 |
 |
 |
|
|
|
[Sponsors] | ||||
[snappyHexMesh] Multiple sets of layers to gain more control over distribution |
 |
|
|
LinkBack | Thread Tools | Search this Thread | Display Modes |
April 16, 2021, 11:14
|
Multiple sets of layers to gain more control over distribution
|
#1 |
|
Member
Mike Worth
Join Date: Jun 2019
Posts: 45
Rep Power: 6  |
I'm trying to insert layers that have a thickness not described by a geometric
series. (I want an initial exponential growth, a plateau, then another exponential growth to finally meet the background cell size) I thought I had it working by running SHM multiple times in a row, with a series of slightly altered snappyHexMeshDicts, but now everything is crashing when I come to run interFoam (velocities near the interface go very high). I've stripped things back to try and work out the problem, and have what should be two ways of making essentially the same mesh. This seems to show that running SHM twice is the problem; could anyone suggest why this matters? Is there something subtle going on in the way that SHM pushes the old layers back that might be making a substantial difference? Nothing jumps out at me from checkMesh (although I might not be looking in the right place) The 'normal' approach like this works: Code:
addLayersControls
{
relativeSizes false;
layers
{
"movingpart.*"
{
nSurfaceLayers 14;//Calculated using https://openfoamwiki.net/index.php/Scripts/blockMesh_grading_calculation
}
}
// Expansion factor for layer mesh
expansionRatio 1.258498951;
firstLayerThickness 5e-6;
Outer layers (first) Code:
addLayersControls
{
relativeSizes false;
layers
{
"movingpart.*"
{
//Calculated using https://openfoamwiki.net/index.php/Scripts/blockMesh_grading_calculation
nSurfaceLayers 10;
}
}
// Expansion factor for layer mesh
expansionRatio 1.258498951;
firstLayerThickness 12.50e-6;
Code:
addLayersControls
{
relativeSizes false;
layers
{
"movingpart.*"
{
//Calculated using https://openfoamwiki.net/index.php/Scripts/blockMesh_grading_calculation
nSurfaceLayers 5;
}
}
// Expansion factor for layer mesh
expansionRatio 1.258498951;
firstLayerThickness 5e-6;
[Apologies if this counts as cross posting - I've discovered this issue while troubleshooting something I was posting about in the solving sub-forum. A post here on SHM seems much more applicable with what I've now figured out.] Thanks, Mike |
|
|
|
|
|
April 19, 2021, 08:20
|
|
#2 |
|
Member
Mike Worth
Join Date: Jun 2019
Posts: 45
Rep Power: 6 |
I've been looking into this further, and have reduced the inner-layers to 4 and the intermediate cell size to 1.258498951; this should notionally create an identical mesh to the 1-step method. Having visually inspected it, it does look incredibly similar, with just subtle and minor differences in the warping as cells are moved away from the surface.
Somehow it still gives very different performance, with artifically high velocities hampering the solution. I've diffed the checkMesh output, which I include below. The differences here all seem incredibly minor - can anyone suggest what might be going on? Code:
--- ../checkMesh.SHMx1 2021-04-19 09:15:56.378157603 +0000
+++ ../checkMesh.SHMx2 2021-04-19 09:53:57.548388806 +0000
@@ -1,142 +1,142 @@
/*---------------------------------------------------------------------------*\
| ========= | |
| \\ / F ield | OpenFOAM: The Open Source CFD Toolbox |
| \\ / O peration | Version: v1912 |
| \\ / A nd | Website: www.openfoam.com |
| \\/ M anipulation | |
\*---------------------------------------------------------------------------*/
Build : _f3950763fe-20191219 OPENFOAM=1912
Arch : "LSB;label=32;scalar=64"
Exec : checkMesh -allGeometry
Date : Apr 19 2021
-Time : 09:15:54
+Time : 09:53:56
Host : ip-172-31-8-169
-PID : 5238
+PID : 20125
I/O : uncollated
Case : /home/ubuntu/casename
nProcs : 1
trapFpe: Floating point exception trapping enabled (FOAM_SIGFPE).
fileModificationChecking : Monitoring run-time modified files using timeStampMaster (fileModificationSkew 10)
allowSystemOperations : Allowing user-supplied system call operations
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //
Create time
Using #calc at line 69 in file "/home/ubuntu/casename/system/controlDict"
Using #codeStream with "/home/ubuntu/casename/dynamicCode/platforms/linux64GccDPInt32Opt/lib/libcodeStream_90540247e08e92844c35514f8198f20ebc47281f.so"
Using #calc at line 70 in file "/home/ubuntu/casename/system/controlDict"
Using #codeStream with "/home/ubuntu/casename/dynamicCode/platforms/linux64GccDPInt32Opt/lib/libcodeStream_fae589c14ad73a95fb40be0cd9ce4cbb1179cdeb.so"
Using #calc at line 72 in file "/home/ubuntu/casename/system/controlDict"
Using #calc at line 73 in file "/home/ubuntu/casename/system/controlDict"
Using #codeStream with "/home/ubuntu/casename/dynamicCode/platforms/linux64GccDPInt32Opt/lib/libcodeStream_c5283eb0bd727a9daf6cb00bc145aac9a2439283.so"
Create mesh for time = 0
Enabling all geometry checks.
Time = 0
Mesh stats
points: 187080
internal points: 0
faces: 365251
internal faces: 179277
cells: 90642
faces per cell: 6.00745791134
boundary patches: 10
point zones: 0
face zones: 4
cell zones: 4
Overall number of cells of each type:
hexahedra: 88815
prisms: 626
wedges: 0
pyramids: 0
tet wedges: 0
tetrahedra: 0
polyhedra: 1201
Breakdown of polyhedra by number of faces:
faces number of cells
6 6
7 1088
8 107
Checking topology...
Boundary definition OK.
Cell to face addressing OK.
Point usage OK.
Upper triangular ordering OK.
Face vertices OK.
*Number of regions: 2
The mesh has multiple regions which are not connected by any face.
<<Writing region information to "0/cellToRegion"
<<Writing region 0 with 7200 cells to cellSet region0
<<Writing region 1 with 83442 cells to cellSet region1
Checking patch topology for multiply connected surfaces...
Patch Faces Points Surface topology Bounding box
outerAMI 600 1202 ok (non-closed singly connected) (0.0249762055395 0 -0.00109048468413) (0.0249762055395 0.6 0.00109048468413)
top 37 77 ok (non-closed singly connected) (1.0597848687e-75 0.6 -0.0016357270262) (0.0374643083093 0.6 0.0016357270262)
bottom 37 77 ok (non-closed singly connected) (0 0 -0.0016357270262) (0.0374643083093 0 0.0016357270262)
back 90642 93755 ok (non-closed singly connected) (-4.83868159767e-40 0 -0.0016357270262) (0.0374643083093 0.6 1.00830515897e-43)
front 90642 93755 ok (non-closed singly connected) (-4.83868159767e-40 0 -2.11261400988e-41) (0.0374643083093 0.6 0.0016357270262)
innerAMI 1348 2698 ok (non-closed singly connected) (0.0249762055395 0 -0.00109048468413) (0.0249762055395 0.6 0.00109048468413)
thing 441 883 ok (non-closed singly connected) (0 0.3 -0.000785148938775) (0.0179828672143 0.340000003576 0.000785148938775)
thing_wetted 1627 3255 ok (non-closed singly connected) (0 0.10000000149 -0.000959626516838) (0.0219790607557 0.3 0.000959626516838)
bathWall 325 652 ok (non-closed singly connected) (0.0374643083093 0 -0.0016357270262) (0.0374643083093 0.325 0.0016357270262)
atmosphereSide 275 552 ok (non-closed singly connected) (0.0374643083093 0.325 -0.0016357270262) (0.0374643083093 0.6 0.0016357270262)
Checking faceZone topology for multiply connected surfaces...
FaceZone Faces Points Surface topology Bounding box
top 25 51 ok (non-closed singly connected) (1.0597848687e-75 0.6 -0.00109048468413) (0.0249762055395 0.6 0.00109048468413)
- centralTopFaces 102 205 ok (non-closed singly connected) (0 0.342870999563 -0.00109048468413) (0.0249762055395 0.343124069787 0.00109048468413)
- centralBottomFaces 102 205 ok (non-closed singly connected) (0 0.0968728834598 -0.00109048468413) (0.0249762055395 0.0971234991189 0.00109048468413)
+ centralTopFaces 102 205 ok (non-closed singly connected) (0 0.342870810231 -0.00109048468413) (0.0249762055395 0.343123893846 0.00109048468413)
+ centralBottomFaces 102 205 ok (non-closed singly connected) (0 0.0968732213925 -0.00109048468413) (0.0249762055395 0.0971243427523 0.00109048468413)
bottom 25 51 ok (non-closed singly connected) (0 0 -0.00109048468413) (0.0249762055395 0 0.00109048468413)
Checking basic cellZone addressing...
CellZone Cells Points VolumeBoundingBox
background 7200 15626 2.04271272065e-05 (0.0249762055395 0 -0.0016357270262) (0.0374643083093 0.6 0.0016357270262)
- topBlockCells 7033 14532 6.99977663975e-06 (-4.83868159767e-40 0.342870999563 -0.00109048468413) (0.0249762055395 0.6 0.00109048468413)
- centralBlockCells 73514 151236 3.07632624824e-06 (0 0.0968728834598 -0.00109048468413) (0.0249762055395 0.343124069787 0.00109048468413)
- bottomBlockCells 2895 6096 2.64138591764e-06 (0 0 -0.00109048468413) (0.0249762055395 0.0971234991189 0.00109048468413)
+ topBlockCells 7032 14530 6.99971599829e-06 (-4.83868159767e-40 0.342870810231 -0.00109048468413) (0.0249762055395 0.6 0.00109048468413)
+ centralBlockCells 73515 151238 3.07638209319e-06 (0 0.0968732213925 -0.00109048468413) (0.0249762055395 0.343123893846 0.00109048468413)
+ bottomBlockCells 2895 6096 2.64139071415e-06 (0 0 -0.00109048468413) (0.0249762055395 0.0971243427523 0.00109048468413)
Checking geometry...
Overall domain bounding box (-4.83868159767e-40 0 -0.0016357270262) (0.0374643083093 0.6 0.0016357270262)
Mesh has 2 geometric (non-empty/wedge) directions (1 1 0)
Mesh has 3 solution (non-empty) directions (1 1 1)
- Wedge back with angle 2.50000000138 degrees
- Wedge front with angle 2.50000000138 degrees
+ Wedge back with angle 2.50000000133 degrees
+ Wedge front with angle 2.50000000133 degrees
All edges aligned with or perpendicular to non-empty directions.
- Boundary openness (1.64702991369e-14 -8.71103408019e-19 2.62100096573e-14) OK.
+ Boundary openness (1.85672186208e-14 -8.50625678685e-19 6.18299811514e-15) OK.
Max cell openness = 1.53490106268e-15 OK.
Max aspect ratio = 25.0289312037 OK.
Minimum face area = 5.44039160854e-11. Maximum face area = 3.35723618435e-06. Face area magnitudes OK.
Min volume = 3.38966568599e-15. Max volume = 3.35723618435e-09. Total volume = 3.31446160121e-05. Cell volumes OK.
- Mesh non-orthogonality Max: 25.3896599773 average: 2.69630443086
+ Mesh non-orthogonality Max: 25.386544632 average: 2.69326805544
Non-orthogonality check OK.
Face pyramids OK.
- Max skewness = 0.465880892795 OK.
+ Max skewness = 0.465999415964 OK.
Coupled point location match (average 0) OK.
Face tets OK.
Min/max edge length = 4.99999951583e-06 0.0032714540524 OK.
<<Writing 488 near (closer than 6.01177408764e-07 apart) points to set nearPoints
All angles in faces OK.
Face flatness (1 = flat, 0 = butterfly) : min = 1 average = 1
All face flatness OK.
- Cell determinant (wellposedness) : minimum: 0.00809562991112 average: 0.396592299529
+ Cell determinant (wellposedness) : minimum: 0.00809562991112 average: 0.396372492196
Cell determinant check OK.
- ***Concave cells (using face planes) found, number of cells: 360
- <<Writing 360 concave cells to set concaveCells
- Face interpolation weight : minimum: 0.294523453407 average: 0.483772637659
+ ***Concave cells (using face planes) found, number of cells: 358
+ <<Writing 358 concave cells to set concaveCells
+ Face interpolation weight : minimum: 0.294516579567 average: 0.483787294555
Face interpolation weight check OK.
- Face volume ratio : minimum: 0.12468161947 average: 0.932602925846
+ Face volume ratio : minimum: 0.124681118055 average: 0.932634914103
Face volume ratio check OK.
Calculating AMI weights between owner patch: outerAMI and neighbour patch: innerAMI
AMI: Creating addressing and weights between 600 source faces and 1348 target faces
AMI: Patch source sum(weights) min:0.999999999996 max:1 average:1
AMI: Patch target sum(weights) min:0.999996339997 max:1.0000039496 average:1.00000000146
Failed 1 mesh checks.
End
Mike |
|
|
|
|
|
|
|
|
Similar Threads
|
||||
| Thread | Thread Starter | Forum | Replies | Last Post |
| 3D Multiple layers model (VIV) | jordan.c | STAR-CCM+ | 1 | April 10, 2015 07:09 |
| [snappyHexMesh] Layers in snappy collapse in some places of my solid | Yuby | OpenFOAM Meshing & Mesh Conversion | 0 | March 8, 2015 15:44 |
| [snappyHexMesh] SnappyHexMesh no layers and no decent mesh for complex geometry | pizzaspinate | OpenFOAM Meshing & Mesh Conversion | 1 | February 25, 2015 07:05 |
| [ANSYS Meshing] Inflation Layers around a 90 degree bend that is separated into multiple parts | Kaaji1359 | ANSYS Meshing & Geometry | 2 | August 18, 2013 18:57 |
| Inserting Boundary Layers in multiple objects in Gambit for use in FLUENT | 3quarters | FLUENT | 0 | July 16, 2010 08:55 |
Linear Mode
