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 The information provided may be out of date.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.
Plotting an antenna's far field
Posted 6 giu 2010, 11:47 GMT-4 RF & Microwave Engineering Version 4.0 6 Replies
Please login with a confirmed email address before reporting spam
Using COMSOL4.0 I modeled an antenna using the emw-Interface and the "Far Field Calculation"-node.
My far-field variable is called "Efar", as in the documentation, the equation view shows emw.Efar{x,y,z} and emw.normEfar{dB}.
From the solution I do a selection of the outer sphere boundary, the auto-generated surface plot shows the near field, the default expression is: "emw.normE". When changing that to "emw.normEfar" I get the error: "cannot evaluate expression: `mod1.emw.normEfar`: Failed to eaulate variable normEfar". Changing the expression to "emw.normEfar(x,y,z)" gives me "Unknown function normEfar". -- But "emw.Efarx(x,y,z)" works, just not the norm/normdB-variants.
The other problem I run into: the plot takes literally hours to compute. It seems to me, as if the far field was not part of the solution, and is being re-calculated every time. (And not just for refinement: plotting the far field not in 3D but on a plane on a rough circle of ~36 points takes 5min)
Is there a way to have the far field calculated at a set of angles (for example using the vertices of the outer sphere's mesh for directions), cache that result and plot in in a few seconds?
This wouldn't be a huge problem, if COMSOL wouldn't re-plot at the drop of a hat: adding a new element to a plot, for example an arrow surface to the existing surface, yields an automatic re-plot of the surface!)
I guess there is a node I can add to force evaluation and storage of the far-field?
Sadly, the "Far Field Postprocessing" chapter doesn't mention it, and neither do the tutorial examples, because they are simple enough to not depend on this…
Thanks for any pointers!
Please login with a confirmed email address before reporting spam
I agree that "ages" to calculate is not nice, but I do not understand why either, apart that I believe far field is not part of the solution by default and is recalculated on demand. For me this seems that you are hitting a particular issue that is not fully debugged yet in V4, ideally remake it in 3.5 and check. In anycase such problems I typically send to support, documented with the model so they can check. They have always provided me with very good support :)
I havent had time yet to test ACDC and far field calculations in V4, so I cannot tell more
Good luck
Ivar
Please login with a confirmed email address before reporting spam
The problem is, I don't have access to 3.5, only 4.0.
Is there a way to force storage?
For example, evaluate all far fields at a sphere's vertices into a table and plot that table?
(or even: export it, to plot in another software. I haven't figured out how to automatically evaluate arbitrary points [like: far field in 5° steps] and store them in a file)
Please login with a confirmed email address before reporting spam
i really do not know, for me typically a issue for support, on the oter hand perhaps its usefull to work with 3.5a, still some weeks ?
Good luck
Ivar
Please login with a confirmed email address before reporting spam
I have the same problem about the far field plot in Comsol 4: it takes a lot of time.
Any progress in the solution of this problem?
Please login with a confirmed email address before reporting spam
Please login with a confirmed email address before reporting spam