Learning Forum

[Xiao.yu]

Hi, 

I would like to ask if it is possible to use Maxwell to simulate a capacitor simultaneously considering eddy currents in the plates and displacement current in the dielectric material.

Please refer to the picture for an example.

The current flows from the left side of the top conductor, passes through the dielectric material into the lower conductor, and then flows out from the right side of the lower conductor.

I would like to visualize the field overlay for both the conductors and the dielectric. Eddy current effects should be considered in the conductors, while the capacitive current in the dielectric material should also be taken into account.

I have tried different solvers, such as 2D and 3D eddy current, as well as 2D and 3D AC conduction, but none seem to handle this problem adequately.

The problem of 2D and 3D AC conduction is that they ignore the magnetic field (Maxwell help: p.3372) which results a frequency and location independent current distribution inside the conductor.

The 2D eddy current solver doesn't support the required current flow, while the 3D eddy current solver allows neither current excitations touching each other nor single excitation for a conductor.

Should I add boundary conditions? 

Can the transient solver handle this problem, or should I use multi-physics coupling? 

Any ideas and/or suggestions would be greatly appreciated!

[HDLI]

Hello Xiao,

      Please use 3D transient A-phi solution, or 3D eddy current A-phi solution of beta version if you have 24R2.

      In A-phi solver, tangential flux boudary is required.

HDLI 

 

[Xiao.yu]

Hi, HDLI,

Thank you very much for your kind reply. After posting this thread, I tried using the 3D Eddy Current A-Phi solver. I attempted to assign current inlet and outlet on the faces described above, but it didn’t work, since the displacement current was activated, ansys needs at least one potential for that and this potential should also only be on an external terminal.

I also tried only using voltage excitations, which worked, though the currents were not continuous through the surfaces along the current path.

After reading your answer, I tried again, this time using mixed external excitations: one current inlet and a ground with zero potential. Insulator threshold has been reduced to 1e-20.

Initially, I chose a dielectric with a permittivity of 4, but Ansys returned an error message, ‘matrix may be badly conditioned.’ I then tried adding the boundary condition you suggested to the outer surface of the conductors, but the error persisted. During my search, I found an example showing this feature where a permittivity of 1e12 was used, which inspired me to increase the permittivity from 4 to 500. With this change, the simulation worked even without any boundary conditions.

I checked the current continuity, and it seems correct. However, the expected skin effect is not visible.

I would like to know if you could give me some advises where I should assign the trangential flux boundary condition. 

Thanks again for your help.

Xiao