Mesh & Questions

How to use openEMS. Discussion on examples, tutorials etc

Moderator: thorsten

Post Reply
aris12
Posts: 1
Joined: Sun 04 Jul 2021, 20:05

Mesh & Questions

Post by aris12 » Wed 01 Sep 2021, 11:15

Hi all, my name is Aris and i study electronics engineer. I learn how to use openEMS for 3 weeks (I hadn't experience in EM solve) and I understand that a very good job has been done, so kudos to developers 8-) .
I used it for 3 experiments:
a) make antenna dipole.
b) make antenna dipole with stl (with FreeCAD).
c) make antenna dipole with pocketQube (satellite) with stl.
The first thing i wanted to see was if i had the same results from (a) and (b), and I concluded that it is not because of mesh. I used DetectEdges for stl but it was wrong, DetectEdges isn't for stl format. I searched for how I can generate Mesh and i have found MeshLab and Aegmesher but without success. I think that I have not understood how to make mesh. Ιn short I try to find a way to get away from meshing :lol: .
For pocketQube i design FR4 and ground plane but i think it is also correct to put only ground because around of satellite it has solar panel, what do you say?
Τhe reason we deal with the dipole and pocketQube is because i'm contributor in Libre Space Foundation and i want to see with openEMS the radiation pattern from satellite Qubik ( see https://gitlab.com/librespacefoundation/qubik ). I will gradually upload how I go so that new people who want to use it can also see and in the end i upload here the finish code & in GitLab. Also i will create account in wiki and i will put all this as tutorial (when there will be full understanding) and documentation in Overview for some function.
Cheers
Attachments
PocketQube_Antenna.zip
If anyone want to see the code don't forget to change filename from STL .
(1.39 MiB) Downloaded 85 times

thorsten
Posts: 1475
Joined: Mon 27 Jun 2011, 12:26

Re: Mesh & Questions

Post by thorsten » Thu 02 Sep 2021, 20:08

Hi,

well meshing is always the most important and in many cases the most difficult to get right.

But if you understand the ideas behind it, it is not that hard.
You just can e.g. manually define mesh lines in all directions. For a very thin dipole a single line through the length and lines at the end and port (in the other directions) should be sufficient. Just look at the dipole example and what the resulting mesh looks like.

If the real antenna experiences a ground the simulation needs to model that of course.

regards
Thorsten

Post Reply