CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   ANSYS Meshing & Geometry (http://www.cfd-online.com/Forums/ansys-meshing/)
-   -   [ICEM] Can I define periodic boundaries in an unstructured mesh? (http://www.cfd-online.com/Forums/ansys-meshing/102560-can-i-define-periodic-boundaries-unstructured-mesh.html)

Aoki May 28, 2012 13:19

Can I define periodic boundaries in an unstructured mesh?
 
Hi all. I'm modelling a tidal stream turbine. To reduce the size I only built 1/3 of the turbine with periodic boundaries. I defined periodicity in global mesh parameters and generated the mesh. But when I read it in FLUENT it can only recognise one of the periodic boundaries.

I've read a few threads regarding to periodic boundaries, and it seems that to define a periodic boundary I would need to define the periodic vertices from the corresponding block after I define periodicity in global mesh parameters. Is that the reason why my mesh doesn't work properly? However, I'm currently doing an unstructured mesh and it doesn't have any blocking. Does it mean that I cannot define periodic boundaries on an unstructured mesh? Are there other ways to work round it?

Thanks in advance.

yonchong June 13, 2012 10:01

I am not sure what you are trying to do but you can do periodic boundary with unstructed mesh.

Aoki June 13, 2012 10:38

Quote:

Originally Posted by yonchong (Post 366251)
I am not sure what you are trying to do but you can do periodic boundary with unstructed mesh.

Thanks. Well, this is what I was trying to mesh.

http://i47.tinypic.com/90bua1.png
http://i46.tinypic.com/23h40bc.png

I don't know what's wrong with it. I checked the mesh, which doesn't seem to have any problem. But when I try to solve it in fluent it diverge straight away. I was told that there should be two faces for the periodicity but when I important it into fluent it could found recognise the top one.

yonchong June 13, 2012 10:50

Recognizing the only one side is normal.

Why don't you run one iteration and check the solution.

You could also try the fmg-initialization and check the solution. You have to refer to the manual for this as this option is only available through Text User Interface. However, it could give you very good initial guess of the solution. If the fmg-initalization fails I would check your boundary condition again including the periodic axis definition in fluid domains. Fluid might be cyclic in an wrong axis.

hadikhayyamian June 16, 2012 15:21

Hi,
If you are working with ICEM go to view menu and Mirrors and replicates, and generate the 2nd part of the mesh and you can check manually that your nodes are matching (i.e. conformal mesh)
Also check your mesh in the edit mesh tab . this is very useful!

In your fluent did you set periodic boundary condition?

Aoki June 19, 2012 11:24

Quote:

Originally Posted by hadikhayyamian (Post 366817)
Hi,
If you are working with ICEM go to view menu and Mirrors and replicates, and generate the 2nd part of the mesh and you can check manually that your nodes are matching (i.e. conformal mesh)
Also check your mesh in the edit mesh tab . this is very useful!

In your fluent did you set periodic boundary condition?

In ICEM I have checked the mesh and I didn't find any problem.

In FLUENT I only set periodicity as rotational at boundary condition. Are there anything else that I need to set up in FLUENT? This is the first time for me to set up a periodic boundary condition, I might have missed something that I didn't know about.

hadikhayyamian June 19, 2012 11:35

However there are other advanced settings, I guess thats fine. now, what is your problem?

Aoki June 19, 2012 11:43

Quote:

Originally Posted by hadikhayyamian (Post 367237)
However there are other advanced settings, I guess thats fine. now, what is your problem?

My problem is that it deverged straight away in the first iteration.

Divergence detected in AMG solver: pressure correction
Divergence detected in AMG solver: k
Primitive Error at Node 0: floating point exception
Primitive Error at Node 1: floating point exception
Primitive Error at Node 2: floating point exception
Primitive Error at Node 3: floating point exception
Error: floating point exception
Error Object: #f

I've never seen something like this happened before so I don't even know which bit of it is wrong: whether is the mesh or the setting in FLUENT.

yonchong June 19, 2012 12:38

You need to set the rotational-axis direction.

Cell Zone Conditions -> Zone, Edit

As I said, stop or write out after few interations to check the boundary conditions.

Aoki June 19, 2012 13:34

Hi guys, thanks very much for your help and information.
The model is working now. Probably because I missed defining the axis.
Thanks again.

rijas August 1, 2013 22:57

Translational periodic condition in icem cfd
 
Hi could any one tell me how to set translatinal periodic condition for a cascade airfoil?


All times are GMT -4. The time now is 11:53.