Symmetry boundary considerations

mrwts · August 22, 2016, 04:37

Hello everybody,
A couple of questions regarding symmetry boundaries and mesh with mirror planes.
I am running multiple simulations to determine the best location of the engine and the wing for an amphibian aircraft (considering Cd and Cl values for many points within a grid)

1) My geometric model is symmetrical relative to the XY plane, would it be physically correct to apply a Symmetry Boundary in that plane and therefore only work with half the geometry/mesh?
As far as I understand, the symmetry boundary does not allow any diffusion flux across a symmetry plane. So, since I am expecting 3D swirling vortices right behind the engine and the pylon, I think I may be limiting the 3D flows of the vortices accros the symmetry plane. Wouldn't I?

2) Is there any possibility to cut a mesh in FLUENT along a symmetry plane?

Thanks in advance for your insights.

FMDenaro · August 22, 2016, 07:40

Quote:

Originally Posted by mrwts

Hello everybody,
A couple of questions regarding symmetry boundaries and mesh with mirror planes.
I am running multiple simulations to determine the best location of the engine and the wing for an amphibian aircraft (considering Cd and Cl values for many points within a grid)

1) My geometric model is symmetrical relative to the XY plane, would it be physically correct to apply a Symmetry Boundary in that plane and therefore only work with half the geometry/mesh?
As far as I understand, the symmetry boundary does not allow any diffusion flux across a symmetry plane. So, since I am expecting 3D swirling vortices right behind the engine and the pylon, I think I may be limiting the 3D flows of the vortices accros the symmetry plane. Wouldn't I?

2) Is there any possibility to cut a mesh in FLUENT along a symmetry plane?

Thanks in advance for your insights.

imposing symmetry is a strong assumption on the physics of the problem that can be acceptable only in particular cases... what kind of formulation are you using? give more details..

agd · August 22, 2016, 08:57

To add to FMDenaro's answer, the use of a symmetry plane implies that you have mirror-image flow across the plane. Thus, any asymmetric 3D behavior will be lost. At low angles of attack, such an assumption is not unreasonable. But at high angles of attack asymmetric shedding becomes very pronounced, even at zero yaw. Any mutual vortex induction between the pylon/engine and its mirror is also affected by the use of a symmetry boundary condition, so if that is a large effect (even at small angles of attack) a symmetry condition may not be a good choice. It's one of those areas that requires some experience and judgment - I would suggest trying both a symmetry plane and full 3D and comparing the results. That's really one of the best ways to answer your question.

sbaffini · August 23, 2016, 04:45

I try to add a more practical perspective to the previous correct answers.

The feasibility of a boundary condition is just dependent from the final solution. In the specific case of symmetry, it simply depends from a single question:

for a full simulation (no symmetry), do you expect the final solution to be symmetric across that plane?

If the geometry is symmetric and all the remaining boundary conditions are symmetric with respect to that plane than what really remains as possible source of asymmetry are the equations.

Do you expect the equations to produce asymmetry under symmetric conditions? RANS equations are not expected to do so, so symmetry is ok for them. This is also true for any experimental campaign satisfying such simmetry conditions (even if, in practice, this might be impossible to achieve), the statistics extracted from the experimental data are expected to satisfy the symmetry within the experimental uncertainty.

Any vortex in a RANS solution is a an average feature of the flow and, if everything else is symmetric, an equal counter-rotating vortex is expected from the other side of the plane, so that their net effect is null on the symmetry plane.

The matter is different for unsteady simulations (or even steady laminar ones), where the objections of the previous posters effectively apply.

August 22, 2016, 04:37	Symmetry boundary considerations	#1
mrwts New Member Mario Alberto Join Date: Sep 2015 Location: Russian Federation Posts: 10 Rep Power: 10	Hello everybody, A couple of questions regarding symmetry boundaries and mesh with mirror planes. I am running multiple simulations to determine the best location of the engine and the wing for an amphibian aircraft (considering Cd and Cl values for many points within a grid) 1) My geometric model is symmetrical relative to the XY plane, would it be physically correct to apply a Symmetry Boundary in that plane and therefore only work with half the geometry/mesh? As far as I understand, the symmetry boundary does not allow any diffusion flux across a symmetry plane. So, since I am expecting 3D swirling vortices right behind the engine and the pylon, I think I may be limiting the 3D flows of the vortices accros the symmetry plane. Wouldn't I? 2) Is there any possibility to cut a mesh in FLUENT along a symmetry plane? Thanks in advance for your insights.

August 22, 2016, 08:57		#3
agd Senior Member Join Date: Jul 2009 Posts: 351 Rep Power: 18	To add to FMDenaro's answer, the use of a symmetry plane implies that you have mirror-image flow across the plane. Thus, any asymmetric 3D behavior will be lost. At low angles of attack, such an assumption is not unreasonable. But at high angles of attack asymmetric shedding becomes very pronounced, even at zero yaw. Any mutual vortex induction between the pylon/engine and its mirror is also affected by the use of a symmetry boundary condition, so if that is a large effect (even at small angles of attack) a symmetry condition may not be a good choice. It's one of those areas that requires some experience and judgment - I would suggest trying both a symmetry plane and full 3D and comparing the results. That's really one of the best ways to answer your question. FMDenaro likes this.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
3D Windturbine simulation in SU2	k.vimalakanthan	SU2	15	October 12, 2023 05:53
Basic Nozzle-Expander Design	karmavatar	CFX	20	March 20, 2016 08:44
Problem in setting Boundary Condition	Madhatter92	CFX	12	January 12, 2016 04:39
Difficulty In Setting Boundary Conditions	Moinul Haque	CFX	4	November 25, 2014 17:30
[Gmsh] Import problem	ARC	OpenFOAM Meshing & Mesh Conversion	0	February 27, 2010 10:56

August 23, 2016, 04:45		#4
sbaffini Senior Member Paolo Lampitella Join Date: Mar 2009 Location: Italy Posts: 2,152 Blog Entries: 29 Rep Power: 39	I try to add a more practical perspective to the previous correct answers. The feasibility of a boundary condition is just dependent from the final solution. In the specific case of symmetry, it simply depends from a single question: for a full simulation (no symmetry), do you expect the final solution to be symmetric across that plane? If the geometry is symmetric and all the remaining boundary conditions are symmetric with respect to that plane than what really remains as possible source of asymmetry are the equations. Do you expect the equations to produce asymmetry under symmetric conditions? RANS equations are not expected to do so, so symmetry is ok for them. This is also true for any experimental campaign satisfying such simmetry conditions (even if, in practice, this might be impossible to achieve), the statistics extracted from the experimental data are expected to satisfy the symmetry within the experimental uncertainty. Any vortex in a RANS solution is a an average feature of the flow and, if everything else is symmetric, an equal counter-rotating vortex is expected from the other side of the plane, so that their net effect is null on the symmetry plane. The matter is different for unsteady simulations (or even steady laminar ones), where the objections of the previous posters effectively apply.