CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   ANSYS Meshing & Geometry (https://www.cfd-online.com/Forums/ansys-meshing/)
-   -   [ICEM] How to mesh 2 cylinders that are perpendicular? (https://www.cfd-online.com/Forums/ansys-meshing/102850-how-mesh-2-cylinders-perpendicular.html)

despaired student June 4, 2012 10:50

How to mesh 2 cylinders that are perpendicular?
 
1 Attachment(s)
Hi,

I've a "simple" question. How to best mesh 2 cylinders that are perpendicular towards eachother?

Condition is that the top of the one cylinder is in line with the other cylinder.
A .tin file is attached.


Kind regards

BrolY June 6, 2012 11:20

I guess you plan to use blocking ?

Once you have done the splits to capture the footprint of the perpendicular tube, you can use the option blocking -> create block -> extrude face and select "fixed distance" for the method.
It's pretty easy to use.

despaired student June 6, 2012 14:31

Thank you - I'll try this.

BrolY June 7, 2012 03:43

1 Attachment(s)
I have attached an example with your project.

despaired student June 7, 2012 07:53

I tried it but it didn't work ...yet..
Nevertheless thank you again for helping me.


The problem is the point where the cricle of the tube touches the top respectively the bottom curve that describes the circumference. Here one must decide whether to associate the edge to one of these curves and that results in an incorrect mesh here.

My idea was to make a small layer on top respectively on the bottom.
This means I have the edges of layer 1 to associate to the top-curve. In layer 2 I have curves that encircle along the surface of the main-cylinder and I create a "hole" where I connect the block for the tube. This block is smaller than the second layer but by associate edge to curve I mesch the perpendicular tube correctly.
But by doing so the cells below and above my "tube-block" are squeezed/compressed. The cells at the top/bottom are so heavyly compressed that here the det 3x3x3 quality-criterion is at 0.1.

BrolY June 7, 2012 09:21

2 Attachment(s)
You need to use 1/4 of O-grid to sort this out.
I have attached a project with an example.
But with this kind of geometry, dont't expect to have a good quality :S

despaired student June 7, 2012 10:07

...yes, O-grid and Y-Grid is a good idea, I think, but I couldn't get good results. Maybe an tetra-Mesh or an inflow/outflow profile at the connection-area of cylinder and pipes instead of the pipes...

Thank you for your try!


All times are GMT -4. The time now is 16:08.