[ICEM] A problem in blockinhg

TommyLui · December 21, 2011, 23:30

HELLO, I got a problem in build the mesh of my geometry by using the blocking. My geometry is the attachment "Geometry.gif". Then my target is using the blocking to mesh the cross tunnel. which is the middle part of the geometry. The attachemnt "NAME" shows the name of the side of block. if A is 4, B is 2 and C is 20 will output as attachment 4220.gif. ifA is 4, B is 4 and C is 20 will output as attachment 4420.gif. However 4420 is require but the output is not desire. How can I do to fix this problem?

PSYMN · December 22, 2011, 11:43

You are probably starting with the wrong approach if you want to block the whole model... ICEM CFD is a top down blocking tool, but you are trying to block just the middle part first... So if you were thinking that you would then move on to blocking the two sides, you are really due for a rethink. That is how many other blocking tools work, but not ICEM CFD Hexa. Try some tutorials...

Now specific to your problem here, you have blocked the central portion of a larger model. This portion has walls on 4 of the 6 sides. The remaining two sides are open. That distortion you are seeing is really just those sides trying to project to the nearest wall.

There are several solutions.

1) If you plan to block the entire model, you should really just start again and do a top down blocking. Then you won't have this problem at all. You would split the larger block twice in each direction to separate out the bridge. THen delete the blocks that would be full of empty space. Then associate the edges to curves and move things into place. Then put an Ogrid thru the entire model with Faces on the ends of the pipes. It is a 5 minute Job once you know how (Knowing how will take a little practice).

2) If you really want to block just the central portion, you could close the ends with surfaces.

3) You could turn off the projection of the side faces. This is done with the association menu. You can either just turn it off, in which case you would get not shells, or you could set it to "Interpolate" in which case the end wall will just be stretched across the space defined by the perimeter.

Have fun.

TommyLui · December 22, 2011, 20:51

The mesh becomes better, however the quality still not good. Some of the quality still have negative value. When I choose the negative quality, I see some of the mesh is very thin and lying on the side of the bridge. How can I improve the problem?

PSYMN · December 23, 2011, 15:38

Picture Please... And maybe a little more explanation about what you ended up doing...

December 22, 2011, 11:43	Projecting to the nearest surface...	#2
PSYMN Senior Member Simon Pereira Join Date: Mar 2009 Location: Ann Arbor, MI Posts: 2,663 Blog Entries: 1 Rep Power: 47	You are probably starting with the wrong approach if you want to block the whole model... ICEM CFD is a top down blocking tool, but you are trying to block just the middle part first... So if you were thinking that you would then move on to blocking the two sides, you are really due for a rethink. That is how many other blocking tools work, but not ICEM CFD Hexa. Try some tutorials... Now specific to your problem here, you have blocked the central portion of a larger model. This portion has walls on 4 of the 6 sides. The remaining two sides are open. That distortion you are seeing is really just those sides trying to project to the nearest wall. There are several solutions. 1) If you plan to block the entire model, you should really just start again and do a top down blocking. Then you won't have this problem at all. You would split the larger block twice in each direction to separate out the bridge. THen delete the blocks that would be full of empty space. Then associate the edges to curves and move things into place. Then put an Ogrid thru the entire model with Faces on the ends of the pipes. It is a 5 minute Job once you know how (Knowing how will take a little practice). 2) If you really want to block just the central portion, you could close the ends with surfaces. 3) You could turn off the projection of the side faces. This is done with the association menu. You can either just turn it off, in which case you would get not shells, or you could set it to "Interpolate" in which case the end wall will just be stretched across the space defined by the perimeter. Have fun. __________________ ----------------------------------------- Please help guide development at ANSYS by filling in these surveys Public ANSYS ICEM CFD Users Survey This second one is more general (Gambit, TGrid and ANSYS Meshing users welcome)... CFD Online Users Survey

December 23, 2011, 15:38		#4
PSYMN Senior Member Simon Pereira Join Date: Mar 2009 Location: Ann Arbor, MI Posts: 2,663 Blog Entries: 1 Rep Power: 47	Picture Please... And maybe a little more explanation about what you ended up doing... __________________ ----------------------------------------- Please help guide development at ANSYS by filling in these surveys Public ANSYS ICEM CFD Users Survey This second one is more general (Gambit, TGrid and ANSYS Meshing users welcome)... CFD Online Users Survey

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
UDF compiling problem	Wouter	Fluent UDF and Scheme Programming	6	June 6, 2012 04:43
Gambit - meshing over airfoil wrapping (?) problem	JFDC	FLUENT	1	July 11, 2011 05:59
natural convection problem for a CHT problem	Se-Hee	CFX	2	June 10, 2007 06:29
Adiabatic and Rotating wall (Convection problem)	ParodDav	CFX	5	April 29, 2007 19:13
Is this problem well posed?	Thomas P. Abraham	Main CFD Forum	5	September 8, 1999 14:52

December 22, 2011, 20:51		#3
TommyLui New Member Tommy Lui Join Date: Sep 2011 Posts: 9 Rep Power: 14	The mesh becomes better, however the quality still not good. Some of the quality still have negative value. When I choose the negative quality, I see some of the mesh is very thin and lying on the side of the bridge. How can I improve the problem?