[mattijs]

Have you had a look at those faces with zero area? (using e.g. foamToVTK) Are they at the axis?

Have a look at the collapseEdges utility. I think it will remove your 0 area face at the axis. (if they are quads)

[vatant]

Hello Mattijs:
What is the fundamental difference betweeen an wedge type and cyclic type? If I consider an axi-symmetric grid (one cell thick) and impose wedge type condition, does OF compute in the rotational direction? For the same case, why doesnt a cyclic condition work?

Regards,

Vatant

[gfcoppola]

Hi all,

I am new to OF so my question may be stupid, anyway, I am trying to simulate an axisymmetric opposed jet configuration, where each jet has a circular flange flat with the exit section. I am using icoFoam from FoamX and blockMesh to generate the mesh. Below is the blockMeshDict file:

arguments "/home/gcoppola/OpenFOAM/gcoppola-1.3/run/tutorials/icoFoam/counterflow" off;

convertToMeters 0.001;

vertices
(
(-14.5852 0 0)
(-0.297656 0 0)
(-0.297656 1.19063 -0.051984)
(-14.5852 1.19063 -0.051984)
(0.297656 0 0)
(0.297656 1.19063 -0.051984)
(14.5852 0 0)
(14.5852 1.19063 -0.051984)
(0.297656 100 -4.366094)
(-0.297656 100 -4.366094)
(-0.297656 1.19063 0.051984)
(-14.5852 1.19063 0.051984)
(0.297656 1.19063 0.051984)
(14.5852 1.19063 0.051984)
(0.297656 100 4.366094)
(-0.297656 100 4.366094)
);

blocks
(
hex (0 1 2 3 0 1 10 11) (39 20 1) simpleGrading (0.04 0.3 1)
hex (1 4 5 2 1 4 12 10) (10 20 1) simpleGrading (1 0.3 1)
hex (4 6 7 5 4 6 13 12) (39 20 1) simpleGrading (25 0.3 1)
hex (2 5 8 9 10 12 14 15) (10 395 1) simpleGrading (1 25 1)
);

edges
(
);
patches
(
empty axis
(
(0 1 1 0)
(1 4 4 1)
(4 6 6 4)
)

wedge back
(
(0 1 2 3)
(0 1 10 11)
(1 4 5 2)
(1 4 12 10)
)

wedge front
(
(4 6 7 5)
(4 6 13 12)
(2 5 8 9)
(10 12 14 15)
)

patch lx_inlet
(
(0 0 11 3)
)

patch rx_inlet
(
(6 6 7 13)
)

wall top_flange
(
(2 3 11 10)
(9 2 10 15)
)

wall btm_flange
(
(7 5 12 13)
(5 8 14 12)
)
patch outlet
(
(8 9 15 14)
)

);

mergePatchPairs
(
);


The axis of symmetry is along X, the radius along Y and the one cell thickness along Z. The mesh is quite corse and I do get the 'Severe skewness' warning.
When I run the simulation, the result seem to be much like the flow between 2D flanged jets, as the velocity between the flanges seem to be constant along the radius.
I also used another file, which gives the same result, but the FRONT and BACK (wedge) palnes are parallel for the flanges and at an angle for the jets.
I am very likely doing some very stupid mistake, and would appreciate if you could help.

Regards,
Gianfilippo

[sek]

In my setup, the mesh is cylindrical, covering the 180 deg sector in the azimuthal direction. As a result, I have wedge elements touching the axis of symmetry at r = 0. The axis is not a boundary in my setup. The solutions seems to converge fine. But the solutions near the axis are not quite what I expected, showing high velocity in the body wake where I expect typical wake velocity profiles. I used starToFoam to convert the mesh generated using GridGen. Has anyone had similar experience?

[egp]

Sung-Eun,

Did you solve your problem? If your domain is 180 deg, I don't think the wedge BC is appropriate. I would use the cyclic condition instead.

Eric

[sek]

I haven;t reolved the problem yet. The solution converges and looks fine in the boundary layer. The velocity field in the wake at and near the axis looks wrong.

I'm using symmetry boundary for 0 and 180 deg plane. My question was about the treatments at r = 0. Apparently, the axis (r=0) is not a boundary in cell-centered unstructured FV . Yet, the solution behavior at and near the axis made me suspicious of whether it has anything to do with the treatments of the wedge-shaped cells touching the axis.

[T.D.]

Hi guys,
Anyone has ideas about how to setup flow between two parallel disks, as an axi-symmetric flow problem.

i tried 5° wedge geometry, and applied a rotatingWallVelocity for Upper disk and zero fixedValue for lower and for the Outer side i applied a zero Gradient.
For pressure i applied all zeroGradient except for Outer i put fixedValue 0.
for the empty axis i put in U, p a zeroGradient also.

when i run with icoFoam, i got extraordinary nonlogic results. it didn't work!

Any Ideas?
help please

thanks a lot

[diwakaranant]

Hi

I am solving axisymmetric flow problem of flow between two rotating cylinders using rhoCentralFoam. I generated the mesh in GMSH and converted it into openFoam using gmshToFoam utility.
But when I am running the simulation, I am getting non-zero radial velocity
near the outer rotating cylinder, which ideally should be zero.

The Uy and the velocity vector plots near the outer rotating wall
are attached.

The rotation is about X axis. Hence Uy is the radial velocity component.

Can some one tell why is this happening ?

Thanks
Anant

[diwakaranant]

Hi

I am solving axisymmetric flow problem of flow between two rotating cylinders using rhoCentralFoam. I generated the mesh in GMSH and converted it into openFoam using gmshToFoam utility.
But when I am running the simulation, I am getting non-zero radial velocity
near the outer rotating cylinder, which ideally should be zero.

The Uy and the velocity vector plots near the outer rotating wall
are attached.

The rotation is about X axis. Hence Uy is the radial velocity component.

Can some one tell why is this happening ?

Thanks
Anant

[anshul bansal]

anyone please tell me how i put my wedge patch for back and front face in 0/p,0/T,0/U files.

means i put like this
back
{
type wedge;
}

front
{
type wedge;
}

is it correct.?


actually i find this error when i run my c-d nozzle with rhoCentralFoam solver


--> FOAM FATAL IO ERROR:

patch type 'genericPatch' not constraint type 'wedge'
for patch back of field p in file "/home/anshu/OpenFOAM/anshu-2.2.1/run/tutorials/compressible/rhoCentralFoam/with_axis/0/p"

file: /home/anshu/OpenFOAM/anshu-2.2.1/run/tutorials/compressible/rhoCentralFoam/with_axis/0/p.boundaryField.back from line 41 to line 41.

From function wedgeFvPatchField<Type>::wedgeFvPatchField
(
const fvPatch& p,
const Field<Type>& field,
dictionary& dict
)

in file fields/fvPatchFields/constraint/wedge/wedgeFvPatchField.C at line 99.

FOAM exiting

[T.D.]

Hi,

This should help you:

http://openfoamwiki.net/index.php/Ma...s/AxiSymmetric

Regards,

T.D.

[anshul bansal]

thanx for ur reply.
first i have problem in checkMesh of my geometry.
i got this problem-

***Number of edges not aligned with or perpendicular to non-empty directions: 48277
<<Writing 20200 points on non-aligned edges to set nonAlignedEdges

what should i do?
i can send u my geometry file in your email .
*(i made my geometry in gambit and import in openfoam)

[anshul bansal]

Can anybody help me to put my boundary condition and patch field in c-d nozzle problem.

my boundary file is-
(
top
{
type wall;
nFaces 1000;
startFace 49825;
}
inlet
{
type patch;
nFaces 175;
startFace 50825;
}
outlet
{
type patch;
nFaces 175;
startFace 51000;
}
front
{
type wedge;
nFaces 2000;
startFace 51175;
}
back
{
type wedge;
nFaces 2000;
startFace 53175;
}
)

and intial (0 file) for pressure is --
internalField uniform 101325;

boundaryField
{
inlet
{
type fixedValue;
value uniform 235000;
}

outlet
{
type zeroGradient;
}

top
{
type zeroGradient;
}


back
{
type wedge;
}
front
{
type wedge;
}


}

[Rokerda]

Originally Posted by hjasak

Please make sure that the front and back of the wedge are in two separate patches.

Enjoy,

Hrv

Thank you very much for this information. I have been struggling with my code for a few days now

[openfoammaofnepo]

Hi All,

For the boundary condition "wedge" in OpenFOAM, how is it treated when the numerical fluxes are computed? Are they actually assumed that there are no any fluxes (i.e. convection, molecular and turbulent diffusions) across the wedge face?

I checked the source file as follows, but did not find any information in it about how the numerical flues on them.

https://github.com/OpenFOAM/OpenFOAM...FvPatchField.C

Do you kow where I can find the relevant information in the source files? Thank you.

OFFO

[dzordz]

Originally Posted by anshul bansal

thanx for ur reply.
first i have problem in checkMesh of my geometry.
i got this problem-

***Number of edges not aligned with or perpendicular to non-empty directions: 48277
<<Writing 20200 points on non-aligned edges to set nonAlignedEdges

what should i do?
i can send u my geometry file in your email .
*(i made my geometry in gambit and import in openfoam)

Hi Anshul,

what I found out is that when I create a geometry with snappyHexMesh (don't know if it is the same with Gambit) the front and back are not identical! So I get the same error when using extrude wedge. So what I do first is a linear extrude in one direction, so the both front and back patches are identical. Then you do wedge extrude and this should no longer be a problem.

Cheers.

[openfoammaofnepo]

Dear Eric,

Have you got the idea about the wedge boundary conditions implemented in OpenFOAM? I mean the governing equation in Cartersian coordinate system is solver, but the transformation for the BC is applied and then this small wedge can be used to represent the axisymmetric problem. Could you please comment the transformation implemented in OF? Thank you so much.

Originally Posted by egp

OF community,

I've recently begun experimenting with OF, and am posting my first question to the message board.

When simulating flow over an axisymmetric body, the User Manual (U-145, U-146) specifies that the wedge-type cells must be 5 degrees, and must straddle one of the coordinate planes.

What is the reason for this restriction?

So that I can understand the implementation, where would I look in the source?

Thank you,

Eric Paterson
Penn State Univ
State College, PA USA

[jiejie]

Originally Posted by openfoammaofnepo

Dear Eric,

Have you got the idea about the wedge boundary conditions implemented in OpenFOAM? I mean the governing equation in Cartersian coordinate system is solver, but the transformation for the BC is applied and then this small wedge can be used to represent the axisymmetric problem. Could you please comment the transformation implemented in OF? Thank you so much.

Interested to know the how the governing equation in Cartesian is handled for the wedge boundary conditions?