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 The information provided may be out of date.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.
2D line constraints in 3D model - Electrostatics
Posted 18.03.2011, 04:27 GMT-4 Geometry, Results & Visualization, Studies & Solvers Version 3.5a 4 Replies
Please login with a confirmed email address before reporting spam
I'm quite a newbie in Comsol, so please forgive me if these questions are not very accurate.
I have developed a 3D model for the study of the electric field distribution over a volume. This volume is a cylinder with a sphere over its top. Once I create the volume, I get it substracted from a cilinder that determines the boundaries I want to work with. When I apply the boundary settings (zero charge/symmetry for the walls of the cylinder, Potential for the top and ground for the bottom geometry) I get the problem solved and the electric field distribution obtained does make sense.
Now, I would like to analyse the same volume with the addition of a 2D vertical line at the top of the sphere. This would represent a very thin element, such as a wire. I have tried to define it as a very thin cylinder and the substracting it the same way as the general study volume, but I have trouble with the meshing of such an element. I guess this is because the difference in size in comparison with the sphere (2.3 m of radius for the sphere vs 3 cm radius for the wire).
If I define a 2D line as I have commented before, I try to set the V constraint in Physics>Equation System>Edge Settings> constr. This would equal V to zero for that edge. However, when I mesh and solve the entire system, the 2D line is not considered and the solution is the same as in the previous case. The soultion I'm seeking would look similar to a lightning rod effect. I guess I'm missing some other constraint that leads Comsol to ommit the 2D line.
Thanks in advance for all your suggestions
Regards,
Javier
Please login with a confirmed email address before reporting spam
indeed in 3D you cannot add in a line/edge that is not related to some boundaries. But what you can do, is to split your sphere into 4 sections and then you get a common line onto which you might apply an edge current
--
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Thanks for this reply Ivar. I have attached to this new post a sample model of what I'm trying to work with. Let's say it's a single electrode to which I need to add a very thin vertical element at the top of the sphere. As I related before, a very thin cylinder will cause meshing problems, and single lines seem to be impossible to add to a 3D model, as you say.
Is there any way to avoid the meshing problems? The steps followed in order to work with the model are as follows:
Case one: complete meshing of the entire volume. Electrode + thin cylinder substracted from the "environment" cylinder. Crash during the meshing process.
Case two: tried to mesh independently points, edges, boundaries and subdomain. Crash in the subdomain meshing.
Any help on this would be kindly appreciated.
Best regards
Javier
Please login with a confirmed email address before reporting spam
I'll try it back later
Regards,
Javier
Please login with a confirmed email address before reporting spam
when you have extreme 1:10'000 ratios the mesher will have problems to converge, there are ways arounf, such as starting with the detaile items in edge/surface mesh mode and then expand, or even to surround a "thin" cylinder with another cylinder some 50-100 bigger in diameter to make an intermediate domain, than one can mesh with a smaller ratio between items. Then mesh from inside and out.
clear you model before you uplaod it, uploading is limited in size
--
Good luck
Ivar