[Solved] Wrong geometry in radiation problem

[Remy]

Hello

I am trying to solve heat transport problem with radiation. Radiation is assumed to be on the inner surface. I have got following error suggesting wrong geometry (I have done some checks and nothing wrong was found):

ElmerAsciiMesh: Base mesh name: ./g4
ViewFactors: Computing view factors for radiation body 1
ViewFactors: Number of surfaces participating in radiation: 46
ViewFactors: Computing viewfactors...
Combining original boundary elements for shading
The combined set includes 1 line segments (vs. 46)
surfs: 46, min(1)=0.00, max(1)=0.00, avg=0.00
ViewFactors: View factors computed in time (s): 0.00 0.00
ViewFactors:
ViewFactors: Viewfactors before manipulation:
ViewFactors: Minimum row sum: 0.000E+00
ViewFactors: Maximum row sum: 0.000E+00
ViewFactors: Number of rowsums below 0.5 is: 46 (out of 46)
ViewFactors: Normalizaing Factors...
ERROR:: ViewFactors: Invalied view factors for normalization, check your geometry!
STOP 1

Geometry is attached in the .png file. The material definitions are:

Code: Select all

Material 1
  Name = "a"
  Heat Conductivity = 0.3
  Density = 2000.0
  Emissivity = 0.9
End

Material 2
  Name = "b"
  Density = 30.0
  Heat Conductivity  = 0.001
  Emissivity = 0.9
End

Material 3
  Name = "air"
  Heat Conductivity = 0.025
  Density = 1.2
End

Radiation boundry condition is:

Code: Select all

Boundary Condition 1
  Target Boundaries(1) = 1
  Name = "cavities"
  Heat Flux BC = True
  Radiation = Diffuse Gray
!  Radiation Boundary = 1
  Save Scalars = true
End

Solver configuration:

Code: Select all

Solver 1
  Equation = Heat Equation
  Variable = -dofs 1 Temperature
  Procedure = "HeatSolve" "HeatSolver"
  Exec Solver = Always
  Stabilize = True
  Bubbles = False
  Lumped Mass Matrix = False
  Optimize Bandwidth = True
  Steady State Convergence Tolerance = 1.0e-10
  Nonlinear System Convergence Tolerance = 1.0e-11
  Nonlinear System Max Iterations = 200000
  Nonlinear System Newton After Iterations = 3
  Nonlinear System Newton After Tolerance = 1.0e-5
  Nonlinear System Relaxation Factor = .9
  Linear System Solver = Iterative
  Linear System Iterative Method = BiCGStab
  Linear System Max Iterations = 5000000
  Linear System Convergence Tolerance = 1.0e-11
  Linear System Preconditioning = ILU0
  Linear System ILUT Tolerance = 1.0e-11
  Linear System Abort Not Converged = True
  Linear System Residual Output = 1
  Linear System Precondition Recompute = 1

  Calculate Loads = True

  Gebhardt Factors Solver Iterative = True
  Viewfactor Binary Output = Logical True

End

Any idea what is wrong?

[Remy]

PS
In the case if no radiation is present the problem is solved correctly.

[kevinarden]

Difficult to say without seeing the actual mesh. Are those vertical lines top and bottom boundaries?

[Remy]

Vertical are adiabatic. Horizontal are with convective exchange.
Mesh attached.

[Rich_B]

Hello, welcome to the Elmer Forum.

Since this is your first time here, a few pointers may be helpful.

Attaching a complete Elmer example, including sif file, geometry input files, and optionally a solver log, will help obtain the most help. There are so many ways for an Elmer project to go wrong, and sometimes just in small details, that attaching an example will usually return the best answer.

The forum has a three attachments limitation, and the largest allowable attachment size is one Megabyte. Archiving attachments using a compression program, zip, tar, gz, etc., is useful for reducing file size.

If your Elmer mesh folder is too large to attach, try reducing the geometry size, or post the input files that generated the mesh. For example, you created a gmsh .geo input file and meshed the file with gmsh. If the generated gmsh msh file is too large, then post the gmsh .geo file with instructions on how to create your mesh.

Notifications from the forum don't work. Yes, you can set up notifications when signing up, but, no, you won't get any notifications. A suggestion is to create a browser bookmark to your post, and once a day or so, click on your bookmark to check for responses.

Lastly, you are asking other Elmer users for help. If investigating your question takes more than about five minutes, (five minutes to download and extract your archive and run your example), then you may not get a timely answer.

Thanks, Rich.

[Remy]

Rich, thank you for the invitations and hints.
Kevin, thanks for response. I have done remeshing and radiation works well.