symmetryPlane vs slip and zeroGradient (possible bug)

feymark · September 11, 2009, 03:26

Hi,

I'm currently running a LES simulation of a cylinder, with a prescribed sinusoidal movement, in a steady stream. The mesh boundary has the shape of a cuboid. I have employed the following boundary conditions in the pointDisplacement file, top/bottom [fixedValue (0,0,0)], left-side/right-side [symmetryPlane], Inlet/Outlet [fixedValue (0,0,0)] and for the cylinder [oscillatingDisplacement, amplitude(0.02435 0 0),omega 1.759].

The problem arises when I increase the grid resolution. I then get a perturbation in the pointDisplacement field (in the y and z-direction! according to the boundary conditions the analytical solution should be zero everywhere!). It starts on the symmetryPlane boundary and spreads and in the end makes the simulation crash. Changing the BC to be slip (vectors) and zeroGradient (scalars) works!

Should there really be a difference?

Best regards,
Andreas FEYMARK

henry · September 16, 2009, 10:56

Could you send your case or better still a small case which reproduces the problem?

H

September 11, 2009, 03:26	symmetryPlane vs slip and zeroGradient (possible bug)	#1
feymark New Member Andreas Feymark Join Date: Mar 2009 Location: Gothenburg, Sweden Posts: 18 Rep Power: 17	Hi, I'm currently running a LES simulation of a cylinder, with a prescribed sinusoidal movement, in a steady stream. The mesh boundary has the shape of a cuboid. I have employed the following boundary conditions in the pointDisplacement file, top/bottom [fixedValue (0,0,0)], left-side/right-side [symmetryPlane], Inlet/Outlet [fixedValue (0,0,0)] and for the cylinder [oscillatingDisplacement, amplitude(0.02435 0 0),omega 1.759]. The problem arises when I increase the grid resolution. I then get a perturbation in the pointDisplacement field (in the y and z-direction! according to the boundary conditions the analytical solution should be zero everywhere!). It starts on the symmetryPlane boundary and spreads and in the end makes the simulation crash. Changing the BC to be slip (vectors) and zeroGradient (scalars) works! Should there really be a difference? Best regards, Andreas FEYMARK

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Convection discretization schemes for LES	sek	OpenFOAM Running, Solving & CFD	38	July 31, 2017 14:30
Flow Around a Cylinder	ronaldo	OpenFOAM	5	September 18, 2009 08:13
Waves in simple channel	taranov	OpenFOAM Running, Solving & CFD	0	July 17, 2009 06:59

September 16, 2009, 10:56		#2
henry Senior Member Join Date: Mar 2009 Posts: 854 Rep Power: 22	Could you send your case or better still a small case which reproduces the problem? H