thermic analisys with adiabatic surfaces

[teofilBitron]

Hi all,
may be I did some errors but I can not understaind my results.
On some bouldaries I impose an adiabatic condition like:

Boundary Condition 8
Target Boundaries = 19
Name = IntConveMidi

Heat Flux BC = True
Heat Flux = 0
Save Scalars = Logical True
End


but if I check the outgoing convective flux I foud something different from 0.
The convective flux is request by this card:

Solver 4
Equation = SaveScalars
Procedure = File "SaveData" "SaveScalars"
Exec Solver = after timestep
Filename = Tint.dat
Variable 1 = Temperature
Operator 1 = boundary mean
Operator 2 = convective flux
End

Someone has an advice?
In attached the sif and mesh files.

[raback]

Hi

The "calculate flux" operator is not the most accurate one since it uses the integration point of the surface, and not of the bulk which would be best for consistency. Integrating over the boundary has the advantage that it is easier to separate different regions.

Now there is another confusing strategy which includes computing the nodal load and taking a "boundary sum" of them. This is consistent with the solution of the linear system.

-Peter