|November 11, 2010, 08:04||
About joint face in implicit method
Join Date: Nov 2010
Posts: 5Rep Power: 7
I have a question when I implementing LUSGS in the multiblock framwork: how to handle the joint interface.
In my implementation, two levels of ghost cells out of a joint interface are applied. The flow states, W, in the ghost cells come from the neighbor blocks. When applying LUSGS scaning, the unknown, delta W, in the ghost cells are assumed as zero. But this decreases convergence, especially in case of the cells are of high aspect ratio, such as the wake cut of a C-grid of a airfoil. So, the resulted Cp contour lines are not smooth across the joint face.
Is there any easy method to augment the joint face when applying implicit method in a multiblock grid?
|Thread||Thread Starter||Forum||Replies||Last Post|
|Semi Implicit method||mnabi||Main CFD Forum||0||August 21, 2009 16:53|
|determine face position||Ralf Schmidt||FLUENT||1||April 24, 2009 09:37|
|Implicit, fractional step method!!||Frederic Felten||Main CFD Forum||2||November 15, 2002 08:06|
|implicit method on unstructured grid||Koo||Main CFD Forum||5||July 29, 2001 01:43|
|Physical Reason for stability of Implicit Schemes?||radhakrishnan||Main CFD Forum||25||December 7, 2000 16:25|