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.
Basic capacitance simulation: non physical variation of results with minor modification of mesh
Posted Apr 13, 2012, 5:08 a.m. EDT 2 Replies
Please login with a confirmed email address before reporting spam
Hi all,
I am trying to simulate capacitance between two electrodes with the electrostatics module of Comsol 4.2a using a model succinctly described in the attached picture.
I want later to investigate the influence of a "trench" (rectangle R in the picture) in material 2, i.e. the rectangle R will have the properties of material 1.
Before doing this, I checked what happen when I insert the rectangle R in the model but still apply the properties of material 2 to it. And here comes the unphysical results: the predicted capacitance value vary (quite importantly) with the height d of the rectangle R. This does not make sense from a physical point of view . I understand that the mesh is modified each time that the value of d is changed and that this discrepancy may be related to that. But this is very annoying to me as I feel that I will not be able to make the difference between numerical artifacts and real capacitance variation when I will investigate the influence of trenches in material 2.
Any suggestion to solve this issue would be appreciated.
Thanks
I am trying to simulate capacitance between two electrodes with the electrostatics module of Comsol 4.2a using a model succinctly described in the attached picture.
I want later to investigate the influence of a "trench" (rectangle R in the picture) in material 2, i.e. the rectangle R will have the properties of material 1.
Before doing this, I checked what happen when I insert the rectangle R in the model but still apply the properties of material 2 to it. And here comes the unphysical results: the predicted capacitance value vary (quite importantly) with the height d of the rectangle R. This does not make sense from a physical point of view . I understand that the mesh is modified each time that the value of d is changed and that this discrepancy may be related to that. But this is very annoying to me as I feel that I will not be able to make the difference between numerical artifacts and real capacitance variation when I will investigate the influence of trenches in material 2.
Any suggestion to solve this issue would be appreciated.
Thanks
Attachments:
2 Replies Last Post Apr 13, 2012, 10:27 a.m. EDT
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi Michael Renaud,
The answer is in your question, I think that maybe you will use Moving Mesh (ALE) application mode to make a frame to allow the mesh to move. Can you attach your model to take a look.
The answer is in your question, I think that maybe you will use Moving Mesh (ALE) application mode to make a frame to allow the mesh to move. Can you attach your model to take a look.
I understand that the mesh is modified each time that the value of d is changed and that this discrepancy may be related to that.
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Thanks for your answer Amine.
I found the problem: what made the mess was just the "infinite" elements that I used in my model. If I replace them by large surface without the "infinite" property, I do not observe the problem mentioned before.
I found the problem: what made the mess was just the "infinite" elements that I used in my model. If I replace them by large surface without the "infinite" property, I do not observe the problem mentioned before.