Discussion Forum

Hi,

I am going to model a magnet moving over a long copper (or iron, later) plate to study eddy current. I got some questions about the concept of this problem in COMSOL and would be thankful if someone helps me to know how COMSOL treats this kind of problems:

1. As I reviewed the "Magnetic Fields" interface in the help document, it does not include the electric field term in the Ohm's law for moving conductors, J=sigma(vxB) and E=0. Thus, it can not consider the induced charge on the surfaces of the conductive plate! So, it is more accurate we use the "Magnetic and Electric Fields" interface if charge is a matter of concern in our problem! am I right?

2. It is well-known that induced current normal to the boundary surfaces of conductive plates is equal to zero, J.n=0. How can we apply this B.C to our model in COMSOL? I know there is a "Magnetic Insulation" B.C in both "Magnetic Fields" and "Magnetic and Electric Fields" interfaces, but I am not sure if it can satisfy this condition. It says that tangential component of the magnetic potential is zero, nxA=0. I saw this B.C used in some examples of COMSOL Model Gallery, those are concerned about the eddy current. Is this B.C enough to apply J.n=0?

3. When we have a copper or aluminum plate (non-magnetic), applying the Lorentz force in form of f=JxB (without E term) is enough to get accurate results, as these materials have good conductivity to reach to the charge equilibrium condition quickly. What about magnetic materials like Iron though it is also a good conductor? is there any difference between these two materials in this case? Is it not better we use Maxwell tensor which is the most accurate way to calculate magnetic force as it originally consider the complete form of Lorentz force f=JxB+qE! Should I expect difference between my results when I use these two methods of force calculation for the materials mentioned?

Thank you