|
[Sponsors] | |||||
|
|
|
#1 | |
|
Member
William
Join Date: Feb 2011
Location: Minnesota USA
Posts: 33
Rep Power: 17 ![]() |
Hey,
I'm having a hard time getting the wedge to look right. Can somebody tell me what I am doing wrong. Quote:
|
||
|
|
|
||
|
|
|
#2 |
|
New Member
Radko Bankras
Join Date: Jul 2010
Location: Almere, The Netherlands
Posts: 5
Rep Power: 17 ![]() |
billynoe,
Your wedge looks better if you don't split up your blocks, so: Code:
hex (3 31 45 3 0 30 44 0) (1 1 1) simpleGrading (1 1 1)
hex (2 18 32 2 1 19 33 1) (1 1 1) simpleGrading (1 1 1)
hex (21 25 39 35 20 22 36 34) (1 1 1) simpleGrading (1 1 1)
hex (27 41 35 21 29 43 39 25) (1 1 1) simpleGrading (1 1 1)
hex (38 42 40 37 24 28 26 23) (1 1 1) simpleGrading (1 1 1)
I guess the AER patch defines internal walls in the wedge of your cylinder. In that case I would define only the blocks which form the volume of your wedge together. This way it is easier to increase the grading in X and Y direction, and you can keep your wedge 1 cell thick. Kind regards, Radko Last edited by Radko; February 12, 2011 at 07:25. Reason: style |
|
|
|
|
|
|
|
|
#3 |
|
Member
William
Join Date: Feb 2011
Location: Minnesota USA
Posts: 33
Rep Power: 17 ![]() |
Yes that definitely fixed the shape I changed the faces to (1 1 1) on the AER block and input output and (15 1 5) on the wedge. I am getting errors in checkMesh from the internal walls. I think I will need to use snappyHex to join the blocks is this correct?
|
|
|
|
|
|
|
|
|
#4 |
|
New Member
Radko Bankras
Join Date: Jul 2010
Location: Almere, The Netherlands
Posts: 5
Rep Power: 17 ![]() |
Hi,
Any other mesher would be easier to use for your design. I suggest to have a look at Salome for drawing, meshing, and exporting to Foam via ideasUnvToFoam. This will allow you to perform the logical actions of subtracting two solids, like I guess you are trying to achieve. As an example of how the structure should be defined in blockMesh (for all I know), I have modified your file and added some extra points. Note that your AER structure has not been put in yet. It will take a few more hours to do so. Make sure the patches are defined in clock-wise direction when looking from inside the mesh. Code:
/*--------------------------------*- C++ -*----------------------------------*\
| ========= | |
| \\ / F ield | OpenFOAM: The Open Source CFD Toolbox |
| \\ / O peration | Version: 1.7.1 |
| \\ / A nd | Web: www.OpenFOAM.com |
| \\/ M anipulation | |
\*---------------------------------------------------------------------------*/
FoamFile
{
version 2.0;
format ascii;
class dictionary;
object blockMeshDict;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //
convertToMeters 0.146;
vertices
(
(0 241.3 0) //0
(0 0 0) //1
(0 -12.7 0) //2
(0 -1524 0) //3
(279.4 -12.7 0) //4
(279.4 0 0) //5
(292.1 38.1 0) //6
(292.1 -313.6926357352644 0) //7
(304.8 38.1 0) //8
(304.8 13.51505771044411 0) //9
(304.8 0 0) //10
(304.8 -304.8 0) //11
(863.6 216.9016246183983 0) //12
(863.6 -521.7016246183977 0) //13
(867.943655820235 204.9675283344171 0) //14
(867.943655820235 -509.767528334418 0) //15
(7620 241.3 0) //16
(7620 -1524 0) //17
(279.1341 -12.7 12.1873) //18
(279.1341 0 12.1873) //19
(291.822 38.1 12.7412) //20
(291.822 -313.6926357352644 12.7412) //21
(304.5099 38.1 13.2952) //22
(304.5099 13.51505771044411 13.952) //23
(304.5099 0 13.29522) //24
(304.5099 -304.8 13.2952) //25
(862.7780 216.9016246183983 37.6697) //26
(862.7780 -521.7016246183977 37.6697) //27
(867.1176 204.9675283344171 37.8592) //28
(867.1176 -509.767528334418 37.8592) //29
(7612.7474 241.3 332.3797) //30
(7612.7474 -1524 332.3797) //31
(279.1341 -12.7 -12.1873) //32
(279.1341 0 -12.1873) //33
(291.822 38.1 -12.7412) //34
(291.822 -313.6926357352644 -12.7412) //35
(304.5099 38.1 -13.2952) //36
(304.5099 13.51505771044411 -13.2952) //37
(304.5099 0 -13.2952) //38
(304.5099 -304.8 -13.2952) //39
(862.7780 216.9016246183983 -37.6697) //40
(862.7780 -521.7016246183977 -37.6697) //41
(867.1176 204.9675283344171 -37.8592) //42
(867.1176 -509.767528334418 -37.8592) //43
(7612.7474 241.3 -332.3797) //44
(7612.7474 -1524 -332.3797) //45
(279.1341 241.3 12.1873) //46 new
(279.1341 241.3 -12.1873) //47 new
(279.1341 -1524 12.1873) //48 new
(279.1341 -1524 -12.1873) //49 new
(7612.7474 0 332.3797) //50 new
(7612.7474 0 -332.3797) //51 new
(7612.7474 -12.7 332.3797) //52 new
(7612.7474 -12.7 -332.3797) //53 new
);
blocks
(
hex (3 48 49 3 2 18 32 2) (20 1 20) simpleGrading (1 1 1) //0
hex (1 19 33 1 0 46 47 0) (20 1 10) simpleGrading (1 1 1) //1
hex (48 31 45 49 18 52 53 32) (100 1 20) simpleGrading (1 1 1) //2
hex (18 52 53 32 19 50 51 33) (100 1 5) simpleGrading (1 1 1) //3
hex (19 50 51 33 46 30 44 47) (100 1 10) simpleGrading (1 1 1) //4
);
patches
(
wall Top
(
(0 46 47 0)
(46 30 44 47)
)
wall bottom
(
(3 49 48 3)
(48 49 45 31)
)
wall back
(
(31 45 53 52)
(52 53 51 50)
(50 51 44 30)
)
patch inlet
(
(2 18 32 2)
)
patch outlet
(
(1 33 19 1)
)
wall nub
(
(18 19 33 32)
)
wedge side1
(
(3 2 18 48)
(1 0 46 19)
(48 18 52 31)
(18 19 50 52)
(19 46 30 50)
)
wedge side2
(
(3 2 32 49)
(1 0 47 33)
(49 32 53 45)
(32 33 51 53)
(33 47 44 51)
)
empty zoinks
(
(3 2 2 3)
(1 0 0 1)
)
);
mergePatchPairs
(
);
Last edited by Radko; February 15, 2011 at 17:43. Reason: no comments |
|
|
|
|
|
|
|
|
#5 |
|
Member
William
Join Date: Feb 2011
Location: Minnesota USA
Posts: 33
Rep Power: 17 ![]() |
Thanks for the help Radko.
I have been using gmsh for converting and meshing iges files which is ok, but I have been having problems with makeAxialMesh specifically collapseEdges. I also tried just reassigning the generic patches created in gmshToFoam into wedge type which did not work. Do you know of another way to make wedge geometry? Beyond that I tried a 3D test case icoFoam using the gmshToFoam mesh assigning the sides of the wedge as "walls" as well as the other patches. Except output as constant velocity 1.3 m/s and input as 0 pressure just to see what would happen and the solution wouldn't converge after about 10 steps. Maybe I should set it as a negative pressure on input? I am basically trying to model a vertical mixer I know the velocity in the volute due to an impeller, which I am simplifying. Eventually I will need to model this as interfoam as the top surface is exposed to air. So this looks like it is going to be a long time before I get any data that is worth anything. I will try out the revised geometry tomorrow and I'll let you know how it goes. |
|
|
|
|
|
![]() |
| Thread Tools | Search this Thread |
| Display Modes | |
|
|
Similar Threads
|
||||
| Thread | Thread Starter | Forum | Replies | Last Post |
| 3D Hybrid Mesh Errors | DarrenC | ANSYS Meshing & Geometry | 11 | August 5, 2013 07:42 |
| blockMesh: block with 6 vertexes | dani | OpenFOAM | 3 | June 25, 2009 14:13 |
| Bug in blockMesh | feymark | OpenFOAM Bugs | 8 | March 24, 2009 23:11 |
| basic of mesh refinement | arya | CFX | 4 | June 19, 2007 13:21 |
| Icemcfd 11: Loss of mesh from surface mesh option? | Joe | CFX | 2 | March 26, 2007 19:10 |