it is not well suited to an ordinary 2D model either. However, you could use a 3D RF model with a symmetry plane. Alternatively, if you wish to pursue a simpler 2D axi-symmetric model of the horn alone, you can represent the azimuthal dependence of the TE11 mode *analytically*. This hybrid approach is illustrated in examples provided by Comsol in the Application Library. See https://www.comsol.com/model/fast-numerical-modeling-of-a-conical-horn-lens-antenna-18695 and also https://www.comsol.com/model/corrugated-circular-horn-antenna-15677 . >FYI, your "nozzle" is referred to in RF literature as a "conical horn." I have it set up in 3D, I was just hoping for computational simplicity to be able to do this in 2D as well; what is the reason it's not well suited to a 2D model? I will take a look at that example.. However, I suppose I should have clarified in my post that the conical structure will in fact be a nozzle for an electric propulsion thruster, but does in a way function like a receiving horn antenna. The crux of my problem lies in the fact that the target area is small relative to the wavelength, and unfortunately we do not have power sources at higher frequencies or the capability to increase the nozzle size, therefore I am trying to come up with a creative solution.
Robert Koslover Certified Consultant
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 3 days ago Mar 28, 2025, 4:32 p.m. EDT
Updated: 3 days ago Mar 28, 2025, 4:34 p.m. EDT

For an ordinary 2D model (Cartesian, in x,y coordinates) to be representative of 3D reality, there must be no important dependences on directions perpendicular to the plane (e.g., dependence upon z). That requirement clearly isn't satisfied here. Simiarly, in a 2D axisymmetric model, there should be no important dependence on the azimuthal direction (phi), EXCEPT that in the rather-clever example models that I cited, one can address this problem by means of an analytic extension that represents the phi dependence of the mode(s), while using Comsol's finite elements-based computations to solve for only rho and z dependences. Now, if you were to replace your "offset" reflector by an axisymmetric reflector, you could in principle make use of the same trick yourself, since once again, the azimuthal dependence could be handled analytically. This would greatly reduce the computational size of your problem compared to a 3D treatment, so you might want to consider doing that. (If so, study the examples carefully and make sure you understand exactly how this trick was implemented.)

-------------------
Scientific Applications & Research Associates (SARA) Inc.
www.comsol.com/partners-consultants/certified-consultants/sara
For an ordinary 2D model (Cartesian, in x,y coordinates) to be representative of 3D reality, there must be no important dependences on directions perpendicular to the plane (e.g., dependence upon z). That requirement clearly isn't satisfied here. Simiarly, in a 2D axisymmetric model, there should be no important dependence on the azimuthal direction (phi), EXCEPT that in the rather-clever example models that I cited, one can address this problem by means of an *analytic* extension that represents the phi dependence of the mode(s), while using Comsol's finite elements-based computations to solve for only rho and z dependences. Now, if you were to *replace* your "offset" reflector by an axisymmetric reflector, you could in principle make use of the same trick yourself, since once again, the azimuthal dependence could be handled analytically. This would greatly reduce the computational size of your problem compared to a 3D treatment, so you might want to consider doing that. (If so, study the examples carefully and make sure you understand exactly how this trick was implemented.)
Edgar J. Kaiser Certified Consultant
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 3 days ago Mar 28, 2025, 5:03 p.m. EDT

However, I suppose I should have clarified in my post that the conical structure will in fact be a nozzle for an electric propulsion thruster, but does in a way function like a receiving horn antenna.

What I think I understand is that you expand the gas at the tip of the nozzle into the cone and you want to ignite a plasma by focused microwaves. The plasma then expands into the vacuum and produces thrust? The cone is supposed to efficiently pick up the power from the microwave focus? If my assumptions are correct you will likely need to achieve a match of cone size and wavelength. The waves don't leave much space for creativity I am afraid.

It is possible to confine RF energy in structures that are much smaller than a wavelength by using lumped components. Good old L-C resonant circuits made up by a coil and a capacitor are an example.

-------------------
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
>However, I suppose I should have clarified in my post that the conical structure will in fact be a nozzle for an electric propulsion thruster, but does in a way function like a receiving horn antenna. What I think I understand is that you expand the gas at the tip of the nozzle into the cone and you want to ignite a plasma by focused microwaves. The plasma then expands into the vacuum and produces thrust? The cone is supposed to efficiently pick up the power from the microwave focus? If my assumptions are correct you will likely need to achieve a match of cone size and wavelength. The waves don't leave much space for creativity I am afraid. It is possible to confine RF energy in structures that are much smaller than a wavelength by using lumped components. Good old L-C resonant circuits made up by a coil and a capacitor are an example.

Reply

Please read the discussion forum rules before posting.

Please log in to post a reply.

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.

Suggested Content