Modeling of resistance in a rectangular sheet with COMSOL

  • Thread starter sunshine09
  • Start date

I have got a simple problem which makes me struggle.
I have a rectangular thin conductive sheet (2D) of width w, height h
and conductivity suspended in air.
I would like to calculate with COMSOL/Femlab the electrical resistance between one of the corners of the conductive sheet and any other point on the sheet?
I selected the 'conductive media DC' mode in 2D. I am not sure what
boundary condition I should choose and how to apply a potential to one
of the corners.
Since the sheet is suspended in air I thought of putting isolating
boundary conditions on all 4 sides, but then how will I apply the
potential to the corner? Somehow it does not give meaningful results.

Your help is greatly appreciated.



Science Advisor
Gold Member
I don't know COMSOL so I can't help with your numerical simulation. I can tell you that the potential becomes singular at the infinitesimal points where your are introducing/removing current. Suggest using finite-sized electrodes to make the numerical solver better behaved. A small circular electrode on the plate and a semi-circular electrode at the corner, both small compared to the rectangle but enclosing multiple meshes, should do it.
you should be able to set the potential along the boundaries. Make sure youve got a ground somewhere on ther, and have you tried a point condition?
@Marcusl: Thanks for the suggestion. makes sense to me.

@Nick: Yes, it is possible to set the potential along a boundary. I could for example define a small square electrode whose boundary I set to a potential of say 5V. I am also able to set the border of the big rectangle to isolating. Now I still need to get a ground somewhere, but where. I think if I want to calculate the resistance between one corner and a point (X,Y) then I have to set the corner to 5V and this point (X,Y) to ground and then somehow calculate the current. But this calculation I would have to code in a script and repeat 10000 times, if I assume a 100 x 100 grid.
I am wondering if there is a faster solution.
I have looked through all Femlab examples and could not find the solution.

Try and contact Mina with Comsol on their website. shes a big help.

Physics Forums Values

We Value Quality
• Topics based on mainstream science
• Proper English grammar and spelling
We Value Civility
• Positive and compassionate attitudes
• Patience while debating
We Value Productivity
• Disciplined to remain on-topic
• Recognition of own weaknesses
• Solo and co-op problem solving

Hot Threads