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.
Meshing Across Many Orders of Magnitude
Posted 20.02.2015, 09:20 GMT-5 MEMS & Nanotechnology, RF & Microwave Engineering, MEMS & Piezoelectric Devices, Mesh Version 5.0 1 Reply
Please login with a confirmed email address before reporting spam
Hello, I am quite new to COMSOL and I am having difficulty creating a mesh for my system which has components which scale from 60[mm]-[150nm]. I started with two coupled cavities which are [mm] scale with a small dielectric chip inside one of the cavities which is 450[um] thick and a few [mm] wide and tall. Everything was fine up to this point.
Now I am trying to add a square metal element 150[nm] thick and 500[um] sides onto the substrate. I first build a quadmesh onto the metal square with 20[um] elements and sweep it through the 150nm thickness. I then do a quadmesh on the substrate surface which starts out 20[um] minimum element size to [1mm] maximum with growth rate of 50. sweeping this and converting to tets all works fine except sometimes I get warnings
"Face is (or has a narrow region that is) much smaller than the specified minimum element size."
This is just a warning and not a breakout error of the meshing algo and I don't think its that big a deal. Now, when I try to do a tetra mesh on the outside cavity, I always get the error
"Failed to respect boundary element edge on geometry face."
and the mesh fails. The face it points to is usually the substrate face with the metal pad on it, though sometimes its the metal square face and sometimes its the thickness face of the metal square. This is somehow related to the fact that the inner substrate/metal contact mesh is very fine compared to the much coarser cavity mesh. I have tried to put an intermediate vaccum cube between the substrate and the cavity walls to try and slowly grow the mesh size outwards, but I still always get the same error, even when the intermediate mesh has identical mesh element size as the substrate mesh size. I searched through the internet for a solution but found nothing that worked. Any help would be greatly appreciated.
Now I am trying to add a square metal element 150[nm] thick and 500[um] sides onto the substrate. I first build a quadmesh onto the metal square with 20[um] elements and sweep it through the 150nm thickness. I then do a quadmesh on the substrate surface which starts out 20[um] minimum element size to [1mm] maximum with growth rate of 50. sweeping this and converting to tets all works fine except sometimes I get warnings
"Face is (or has a narrow region that is) much smaller than the specified minimum element size."
This is just a warning and not a breakout error of the meshing algo and I don't think its that big a deal. Now, when I try to do a tetra mesh on the outside cavity, I always get the error
"Failed to respect boundary element edge on geometry face."
and the mesh fails. The face it points to is usually the substrate face with the metal pad on it, though sometimes its the metal square face and sometimes its the thickness face of the metal square. This is somehow related to the fact that the inner substrate/metal contact mesh is very fine compared to the much coarser cavity mesh. I have tried to put an intermediate vaccum cube between the substrate and the cavity walls to try and slowly grow the mesh size outwards, but I still always get the same error, even when the intermediate mesh has identical mesh element size as the substrate mesh size. I searched through the internet for a solution but found nothing that worked. Any help would be greatly appreciated.
1 Reply Last Post 24.02.2015, 04:25 GMT-5