CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > Software User Forums > OpenFOAM > OpenFOAM Meshing & Mesh Conversion

[snappyHexMesh] layer addition problem

Register Blogs Community New Posts Updated Threads Search

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   March 5, 2018, 07:05
Default layer addition problem
  #1
New Member
 
krishnan
Join Date: Feb 2018
Posts: 16
Rep Power: 8
tankerp is on a distinguished road
Dear Foamers....

Im having a problem adding layers to my geometry. The geometry is a series of pipes and am planning to perform multiphase internal flow simulations.

Attached is an image of the mesh at one of the corners. While the snapping process and layer addition seems fine in some regions, layers do not propogate along the 90 degree bends.

Any suggestions on how to remove this problem are welcome. I have spent a considerable amount of time trying to fix this problem by tweaking various parameters in SHM. This problem is now causing a serious delay to the entire project.

Attached is a snapshot of the mesh at the corners. Thanks in advance.
Attached Images
File Type: png msh_img.png (8.5 KB, 115 views)
tankerp is offline   Reply With Quote

Old   March 19, 2018, 11:38
Default
  #2
Senior Member
 
Join Date: Nov 2010
Location: USA
Posts: 1,232
Rep Power: 25
me3840 is on a distinguished road
Your background mesh is not very isotropic. If you make it isotropic the layer generation on the smaller walls will work better.
me3840 is offline   Reply With Quote

Old   March 19, 2018, 12:02
Default
  #3
New Member
 
krishnan
Join Date: Feb 2018
Posts: 16
Rep Power: 8
tankerp is on a distinguished road
Thank you for replying. What do you mean by making the background more isotropic?
tankerp is offline   Reply With Quote

Old   March 19, 2018, 13:14
Default
  #4
Senior Member
 
piu58's Avatar
 
Uwe Pilz
Join Date: Feb 2017
Location: Leipzig, Germany
Posts: 744
Rep Power: 15
piu58 is on a distinguished road
How do you make three nodes form two in a hex mesh? That is not an easy task. One solution is to put tringular elements in between. One of the solution is what you see in your corners.

I downt assume that your geometry is twisted (no isotropic).
__________________
Uwe Pilz
--
Die der Hauptbewegung überlagerte Schwankungsbewegung ist in ihren Einzelheiten so hoffnungslos kompliziert, daß ihre theoretische Berechnung aussichtslos erscheint. (Hermann Schlichting, 1950)
piu58 is offline   Reply With Quote

Old   March 19, 2018, 13:58
Default
  #5
New Member
 
krishnan
Join Date: Feb 2018
Posts: 16
Rep Power: 8
tankerp is on a distinguished road
Hi Uwe,

When you say that the geometry is twisted, do you mean that the geometry can be twisted if the snapping process is incorrect or just the geometry itself is incorrect?

Btw, I have an update. Since Simscale uses snappyHexMesh, I tried meshing the geometry using the default settings in simscale. Again the same problem persists at the corners. Do you have any suggestions? Could it be the geometry?

About the Geometry, I generated it using the following steps:
1. Created the geometry in FreeCAD and exported the STL file to Salome.
2. Generated a 2D surface mesh using NetGen2D before importing to Snappy.

I hope it is not too much to ask for, but I have also attached the SnappyHexMeshDict in this post. Could anybody check on it for any suggestions/changes in the settings for snapping/layer addition?
Attached Files
File Type: txt snappyHexMeshDict.txt (8.8 KB, 10 views)
tankerp is offline   Reply With Quote

Old   March 20, 2018, 00:23
Default
  #6
Senior Member
 
Join Date: Nov 2010
Location: USA
Posts: 1,232
Rep Power: 25
me3840 is on a distinguished road
Look at the size of the cells in the middle of the channel. They are long in one direction and short in another. They should be nearly identical in lengths. Snappy doesn't work very well with anisotropic meshes like the one you have; you need to modify the blockMesh.
me3840 is offline   Reply With Quote

Old   March 20, 2018, 01:50
Default
  #7
Senior Member
 
piu58's Avatar
 
Uwe Pilz
Join Date: Feb 2017
Location: Leipzig, Germany
Posts: 744
Rep Power: 15
piu58 is on a distinguished road
Dear Krishnan,

your mesh seems to be simple enough for blockmesh. Make your own meshes! They are much better than generated ones.

In your case: Is ist really necessary to change the node structure at the corners? I would let it as ist is.

Denser node packing is not easy. In one direction you may use denser layer of nodes which is supported by blockmesh. But what do yo do in the other coordinate(s)? One solution is to insert a layer of triangular elements. If you do that in a region where the solution is not important for you, that is a way.

I give you another way with only hex elements. There ist a layer with slightly distort elements, that is unavoidable:

Code:
 _______
|   |   |
|___|___|
|__/ \__|
| |\ /| |
| | | | |
__________________
Uwe Pilz
--
Die der Hauptbewegung überlagerte Schwankungsbewegung ist in ihren Einzelheiten so hoffnungslos kompliziert, daß ihre theoretische Berechnung aussichtslos erscheint. (Hermann Schlichting, 1950)
piu58 is offline   Reply With Quote

Reply


Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
[snappyHexMesh] SnappyHexMesh - no layer added bejbro OpenFOAM Meshing & Mesh Conversion 5 February 1, 2020 21:05
Topology changes - problem when adding a layer for the second time mike.franky OpenFOAM Running, Solving & CFD 0 September 30, 2017 09:26
[snappyHexMesh] Problem in layer addition at complex/bend shapes vineet_bhola OpenFOAM Meshing & Mesh Conversion 0 November 6, 2015 10:48
[snappyHexMesh] Problem with snappyHexMesh samiam1000 OpenFOAM Meshing & Mesh Conversion 2 December 13, 2012 09:41
[snappyHexMesh] snappy hangs while layer generation bastil OpenFOAM Meshing & Mesh Conversion 3 October 29, 2010 08:19


All times are GMT -4. The time now is 11:20.