Discussion Forum

I have a Magnetic Fields model where an infinite line current induces current in a nearby passive loop. I have attached a toy model that illustrates this scenario, and image of the geometry.

I define the driving current as an edge current on a straight line segment, and I mesh the actual geometry of the passive loop as a single-turn coil. This kind of model runs perfectly fine if that edge current is a closed loop, but I cannot get the solver to converge for an open loop like the line current shown here. Obviously, an open loop for an edge current is not physically realistic, but there must be some way to get it to work when representing an infinite line current in 3D space. (The attached model has the edge current floating in space within the domain just because the meshing was easier that way. I've tried terminating the edge current to the domain boundary, but that fails in the same way.)

The issue seems tied to the use of infinite element domains on the boundary. If I do not use any infinite element domains, then the solver converges just fine even for the line current, and I get the results you'd expect for that scenario. I do want the infinite element domain included, however, as it's important to the physics of my problem.

Any advice would be appreciated. Thanks!