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.
Issues at connecting one coarser mesh and an extremely thin mesh
Posted 05.04.2019, 09:27 GMT-4 Low-Frequency Electromagnetics, Mesh Version 5.3a 3 Replies
Please login with a confirmed email address before reporting spam
Hi everyone,
Here is my problem, I got a really thin (1mm) frame (and 1.5mx0.5m) inside a big box (2mx2mx1m) and I want to have an extremelly fine mesh in the frame and a coarser one outside. Inside the frame I want exactly 10 layers of prisms, the thickness of each layer is therefore 0.1mm. Please look at the fille attached !
I first tried to use a surface mesh (free triangular mesh) on one face of the frame and swept it through the frame with a distribution of 10 layers, the remaining domains were meshed using free tetrahedral with a coarser mesh. But unfortunately I had some many errors saying that the boundaries of the two meshes couldn't be connected or that there was an error in the meshing domain...
Then I did exactly the same thing but using the boundary layer feature on the frame with 10 layers of 0.1mm and the boundaries being only one face (I also did 5 layers with the two faces) and I got the same kind of error.
Initially I wanted to mesh so thin in the frame to observe the skin effect (I'm working at 85kHz with conducitivites up to 10^7 S/m)...
Does anyone know how to do this ?
Cheers !
Attachments:
Please login with a confirmed email address before reporting spam
Hi Paul,
I modified your mesh so it meshes now. It still throws a warning because the aspect ratio in the thin layer is still really bad. You may want to adjust that, but may well be it doesn't matter.
Good luck Edgar
-------------------Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
Attachments:
Please login with a confirmed email address before reporting spam
Hi Edgar, Thanks for trying but unfortunately I'm on COMSOL 5.3a and it's not working with your 5.4 file :( ! Cheers
Please login with a confirmed email address before reporting spam
Hi Paul,
I should have looked to your version information. Seems the mesher really improved in 5.4. I implemented the same sequence in 5.3a and it fails. I played a little with the size settings, but no success.
You may consider helper boundaries, i.e a sequence of boxes around the sheet. This may help the mesher.
I send the model with the sequence that works in 5.4.
Good luck Edgar
-------------------Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
Attachments: