Large residual distributed at uniform flow region

Anna Tian · August 16, 2015, 23:10

As you can see in the attached pictures, the region with vortex doesn’t have large residual over there but the region with quite uniform flow is distributed with large residual which causes convergence problem. Could anyone explain that’s why and how to solve this problem?

ghorrocks · August 17, 2015, 02:54

You have very high aspect ratio elements in the wake region. It is always going to have a hard time resolving the wake when the elements are that long and skinny. You will need to reduce the aspect ratio of those elements.

Anna Tian · August 19, 2015, 00:45

Quote:

Originally Posted by ghorrocks

You have very high aspect ratio elements in the wake region. It is always going to have a hard time resolving the wake when the elements are that long and skinny. You will need to reduce the aspect ratio of those elements.

I improve the aspect ratio over there but convergence doesn’t become any better. I think coarsening the grids over the vortex region at the trailing edge is the general way to improve the convergence for this kind of problem (of course not too coarse).

ghorrocks · August 19, 2015, 01:31

Refining the mesh results in less numerical dissipation which means the inherent instabilities in shear layers, wakes and similar features become more difficult to resolve. So yes, I would expect convergence to get more difficult as mesh size gets smaller.

Be careful coarsening the grid as you will be loosing accuracy by doing it. Make sure it is not a critical part of the flow, and do a sensitivity analysis on it to be sure.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
a problem with convergence in buoyantSimpleFoam	skuznet	OpenFOAM Running, Solving & CFD	6	November 15, 2017 12:12
Cannot run the code properly: very large time step continuity error	crst15	OpenFOAM Running, Solving & CFD	9	December 14, 2014 18:17
Courant-number explodes after a lon while (icoFoam)	Rody-	OpenFOAM Running, Solving & CFD	6	January 29, 2014 04:27
Unstabil Simulation with chtMultiRegionFoam	mbay101	OpenFOAM Running, Solving & CFD	13	December 28, 2013 13:12
Unknown error	sivakumar	OpenFOAM Pre-Processing	9	September 9, 2008 12:53

August 17, 2015, 02:54		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	You have very high aspect ratio elements in the wake region. It is always going to have a hard time resolving the wake when the elements are that long and skinny. You will need to reduce the aspect ratio of those elements.

August 19, 2015, 01:31		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	Refining the mesh results in less numerical dissipation which means the inherent instabilities in shear layers, wakes and similar features become more difficult to resolve. So yes, I would expect convergence to get more difficult as mesh size gets smaller. Be careful coarsening the grid as you will be loosing accuracy by doing it. Make sure it is not a critical part of the flow, and do a sensitivity analysis on it to be sure.