[julieng]

Hello,


I am stuck with one problem I have since 1 week .
I try to reproduce results obtained with blockmesh mesh and chtMultiRegionSimpleFoam of a 2D simple case (1 solid domain enclosed with 2 air domains) with Salome mesh. The blockMesh case runs very fine. Results are in accordance with other software.



















My BC are the same strictly the same. But the topAir domain doesn't converge (p_rgh and U error value of 0.01-0.015 after 3000 iter). And I see some difference in Temperature field other parameters, pressure velocity. And I don't know why. Meshes is identical.


with blockMesh





with Salome








I don't know what happened at the outlet of the topAir region close to the wall of the plate.

The only little difference in the 2 cases is that write by hand the empty condition on the front and bottom faces for each domains (I need to write it in the all cellToRegion files each time I redo the mesh).


The Salome geometry is not oriented and not centered on the point (0 0 0) as the geometry of blockMesh, so I thank that it could a proble of pRefCell 0 on a coupled boundary but without pRefCell I have the same problem.

I think that it could be a problem pRefCell. Orientation of the gravity is not the problem also.



I cannot join my Salome file it is too large. If I clear mesh data and recompute mesh, I break the link of the groups creation. Maybe I can upload my mesh via a different way?


Please I need any idea to understand what is wrong


Best regards



Julien

[julieng]

There is attached the 0 and system files for the blockMesh and salome cases.


I use few BC names with the blockMesh case (minX maxX ... ), names shared with all the domains. With the Salome case, I have to name independently all domains boundaries because I have conflicts in ParaFoam if I do not.



PS:I use a lot of postprocess operations in the controlDict files for power, temp average, massflow etc..



Cases run with Openfoam 5 on blueCFDcore windows 10.

[Bloerb]

Why is the minZ_botAir and maxZ_topAir defined as an outlet and not a wall for T,U and p_rgh boundaries? Is that correct?


Please run this as well to be on the safe side.

checkMesh -region topAir
 checkMesh -region botAir

[julieng]

Thank you for your answer.


Yes min and max Z are "open" boundaries not walls, the same defined with the blockMesh case.


I run the commands for the topAir domain

checkMesh -region topAir
/*---------------------------------------------------------------------------*\
| ========= | |
| \\ / F ield | OpenFOAM: The Open Source CFD Toolbox |
| \\ / O peration | Version: 5.x |
| \\ / A nd | Web: www.OpenFOAM.org |
| \\/ M anipulation | |
\*---------------------------------------------------------------------------*/
/* Windows 32 and 64 bit porting by blueCAPE: http://www.bluecape.com.pt *\
| Based on Windows porting (2.0.x v4) by Symscape: http://www.symscape.com |
\*---------------------------------------------------------------------------*/
Build : 5.x-963176928289
Exec : C:/PROGRA~1/BLUECF~1/OpenFOAM-5.x/platforms/mingw_w64GccDPInt32Opt/bin/checkMesh.exe -region topAir
Date : Feb 16 2019
Time : 13:55:20
Host : "PC_JULIEN"
PID : 9912
I/O : uncollated
Case : C:/PROGRA~1/BLUECF~1/OFUSER~1/run/CHT_1S~2
nProcs : 1
SigFpe : Enabling floating point exception trapping (FOAM_SIGFPE).
fileModificationChecking : Monitoring run-time modified files using timeStampMaster (fileModificationSkew 10)
allowSystemOperations : Allowing user-supplied system call operations

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

Create polyMesh topAir for time = 0

Time = 0

Mesh stats
points: 20402
internal points: 0
faces: 40200
internal faces: 19800
cells: 10000
faces per cell: 6
boundary patches: 5
point zones: 0
face zones: 4
cell zones: 3

Overall number of cells of each type:
hexahedra: 10000
prisms: 0
wedges: 0
pyramids: 0
tet wedges: 0
tetrahedra: 0
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_topAir 100 202 ok (non-closed singly connected)
maxZ_topAir 100 202 ok (non-closed singly connected)
outlet_topAir 100 202 ok (non-closed singly connected)
top_bottom_topAir 20000 20402 ok (non-closed singly connected)
topAir_to_plate 100 202 ok (non-closed singly connected)

Checking geometry...
Overall domain bounding box (0 0 0.27) (0.5 1 0.5)
Mesh has 2 geometric (non-empty/wedge) directions (1 0 1)
Mesh has 2 solution (non-empty) directions (1 0 1)
All edges aligned with or perpendicular to non-empty directions.
Boundary openness (0 -1.10681645e-016 -1.3138734e-016) OK.
Max cell openness = 1.19998264e-016 OK.
Max aspect ratio = 8.53204465 OK.
Minimum face area = 2.93013e-006. Maximum face area = 0.005860263. Face area magnitudes OK.
Min volume = 2.93013e-006. Max volume = 2.9301315e-005. Total volume = 0.115. Cell volumes OK.
Mesh non-orthogonality Max: 0 average: 0
Non-orthogonality check OK.
Face pyramids OK.
Max skewness = 1.66533454e-013 OK.
Coupled point location match (average 0) OK.

Mesh OK.

End


And for the botAir domain


checkMesh -region botAir
/*---------------------------------------------------------------------------*\
| ========= | |
| \\ / F ield | OpenFOAM: The Open Source CFD Toolbox |
| \\ / O peration | Version: 5.x |
| \\ / A nd | Web: www.OpenFOAM.org |
| \\/ M anipulation | |
\*---------------------------------------------------------------------------*/
/* Windows 32 and 64 bit porting by blueCAPE: http://www.bluecape.com.pt *\
| Based on Windows porting (2.0.x v4) by Symscape: http://www.symscape.com |
\*---------------------------------------------------------------------------*/
Build : 5.x-963176928289
Exec : C:/PROGRA~1/BLUECF~1/OpenFOAM-5.x/platforms/mingw_w64GccDPInt32Opt/bin/checkMesh.exe -region botAir
Date : Feb 16 2019
Time : 13:58:47
Host : "PC_JULIEN"
PID : 10196
I/O : uncollated
Case : C:/PROGRA~1/BLUECF~1/OFUSER~1/run/CHT_1S~2
nProcs : 1
SigFpe : Enabling floating point exception trapping (FOAM_SIGFPE).
fileModificationChecking : Monitoring run-time modified files using timeStampMaster (fileModificationSkew 10)
allowSystemOperations : Allowing user-supplied system call operations

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

Create polyMesh botAir for time = 0

Time = 0

Mesh stats
points: 20402
internal points: 0
faces: 40200
internal faces: 19800
cells: 10000
faces per cell: 6
boundary patches: 5
point zones: 0
face zones: 4
cell zones: 3

Overall number of cells of each type:
hexahedra: 10000
prisms: 0
wedges: 0
pyramids: 0
tet wedges: 0
tetrahedra: 0
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_botAir 100 202 ok (non-closed singly connected)
minZ_botAir 100 202 ok (non-closed singly connected)
outlet_botAir 100 202 ok (non-closed singly connected)
top_bottom_botAir 20000 20402 ok (non-closed singly connected)
botAir_to_plate 100 202 ok (non-closed singly connected)

Checking geometry...
Overall domain bounding box (0 0 0) (0.5 1 0.23)
Mesh has 2 geometric (non-empty/wedge) directions (1 0 1)
Mesh has 2 solution (non-empty) directions (1 0 1)
All edges aligned with or perpendicular to non-empty directions.
Boundary openness (0 1.05954299e-017 0) OK.
Max cell openness = 1.26904551e-016 OK.
Max aspect ratio = 8.53204465 OK.
Minimum face area = 2.93013e-006. Maximum face area = 0.00586026339. Face area magnitudes OK.
Min volume = 2.93013e-006. Max volume = 2.9301317e-005. Total volume = 0.115. Cell volumes OK.
Mesh non-orthogonality Max: 0 average: 0
Non-orthogonality check OK.
Face pyramids OK.
Max skewness = 8.32678562e-014 OK.
Coupled point location match (average 0) OK.

Mesh OK.

End



Seems ok for me.



Best regards

[julieng]

Hello,


I think that my outlet BC is a problem. Have a look at the flow fields and convergence curves for different inlet fixed values. I have also modify the maxZ BC to a symmetryPlane; the velocity is more stable in the top domain. Whitout symmetryPlane condition it is a lot more perturbated.



Vinlet= 1m/s














V=0.1 m/s










I think that my case is not stationnary due to buyoancy forces, so I try to put the g gravity in the x direction (-9.81 0 0) with 0.1 m/s at the inlet. But I have still the same behaviour with a perturbated outlet.


I try also with different outlet BC for T



0/ T


outlet_topAir
{
type zeroGradient;
value uniform 400;



instead of



outlet_topAir
{
type inletOutlet;
value uniform 400;
inletValue uniform 400;


Still have a non natural behaviour at the outlet. It seems that the outlet is twisted, wrong orientation.








These new elements don't explain my first question about why I have different results with the same meshes.


Is it possible that my outlet boundary is not well oriented ? Does the order of the created points defining the geometry can have an influence ?
I know that defining the mesh in Salome, the grading on edge depends of the order of created points.


A second question: If I add a symmetry plane like I did in this case, I increase drastically the Reynolds number, because now, the external flow becomes an internal flow between to plates ?



Best regards


Julien