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.

Refectance in microwaveguide

Please login with a confirmed email address before reporting spam


I am trying to use the 3D RF modules to calculate the reflection of a T junction waveguide. The first step was to calculate the Reflectance of the T Junction waveguide filled with dielectric constant and the results are fine.

The second step is to calculating the reflectance of TiO2 semiconductor, with the thickness of 13 micro meter between two 1mm thick glass. My problem is I cannot generate the mesh on the structure and I think its because of the thickness of the TiO2. As the TiO2 is between two layer which are far more thicker that it can it be the reason which, I cannot generate the mesh? I am using the mesh as Normal and I am also trying to change the meshing to see if it will make difference, (from normal to coarse, extra coarse and extremely coarse

To try to solve the problem I have used 130 micron thickness of the TiO2 and the meshing was fine, but the results of the reflectance which I am getting is not the same as I am expecting. I am changing the conductivity of the TiO2 from 1e-9 to 1e9, so I am expecting to get the similar reflectance value at the first few values from 1e-9 and the last few values towards 1e9. From what I have realized the model behave well at high conductivity and its not at low conductivity.
My question is that is it because of the meshing problem I experienced above or there is another explanation?

Thanks
Eric

3 Replies Last Post 8 mar 2010, 09:31 GMT-5
Robert Koslover Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 7 mar 2010, 20:28 GMT-5
Yes, this sounds like a meshing problem.

Any chance that you are modeling an H-plane T? If so, depending on the geometry of your inserted parts (the glass and TiO2) you may be able to run the simulation in 2D. If so, you can use a much finer mesh.

Alternatively, you need to get more skilled with meshing. Most importantly, you can mesh your problem one subdomain at a time, rather than all at once. Go to Mesh --> Free Mesh Parameters and choose the subdomain tab. Set the max element size for each subdomain appropriately. On the advanced tab, you can choose different scale factors in x,y, and z dimensions. Again, choose thoughtfully and do some experimenting with those settings, going back to the subdomain tab and choosing Mesh Selected, not just Remesh, so you can mesh one subdomain at a time. You may have to clear the whole mesh (choose File --> Reset Model) and start meshing over again, in order to fix any mistakes.

Good luck.
Yes, this sounds like a meshing problem. Any chance that you are modeling an H-plane T? If so, depending on the geometry of your inserted parts (the glass and TiO2) you may be able to run the simulation in 2D. If so, you can use a much finer mesh. Alternatively, you need to get more skilled with meshing. Most importantly, you can mesh your problem one subdomain at a time, rather than all at once. Go to Mesh --> Free Mesh Parameters and choose the subdomain tab. Set the max element size for each subdomain appropriately. On the advanced tab, you can choose different scale factors in x,y, and z dimensions. Again, choose thoughtfully and do some experimenting with those settings, going back to the subdomain tab and choosing Mesh Selected, not just Remesh, so you can mesh one subdomain at a time. You may have to clear the whole mesh (choose File --> Reset Model) and start meshing over again, in order to fix any mistakes. Good luck.

Robert Koslover Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 7 mar 2010, 20:31 GMT-5
I should also have asked: "Why are you "changing the conductivity of the TiO2 from 1e-9 to 1e9" ?? Either you know it's conductivity or you don't, right? So that seems strange to me.
I should also have asked: "Why are you "changing the conductivity of the TiO2 from 1e-9 to 1e9" ?? Either you know it's conductivity or you don't, right? So that seems strange to me.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 8 mar 2010, 09:31 GMT-5

I should also have asked: "Why are you "changing the conductivity of the TiO2 from 1e-9 to 1e9" ?? Either you know it's conductivity or you don't, right? So that seems strange to me.


Thanks for the information, I will try to do that and hope it will solve my problem.

The explanation to your question.
I am changing the conductivity to know what is the limit value of (low) dark conductivity where the reflectance is independent of conductivity. My calculation are based on the understanding that at low conductivity , the reflectance is independent of the conductivity. I am expecting that after the limit value of the conductivity the reflectance will start to change.
[QUOTE] I should also have asked: "Why are you "changing the conductivity of the TiO2 from 1e-9 to 1e9" ?? Either you know it's conductivity or you don't, right? So that seems strange to me. [/QUOTE] Thanks for the information, I will try to do that and hope it will solve my problem. The explanation to your question. I am changing the conductivity to know what is the limit value of (low) dark conductivity where the reflectance is independent of conductivity. My calculation are based on the understanding that at low conductivity , the reflectance is independent of the conductivity. I am expecting that after the limit value of the conductivity the reflectance will start to change.

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.