CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   Main CFD Forum (https://www.cfd-online.com/Forums/main/)
-   -   ICEM mesh problems (https://www.cfd-online.com/Forums/main/94842-icem-mesh-problems.html)

luxingzhe November 28, 2011 05:13

ICEM mesh problems
 
1 Attachment(s)
Hi, guys

I am meshing my wind turbine in ICEM, however, there are some bad elements around the blade tip and the trailing edge. Can anyone provide good strategies on this?

CapSizer November 28, 2011 06:53

O & o
 
Ah ... that's a pain to sort out. My strategy is to do an O-grid for a blade like that, but split in the middle and aligned with the trailing edge. That way you have a C-grid about the airfoil, which goes downstream again before closing in again as a reversed C-grid. This lets you capture the wake without having to take the wake resolution miles downstream. Then do another C-grid for the tip. Can even move that C-grid further away radially, to avoid skewness close to the tip.

luxingzhe November 28, 2011 17:33

Quote:

Originally Posted by CapSizer (Post 333791)
Ah ... that's a pain to sort out. My strategy is to do an O-grid for a blade like that, but split in the middle and aligned with the trailing edge. That way you have a C-grid about the airfoil, which goes downstream again before closing in again as a reversed C-grid. This lets you capture the wake without having to take the wake resolution miles downstream. Then do another C-grid for the tip. Can even move that C-grid further away radially, to avoid skewness close to the tip.

I tried several days trying to refine it, they all turn out in vain. Do you think i can force running the simulation with this quality of mesh?

CapSizer November 29, 2011 01:11

Fix it mate!
 
Having gone to the trouble of building a hexa grid, there probably isn't that much point in trying to run with it unless you can fix those issues. Some solvers (Fluent comes to mind) will not be happy with such nasty cells. Keep thinking how you can change that tip blocking topology using O-splits.

Quote:

Originally Posted by luxingzhe (Post 333878)
I tried several days trying to refine it, they all turn out in vain. Do you think i can force running the simulation with this quality of mesh?


aero_suresh November 29, 2011 12:54

Quote:

Originally Posted by luxingzhe (Post 333878)
I tried several days trying to refine it, they all turn out in vain. Do you think i can force running the simulation with this quality of mesh?

Mesh quality of less than 0.2 itself is very bad, but here you are having approx. 2700 cells with less than 0.1. So don't try to run it forcefully without removing bad cells.

Please note that "Meshing with fine quality cells is like an art, rather than engineering".

luxingzhe November 29, 2011 18:06

Quote:

Originally Posted by aero_suresh (Post 334005)
Mesh quality of less than 0.2 itself is very bad, but here you are having approx. 2700 cells with less than 0.1. So don't try to run it forcefully without removing bad cells.

Please note that "Meshing with fine quality cells is like an art, rather than engineering".

Thanks for the kind remind, however, do you have any suggestions on how to refine it, or namely converting the bad elements?

Best regards,


All times are GMT -4. The time now is 06:10.