[Other] Can't Shake Erros: patch type 'patch' not constraint type 'empty'

BrendaEM · October 15, 2014, 22:54

Hi,

I am trying to make a naca duct, meshed with Solome/Netgen in tetrahedrons.
I've seen other threads about non-othogonal issues, but the mesh seems to check out.

I keep getting these in decomposePar, using scotch:

Code:

--> FOAM FATAL IO ERROR: 

    patch type 'patch' not constraint type 'empty'

:

The Mesh seems okay

Code:

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

Create polyMesh for time = 0

Time = 0

Mesh stats
    points:           2863229
    faces:            32727449
    internal faces:   32248843
    cells:            16244073
    faces per cell:   4
    boundary patches: 5
    point zones:      0
    face zones:       0
    cell zones:       0

Overall number of cells of each type:
    hexahedra:     0
    prisms:        0
    wedges:        0
    pyramids:      0
    tet wedges:    0
    tetrahedra:    16244073
    polyhedra:     0

Checking topology...
    Boundary definition OK.
    Cell to face addressing OK.
    Point usage OK.
    Upper triangular ordering OK.
    Face vertices OK.
    Number of regions: 1 (OK).

Checking patch topology for multiply connected surfaces...
    Patch               Faces    Points   Surface topology                  
    inlet               28210    14366    ok (non-closed singly connected)  
    nowall              244160   122731   ok (non-closed singly connected)  
    outlet1             27442    13982    ok (non-closed singly connected)  
    outlet2             2269     1235     ok (non-closed singly connected)  
    wall                176525   88892    ok (non-closed singly connected)  

Checking geometry...
    Overall domain bounding box (-0.16 -0.1 -0.0150429) (0.17 0.1 0.06)
    Mesh (non-empty, non-wedge) directions (1 1 1)
    Mesh (non-empty) directions (1 1 1)
    Boundary openness (-5.2313e-15 4.00333e-15 -7.32111e-14) OK.
    Max cell openness = 3.08746e-16 OK.
    Max aspect ratio = 5.19132 OK.
    Minimum face area = 1.22756e-08. Maximum face area = 2.5373e-06.  Face area magnitudes OK.
    Min volume = 6.91365e-13. Max volume = 1.2265e-09.  Total volume = 0.00409514.  Cell volumes OK.
    Mesh non-orthogonality Max: 56.1469 average: 15.1307
    Non-orthogonality check OK.
    Face pyramids OK.
    Max skewness = 0.734916 OK.
    Coupled point location match (average 0) OK.

Mesh OK.

End

I've looked at snappyhexmesh, but the setup looks really complicated.
I built from 2.3.0 source. Was anything fixed in 2.3.x that would affect this?
Does OpenFoam just not play well with tetrahedrons?

I am pretty disappointed. Any help would be appreciated.

BrendaEM · October 17, 2014, 16:21

I am considering opening a bug report.

deepsterblue · October 17, 2014, 16:41

You have a tetrahedral mesh, but use an empty patch type? Shouldn't that be restricted to 2D meshes alone?

You should post your entire decomposePar output

BrendaEM · October 17, 2014, 17:50

I wondered that too, that's why I (not cross) posted this in the appropriate sub forum.

Could you give me a hint, as to which type I could use for the truncated surfaces of a solve that is not necessarily symmetrical, nor 2.5D Extruded?

BTW, I've found that both decomposePar and IcoFoam both error, so I suspect that the problem is not in the decomposePar. I posted the icoFoam output at the end. It's pretty much the same error.

There just aren't enough small 3D samples in the tutorials, and the motorbike is so included and detailed as to make it formidable for a beginner tutorial.

p

Code:

/*--------------------------------*- C++ -*----------------------------------*\
| =========                 |                                                 |
| \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox           |
|  \\    /   O peration     | Version:  2.3.0                                 |
|   \\  /    A nd           | Web:      www.OpenFOAM.org                      |
|    \\/     M anipulation  |                                                 |
\*---------------------------------------------------------------------------*/
FoamFile
{
    version     2.0;
    format      ascii;
    class       volScalarField;
    object      p;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

dimensions      [0 2 -2 0 0 0 0];

internalField   uniform 0;

boundaryField
{
    inlet
    {
        type            zeroGradient;
    }

    outlet1
    {
        type            fixedValue;
        value           uniform 0;
    }

    outlet2
    {
        type            fixedValue;
        value           uniform 0;
    }

    wall
    {
        type            zeroGradient;
    }
    nowall
    {
        type            empty;
    }
}

// ************************************************************************* //

U:

Code:

/*--------------------------------*- C++ -*----------------------------------*\
| =========                 |                                                 |
| \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox           |
|  \\    /   O peration     | Version:  2.3.0                                 |
|   \\  /    A nd           | Web:      www.OpenFOAM.org                      |
|    \\/     M anipulation  |                                                 |
\*---------------------------------------------------------------------------*/
FoamFile
{
    version     2.0;
    format      ascii;
    class       volVectorField;
    object      U;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //


dimensions      [0 1 -1 0 0 0 0];

internalField   uniform (0 0 0);

boundaryField
{
    wall
    {
        type            fixedValue;
        value           uniform (0 0 0);
    }
    inlet
    {
        type            fixedValue;
        value           uniform (44.704 0 0); // 100MPH along x dimention
    }
    outlet1
    {
        type            zeroGradient;
    }
    outlet2
    {
        type            zeroGradient;
    }
    nowall
    {
        type             empty;
    }
}    

// ************************************************************************* //

icoFoam:

Code:

/*---------------------------------------------------------------------------*\
| =========                 |                                                 |
| \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox           |
|  \\    /   O peration     | Version:  2.3.0                                 |
|   \\  /    A nd           | Web:      www.OpenFOAM.org                      |
|    \\/     M anipulation  |                                                 |
\*---------------------------------------------------------------------------*/
Build  : 2.3.0-f5222ca19ce6
Exec   : icoFoam
Date   : Oct 17 2014
Time   : 14:47:05
Host   : "Brenda-W540"
PID    : 2999
Case   : /home/brenda/OpenFOAM/OpenFOAM-2.3.0/run/naca
nProcs : 1
sigFpe : Enabling floating point exception trapping (FOAM_SIGFPE).
fileModificationChecking : Monitoring run-time modified files using timeStampMaster
allowSystemOperations : Disallowing user-supplied system call operations

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

Create mesh for time = 0

Reading transportProperties

Reading field p



--> FOAM FATAL IO ERROR: 

    patch type 'patch' not constraint type 'empty'
    for patch nowall of field p in file "/home/brenda/OpenFOAM/OpenFOAM-2.3.0/run/naca/0/p"

file: /home/brenda/OpenFOAM/OpenFOAM-2.3.0/run/naca/0/p.boundaryField.nowall from line 46 to line 46.

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

    in file fields/fvPatchFields/constraint/empty/emptyFvPatchField.C at line 100.

FOAM exiting

decomposePar:

Code:

/*---------------------------------------------------------------------------*\ | ========= | | | \\ / F ield | OpenFOAM: The Open Source CFD Toolbox | | \\ / O peration | Version: 2.3.0 | | \\ / A nd | Web: www.OpenFOAM.org | | \\/ M anipulation | | \*---------------------------------------------------------------------------*/ Build : 2.3.0-f5222ca19ce6 Exec : decomposePar Date : Oct 17 2014 Time : 15:48:03 Host : "Brenda-W540" PID : 4482 Case : /home/brenda/OpenFOAM/OpenFOAM-2.3.0/run/naca nProcs : 1 sigFpe : Enabling floating point exception trapping (FOAM_SIGFPE). fileModificationChecking : Monitoring run-time modified files using timeStampMaster allowSystemOperations : Disallowing user-supplied system call operations // * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * // Create time Decomposing mesh region0 Create mesh Calculating distribution of cells Selecting decompositionMethod scotch Finished decomposition in 22.52 s Calculating original mesh data Distributing cells to processors Distributing faces to processors Distributing points to processors Constructing processor meshes Processor 0 Number of cells = 923645 Number of faces shared with processor 1 = 6120 Number of faces shared with processor 2 = 5371 Number of processor patches = 2 Number of processor faces = 11491 Number of boundary faces = 32175 Processor 1 Number of cells = 922718 Number of faces shared with processor 0 = 6120 Number of faces shared with processor 2 = 1531 Number of faces shared with processor 3 = 2722 Number of faces shared with processor 5 = 4224 Number of processor patches = 4 Number of processor faces = 14597 Number of boundary faces = 26623 Processor 2 Number of cells = 927333 Number of faces shared with processor 0 = 5371 Number of faces shared with processor 1 = 1531 Number of faces shared with processor 3 = 5250 Number of processor patches = 3 Number of processor faces = 12152 Number of boundary faces = 30852 Processor 3 Number of cells = 929606 Number of faces shared with processor 1 = 2722 Number of faces shared with processor 2 = 5250 Number of faces shared with processor 4 = 6600 Number of faces shared with processor 5 = 2100 Number of processor patches = 4 Number of processor faces = 16672 Number of boundary faces = 24460 Processor 4 Number of cells = 926893 Number of faces shared with processor 3 = 6600 Number of faces shared with processor 5 = 5008 Number of faces shared with processor 6 = 2124 Number of faces shared with processor 7 = 6303 Number of processor patches = 4 Number of processor faces = 20035 Number of boundary faces = 25977 Processor 5 Number of cells = 925384 Number of faces shared with processor 1 = 4224 Number of faces shared with processor 3 = 2100 Number of faces shared with processor 4 = 5008 Number of faces shared with processor 6 = 4326 Number of processor patches = 4 Number of processor faces = 15658 Number of boundary faces = 23410 Processor 6 Number of cells = 922515 Number of faces shared with processor 4 = 2124 Number of faces shared with processor 5 = 4326 Number of faces shared with processor 7 = 3650 Number of processor patches = 3 Number of processor faces = 10100 Number of boundary faces = 28170 Processor 7 Number of cells = 922298 Number of faces shared with processor 4 = 6303 Number of faces shared with processor 6 = 3650 Number of processor patches = 2 Number of processor faces = 9953 Number of boundary faces = 29187 Number of processor faces = 55329 Max number of cells = 929606 (0.492623% above average 925049) Max number of processor patches = 4 (23.0769% above average 3.25) Max number of faces between processors = 20035 (44.8427% above average 13832.2) Time = 0 --> FOAM FATAL IO ERROR: patch type 'patch' not constraint type 'empty' for patch nowall of field p in file "/home/brenda/OpenFOAM/OpenFOAM-2.3.0/run/naca/0/p" file: /home/brenda/OpenFOAM/OpenFOAM-2.3.0/run/naca/0/p.boundaryField.nowall from line 46 to line 46. From function emptyFvPatchField<Type>::emptyFvPatchField ( const fvPatch& p, const Field<Type>& field, const dictionary& dict ) in file fields/fvPatchFields/constraint/empty/emptyFvPatchField.C at line 100. FOAM exiting

Thank you for your reply : )

deepsterblue · October 17, 2014, 21:41

You may want to use a symmetry type boundary instead of empty if you're trying to achieve a pseudo 2D result with a 3D mesh. Otherwise, take a 2D mesh and extrude in the third direction by one cell to get a mesh that's truly 2D. Also ensure that the constant/polyMesh/boundary file is consistent with all your other fields like p, U, etc. That's probably the reason for the error that you're seeing.

BrendaEM · October 18, 2014, 12:56

Actually, I do not want a pseudo 2D solve. I am trying to do a 3D one, and so symmetry seem inappropriate. I don't want to confine the air near at the top and sides, so slip seems wrong as well.

From the manual, as in RTFM, I do not see an appropriate boundary condition for the truncated sides 3D simulations, whereas the air/fluid touching a face will cease/become extinct.

Thank you for your reply.

BrendaEM · October 18, 2014, 14:57

Interesting, I tried:

{
type fixedValue;
value uniform 0;
}

Which gave me an expecting "(" syntax-type error.

Looking at the motorbike example, I tried:

nowall
{
type slip;
}

...and the decomposePar worked, though I cannot find documentation explain whether or not "slip" will negate pressure. In other words, I don't know if there are fluid effects that send effects all the way to the boundaries of my solve/simulation, I do not want them ever coming back into the solution.

If I can get this case going, I will document it, make a video, and share it.

BrendaEM · October 18, 2014, 17:52

OpenFoam exited with an exception. It failed in So6. I don't know if it is is a currant number issue, or from my boundaries, but I should think that it OpenFoam should not crash with exceptions.

Please ignore:

Code:

Sadly, I am finding OpenFoam, cryptic, buggy, and poorly documented. It's shouldn't be this hard to set up and run a case.

Except for where timesteps and mesh are concerned, all the condition settings should live in the same directory/folder. The setup files all should have a common extension, such as .cfg. There are too many different configuration files.

OpenFoam's utilities and solvers should be stripped of inConSistant CamelCase, such as icoFOAM, ideasToFoam.

There needs to be better documentation for 3D examples. Every line in every tutorial configuration file should be commented.

The best tools have no pointy nails driven through their handles.

And lastly, I have lingering fantasies of this whole project being ported to Julia, for its multi-processor and clustering ease.

I feel better now.

deepsterblue · October 18, 2014, 18:14

I suspect you'll benefit from this:
http://www.sourceflux.de
OpenFOAM is not too bad once you get going. It's the basis for the PhD theses of plenty of folks, including mine.

BrendaEM · October 20, 2014, 14:23

I have seen OpenFoam throw exceptions, when it should have trapped and reported problems, but I also seen comparison with other solvers, whereas it seems too close to call which speaks the truth. Still the provided documentation is...interesting.

The book is interesting. I am looking for a lot of case setups using real world 3D examples, explanations and proper use of boundary conditions for 3D unsymmetrical solves--the kind of stuff I've been asking for on the forum. I though that PyFoam's development was in question, as it seems to be Pyflu, now, is not up to 2.3.0.

kmooney · November 5, 2014, 21:14

Quote:

{
type fixedValue;
value uniform 0;
}

Which gave me an expecting "(" syntax-type error.

Looking at the motorbike example, I tried:

nowall
{
type slip;
}

It looks like your trying to assign a scalar value to a vector field here.

Try this instead:

Code:

{
type fixedValue;
value uniform (0 0 0);
}

tmik · August 24, 2018, 12:03

It might be that you haven't defined the boundary file correctly. This is an easy mistake to make, and can cause this error.

To use the empty boundary condition using a mesh made in Salome, start with a 2D mesh first. Extrude this to create a single-element 3d mesh.

after you use

Code:

ideasUnvToFoam filename.unv

run

Code:

checkMesh

A key thing you'll see is the line below:

Code:

Checking geometry...
    Overall domain bounding box (-0.16 -0.1 -0.0150429) (0.17 0.1 0.06)
    Mesh (non-empty, non-wedge) directions (1 1 1)
    Mesh (non-empty) directions (1 1 1)

Even though the mesh was imported as 3D, we can define the faces which should be ignored - forcing OpenFOAM to see it as a 2D mesh.
Open the boundary file. Assuming 'frontAndback' is the group which contains the 2D faces, use the following:

Code:

    frontAndback
    {
        type            empty;
        nFaces          50269;
        startFace       85150;
    }

After you've done this, change each of the input files (U, P, etc):

Code:

    frontAndback
    {
        type            empty;
    }

Hope this helps.

Saeb-sahand · April 3, 2022, 18:32

To solve the the following error

"patch type 'wall' not constraint type 'symmetry' "

Edit the type of corresponding boundary in constant/polyMesh/boundary
to be symmetry rather than wall e.g.

boundary_name
{
type symmetry;
inGroups List<word> 1(symmetry);
nFaces 8117;
startFace 813891;
}

The following link is also useful

https://discourse.ladybug.tools/t/ho...tric-case/9563

October 17, 2014, 16:41		#3
deepsterblue Senior Member Sandeep Menon Join Date: Mar 2009 Location: Amherst, MA Posts: 403 Rep Power: 25	You have a tetrahedral mesh, but use an empty patch type? Shouldn't that be restricted to 2D meshes alone? You should post your entire decomposePar output __________________ Sandeep Menon University of Massachusetts Amherst https://github.com/smenon

October 17, 2014, 21:41		#5
deepsterblue Senior Member Sandeep Menon Join Date: Mar 2009 Location: Amherst, MA Posts: 403 Rep Power: 25	You may want to use a symmetry type boundary instead of empty if you're trying to achieve a pseudo 2D result with a 3D mesh. Otherwise, take a 2D mesh and extrude in the third direction by one cell to get a mesh that's truly 2D. Also ensure that the constant/polyMesh/boundary file is consistent with all your other fields like p, U, etc. That's probably the reason for the error that you're seeing. luoyang likes this. __________________ Sandeep Menon University of Massachusetts Amherst https://github.com/smenon

October 18, 2014, 12:56		#6
BrendaEM Member Brenda EM Join Date: Jan 2012 Posts: 38 Rep Power: 14	Actually, I do not want a pseudo 2D solve. I am trying to do a 3D one, and so symmetry seem inappropriate. I don't want to confine the air near at the top and sides, so slip seems wrong as well. From the manual, as in RTFM, I do not see an appropriate boundary condition for the truncated sides 3D simulations, whereas the air/fluid touching a face will cease/become extinct. Thank you for your reply. Last edited by BrendaEM; October 18, 2014 at 14:33.

October 18, 2014, 18:14		#9
deepsterblue Senior Member Sandeep Menon Join Date: Mar 2009 Location: Amherst, MA Posts: 403 Rep Power: 25	I suspect you'll benefit from this: http://www.sourceflux.de OpenFOAM is not too bad once you get going. It's the basis for the PhD theses of plenty of folks, including mine. __________________ Sandeep Menon University of Massachusetts Amherst https://github.com/smenon

August 24, 2018, 12:03		#12
tmik New Member Thomas M Join Date: Aug 2018 Posts: 20 Rep Power: 7	It might be that you haven't defined the boundary file correctly. This is an easy mistake to make, and can cause this error. To use the empty boundary condition using a mesh made in Salome, start with a 2D mesh first. Extrude this to create a single-element 3d mesh. after you use Code: ideasUnvToFoam filename.unv run Code: checkMesh A key thing you'll see is the line below: Code: Checking geometry... Overall domain bounding box (-0.16 -0.1 -0.0150429) (0.17 0.1 0.06) Mesh (non-empty, non-wedge) directions (1 1 1) Mesh (non-empty) directions (1 1 1) Even though the mesh was imported as 3D, we can define the faces which should be ignored - forcing OpenFOAM to see it as a 2D mesh. Open the boundary file. Assuming 'frontAndback' is the group which contains the 2D faces, use the following: Code: frontAndback { type empty; nFaces 50269; startFace 85150; } After you've done this, change each of the input files (U, P, etc): Code: frontAndback { type empty; } Hope this helps. Last edited by tmik; August 24, 2018 at 13:26.

October 17, 2014, 16:21		#2
BrendaEM Member Brenda EM Join Date: Jan 2012 Posts: 38 Rep Power: 14	I am considering opening a bug report.

October 18, 2014, 14:57		#7
BrendaEM Member Brenda EM Join Date: Jan 2012 Posts: 38 Rep Power: 14	Interesting, I tried: { type fixedValue; value uniform 0; } Which gave me an expecting "(" syntax-type error. Looking at the motorbike example, I tried: nowall { type slip; } ...and the decomposePar worked, though I cannot find documentation explain whether or not "slip" will negate pressure. In other words, I don't know if there are fluid effects that send effects all the way to the boundaries of my solve/simulation, I do not want them ever coming back into the solution. If I can get this case going, I will document it, make a video, and share it.

October 20, 2014, 14:23		#10
BrendaEM Member Brenda EM Join Date: Jan 2012 Posts: 38 Rep Power: 14	I have seen OpenFoam throw exceptions, when it should have trapped and reported problems, but I also seen comparison with other solvers, whereas it seems too close to call which speaks the truth. Still the provided documentation is...interesting. The book is interesting. I am looking for a lot of case setups using real world 3D examples, explanations and proper use of boundary conditions for 3D unsymmetrical solves--the kind of stuff I've been asking for on the forum. I though that PyFoam's development was in question, as it seems to be Pyflu, now, is not up to 2.3.0.

April 3, 2022, 18:32		#13
Saeb-sahand New Member zurich Join Date: Apr 2022 Posts: 3 Rep Power: 4	To solve the the following error "patch type 'wall' not constraint type 'symmetry' " Edit the type of corresponding boundary in constant/polyMesh/boundary to be symmetry rather than wall e.g. boundary_name { type symmetry; inGroups List<word> 1(symmetry); nFaces 8117; startFace 813891; } The following link is also useful https://discourse.ladybug.tools/t/ho...tric-case/9563 dkk_BO likes this.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
activeBaffleVelocity boundary condition ?	om3ro	OpenFOAM Programming & Development	10	November 16, 2020 23:26
[Commercial meshers] Fluent3DMeshToFoam	simvun	OpenFOAM Meshing & Mesh Conversion	50	January 19, 2020 15:33
[Commercial meshers] Mesh conversion problem (fluent3DMeshToFoam)	Aadhavan	OpenFOAM Meshing & Mesh Conversion	2	March 8, 2018 01:47
Near wall treatment in k-omega SST	Arnoldinho	OpenFOAM Running, Solving & CFD	38	March 8, 2017 13:48
Divergent temperature in chtMultiRegion(Simple)Foam	akrasemann	OpenFOAM Running, Solving & CFD	13	March 24, 2014 02:54