Discussion Forum

Hi Bahar

For integration in COMSOL it all depends, what you want, the units are there to give a good hint.

i.e. in 3D you normally look for total current passing a boundary (2D surface for a 3D model) so you do a surface integration of a current density in [A/m^2] (*dx*dy) which ends up in [A] (note the *dx*dy is implicit in a 2D integration

in 2D and 2D axi you have 2 cases, either the floe is going "out of the paper/screen" then you do a domain integration in 2D (a 2D surface again as above), or you have a current flowing in the screen plane crossing a boundary (a 1D line in 2D). But as COMSOL always calculates in 3D assuming generally by default a 1[m] deep geometry, you current flowing through a boundary in 2D (=1D line) is also in [A/m^2] and when you do a line integration on this boundary you end up with a current density [A/m] (from the [A/m^2](*dx)=[A/m]. and this tells you its "per meter". To get an absolute value you need to do the line integration of [A/m^2]*1[m]*dx = [A]

Now in certain physics you can decide the true thickness and use a different value than the default 1[m], in which case you must use this thickness often referred to by the variable name "d" with the physics prefix

I hope I made myself clear, have fun Comsoling

--
Good luck
Ivar

Dear Ivar

I appreciate you full, and helpful reply. That is exactly the hints that I was looking for.

Best
Bahareh

Hi
that is nice to hear, but you know, you can also try it out with a simple example, i.e. calculate the area of a cylinder in 3D and of a cylinder in 2D-axi view and compare. Some time you discover sum "bug" to like that ;)

--
Good luck
Ivar

Thanks I will follow your advice :)

Dear Ivar and Bhar,

I have raised a question today in these lines here:

www.comsol.com/community/forums/general/thread/26157/

Could you please give me your inputs ?

Thanks
Murali

Hi Bahare

Have you ever tried comsol in 3D? or ever simulated with frequency? if so Id be glad to have your help



Best Regards