Menu

Discussion Forum

New Discussion
Filter

Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.

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.

2D axisymmetric numeric port settings

Posted May 23, 2011, 8:25 a.m. EDT RF & Microwave Engineering, Geometry Version 4.1, Version 4.3b 6 Replies

Boris Le Feber
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam


Dear all,

I am trying to model EM wave propagation through a tapered step-index fiber. Because of the large size of my geometry and the symmetry of the problem I use the 2D axisymmetric settings. My fiber has a numeric port on one end and a small opening at the other. My problem is that the mode index and the e-field distribution at the port side i find with respectively the boundary mode analysis and the frequency domain study do not match the results I find when I run a mode analysis study in regular 2D on the port side of the geometry.
For example in the axisymmetric case I find a lower effective mode index from the boundary mode analysis study and I find a much more spread out mode with. Does anybody know how to setup a numeric port in the 2d axisymmetric case? Or how to properly compute the effective mode index for what is basically a line in the 2 axisymmetric case using the boundary mode analysis study?

Cheers!

Boris
6 Replies Last Post Jan 12, 2014, 12:04 p.m. EST
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago May 23, 2011, 8:57 a.m. EDT
Hi

I do not follow you when you say that the 2D-axi and the standard 2D differs, for me this is normal, as 2D has an "infinite" or "per meter " depth along Z, while the 2D axis has a loop length worth 2*pi*r. This is a radically different topology , hence different mode shapes

normally the 2D-axi takes care of the r=0 (by multiplying many of the formulas by "r" and then normalising back the results. But one should not forget that a 2D-axi symmetry means really that the axis remains fixed and that all phenomena are radially symmetric

--
Good luck
Ivar
Hi I do not follow you when you say that the 2D-axi and the standard 2D differs, for me this is normal, as 2D has an "infinite" or "per meter " depth along Z, while the 2D axis has a loop length worth 2*pi*r. This is a radically different topology , hence different mode shapes normally the 2D-axi takes care of the r=0 (by multiplying many of the formulas by "r" and then normalising back the results. But one should not forget that a 2D-axi symmetry means really that the axis remains fixed and that all phenomena are radially symmetric -- Good luck Ivar
Boris Le Feber
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago May 23, 2011, 9:58 a.m. EDT


Dear Ivar,

Apologies for not phrasing the problem clearly and thanks for your fast reply! In the regular 2d case I study a circle that should have the same dimensions as the the port that I defined as a line from r=0 in the 2D axisymmetric case (if I unstand the 2D axisymmetry correcly).
Do I not correctly assume that applying a mode analysis to the regular 2D circle should yield the same effective mode index as the a boundary mode analysis to this 'line' from r=0 in the 2D axisymmetric case?

Boris
Dear Ivar, Apologies for not phrasing the problem clearly and thanks for your fast reply! In the regular 2d case I study a circle that should have the same dimensions as the the port that I defined as a line from r=0 in the 2D axisymmetric case (if I unstand the 2D axisymmetry correcly). Do I not correctly assume that applying a mode analysis to the regular 2D circle should yield the same effective mode index as the a boundary mode analysis to this 'line' from r=0 in the 2D axisymmetric case? Boris
Boris Le Feber
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago May 23, 2011, 10:19 a.m. EDT
For clarity, these are the geometries I use (in the 2D axisymmetric case my port is at z=0)
For clarity, these are the geometries I use (in the 2D axisymmetric case my port is at z=0)

Attachments:

Alexandros Kapsalis
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Feb 28, 2012, 6:31 a.m. EST
Shouldn't the 'dif' half circles on the top of your 2D-axi actually be rectangle-shaped? Because this revolves around the r=0 axis and the circular fiber profile would be created automatically...
Shouldn't the 'dif' half circles on the top of your 2D-axi actually be rectangle-shaped? Because this revolves around the r=0 axis and the circular fiber profile would be created automatically...
Clenilson Rodrigues da Silveira
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jul 19, 2012, 3:09 p.m. EDT
Hi! I'm a new user of Comsol. I got an evaluation trial license. I need to model a tapered fiber. So firstly, I decided to run simple examples contained in model library. I ran RF/step_index_fiber. I tried to follow the same steps contained in this example but in 2D axisymmetric and 3D model. But unfortunately I did not get the same results. In 3D model I couldn't get any distribution of electrical field. I realized that it is a common problem but anyone was able to appropriately answer those kind of questions. Is it a problem related to excitation? Thanks in advance.
Hi! I'm a new user of Comsol. I got an evaluation trial license. I need to model a tapered fiber. So firstly, I decided to run simple examples contained in model library. I ran RF/step_index_fiber. I tried to follow the same steps contained in this example but in 2D axisymmetric and 3D model. But unfortunately I did not get the same results. In 3D model I couldn't get any distribution of electrical field. I realized that it is a common problem but anyone was able to appropriately answer those kind of questions. Is it a problem related to excitation? Thanks in advance.
Wang ZQ
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jan 12, 2014, 12:04 p.m. EST
I've come across exactly the same question just as you did. I want to simulate a model that a continuous wave pass through a fiber taper with different diameters at the two ends. Firstly, I want to simulate a Single Mode Fiber as a test, but it doesn't work either. Have you solve it yet?
I've come across exactly the same question just as you did. I want to simulate a model that a continuous wave pass through a fiber taper with different diameters at the two ends. Firstly, I want to simulate a Single Mode Fiber as a test, but it doesn't work either. Have you solve it yet?

Attachments:

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