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/)
-   -   [ANSYS Meshing] Meshing complex geometry (http://www.cfd-online.com/Forums/ansys-meshing/98707-meshing-complex-geometry.html)

cfdhydraulic March 16, 2012 15:26

Meshing complex geometry
 
2 Attachment(s)
Hi everyone
I am struggling to mesh a complex geometry. I have tried almost everything without success,
I would appreciate if some of you could give me a hand on it
Attach you will find such a model

Any idea about how to approach this problem?
this is a domain of a fluid.

Thanks

tauqirnawaz March 19, 2012 09:49

Quote:

Originally Posted by cfdhydraulic (Post 349905)
Hi everyone
I am struggling to mesh a complex geometry. I have tried almost everything without success,
I would appreciate if some of you could give me a hand on it
Attach you will find such a model

Any idea about how to approach this problem?
this is a domain of a fluid.

Thanks

Can you attach your geometry file please.

cfdhydraulic March 19, 2012 09:59

Thanks for you quick response.
This site doesn't allow to upload a 3MB file.
Could you please give me your email so i can send to you?
you can email me at andres.rperez@hotmail.com

Thanks again

tauqirnawaz March 19, 2012 10:13

Quote:

Originally Posted by cfdhydraulic (Post 350209)
Thanks for you quick response.
This site doesn't allow to upload a 3MB file.
Could you please give me your email so i can send to you?
you can email me at andres.rperez@hotmail.com

Thanks again

I have sent you an email. Please respond to that. Why are you using all quad mesh by the way?

cfdhydraulic March 19, 2012 10:14

I tried everything... not only quad mesh... just everything and nothing worked...
I will check my email

cfdhydraulic March 19, 2012 10:17

still waiting... no new emails yet

cfdhydraulic March 19, 2012 10:28

Could you please send an email again?

cfdhydraulic March 19, 2012 10:29

fine! I got it!

sarangdalne March 19, 2012 23:37

can you send geometry to sarang.dalne@gmail.com?
What type of analysis you want to perform on this model? This will drive what type of mesh to be used.
Did you try thin sweep with this model?
I guess you have to try mixing couple of methods to get the desired mesh quality here.

tauqirnawaz March 20, 2012 04:31

Quote:

Originally Posted by cfdhydraulic (Post 350212)
I tried everything... not only quad mesh... just everything and nothing worked...
I will check my email

Andres,

Have just checked the IGES file sent by you. One quick issue is that you did create the hydraulic model in a way that some additional solid bodies were also created. You need to suppress the solid bodies in solidworks or you can bring in the solid geometry in WorkBench and create hydraulic model by fill tool.

If you just want to continue with the iges file that you have sent to me then suppress all the bodies in DM except the first one. Then try refined mesh using curvature as advanced sizing Function and defining appropriate sizing criteria.

cfdhydraulic March 20, 2012 19:59

convergence
 
1 Attachment(s)
Thanks I will try to do that. But with IGES file no problems were found.
However I have a problem with convergence

I am studying a transient state study (unsteady).
The first iteration is correct and it achieves a perfect convergence but, the second one... it doesn't converge at all

that's the problem?? is there anything I would change?
I decreased the time step to 1E-05 but it didn't converge whatsoever

any idea??

thank you :)

tauqirnawaz March 21, 2012 04:06

This is because of the mesh, you need to refine it and before that delete the bodies except the first, otherwise the mesher creates two meshes on same part.

cfdhydraulic March 21, 2012 09:24

but actually I modified the model and I don't have 8 bodies anymore but just one.
I am going to refine the mesh anyway. However if I refine it I will get smaller cells which will decrease my time step right? I am working with a 0.00001 time step... if I use lower values it going to take forever to simulate 1 second simulation!

any suggestions?
thanks


All times are GMT -4. The time now is 22:39.