[helmut]

Hi everyone,

I'm trying to implement Henry Weller's earlier suggestion that "Another two-mesh approach would be to have one mesh for the fluid and one mesh for the thermal equation, i.e. a mesh of everything, and transfer fields between the two. This gives implicit coupling between the regions for the thermal equation but sufferes from a modest storage overhead and the need for quite a bit of mapping code but is also possible in OpenFOAM but has not yet been tried."

I've made some progress with this, but I'm still having difficulties. Please refer to my post of 2006 August 18, this thread, for more details on my test case.

The very high temperatures at the outlets reported in that aforementioned post apparently resulted from my failure to correctBoundaryConditions() of the whole-mesh velocity field after mapping the computed velocities from the submeshes to the whole mesh.

The temperature boundary conditions demand a solution in the range [300,400], but the actual solution is now in the range (292.1, 401.3). The too-high values are near the outer boundaries along the material interfaces, and the too-low values are near the outer boundaries at the inlet.

I'm at a loss for where to go from here ...

Thanks and best regards,
Helmut

[helmut]

Further to my question of September 12, I think the problem might be that the temperature interface condition, k1*gradT on the region1 side of the interface = k2*gradT on the region2 side of the interface, is not represented in the temperature equation. The exterior temperature boundary conditions are set in the boundary condition section of the T field. Is there some way in OpenFOAM to set the interface condition so that it gets into the matrix equation?

[khleitz]

Hallo,
I want to implement a local heat source in OpenFOAM.
Can anyone give me a useful tip how to do that?

[oscar_j]

Hello to All,

I'm a new OF user and I'm interesed in to simulate transfer heat by the three process: Conduction, Convection and Radiation but I haven't found info about that. Is it possible? What about examples and/or tutorials?

I have the impression that isn't so simple and it's scarce the info about this subject.

I appreciate any help.

Thanks!

[oscar_j]

hi,

I am trying to get the compilation of the laplacianFoam solver with wmake. But some errors happens.

I type within $FOAM_APP/solvers/basic/laplacianFoam directory (OpenFOAM-1.4) simply 'wmake' but I have error messages like these:

/home/oscar/OpenFOAM/linux/gcc-4.1.2/bin/../lib/gcc/i686-pc-linux-gnu/4.1.2/../. ./../../include/c++/4.1.2/bits/locale_facets.h:3916: error: variable or field '_M_initialize_moneypunct' declared void
/home/oscar/OpenFOAM/linux/gcc-4.1.2/bin/../lib/gcc/i686-pc-linux-gnu/4.1.2/../. ./../../include/c++/4.1.2/bits/locale_facets.h:3916: error: 'int std::moneypunct<wchar_t,>::_M_initialize_moneypunc t' is not a static member of 'class std::moneypunct<wchar_t,>'
/home/oscar/OpenFOAM/linux/gcc-4.1.2/bin/../lib/gcc/i686-pc-linux-gnu/4.1.2/../. ./../../include/c++/4.1.2/bits/locale_facets.h:3916: error: '__c_locale' was not declared in this scope
/home/oscar/OpenFOAM/linux/gcc-4.1.2/bin/../lib/gcc/i686-pc-linux-gnu/4.1.2/../. ./../../include/c++/4.1.2/bits/locale_facets.h:3917: error: expected primary-expression before 'const'
/home/oscar/OpenFOAM/linux/gcc-4.1.2/bin/../lib/gcc/i686-pc-linux-gnu/4.1.2/../. ./../../include/c++/4.1.2/bits/locale_facets.h:3917: error: initializer expression list treated as compound expression
/home/oscar/OpenFOAM/linux/gcc-4.1.2/bin/../lib/gcc/i686-pc-linux-gnu/4.1.2/../. ./../../include/c++/4.1.2/bits/locale_facets.h:3921: error: variable or field '_M_initialize_moneypunct' declared void



What's my mistake?

Thanks!
Oscar

[stephan]

hi,
i have problems converting to VTK.
it workes with -mesh-option but only for one timestep...
Any hints?
greetings
stephan

[sradl]

Dear Stephan

use the PyFoam script of Bernhard:
http://openfoamwiki.net/index.php/Ho...rocMultiregion

It's really great and is much faster than converting to VTK.

br
Stefan

[stephan]

hi,

thanx for the hint!
regards
stephan

[oscar_j]

Hi everybody

I'm interested in to simulate heat transfer between many regions, so I'm working on conjugated2 solver posted at Friday, August 26, 2005 by Daniele but when I compile some errors happen:

readSIMPLEControls1.H: In function 'int main(int, char**)':
readSIMPLEControls1.H:33: error: 'findRefCell' was not declared in this scope
readSIMPLEControls1.H:39: warning: use of old-style cast
readSIMPLEControls1.H:40: warning: use of old-style cast
readSIMPLEControls1.H:39: warning: use of old-style cast
readSIMPLEControls1.H:40: warning: use of old-style cast
make: *** [Make/linuxGcc4DPOpt/conjugateFoam2.o] Error 1

I think that those errors are produced by the following code part in readSIMPLEControls1.H:

label pRefCellNew = findRefCell(mesh1, pRefCell);

if (pRefCellNew != pRefCell)
{
pRefCell = pRefCellNew;

((dictionary&)simple).remove("pRefCell");
((dictionary&)simple).add("pRefCell", pRefCell);
}

I work in OpenFOAM-1.4. What is the findRefCell(mesh1, pRefCell) function for OpenFOAM-1.4 and how can I use it ?

Many Thanks
Oscar G

[panara]

I think is a problem with the OF 1.4
compare the readSIMPLEControls of the 1.4 version with the one used with conjugateFoam and change mesh with mesh1 or mesh2 and recompile..

I am not sure that it works, I have got no time even to install OF 1.4..

The other possibility is to use the OF 1.3

Daniele

[oscar_j]

Hi Daniele
Thanks for quick response.

Working in OF-1.4, I tryed having the Make/files file like:

conjugateFoam2.C

EXE = $(FOAM_USER_APPBIN)/conjugateFoam2

and the Make/options like:

EXE_INC = \
-I$(LIB_SRC)/cfdTools/lnInclude \
-I$(LIB_SRC)/finiteVolume/lnInclude \
-I$(LIB_SRC)/finiteVolume/cfdTools/general \
-I$(LIB_SRC)/finiteVolume/cfdTools/general/findRefCell

EXE_LIBS = \
-lcfdTools \
-lfiniteVolume \
-lfindRefCell

but it didnt work, I get the following:

bash-3.1$ wmake
Making dependency list for source file conjugateFoam2.C
SOURCE=conjugateFoam2.C ; g++ -m32 -Dlinux -DDP -Wall -Wno-strict-aliasing -Wextra -Wno-unused-parameter -Wold-style-cast -O3 -DNoRepository -ftemplate-depth-40 -I/home/oscarj/OpenFOAM/OpenFOAM-1.4/src/cfdTools/lnInclude -I/home/oscarj/OpenFOAM/OpenFOAM-1.4/src/finiteVolume/lnInclude -I/home/oscarj/OpenFOAM/OpenFOAM-1.4/src/finiteVolume/cfdTools/general -I/home/oscarj/OpenFOAM/OpenFOAM-1.4/src/finiteVolume/cfdTools/general/findRefCe ll -IlnInclude -I. -I/home/oscarj/OpenFOAM/OpenFOAM-1.4/src/OpenFOAM/lnInclude -fPIC -pthread -c $SOURCE -o Make/linuxGcc4DPOpt/conjugateFoam2.o
createMeshes.H: In function 'int main(int, char**)':
createMeshes.H:15: error: 'class Foam::fvMesh' has no member named 'constructAndClear'
createMeshes.H:31: error: 'class Foam::fvMesh' has no member named 'constructAndClear'
readSIMPLEControls1.H:33: error: 'findRefCell' was not declared in this scope
readSIMPLEControls1.H:39: warning: use of old-style cast
readSIMPLEControls1.H:40: warning: use of old-style cast
make: *** [Make/linuxGcc4DPOpt/conjugateFoam2.o] Error 1

What could I do?...

Beforehand thank you very much for your valuable help.

Have a good day
Oscar G

[ariorus]

Hello,

some months ago I have been playing a little with the coniugate heat application written by Daniele, for compiling it with openFOAM 1.4.

The modified application uses also intepolations so it should not be necessary that solid mesh and fluid mesh be coincident at the solid faces.

I used buoyantFoam as reference.

I'm attaching herewith the application

coniugate_interp.tgz

and a test_case
test_coniugate_interp.tgz
(in the testcase I removed solid and fluid mesh files).

Hope this can be useful, anyway I have to warn you it has not been tested very much.

Ciao.

Rosario.

[kmurphy]

Hi Oscar,

My (limited) understanding is that the call to constructAndClear is no longer needed.

In OpenFOAM 1.3 fvMesh had a method called constructAndClear which according to the code comments performed the tasks:
// Construct all necessary data and clear all storage
// Force creation of geometric data
// Force creation of interpolation weights
(see OpenFOAM-1.3/src/finiteVolume/fvMesh/fvMesh.C)

The method constructAndClear or a similar named method does not appear to exist in the corresponding file in OpenFOAM 1.4.
(see OpenFOAM-1.4/src/finiteVolume/fvMesh/fvMesh.C)


At some point in the past (I think when upgrading to OpenFOAM 1.3) I also removed the call to findRefCell from the readSIMPLEControls.H file. I was sure that I had read a suggestion to this somewhere but a quick search this morning failed to find the reference. However, the call to findRefCell does not also appear in the
/src/finiteVolume/lnInclude/readSIMPLEControls.H from version 1.3 on so maybe I just followed (blindly) what was happening in the standard code.

Hope this helps,
Kieran

[oscar_j]

Hi everybody,
thanks a lot for your help.

I have followed your suggestions and indeed my error messages are due to my Openfoam-1.4 version. These errors are about reorganisations of the source code for each Openfoam version.

http://openfoamwiki.net/index.php/Ho...versions_to_13

I will tell you about the results that I get.

Thank you very much
Oscar

[ariorus]

Hi Oscar,

I had a look at your code for the boundary conditions and, as as far as I understand, it is ok..
I guess it should be possible to improve a little the setting of the boundary conditions on the solid, to make it clearer the choice between heat flux or temperature specification.

Just a general observation: probably it is better to have prisms near the solid faces of the fluid mesh in order to have a good approximation of the normal gradients.



Ciao.

Rosario.

ps: To attach files read this:
http://www.cfd-online.com/cgi-bin/Op...matting#images

[oscar_j]

Hi Rosario,

I have been working in a solver for OF-1.4 that simulates the heat transfer by conduction between two coupled tridimensional solids by one patch. I appreciate any comment.

This is the solver:



and this is the case:



I would want to comparate the simulation with the analytic solution for the two coupled solids by one patch for heat conduction, where could I find it?

Thanks!
Oscar

[oscar_j]

Hi Rosario,

I have been working in a solver for OF-1.4 that simulates the heat transfer by conduction between two coupled tridimensional solids by one patch. I appreciate any comment.

This is the solver:



and this is the case:



I would want to comparate the simulation with the analytic solution for the two coupled solids by one patch for heat conduction, where could I find it?

Thanks!
Oscar

[oscar_j]

solver

[stephan]

hi,

try carslaw and jaeger with "heat conduction in solids"(hope i get the names right?!).
on the other hand crank "mathematics of diffusion" could be nice too...
regards
stephan

[oscar_j]

Hi everybody,

The simulation consists on a 3D case which has two coupled 3D solids with different thermal conductivities.

These are heat conduction solver and case for two coupled 3D solids. Solids are coupled by one patch. I havent comparate the numerical solution with analytic solution. I appreciate any comment and info about the solution analytic for two coupled solids.

two coupled solids Solver:

twomeshes.tar.gz

Test Case:

Twoblocks1patch.tar.gz

Have a good day
Oscar