CFD Online Discussion Forums

CFD Online Discussion Forums (
-   CFX (
-   -   ICEM Blocking (

Joe August 17, 2006 07:50

ICEM Blocking

I have created a mesh around a hull form with the blocks but i believe that the way the blocks are positioned it causes many bad elements around the hull form. I have included a photo and circled the angled blocks that may have affected the elements.

Is there a way to move the vertices of one block without the other trying to reposition eachself like it has happened to me? I have checked the tutorials and i havent found any or even in the manual.

Thanks in Advanced, Joe

Turbine August 17, 2006 08:22

Re: ICEM Blocking

Did you collapse any blocks, or create any blocks manually? I belive something of thissort has happened and resulted in the vertices getting close. There still exists 2 vertices where u SEE them to be one...

Select move vertices optin in ICEM and manually move the vertices at the locations where they appear to be one and later you can align the vertices

Hope this helps


Joe August 17, 2006 09:15

Re: ICEM Blocking
I created the blocks manually and at the beggining the vertices looked as one. I undestand what you are saying but they are not two vertices only one. Is like they are merged together and cannot be unmerged (I can not see any command to do so)..

When I move a vertice then both the adjacent block moves as well.

Ideally I would like to make the side blocks square without any sharp edjes.

Any suggestion?

alex August 17, 2006 09:43

Re: ICEM Blocking
do u try to split your block and delete one of them to separate two block? It could permit to unmerged two block if u make good projections...

Charles August 17, 2006 13:54

Re: ICEM Blocking
Ah yes, the Wigley hull, no? Delete (move to VORFN) the blocks ahead, below and behind the hull. Then associate the vertices of the new "symmetry" faces to points on the corners of the symmetry plane. You are stuck with the wedge cells above the hull, but that shouldn't be a problem. Or send me an e-mail and I will send you a project of how I did it.

Joe August 18, 2006 04:15

Re: ICEM Blocking
I have tried that as well. It still shares the vertices with the existing block.

What do you mean by good projections Alex?

Joe August 29, 2006 09:03

Re: ICEM Blocking
Hi again,

Thanks Charles, what you told me worked until a point. I deleted the blocks that were making the problem but now the blocks above the hull create a problem. These ones I cannot delete them since they take the shape of the hull and the quality of them is quite bad due to the angle they have.

Since they are not in the area of interest I dont mind if they are bad but CFX doesnt allow to proceed. The elements i think they are inverted (normals point to opposite direction).

Any suggestions? Thanks in Advance, Joe

Charles August 29, 2006 13:58

Re: ICEM Blocking
Well, if all you are interested in is the hydrodynamics and the waves, you might as well remove those blocks, since they won't affect that part of the solution. I've had no complaints in principle from either Fluent or CFX about wedge blocks there. It is possible that the blocks might be inverted, especially if you generated them from scratch. Does the pre-mesh quality check report negative volumes there? In any event, it is easy to re-orient the block with your blocking tools, so it should not be a problem to fix it.

Joe August 30, 2006 07:49

Re: ICEM Blocking
This blocks are important because they help mesh the top part of the wigley hull. Without them there would be a hole there. In the pre-mesh quality all the worst elements are in the vertical position at the corners of the wedged blocks. I havent seen anywhere about negative volume (maybe i am not loking in the right place).It is in blocking-check blocks?

When I import it into CFX then it doesnt even run it, something about the symmetry plane - is not symmetrical or not parrallel.

Any ideas? Thanks in advance.

Charles August 31, 2006 08:01

Re: ICEM Blocking
It is not necessarily important to solve the airflow above the hull, especially if you are mostly interested in the waves. You can set CFX's tolerance (under expert settings) as to how flat it wants the symmetry plane. You can do a volume check using the pre-mesh quality check.

All times are GMT -4. The time now is 20:39.