Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

Horn Antenna Simulation - Help

Please login with a confirmed email address before reporting spam

Hello Everyone.

I'm an engineering student and I'm new in the COMSOL world. I am trying to simulate the behavior of a Horn Antenna. I set up my geometry, materials and physics quite a while ago and it's now been 10 days of unsuccessful troubleshooting. Not even my tutor could spot the problem in my model, and the deadline for this deliverable is next Tuesday. I really need some help. Below, I am going to briefly outline what I have done, for an easier understading of the model.

Basically, I first modelled my waveguide ("Flare") and my Horn ("Flare") and made a union of them, "Antenna". Next, I made my feed cable. It is divided it in two parts, the inner part that's sticking out inside the waveguide ("Coax Pin"), and the outer one which is also surrounded by dielectric. What I do next is subtracting all metal parts from the air domain: so it's "Sphere" - "Antenna" - "Coax Pin". Lastly, I take a difference between the cylinder of the dielectric and the cylinder of the outer pin, to get "Coax Insulator".

Now: the boundaries of air in contact with the pin and the antenna become PEC. The inside of the dielectric is also PEC, the outside is also PEC and the bottom face of the dielectric hollow cylinder is selected as coaxial-type lumped port. Of course I set Perfectly Matched Layers, Far-Field domains etc - you can check those in the model.

Everything looks fine to me, and so did to my tutor. However, if I run the study, I can not get any results. My convergence plot shows that the program is not getting anywhere.

Does anyone see the problem? I have also tried different approaches, e.g. starting with a simplified model with no thickness. I have already looked at COMSOL pre-made tutorials (e.g. Double Rigged Horn Antenna, Dipole Antenna, Mononopole Antenna Arrray, Coaxial to Waveguide Coupling), so please avoid just suggesting papers. I don't really have time left!

Regards, Matteo



4 Replies Last Post 30 mar 2020, 21:26 GMT-4
Edgar J. Kaiser Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 5 years ago 28 mar 2020, 14:49 GMT-4

Matteo,

activate the direct solver instead of the iterative one and it solves in seconds.

Cheers Edgar

-------------------
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
Matteo, activate the direct solver instead of the iterative one and it solves in seconds. Cheers Edgar

Please login with a confirmed email address before reporting spam

Posted: 5 years ago 30 mar 2020, 05:21 GMT-4
Updated: 5 years ago 30 mar 2020, 05:33 GMT-4

Hi Edgar,

I chose the direct solver and it computed. However, if you plot e.g. the E field, it still doesn't look correct. Any idea why?

Regards, Matteo

Hi Edgar, I chose the direct solver and it computed. However, if you plot e.g. the E field, it still doesn't look correct. Any idea why? Regards, Matteo

Edgar J. Kaiser Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 5 years ago 30 mar 2020, 06:50 GMT-4

Matteo,

it seems the model isn't properly excited by the port. Sorry I can't spend more time to analyse your model in detail.

Good luck Edgar

-------------------
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
Matteo, it seems the model isn't properly excited by the port. Sorry I can't spend more time to analyse your model in detail. Good luck Edgar

Robert Koslover Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 5 years ago 30 mar 2020, 21:26 GMT-4

Your mesh is too coarse, especially within the coaxial line and inside the waveguide. You need to mesh your problem in sufficient detail to follow the details of the distribution of the fields, in terms of not just phase, but also the typical scale size for amplitude variations.

-------------------
Scientific Applications & Research Associates (SARA) Inc.
www.comsol.com/partners-consultants/certified-consultants/sara
Your mesh is too coarse, especially within the coaxial line and inside the waveguide. You need to mesh your problem in sufficient detail to follow the details of the distribution of the fields, in terms of not just phase, but also the typical scale size for amplitude variations.

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.