Hello there,

in the moment I try to implement a simulation of an electric arc in Fluent utilizing a UDS for the electric potential.

The mdel consist of two metal electrodes with a fixed high electrical conductivity (specified as diffusivity) and a gaseous arcing zone in between, where the conductivity is set to be highly temperature dependent. At the boundary of the one electrode, a current density is fed into the electrode domain (UDS flux), the boundary of the other electrode is set to zero potential. The interfaces between solid electrodes and gas are set as coupled walls.

Right now, I am facing the following problem: Despite the UDS residuals are converging, it seems, that the electric current (UDS flux) is not crossing the interfaces well. In other words, only a small portoin of the current, that is fed into electrode one is crossing the interface into the fluid zone and an even smaller portion of that is crossing the interface between the fluid zone and the second interface. This leads to a high imbalance between the current fed into electrode one and the current flowing out of electrode two.

Doing some parameter variations, I found out that this issue is somehow related to the big difference in diffusivities between solid and fluid zone. Applying the same diffusivities leads to a current crossing the interfaces easily, But this is not the goal, as in reality, there are quite big differences is conductivity on both sides.

My question now is, how to solve this problem. I am guessing I have to implement some kind of "interface udf" where the current (uds flux) crossing the interface is specified. This should somehow work similar to a coupled heat flux boundary. But I dont really now where to start.

Best regards,

Marvin

PS: As an outlook, it would be grat if some kind of electric contact resistance could be implemented in that UDF as well.