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.
Electric Current Flow in a 2D surface
Posted Apr 16, 2015, 10:00 a.m. EDT Low-Frequency Electromagnetics Version 5.0 3 Replies
Please login with a confirmed email address before reporting spam
Hiii,
I was wondering if I can get the electric current flow on a 2d surface. Say there is a rectangle. I want to give current at one corner and take out the current from the opposite corner and see the current distribution on the rectangular surface. Is it possible to do as such or should I have to go for a 3d model ?
Regards,
Anup
I was wondering if I can get the electric current flow on a 2d surface. Say there is a rectangle. I want to give current at one corner and take out the current from the opposite corner and see the current distribution on the rectangular surface. Is it possible to do as such or should I have to go for a 3d model ?
Regards,
Anup
3 Replies Last Post Apr 20, 2015, 5:34 a.m. EDT
Please login with a confirmed email address before reporting spam
Posted:
10 years ago
This is interesting because conservation of current will result in an infinite current density in the corners.
Assuming radial symmetry near the corner: Sheet current density K @ radius r
K = I / (π r / 2)
K = σ E
E = I / (π r σ / 2)
The potential V is the negative of the integral of E, so:
|V| = − ln r (2 I/ π σ)
so potential is proportional to the natural logarithm of distance from the corner which is singular at the corners.
So I'm not sure COMSOL can actually solve this problem.
Alternately you can round the corner and apply the current across the entire rounded edge.
Assuming radial symmetry near the corner: Sheet current density K @ radius r
K = I / (π r / 2)
K = σ E
E = I / (π r σ / 2)
The potential V is the negative of the integral of E, so:
|V| = − ln r (2 I/ π σ)
so potential is proportional to the natural logarithm of distance from the corner which is singular at the corners.
So I'm not sure COMSOL can actually solve this problem.
Alternately you can round the corner and apply the current across the entire rounded edge.
Please login with a confirmed email address before reporting spam
Posted:
10 years ago
Hi
to define a current density you need a volume, area or length ... there is only absolute (total/integrated) flux values issued from a "point", except in COMSOL in 1D because any point represents a default surface (just as in 2D any point represents a default depth/length).
So normally if you define a total current value on a "point" you can at best talk about a max current density in the mesh element(s) that contains the node point, and this flux density will only increase the smaller the mesh element is made => this indeed leads to a singularity.
Personally I avoid to define any flux on any entity smaller than a "Boundary" in the COMSOL sense (see the doc)
I would also suggest that you replace the sharp edge by a fillet of finite length an apply the flux thereon.
Pls do not forget that the COMSOL plots also interpolates the underlying discretized node points, so in sharp corners like this you may get oscillations on the plots that are not truly physical, such as negative concentrations or temperatures ...
--
Good luck
Ivar
to define a current density you need a volume, area or length ... there is only absolute (total/integrated) flux values issued from a "point", except in COMSOL in 1D because any point represents a default surface (just as in 2D any point represents a default depth/length).
So normally if you define a total current value on a "point" you can at best talk about a max current density in the mesh element(s) that contains the node point, and this flux density will only increase the smaller the mesh element is made => this indeed leads to a singularity.
Personally I avoid to define any flux on any entity smaller than a "Boundary" in the COMSOL sense (see the doc)
I would also suggest that you replace the sharp edge by a fillet of finite length an apply the flux thereon.
Pls do not forget that the COMSOL plots also interpolates the underlying discretized node points, so in sharp corners like this you may get oscillations on the plots that are not truly physical, such as negative concentrations or temperatures ...
--
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
10 years ago
Hi,
I wopuld add that there is a specialized interface for electric currents in shells. This would be applied to very thin conductors that you would not like to mesh in 3D, e.g. a 3 mm PCB with 50um copper on it.
The default boundary setting would here be on an edge but it would account for the thickness of the layer, so you could still safely apply a current without andy singularities.
Additionally, in electric current interafces you also have point sources
which you can safely use - and which make sense in some technical contexts.
ch.comsol.com/model/a-geoelectrical-forward-problem-9636
Best regards,
Sven
I wopuld add that there is a specialized interface for electric currents in shells. This would be applied to very thin conductors that you would not like to mesh in 3D, e.g. a 3 mm PCB with 50um copper on it.
The default boundary setting would here be on an edge but it would account for the thickness of the layer, so you could still safely apply a current without andy singularities.
Additionally, in electric current interafces you also have point sources
which you can safely use - and which make sense in some technical contexts.
ch.comsol.com/model/a-geoelectrical-forward-problem-9636
Best regards,
Sven