[ICEM] bad elements on symmetry surface

Pawel Zdunek · June 15, 2017, 22:26

Hey guys,

what could be the reason of bad elements like the ones on the picture?
All of the bad elements are aligned with edges/borders of blocks, I made sure those blocks faces are associated with correct surface, likewise with edges.

Thanks for any help.

bluebase · June 16, 2017, 05:18

Hi Pawal,

it's hard to see the area of interest.
Can you make a detail screen shot of the worst elements only?
Modify the range of the quality histogramm to -1 to 1 (instead of 0 to 1) and print only two bars, then show only the elements within -1/0 range.

You can also try 'Scan Plan' in the context menu of Pre-Mesh to inspect the problematic areas further. Here you can draw a complete layer of your mesh and see the area in crosssectional view.

A general thing: sometimes curves(,points) and surfaces do not align proberly. So when you have face to surface association and edge to curve association, then you might have a little bump around your curves. Check whether the curves of interest really are at the surface. Try projecting the curves to the surface again.

With regards,
Sebastian

Pawel Zdunek · June 16, 2017, 20:51

Hi,
thank you very much for taking time to help me

I managed to get rid of bad elements aligned on symmetry plane along the outlets (smaller pipes going vertical) by removing those edges' association with symmetry (purple) surface. That was quite surprising for me, I thought those should be projected on symmetry plane, however projecting(associating) them generated bad elements from picture from 1st post. I also improved quality by working on edges spacing, which was really poor in some places and clearly contributed to bad elements (now it's also poor but alittle bit better).

I am afraid that some (most) of bad elements are due to wrong association of given entities - it looks like two meshes -worst elements and ok elements - are projected simultaneously on a symmetry plane.

Pawel Zdunek · June 16, 2017, 21:05

ok it looks like those were the cause of bad elements:
-redundant association of both faces and edges to given surface and curve (is that a common thing in ICEM, because i'm not sure if it's even a thing?)
-wrong association of given edges to given surfaces - that's a bigger problem, some of my edges converge into vertices associated to the same point - that's my big big mistake. Will work on it and update for the future users.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
[snappyHexMesh] Missing elements near surface with snappyHexMesh	jmf	OpenFOAM Meshing & Mesh Conversion	6	November 10, 2017 15:16
[snappyHexMesh] Problem with Sanpper, surface still Rough	Zephiro88	OpenFOAM Meshing & Mesh Conversion	7	November 5, 2014 13:05
Bad surface mesh on detailed car	jpesich	STAR-CCM+	1	September 23, 2014 23:33
Possible Bug in pimpleFoam (or createPatch) (or fluent3DMeshToFoam)	cfdonline2mohsen	OpenFOAM	3	October 21, 2013 10:28
Problems of Duns Codes!	Martin J	Main CFD Forum	8	August 15, 2003 00:19

June 16, 2017, 05:18		#2
bluebase Senior Member Sebastian Engel Join Date: Jun 2011 Location: Germany Posts: 567 Rep Power: 21	Hi Pawal, it's hard to see the area of interest. Can you make a detail screen shot of the worst elements only? Modify the range of the quality histogramm to -1 to 1 (instead of 0 to 1) and print only two bars, then show only the elements within -1/0 range. You can also try 'Scan Plan' in the context menu of Pre-Mesh to inspect the problematic areas further. Here you can draw a complete layer of your mesh and see the area in crosssectional view. A general thing: sometimes curves(,points) and surfaces do not align proberly. So when you have face to surface association and edge to curve association, then you might have a little bump around your curves. Check whether the curves of interest really are at the surface. Try projecting the curves to the surface again. With regards, Sebastian

June 16, 2017, 21:05		#4
Pawel Zdunek New Member Join Date: May 2016 Posts: 20 Rep Power: 10	ok it looks like those were the cause of bad elements: -redundant association of both faces and edges to given surface and curve (is that a common thing in ICEM, because i'm not sure if it's even a thing?) -wrong association of given edges to given surfaces - that's a bigger problem, some of my edges converge into vertices associated to the same point - that's my big big mistake. Will work on it and update for the future users.