Inflation

Jade M · February 17, 2010, 13:58

I am new to CFX. What is the best way to determine whether a mesh is adequate?

In my current internal flow problem, I have tried 3 different meshes (1) tet with patch conforming, (2) tet with patch conforming and inflation, and (3) tet with patch conforming and finer inflation. These 3 meshes give 3 different pressure drops which are (1) 50 Pa, (2) 109 Pa and (3) 218 Pa. What is the problem?

ghorrocks · February 17, 2010, 17:54

This means your mesh is nowhere even close to fine enough, regardless of the method used to produce it. You need to do a mesh sensitivity study until your results are independent of mesh size.

JDA · April 21, 2010, 17:39

The main driver of mesh adequacy is the number of elements. Make several meshes with different element sizes and inflation layers and plot the results as a function of number of elements. There should be a point where decreasing the element size (increasing the number of elements) has little impact on the results (akin to a blower curve, only mirrored). You want to have a mesh that puts you on this flat section.

Jade M · April 21, 2010, 17:42

Great thanks for the info. Being new to CFX, I just wanted to make sure that I understood these options. I've now demonstrated grid independence.

February 17, 2010, 13:58	Inflation	#1
Jade M Senior Member Join Date: Feb 2010 Posts: 148 Rep Power: 17	I am new to CFX. What is the best way to determine whether a mesh is adequate? In my current internal flow problem, I have tried 3 different meshes (1) tet with patch conforming, (2) tet with patch conforming and inflation, and (3) tet with patch conforming and finer inflation. These 3 meshes give 3 different pressure drops which are (1) 50 Pa, (2) 109 Pa and (3) 218 Pa. What is the problem?

February 17, 2010, 17:54		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,870 Rep Power: 144	This means your mesh is nowhere even close to fine enough, regardless of the method used to produce it. You need to do a mesh sensitivity study until your results are independent of mesh size. Torque_Converter and f.kh like this.

April 21, 2010, 17:42	Thanks	#4
Jade M Senior Member Join Date: Feb 2010 Posts: 148 Rep Power: 17	Great thanks for the info. Being new to CFX, I just wanted to make sure that I understood these options. I've now demonstrated grid independence.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
question about inflation for impeller	sandy	CFX	4	August 8, 2007 10:03
Creating inflation layers for elbow with airfoil !	Aplus	CFX	1	March 13, 2006 23:22
Inflation	Gab	CFX	6	August 24, 2005 23:45
inflation	Jennifer Haque	CFX	2	June 9, 2004 19:55
Effects of NOT using mesh inflation	Louwrens	CFX	10	May 7, 2003 17:07

April 21, 2010, 17:39		#3
JDA New Member J. D. Aurand Join Date: Apr 2010 Posts: 13 Rep Power: 16	The main driver of mesh adequacy is the number of elements. Make several meshes with different element sizes and inflation layers and plot the results as a function of number of elements. There should be a point where decreasing the element size (increasing the number of elements) has little impact on the results (akin to a blower curve, only mirrored). You want to have a mesh that puts you on this flat section.