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 The information provided may be out of date.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.

Large mesh with Beam Envelopes module

Please login with a confirmed email address before reporting spam

Dear all,

probably I'm missing something about the use of the Electromagnetic Waves, Beam Envelopes; as I understand well, this module can be used to simulate a long waveguide (some mm or cm long) but without requiring a mesh size smaller than the signal wavelength. I have prepared a simple example of a waveguide that starts with a small width and ends with a bigger width (some kind of taper waveguide); the wave envelope change in this case should be minimal, so I'm expecting that a coarse mesh along the beam propagation direction is enough to describe the wave behaviour; unfortunately this is not the case, and I have to reduce the mesh size to a fraction of wavelength to see the results, as in the typical beam propagation methods.

Any suggestion? Thanks in advance

P.S. The file has the mesh and solutions stripped out in order to reduce the size of the file to the minimum.


4 Replies Last Post 12 mars 2014, 06:48 UTC−4
Walter Frei COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 4 déc. 2013, 17:12 UTC−5
Dear Nicola,

There is one important change you will need to make to your model. In the Beam Envelopes interface, when you specify the wavevector, take the average of the two computed propagation constants: (ewbe.beta_1+ewbe.beta_2)/2

You can now run with a much coarser mesh. You could also use either of two, rather than taking the average. The important thing to keep in mind is that, wherever the local wavevector deviates from the specified wavevector, you will need a finer mesh.

If you want to make this model even smaller, you can also only solve for the out-of-plane electric field (Ez formulation) and also solve for 1/2 of the model with a PMC condition along the centerline.

Best Regards,
Walter
Dear Nicola, There is one important change you will need to make to your model. In the Beam Envelopes interface, when you specify the wavevector, take the average of the two computed propagation constants: (ewbe.beta_1+ewbe.beta_2)/2 You can now run with a much coarser mesh. You could also use either of two, rather than taking the average. The important thing to keep in mind is that, wherever the local wavevector deviates from the specified wavevector, you will need a finer mesh. If you want to make this model even smaller, you can also only solve for the out-of-plane electric field (Ez formulation) and also solve for 1/2 of the model with a PMC condition along the centerline. Best Regards, Walter

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 5 déc. 2013, 09:37 UTC−5
Dear Walter,

thanks a lot, this is the answer I was looking for; now I understand that the beta parameter is a key point of the simulation and it should be chosen close to the "real" propagation constant, which is in most cases unknown; I have obtained similar results with only beta_1, or simply 2*pi*n_core/wavelength, which is a very bare approximation. I'll keep in mind also the other suggestions, even if the submitted example in a really simple structure and I'm planning to use this module in more complex cases.

Best regards.

Nicola
Dear Walter, thanks a lot, this is the answer I was looking for; now I understand that the beta parameter is a key point of the simulation and it should be chosen close to the "real" propagation constant, which is in most cases unknown; I have obtained similar results with only beta_1, or simply 2*pi*n_core/wavelength, which is a very bare approximation. I'll keep in mind also the other suggestions, even if the submitted example in a really simple structure and I'm planning to use this module in more complex cases. Best regards. Nicola

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 18 févr. 2014, 07:27 UTC−5
Hi all,
I'm modeling in 2D two coupled waveguides, by grading properly the refractive index of a single material.
Since the structure is long several wavelengths in the longitudinal direction I'm using the Beam Envelopes interface to solve for the envelope only.
I provided the exact propagation costant since I'm injecting single modes.
However I cannot figure out how to make coarser the mesh in the longitudinal direction, while keeping fine that along the transverse geometry.
Does anybody can help me?
Thank you in advance
Hi all, I'm modeling in 2D two coupled waveguides, by grading properly the refractive index of a single material. Since the structure is long several wavelengths in the longitudinal direction I'm using the Beam Envelopes interface to solve for the envelope only. I provided the exact propagation costant since I'm injecting single modes. However I cannot figure out how to make coarser the mesh in the longitudinal direction, while keeping fine that along the transverse geometry. Does anybody can help me? Thank you in advance

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 12 mars 2014, 06:48 UTC−4
Hi Maria,
Have a look at the Mach-Zehnder Modulator example.
Best regards,
Anton.


Hi Maria, Have a look at the Mach-Zehnder Modulator example. Best regards, Anton.

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.