[ksp1717]

Hi

I am new to openfoam. I am trying to do a cfd analysis of an oven. The schematic of the oven is as shown in the figure attached.

The air inlet and outlets are shown by arrows. the oven is completely close in all the directions. The lower part of the diagram shows a material web. The web is moving and the direction is shown by the arrows appropriatley. There is a set of air inlets and outlets below the web which is not shown here.

I am intending to do a 2D analysis of this setup. I am not sure how to model this for the foam. Should I make the entire 3D model or should i just make a crosssectional model (exactly as shown in the figure given a small thrid dimension into the plane. I tried the crossectional method using the GMSH software. But my mesh is present only on the geometries. There is no mesh in the area between the web and the top.


Can anyone help me please.

Shyam

[ksp1717]

Hi

I have managed to create the model. I gave the boundary conditions at the inlet as

inlet
{
type patch;
physicalType inlet;
nFaces 134;
startFace 21254;
}

will this work. or should I just say
inlet
{
type wall;
physicalType inlet;
nFaces 134;
startFace 21254;
}

I checked the user guide and there is no mention of the physical type in it. but i looked into the programming guide and there it is mentioned to use physical type.

[nakul]

Hi Shyam,

Try making the mesh using blockMesh. For learning how to use blockMesh you can refer the User's guide. This is a simple geometry and it would be lot easier to specify boundary conditions if you use blockMesh although it may be a little tedious.

If you face any problems in that, do feel free to ask!!

[ksp1717]

Hi

I managed to create the model and mesh . I wrote the velocity file in 0 directory.

Now I am writing the temperature T file.

I gave values for my inlet and outlets but i am unable to figure out what the condition(type) for moving wall should be.

The wall starts at 25 C and as it moves through the oven gains temperature.

Can anyone help me in writing this.

Shyam

[nakul]

Hi,
If your problem has viscous forces then for U at the moving wall boundary condition in accordance with no slip condition should be plate velocity only.

You may write that explicitly.

As far as T and p are concerned they would vary througout the solution so a boundary condition of "zeroGradient" would suffice. The effect of moving plate has already been incorporated in U directory.

[ksp1717]

Hi nakul thanks for your reply.

I have finished giving the initializations. I cannot find any info as to where I should enter the properties of my moving wall because my final objective is to find out the temperature at the end of the wall. So my first problem will be to give the material properties.

second when i run i get the following error.

/ * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //
Create time

Create mesh for time = 0


Reading g
Reading thermophysical properties

Reading field T

Reading field p_rgh

Reading field U

Reading/calculating face flux field phi

#0  Foam::error::printStack(Foam::Ostream&) in "/opt/openfoam170/lib/linux64GccDPOpt/libOpenFOAM.so"
#1  Foam::sigFpe::sigFpeHandler(int) in "/opt/openfoam170/lib/linux64GccDPOpt/libOpenFOAM.so"
#2   in "/lib/libc.so.6"
#3  Foam::surfaceInterpolation::makeWeights() const in "/opt/openfoam170/lib/linux64GccDPOpt/libfiniteVolume.so"
#4  Foam::surfaceInterpolation::weights() const in "/opt/openfoam170/lib/linux64GccDPOpt/libfiniteVolume.so"
#5  
 in "/opt/openfoam170/applications/bin/linux64GccDPOpt/buoyantBoussinesqSimpleFoam"
#6  __libc_start_main in "/lib/libc.so.6"
#7  
 in "/opt/openfoam170/applications/bin/linux64GccDPOpt/buoyantBoussinesqSimpleFoam"
Floating point exception

I checked the error and found out that it might be because of an initialization with 0, but I get the same without any 0 initializations. Any idea why i get it and how i can correct it.

Thanks, Shyam

[nakul]

Hi,

For your first problem, I don't think OF presently have any solver which would solve the flow field and provide you with temperature distribution on your wall as well. At most you can get the amount of heat flux at that boundary. So you may stop thinking about feeding material properties of the wall to OF.

Secondly this error is not due to initialisation with 0 but because of division with 0, while calculating the field phi. This occurs due to incorrect case setup. Look into your fvSchemes file as well as your initial conditions in 0 directory.

You may have to do some trial and error to get around this problem.
A;ternatively look into the source code to figure out how phi is being calculated and make necessary changes in your case.