Menu

Discussion Forum

New Discussion
Filtern

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 The information provided may be out of date.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.

Recovering desired eigenfrequencies

Posted 20.12.2012, 11:38 GMT-5 RF & Microwave Engineering, Results & Visualization Version 4.3a 6 Replies

Chris Curwen
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Hi,

I'm performing an eigenfrequency analysis of a model in the Comsol 4.3a RF module. My problem is that, though I ask for only 1 eigenfrequency, Comsol ALWAYS gives me 2 eigenfrequencies: one purely real, and one complex. This is frustrating me because when I try to probe the eigenfrequency, it always probes the wrong eigenfrequecy (the complex one), so the only way I can extract the data for a 50 point sweep is to manually go through and extract the proper eigenfreq. The correct eigenfreq is always at the bottom of the dropdown menu, which in previous versions of Comsol always corresponded to the first solution.

I figure there are two solutions to my problem:

1. Figure out how to make Comsol only give me 1 eigenfrequency (should work if the bottom solution does still correspond to the first one it found).

2. Tell Comsol to probe eigenfrequency 1.

I can't seem to figure out how to execute either of these solutions. I'm attaching a couple screen shots of my model builder to help explain.

Any advice would be greatly appreciated.

Thanks, Chris
6 Replies Last Post 24.06.2015, 18:31 GMT-4
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 20.12.2012, 13:55 GMT-5
Hi

A complex eigenfrequency , for me, means also dephasing, so one need to take the norm for the frequency or transfor it into a phasor amplitude and phase

--
Good luck
Ivar
Hi A complex eigenfrequency , for me, means also dephasing, so one need to take the norm for the frequency or transfor it into a phasor amplitude and phase -- Good luck Ivar
Chris Curwen
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 20.12.2012, 14:46 GMT-5
Thanks Ivar,

I understand the meaning of a complex eigenfrequency, my problem is simply that I ask for 1 eigenfrequency, yet comsol insists on giving me 2 eigenfrequencies, which is a problem for me. I don't know why comsol is doing this, but there is another solution to the problem. Let me phrase it this way:

Right now I have a probe that is set to probe the expression "emw.omega" (see attached screenshot), however, comsol solves for multiple eigenfrequencies, so it is finding several emw.omega values. If I probe emw.omega, it only probes the most recent emw.omega (ex. if I solve for 6 eigenfreq, it will output the 6th eigenfreq). In previous versions of comsol, if you wanted to probe say the 3rd eigenfrequency, you would simply probe "emw.omega(3)". Comsol 4.3a rejects this, so my question is, do you know how to tell the probe which solution to observe?

Thanks,
Chris
Thanks Ivar, I understand the meaning of a complex eigenfrequency, my problem is simply that I ask for 1 eigenfrequency, yet comsol insists on giving me 2 eigenfrequencies, which is a problem for me. I don't know why comsol is doing this, but there is another solution to the problem. Let me phrase it this way: Right now I have a probe that is set to probe the expression "emw.omega" (see attached screenshot), however, comsol solves for multiple eigenfrequencies, so it is finding several emw.omega values. If I probe emw.omega, it only probes the most recent emw.omega (ex. if I solve for 6 eigenfreq, it will output the 6th eigenfreq). In previous versions of comsol, if you wanted to probe say the 3rd eigenfrequency, you would simply probe "emw.omega(3)". Comsol 4.3a rejects this, so my question is, do you know how to tell the probe which solution to observe? Thanks, Chris

Attachments:

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 20.12.2012, 15:13 GMT-5
Hi

have you tries with the with(3,...) operator ? (you also have the at() for time series) there are many new operators in v4.3a, check the help on "operators". This might give you the solution. I do not know why you getr several modes, the best is to ask support for such things ;)

--
Good luck
Ivar
Hi have you tries with the with(3,...) operator ? (you also have the at() for time series) there are many new operators in v4.3a, check the help on "operators". This might give you the solution. I do not know why you getr several modes, the best is to ask support for such things ;) -- Good luck Ivar
Xingsheng Luan
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 13.12.2013, 12:14 GMT-5
Hi, I meet a similar problem. In my optimization process, I would like to maximize one cavity mode's quality factor. But since there is a neighbor mode has higher quality factor. Comsol always optimizes the neighbor mode instead of the one I desired. I also tried to limit the number of eigenfrequency solved for(down to 1), to exclude that neighbor mode. But the situation is Comsol always give me solutions more than I want, making my optimization process failed. Could you tell me how you solved that problem later? Thank you.

Best,
Xingsheng
Hi, I meet a similar problem. In my optimization process, I would like to maximize one cavity mode's quality factor. But since there is a neighbor mode has higher quality factor. Comsol always optimizes the neighbor mode instead of the one I desired. I also tried to limit the number of eigenfrequency solved for(down to 1), to exclude that neighbor mode. But the situation is Comsol always give me solutions more than I want, making my optimization process failed. Could you tell me how you solved that problem later? Thank you. Best, Xingsheng
Pierre Allain
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 9 years ago 27.02.2015, 08:28 GMT-5

Hi, I meet a similar problem. In my optimization process, I would like to maximize one cavity mode's quality factor. But since there is a neighbor mode has higher quality factor. Comsol always optimizes the neighbor mode instead of the one I desired. I also tried to limit the number of eigenfrequency solved for(down to 1), to exclude that neighbor mode. But the situation is Comsol always give me solutions more than I want, making my optimization process failed. Could you tell me how you solved that problem later? Thank you.

Best,
Xingsheng


Dear Xingsheng,

I also have a similar problem. Did you solve your problem?

I am trying to optimize the geometry o my structure in eigenfrequencies analysis to maximize the Q factor.

But like you, some junk neibour modes have higher quality factor and I don't want them. I found a probe result (integration of displacement on one of my surface) that can discriminate the good mode from the junk mode.

Do you know how to implement this displacement criteria in my optimization by any chance?

I use strucutral mechancis and eigenfrequency analysis.

Thanks

Pierre
[QUOTE] Hi, I meet a similar problem. In my optimization process, I would like to maximize one cavity mode's quality factor. But since there is a neighbor mode has higher quality factor. Comsol always optimizes the neighbor mode instead of the one I desired. I also tried to limit the number of eigenfrequency solved for(down to 1), to exclude that neighbor mode. But the situation is Comsol always give me solutions more than I want, making my optimization process failed. Could you tell me how you solved that problem later? Thank you. Best, Xingsheng [/QUOTE] Dear Xingsheng, I also have a similar problem. Did you solve your problem? I am trying to optimize the geometry o my structure in eigenfrequencies analysis to maximize the Q factor. But like you, some junk neibour modes have higher quality factor and I don't want them. I found a probe result (integration of displacement on one of my surface) that can discriminate the good mode from the junk mode. Do you know how to implement this displacement criteria in my optimization by any chance? I use strucutral mechancis and eigenfrequency analysis. Thanks Pierre
AC
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 9 years ago 24.06.2015, 18:31 GMT-4
Comsol gives additional eigen-frequencies when there is no extra computational effort by the solver.
Comsol gives additional eigen-frequencies when there is no extra computational effort by the solver.

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