Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.
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.
3D Capacitance Parametric Sweep problem - Meshing issue - results in sinusoidal non physical results
Posted 27.09.2011, 23:10 GMT-4 Low-Frequency Electromagnetics, Mesh, Modeling Tools & Definitions, Parameters, Variables, & Functions, Studies & Solvers Version 4.0, Version 4.0a, Version 4.1, Version 4.2 1 Reply
Please login with a confirmed email address before reporting spam
I have a very simple problem and I have attached two files.
Basically i have a capacitor (Terminal 1 and 2) and a human (modeled as a Terminal 3) in the AC/DC electrostatics module.
If the human is close to the capacitor some of the fringe field couples to the human and so the capacitance across the two plates actually should drop (C21) while the capacitance of the source plate (Terminal 1) to the human increases (C31) . If the human moves away then the capacitance between the human and the source plate (C31) drops and the capacitance between the source plate and the other capacitor plate (C21) should increase in magnitude.
I use a parametric sweep to model the human moving away from the capacitor plates. If I do a parametric sweep of 1m at a time so (1m,2m,3m,4m,5m) away from the plates - C21 increases in magnitude albeit not in an exponential way as supposed to.
Here is an example of C21 from the cylindrical Ground case
xoffset Capacitance (F)
1 -2.4465e-12
1.5 -2.4627e-12
2 -2.4612e-12
2.5 -2.4652e-12
3 -2.4625e-12
3.5 -2.4746e-12
4 -2.47e-12
4.5 -2.4709e-12
5 -2.4768e-12
Here is the C21 values from the air box case
xoffset Capacitance (F)
1 -2.5273e-12
1.5 -2.5607e-12
2 -2.5593e-12
2.5 -2.5636e-12
3 -2.5653e-12
3.5 -2.5825e-12
4 -2.5848e-12
4.5 -2.6042e-12
5 -2.624e-12
If I do a sweep in 0.5m intervals then I see that the capacitance C21 goes up and down in an oscillatory function.
I have done everything and used a lot of Ivar's suggestions from previous posts like infinite elements, Cylindrical ground (attached) but this problem still persists. I have tried cubic discretization (an improvement) but the oscillation exists still. I think this is a meshing issue - but I use "Extremely Fine Meshing"
Could someone help? Ivar you have talked about this in previous posts - your help will also be greatly appreciated.
Best
Shahriar
Attachments:
Hello Shahriar Khushrushahi
Your Discussion has gone 30 days without a reply. If you still need help with COMSOL and have an on-subscription license, please visit our Support Center for help.
If you do not hold an on-subscription license, you may find an answer in another Discussion or in the Knowledge Base.
Please login with a confirmed email address before reporting spam
I ran your file without any alteration and i ran into some problems and hopefully you can drive those doubts away for me.
You have named four terminals bye numbers from 1 to 4. And after running we get 4x4 matrix of capacitance.
1) Is C11 is self capacitance of terminal 1? but even if you name your terminals differently then still we get 4x4 matrix with C11----C44 elements. Then how do we know which is C11 or C12, i mean capacitance between which terminals?
2) In one case C12: NaN but C21: -8.9TeraF. Why C12 and C21 are not equal? And how can capacitance be negative? What is NaN ? How did you know that C21 is the capacitance between two plates but not C12?