Hello,
Is anyone aware of a simple example where a time varying voltage or current is added to one end of a wire, and the other end is treated as ground, and the propagation of the current/electric/magnetic fields can be observed over time?
I already have the mgdyn_steady_wire test case and I am familiar with how make this a transient solve with the current boundary conditions and body force, but I am not sure how to set up these for the specific case I am asking about.
Thanks
Marc
Current propagation in a wire
-
raback
- Site Admin
- Posts: 4868
- Joined: 22 Aug 2009, 11:57
- Antispam: Yes
- Location: Espoo, Finland
- Contact:
Re: Current propagation in a wire
Hi Marc,
I wonder whether in this case the quasistatic assumption is no longer valid and the solution is more wave like? The AV solver does have the possibility to consider also 2nd derivaties when setting "Electrodynamics model = Logical True" in the simulation section. This was not originally include since the target application was electrical machines.
There is also a dedicated solver for transient wave called EMWaveSolver.F90 that solves the electric field directly. And harmonic version of the same called VectorHelmholtz.F90
Generally wave phonemena are tedious to solve in time domain since accurate description requires that CFL number be <~1. This means a lot of timesteps. And if you have a lot of timestep you may rather use some simple explicit method such as FDTD, rather than implicit FEM.
-Peter
I wonder whether in this case the quasistatic assumption is no longer valid and the solution is more wave like? The AV solver does have the possibility to consider also 2nd derivaties when setting "Electrodynamics model = Logical True" in the simulation section. This was not originally include since the target application was electrical machines.
There is also a dedicated solver for transient wave called EMWaveSolver.F90 that solves the electric field directly. And harmonic version of the same called VectorHelmholtz.F90
Generally wave phonemena are tedious to solve in time domain since accurate description requires that CFL number be <~1. This means a lot of timesteps. And if you have a lot of timestep you may rather use some simple explicit method such as FDTD, rather than implicit FEM.
-Peter
-
WidderElmer
- Posts: 12
- Joined: 30 Jan 2024, 12:29
- Antispam: Yes
Re: Current propagation in a wire
Peter,
Your response was insightful. Before you sent your response, I didn't realize the solver was only Implicit. It's funny because I was planning to calculate the CFL number as a guide for my timestep size. At this point I may just estimate the current flow change over time to see what I get.
Thanks for the response and for the suggestions.
Marc
Your response was insightful. Before you sent your response, I didn't realize the solver was only Implicit. It's funny because I was planning to calculate the CFL number as a guide for my timestep size. At this point I may just estimate the current flow change over time to see what I get.
Thanks for the response and for the suggestions.
Marc