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] Rotational periodic problem (https://www.cfd-online.com/Forums/ansys-meshing/98757-rotational-periodic-problem.html)

m5edr March 18, 2012 11:31

Rotational periodic problem
 
1 Attachment(s)
Iam try to make a two face periodic boundary

Mesh --> Global Mesh Setup- Global Mesh Parameters - Setup periodicity -Rotational Periodicity

Base : "I selected first point in my geometry"
Axis: 0 1 0 ( rotated about y-axis)
Angle: 120 ( sector makes 120 degrees)

volume mesh is tetra/mixed , octree (NO Blocking is used)
shell mesh is All Tri , patch indep.

Icem tell me "Finished compute mesh" But
1- view of two face is STRANGE :confused: ,
2- and massage appears during mesh computation such (can't fix spike point on periodic boundary) and (periodic triangle matches tet but no tri) :confused:

My mesh is o.k or Not ?
any suggestion

rikio March 18, 2012 22:01

Have you checked the mesh for periodic problem?

m5edr March 18, 2012 22:11

thanks for reply

No ,how i do this check?

rikio March 18, 2012 22:23

1 Attachment(s)
Edit Mesh —> Check Mesh

m5edr March 20, 2012 08:50

3 Attachment(s)
I did the periodic check , ICEM created a subset contains 200 error due to Periodic problem (it asked me the faces that contains periodic)

the stranger thing , if i send the mesh to fluent >>> No problem appear in periodic faces and seems good (if i do a check in fluent it massaged a problem)

i attached two image of the periodic subset and two periodic faces

Important Notes:
1-A massage appear "shell () has node () which has no twin"
2-Each Face of periodic contains two face where one BIG face and SMALL face "small face inside the BIG" .. and a straight curves between them

Volume mesh is Octree , shell mesh is ALL TRI/Patch indep.

Any suggestion (do i neglect the error?)
thanks

zeeshu April 14, 2016 22:16

so have you solved your error i have the same problem

saad53 May 2, 2018 03:00

quality improvement
 
4 Attachment(s)
Hi!
I'm trying to mesh my aircraft geometry for my project in icem. but the mesh contains spikes and penetrating elements due to which the mesh is of unacceptable quality. i have attached a few pictures of the mesh. some guidance would be really appreciated. the mesh is generated by octree method with prism layers on the wing.

Gert-Jan May 2, 2018 03:40

It looks like some curves are missing.

After a tet-mesh is created, ICEM projects the nodes of the elements on the geometry. Here ICEM obeys some kind of hierarchy. If present, ICEM projects elements to points. Then to curves. And if none are present, then to surfaces.

In your case (and espcially in figure 4) there is no curve at the end of the surface. Then ICEM projects the nodes to a surface, making the edge of your mesh look spiky.

So you have to make sure that curves are present at the edges of all surfaces.

Good Luck, Gert-Jan

saad53 May 5, 2018 10:38

2 Attachment(s)
Thanks for replying Gert-Jan
i have attached the geometry pictures, i am not an expert at it so the curves look fine to me somehow, if you could kindly take a look and tell me if it is the problem i'm facing. would be kind of you to take a look
please ignore the green curve in the tail as i used it to segment the surface in order to make the elements smaller at that specific area for the spikes to be fixed,but the problem still remains.

Gert-Jan May 6, 2018 05:54

It is difficult to say, but my first impression is that you indeed have curves at the end of the surfaces. But curves are missing where surfaces are intersecting.

in other words, to me it looks like you have the surfaces of the plane body and surface of the wing. These surfaces of the wing are going through the surfaces of the plane body. Where these surfaces intersect, there are no curves.

What you should do is go to Create Curve>Surface Surface Intersections (5th icon). And create curves where surfaces intersect each other. You could opt to tick "Only Different Parts"


All times are GMT -4. The time now is 21:36.