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 The information provided may be out of date.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.
Why can't I solve for more eigenvalues?
Posted 03.03.2011, 17:44 GMT-5 Studies & Solvers Version 4.0a 3 Replies
Please login with a confirmed email address before reporting spam
I have a 3.5a file that I have opened in 4.0a. The file simulates some PDEs that simulate light traveling around the edge of an axisymmetric disk optical resonator. It returns the mode shape of the light traveling around the rim of that disk, in a cross-section perpendicular to the light propagation direction.
However I can only ever get it to give me four eigenvalues! Since it seems to be unstable (only giving correct modes for certain eigenvalue guesses, and giving spurious modes if you change the eigenvalue by just a few percent), I'd like to solve for TONS of eigenmodes (which I know will take a long time but that's ok), and quickly flick through them afterwards to find the non-spurious (i.e. correct) ones.
When I click Model Builder > Study 1 > Step 1: Eigenvalue 1, it says:
- Desired number of eigenvalues
If you enter anything greater than 4 in that box, it still only returns four eigenvalues! If you enter "3" in that box it also still gives me 4 eigenvalues.
The version 4.0a .mph is attached.
Thanks!
P.S. Don't worry about there being both "Solution 1" and "Solution 2"... the former is just the solution from when I was using this file in 3.5a, then I opened it in 4.0a and solved again, giving Solution 2.
Attachments:
Please login with a confirmed email address before reporting spam
my experience is that the translation (because of the numerous changes) does not really work, so I use imported 3.5 files in V4 only to "look at" and to analyse. Then I rebuild the model fully in v4 (even v4.0 to v4.1 translations, I do not trust) you can export/import the geometry to go faster
--
Good luck
Ivar
Please login with a confirmed email address before reporting spam
The reason for this is that your study setting asking for 20 eigenvalues in the Step 1: Eigenvalue 1 node's settings in this model is overridden by a local setting the Solver Configurations>Solver 1>Eigenvalue Solver 1 node's settings: It specifies 4 eigenvalues. Select "Step 1: Eigenvalue" from the "Defined by study step" list and re-solve to compute 20 eigenvalues and corresponding eigenmodes. The default setting for an Eigenvalue Solver node in new models is to take its settings from the study step.
Best regards,
Magnus Ringh, COMSOL
Please login with a confirmed email address before reporting spam
(Note to developers: It'd be great to have notification when this setting is overridden)