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 The information provided may be out of date.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.

Magnetic field line

Please login with a confirmed email address before reporting spam

Hi every body,

I have problem with magnetic Field line (Streamline)
It does not look good
the lines stay in the Space dont go out the sapce and that Theoretically not real

I hope that some one know that to help me

With my Regards

Maher

3 Replies Last Post 25 giu 2012, 04:35 GMT-4

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 22 giu 2012, 07:23 GMT-4
Hi!

I use for field lines the isolines with magnetic vector potential Az; this will give a better image.
Since you use infinite elements for large ring, the outer border is the "end of the universe".
So the loop is correct: field lines will close in infinity.

Ralf
Hi! I use for field lines the isolines with magnetic vector potential Az; this will give a better image. Since you use infinite elements for large ring, the outer border is the "end of the universe". So the loop is correct: field lines will close in infinity. Ralf

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 22 giu 2012, 07:46 GMT-4
Hi

a few comments on your model:

- Your infinite area should have a radius of 1/5 to 1/10 of the total air radius, yours is much bigger.
- its better to use the "circle advance tab layer to generate this INF element region, and have the mesher use a mapped mesh with at least 5 elements radially across
- use rather a cylindrical INF element type AND set the centre such that your model is in the middle and that the centre coordinates of your circle are those of the INF element circle
- Parameter and variable names, try avoid single letter names as these are mostly already used by COMSOL, and will not necessarily flagged as an error, as COMSOL allows us to change the internal variables values, but COMSOL cannot tell if it is on purpose or by "error" (I'm thinking particularly about r, w, H)

Why mix a frequency domain AND a stationary like that without any intermediate STORE solution nor any links between them, both overwriting the same Data Set on output ? You cannot tell where the results come from (probably the stationary, as its the last one.

make either separate studies, and one solver per study, or add a "store solution" in between both (if COMSOL hasn't done it for you) and start with a stationary to use that as initial conditions for the frequency domain perhaps ?)

--
Good luck
Ivar
Hi a few comments on your model: - Your infinite area should have a radius of 1/5 to 1/10 of the total air radius, yours is much bigger. - its better to use the "circle advance tab layer to generate this INF element region, and have the mesher use a mapped mesh with at least 5 elements radially across - use rather a cylindrical INF element type AND set the centre such that your model is in the middle and that the centre coordinates of your circle are those of the INF element circle - Parameter and variable names, try avoid single letter names as these are mostly already used by COMSOL, and will not necessarily flagged as an error, as COMSOL allows us to change the internal variables values, but COMSOL cannot tell if it is on purpose or by "error" (I'm thinking particularly about r, w, H) Why mix a frequency domain AND a stationary like that without any intermediate STORE solution nor any links between them, both overwriting the same Data Set on output ? You cannot tell where the results come from (probably the stationary, as its the last one. make either separate studies, and one solver per study, or add a "store solution" in between both (if COMSOL hasn't done it for you) and start with a stationary to use that as initial conditions for the frequency domain perhaps ?) -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 25 giu 2012, 04:35 GMT-4
Hello,

Thank you very much for ypur reply and for your Useful comments
it was some additional option as a frequency domain and stationary and a big infinte Area for test
for every thing thank you very much

With my Regard
Maher
Hello, Thank you very much for ypur reply and for your Useful comments it was some additional option as a frequency domain and stationary and a big infinte Area for test for every thing thank you very much With my Regard Maher

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.