Menu

Discussion Forum

New Discussion
Filtrer

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 The information provided may be out of date.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.

Can't integrate over a cross-section!

Posted 18 juil. 2012, 21:37 UTC−4 Low-Frequency Electromagnetics, RF & Microwave Engineering Version 4.2 2 Replies

Evan Zalys-Geller
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

I'd like to be able to integrate over an arbitrary cross section of my model. I'm trying to integrate the poynting vector over a hole in a cavity after doing modal analysis. I'm coupling a cavity to a piece of microstrip that sits above. I'm running into some difficulty, though. My first attempt was just just going to derived values->surface integration, and then attempting to click on the hole itself... but there's really no surface defined IN the hole, so I can't select it. So instead, I go and create a new work plane, and create a circular surface inside the hole, and figure I could integrate over this. No go... the circular surface I made disappears when I switch over to define the surface integral! As a last resort, I create a 2D slice under Data Sets of the relevant cross section, and figure I'll just integrate over that, since the power flux I'm interested in is only nonzero in the region of interest... so it works out fine. Unfortunately, when I hit integrate, it just gives NaN! Yaaargh! I tried just picking the max value over the defined 2D slice, and this works just fine... Any ideas what might be going on?

The poynting vector I have defined as real(-Ey*conj(emw.Hx) + Ex*conj(emw.Hy))... I can plut this on a 2D plot surface just fine... comsol just won't integrate over it! Also, I'd love to know why the surface defined inside the hole disappears!
2 Replies Last Post 13 déc. 2012, 07:23 UTC−5
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 19 juil. 2012, 01:37 UTC−4
Hi

I'm not sure I caught you fully, but I can confirm that I often add internal domains / boundaries only for integration purpose, or to use as "easy pick" probe entities, and this works OK. I often use this for splitting up complex domains to improve the meshing control too

--
Good luck
Ivar
Hi I'm not sure I caught you fully, but I can confirm that I often add internal domains / boundaries only for integration purpose, or to use as "easy pick" probe entities, and this works OK. I often use this for splitting up complex domains to improve the meshing control too -- Good luck Ivar
Stefan Jovanovic
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 13 déc. 2012, 07:23 UTC−5

I'd like to be able to integrate over an arbitrary cross section of my model. I'm trying to integrate the poynting vector over a hole in a cavity after doing modal analysis. I'm coupling a cavity to a piece of microstrip that sits above. I'm running into some difficulty, though. My first attempt was just just going to derived values->surface integration, and then attempting to click on the hole itself... but there's really no surface defined IN the hole, so I can't select it. So instead, I go and create a new work plane, and create a circular surface inside the hole, and figure I could integrate over this. No go... the circular surface I made disappears when I switch over to define the surface integral! As a last resort, I create a 2D slice under Data Sets of the relevant cross section, and figure I'll just integrate over that, since the power flux I'm interested in is only nonzero in the region of interest... so it works out fine. Unfortunately, when I hit integrate, it just gives NaN! Yaaargh! I tried just picking the max value over the defined 2D slice, and this works just fine... Any ideas what might be going on?

The poynting vector I have defined as real(-Ey*conj(emw.Hx) + Ex*conj(emw.Hy))... I can plut this on a 2D plot surface just fine... comsol just won't integrate over it! Also, I'd love to know why the surface defined inside the hole disappears!


I am really not expert, but shouldn't you put emw.Ey, emw.Ex.... so same as emw.Hy, emw.Hx...

That's the first thing.

Second thing, did you try to express H component from E.

If you use for example plane wave, maybe you should put equations:

variableHx = (mu0_const*emw.Ezy-emw.Eyz)/i/emw.omega and for Hy, Hz the similar....

I mean, I am really not expert, but maybe you should try this.

Especially I mentioned first about emw.Ey, emw.Ex...



Good luck


Stef





[QUOTE] I'd like to be able to integrate over an arbitrary cross section of my model. I'm trying to integrate the poynting vector over a hole in a cavity after doing modal analysis. I'm coupling a cavity to a piece of microstrip that sits above. I'm running into some difficulty, though. My first attempt was just just going to derived values->surface integration, and then attempting to click on the hole itself... but there's really no surface defined IN the hole, so I can't select it. So instead, I go and create a new work plane, and create a circular surface inside the hole, and figure I could integrate over this. No go... the circular surface I made disappears when I switch over to define the surface integral! As a last resort, I create a 2D slice under Data Sets of the relevant cross section, and figure I'll just integrate over that, since the power flux I'm interested in is only nonzero in the region of interest... so it works out fine. Unfortunately, when I hit integrate, it just gives NaN! Yaaargh! I tried just picking the max value over the defined 2D slice, and this works just fine... Any ideas what might be going on? The poynting vector I have defined as real(-Ey*conj(emw.Hx) + Ex*conj(emw.Hy))... I can plut this on a 2D plot surface just fine... comsol just won't integrate over it! Also, I'd love to know why the surface defined inside the hole disappears! [/QUOTE] I am really not expert, but shouldn't you put emw.Ey, emw.Ex.... so same as emw.Hy, emw.Hx... That's the first thing. Second thing, did you try to express H component from E. If you use for example plane wave, maybe you should put equations: variableHx = (mu0_const*emw.Ezy-emw.Eyz)/i/emw.omega and for Hy, Hz the similar.... I mean, I am really not expert, but maybe you should try this. Especially I mentioned first about emw.Ey, emw.Ex... Good luck Stef

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