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.
Conductive Media DC model with incorrect result after changing boundary condition
Posted 6 nov 2012, 05:10 GMT-5 2 Replies
Please login with a confirmed email address before reporting spam
Hello COMSOL Community,
I´m working on a 3D Conductive Media DC model of a probe which is used for geoelectrical measurements. The probe consists of a 4 electrodes array. At the outer electrodes constant currents of 0.1 A/m^2 respectively -0.1A/m^2 were defined. At the inner electrodes potential is measured. My model consists of the probe itself and a homogeneous medium with constant conductivity around the probe. The dimensions of this medium are large enough to define Ground (V=0) at the outer boundaries. However, the results seem to be correct (see attached Model_1).
Solving the same model with Electric Insulation at the outer boundaries leads to a result, which couldn´t be exact. The streamline plot and other plotted results (current density, electric potential...) seem to be reliable, but the potentials calculated at the potential electrodes are 1000 times higher than using Ground conditions. Furthermore, both potentials are negative instead of having an inverse sign (see attached Model_2).
Has anyone of you an idea, how I can fix this problem? I have allready tried different solvers and preconditioners, a finer mesh near the outer boundary, but nothing has worked better.
How can a boundary, which is that far away of my array, effect the result of the model this much? Usually, I would think, that just a few electric field lines would reach the outer boundaries, so that a change in boundary conditions there never could have an influence like that (if it has an influence at all).
Thanks for any hint how I can fix this prblem,
Daniel
I´m working on a 3D Conductive Media DC model of a probe which is used for geoelectrical measurements. The probe consists of a 4 electrodes array. At the outer electrodes constant currents of 0.1 A/m^2 respectively -0.1A/m^2 were defined. At the inner electrodes potential is measured. My model consists of the probe itself and a homogeneous medium with constant conductivity around the probe. The dimensions of this medium are large enough to define Ground (V=0) at the outer boundaries. However, the results seem to be correct (see attached Model_1).
Solving the same model with Electric Insulation at the outer boundaries leads to a result, which couldn´t be exact. The streamline plot and other plotted results (current density, electric potential...) seem to be reliable, but the potentials calculated at the potential electrodes are 1000 times higher than using Ground conditions. Furthermore, both potentials are negative instead of having an inverse sign (see attached Model_2).
Has anyone of you an idea, how I can fix this problem? I have allready tried different solvers and preconditioners, a finer mesh near the outer boundary, but nothing has worked better.
How can a boundary, which is that far away of my array, effect the result of the model this much? Usually, I would think, that just a few electric field lines would reach the outer boundaries, so that a change in boundary conditions there never could have an influence like that (if it has an influence at all).
Thanks for any hint how I can fix this prblem,
Daniel
Attachments:
2 Replies Last Post 6 nov 2012, 06:19 GMT-5
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
There are no materials assigned in your models. How could it ever solve this way?
Cheers
Edgar
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Of course there are materials in my model. I defined conductivities (sigma-sigma3) in each subdomain. These conductivities are defined in Options > Constants.
Maybe there is a problem with my attached files...
Maybe there is a problem with my attached files...