## Potential & Electric Current just in specific bodies

### Potential & Electric Current just in specific bodies

Hey all,

In brief:
I have multiple Domains (14 bodies) in a simulation and I want to apply the potential/joule heating JUST in three bodies (domains).
However, the heat flux shall propagate through every domain and lead to thermal stresses in the overall structure.

I like to set up the following:
1. potential/electric field evolves an electric current
2. electric current results in liberated heat by joule heating
3. the liberated heat drives the thermal expansion und results in thermal stresses

I do need to say, that I reviewed multiple examples for elmer. [e.g. Tutorial 9: Thermal actuator driven with electrostatic currents, http://www.nic.funet.fi/index/elmer/doc/ElmerTutorials_nonGUI.pdf, 2017.Sept.11]
However, I did not find any example that explains how to.
I know it must be pretty simple - nevertheless I'd appreciate if someone could give me a hint...

Cheers
Manuel

Code: Select all
Header  CHECK KEYWORDS Warn  Mesh DB "." "."  Include Path ""  Results Directory ".\output"EndSimulation  Max Output Level = 5  Coordinate System = Cartesian  Coordinate Mapping(3) = 1 2 3  Simulation Type = Transient  Steady State Max Iterations = 10 !1 is weak coupling, 10 is strong couling  Output Intervals = 1  Timestepping Method = BDF  BDF Order = 2  Timestep intervals = 2   ! is the overall time  Timestep Sizes = 1e-1      ! is the time step size  Solver Input File = case.sif  Post File = case.vtuEndConstants  Gravity(4) = 0 0 0 9.82  Stefan Boltzmann = 5.67e-08  ! Permittivity Of Vacuum = 8.8542e-12 ! C^2/Nm^2  Permittivity Of Vacuum = 1.0 ! manipulation for conducting material  Boltzmann Constant = 1.3807e-23  Unit Charge = 1.602e-19EndBody 1  Target Bodies(3) = 1 2 3  Equation = 1  Material = 1  Body Force = 1  Initial condition = 1EndBody 2  Target Bodies(2) = 7 14  Equation = 1  Material = 2  Initial condition = 2EndBody 3  Target Bodies(3) = 4 5 6  Equation = 1  Material = 3  Initial condition = 2EndBody 4  Target Bodies(6) = 8 9 10 11 12 13  Equation = 1  Material = 4  Initial condition = 2EndMaterial 1  Electric Conductivity = Variable Temperature   Real      298.0 0.62500e5      398.0 0.82500e5   End  Heat expansion Coefficient = 20.0e-6  Heat Conductivity = 3.4  Relative Permittivity = 1.0005  Heat Capacity = 1278.0  Mesh Poisson ratio = 0.35  Density = 2780.0  Poisson ratio = 0.35  Youngs modulus = 7.405e9EndMaterial 2  Electric Conductivity = Variable Temperature   Real      298.0 0.62500e5      398.0 0.82500e5   End  Heat expansion Coefficient = 21.0e-6  Heat Conductivity = 730  Heat Capacity = 921.0  Mesh Poisson ratio = 0.33  Density = 2700.0  Poisson ratio = 0.33  Youngs modulus = 70.0e9EndMaterial 3  Electric Conductivity = Variable Temperature   Real      298.0 0.62500e5      398.0 0.82500e5   End    Heat expansion Coefficient = 21.0e-6  Heat Conductivity = 1.0  Heat Capacity = 921.0  Mesh Poisson ratio = 0.33  Density = 2700.0  Poisson ratio = 0.33  Youngs modulus = 70.0e9EndMaterial 4  Electric Conductivity = Variable Temperature   Real      298.0 0.62500e5      398.0 0.82500e5   End    Heat expansion Coefficient = 3.43e-6 ! e-3 is normal, but looks crazy. change does not effect significantly  Heat Conductivity = 0.0257  Heat Capacity = 1005.0  !Mesh Poisson ratio = 0.33  Density = 1205.0  Poisson ratio = 0.33  Youngs modulus = 1.0e5EndInitial Condition 1  Potential = 0  Displacement 1 = 0  Displacement 2 = 0  Displacement 3 = 0  Temperature = 298EndInitial Condition 2  Displacement 1 = 0  Displacement 2 = 0  Displacement 3 = 0  Temperature = 298EndBody Force 1  Heat Source = Equals Joule HeatingEndBoundary Condition 1  Target Boundaries(2) = 1 2   Name = "Sym"  Heat Flux = 0  Displacement 1 = 0EndBoundary Condition 2  Target Boundaries(3) = 3 5 7  Name = "TopPot"  Potential = 30.7EndBoundary Condition 3  Target Boundaries(3) = 4 6 8   Name = "BotPot"  Potential = 0.0EndBoundary Condition 4  Target Boundaries(2) = 9 10  Name = "Boun"  Displacement 3 = 0  Temperature = 298EndEquation 1  Active Solvers(4) = 1 2 3 4EndSolver 1  Equation = Result Output  Output Format = Vtu  Output File Name = case  Procedure = "ResultOutputSolve" "ResultOutputSolver"  Exec Solver = AlwaysEndSolver 2  Equation = Stat Current Solver  Procedure = "StatCurrentSolve" "StatCurrentSolver"  Variable = Potential  Variable DOFs = 1  Calculate Joule Heating = True  Calculate Electric Conductivity = True  Linear System Solver = Iterative  Linear System Iterative Method = CG  Linear System Preconditioning = ILU3  Linear System Max Iterations = 300  Linear System Convergence Tolerance = 1.0e-8  Nonlinear System Max Iterations = 1  Nonlinear System Convergence Tolerance = 1.0-6  Nonlinear System Newton After Iterations = 3  Nonlinear System Newton After Tolerance = 1.0e-12  Nonlinear System Relaxation Factor = 1.0  Steady State Convergence Tolerance = 1.0e-6EndSolver 3  Equation = Heat Equation  Variable = Temperature  Variable DOFs = 1  !Calculate Joule Heating = True  Linear System Solver = Iterative  Linear System Iterative Method = BiCGStab  Linear System Preconditioning = ILU1  Linear System Max Iterations = 350  Linear System Convergence Tolerance = 1.0e-9  Nonlinear System Max Iterations = 1  Nonlinear System Convergence Tolerance = 1.0e-07  Nonlinear System Newton After Iterations = 3  Nonlinear System Newton After Tolerance = 1.0e-12  Nonlinear System Relaxation Factor = 0.5  Steady State Convergence Tolerance = 1.0e-07EndSolver 4  Equation = Stress Analysis  Linear System Solver = Direct  Variable = Displacement  Variable Dofs = 3  Calculate Stresses = True  Linear System Iterative Method = "BiCGStab"  Linear System Max Iterations = 500  Linear System Convergence Tolerance = 1.0e-08  Linear System Abort Not Converged = True  Linear System Preconditioning = "ILU0"  Linear System Residual Output = 1  Steady State Convergence Tolerance = 1.0e-05  Nonlinear System Convergence Tolerance = 1.0e-05  Nonlinear System Max Iterations = 1  Nonlinear System Newton After Iterations = 3  Nonlinear System Newton After Tolerance = 1.0e-02  Nonlinear System Relaxation Factor = 1  Linear System Precondition Recompute = 1End
MFeMult

Posts: 6
Joined: 11 Sep 2017, 16:36

### Re: Potential & Electric Current just in specific bodies

Hi,

I did not quite get where your problem is. What happens when you run the simulation?

Difficult to say if there is a mistake just from the sif file. But at least, if you have 14 bodies, they must all figure in the sif and have materials assigned.

In general, if you are new to Elmer, you could start with a sif file generated by ElmerGUI and only go on to write the sif file from scratch if you are more experienced and if you want to do things that are not possible with the GUI.

HTH,

Matthias
mzenker

Posts: 1692
Joined: 07 Dec 2009, 11:49
Location: Germany

### Re: Potential & Electric Current just in specific bodies

Hey Matthias,

1. My problem is/was that the differential equation of
$(1) \textbf{E} = -\nabla \phi$\\
$(2) \textbf{j} = \frac{1}{\varrho}\cdot\textbf{E}$
were solved in every body/domain. That was not my intent.
However, I misinterpreted the meaning of Equation, body and interaction with solvers.
After I added a second equation (so that I had two), I solved it correctly.
Now it seems absolutely clear and I don't know why I didn't see it before.

Code: Select all
Equation 1  Name = "elec-therm-mech"  Calculate Stresses = True  Active Solvers(4) = 4 3 2 1EndEquation 2  Name = "therm-mech"  Calculate Stresses = True  Active Solvers(2) = 3 2End

2.
But at least, if you have 14 bodies, they must all figure in the sif and have materials assigned.

Fore sure, that must be the case. However, thnx for pointing out.

3.
In general, if you are new to Elmer, you could start with a sif file generated by ElmerGUI and only go on to write the sif file from scratch
if you are more experienced and if you want to do things that are not possible with the GUI.

Well, yes and no. My main aim is to understand how things work... So I started from the scratch -> however, I do need to say, that your hint helped because it is solved Thnx

Cheers Matthias,
Manuel
MFeMult

Posts: 6
Joined: 11 Sep 2017, 16:36

### Re: Potential & Electric Current just in specific bodies

Hmmm...

If you solver numbering didn't change, the above means that you solver for heat and current everywhere, and for stress only in a part of your domain. And you have result output only where you solver for stress. Is that what you wanted?

Also I would set

Exec Solver = After Timestep

for result output.

Matthias
mzenker

Posts: 1692
Joined: 07 Dec 2009, 11:49
Location: Germany