[blockMesh] best practices?

jferrari · April 10, 2012, 11:12

Ultimately I would like to mesh an airfoil inside of a circle in order to rotate the circle and study the dynamic stall on the airfoil. To get more comfortable with blockMesh, I'm meshing a small circle inside of a larger circle where the two circles have different centers. Just making blocks in the eccentric annulus had non-orthogonal cells, so I made the block boundaries splines. I created these splines by marking a couple of points normal to each wall and let the spline figure it out in between. This seemed to work, but I'm still getting some cells that look kind of strange. I'm thinking this could be because I haven't marked enough points on the spline, but maybe it's something else. In any case, is this the best way to go about this - is there a better way to mesh an airfoil transitioning to a circle other than matching the slopes with splines?

The entire mesh

The cells close to the inner circle that look strange

The cells close to the outer circle that look strange

wiedangel · April 19, 2012, 17:03

Hi Jeferrari,

your mesh looks very nice to me. I am also trying to work on good practices to simulate airfoils. My idea is to use a c.grid imbeded in an o-grid, also to be able to rotate the airfoil to study dynamic stall later. What do you mean by the cells look strange. I will try to pots some pics of my mesh during next week.

Regards

April 10, 2012, 11:12	best practices?	#1
jferrari Member Joe Join Date: Dec 2011 Location: Groton, CT Posts: 69 Rep Power: 14	Ultimately I would like to mesh an airfoil inside of a circle in order to rotate the circle and study the dynamic stall on the airfoil. To get more comfortable with blockMesh, I'm meshing a small circle inside of a larger circle where the two circles have different centers. Just making blocks in the eccentric annulus had non-orthogonal cells, so I made the block boundaries splines. I created these splines by marking a couple of points normal to each wall and let the spline figure it out in between. This seemed to work, but I'm still getting some cells that look kind of strange. I'm thinking this could be because I haven't marked enough points on the spline, but maybe it's something else. In any case, is this the best way to go about this - is there a better way to mesh an airfoil transitioning to a circle other than matching the slopes with splines? The entire mesh The cells close to the inner circle that look strange The cells close to the outer circle that look strange

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
to combine the two practices, i.e. bifurcation with deformable walls	bluesky66	FLUENT	2	April 7, 2018 19:49
Best practices for performing an analysis on a supersonic C-D nozzle in Star-CCM	jonny_b	STAR-CCM+	1	August 9, 2010 11:06
CFD Practices	Isaac Newton	FLUENT	0	December 7, 2008 14:14
One question about the cfx 5 best practices	John	CFX	3	August 18, 2005 18:39
CFD best practices	Dave	Main CFD Forum	0	November 3, 2004 11:33

April 19, 2012, 17:03		#2
wiedangel Member wided Join Date: Jul 2010 Posts: 54 Rep Power: 15	Hi Jeferrari, your mesh looks very nice to me. I am also trying to work on good practices to simulate airfoils. My idea is to use a c.grid imbeded in an o-grid, also to be able to rotate the airfoil to study dynamic stall later. What do you mean by the cells look strange. I will try to pots some pics of my mesh during next week. Regards