[johny_walker]

Hello,



I am conducting a simulation on vehicle aerodynamics. I have a rectangular domain. I dont want to have any wall on the side of the vehicle and on top of domain. The only wall present will be the ground.



Which condition in Ansys Fluent should I give so that it represents a no wall?



Best regards,

[vinerm]

There are at least two options you can choose.

Apply a wall condition but change it to free-slip wall. This essentially means that shear-stress is 0. Wall will act like a wall, i.e., it will confine the flow but it will not offer any resistance to the parallel. Most likely, this is what you want and is used by most people working on external aero. Some people using symmetry, which is correct for side wall but not for top.

Second option is to use pressure-outlet. There will be a lot of flow-reversal, however, if this boundary is far away from the vehicle, then there is nothing to worry about.

[LuckyTran]

There is also a pressure farfield boundary condition. All the BC types though are predicated on your boundaries being far enough away that they don't adversely affect the results. If the boundary is not far enough away, a slip-wall is not that great because there is boundary induced flow acceleration so it is generally preferred to use a farfield BC or pressure inlet/outlet BC because those allow flow to cross the boundary and provides relief. Either way, make your boundary very far away if you can.

[vinerm]

As Lucky mentioned, Pressure Far-Field is another option. However, it is inapplicable in your case since this condition has two particular requirements.

1. Fluid should be treated as ideal gas. This you can address by using ideal gas law for air, however, you will be end up solving an extra equation.

2. This condition should not be applied to a boundary that is adjacent to a wall. This you cannot address since the vehicle is on the ground and ground is a wall.

So, if the side and top boundaries are far away from the vehicle; gap between vehicle top and top boundary should be at least 3 times the height and similarly gap between sides of the vehicle and sides of the corresponding side boundary should be at least 3 times the width, then stress free wall (otherwise called free-slip bc) is perfectly applicable. Pressure outlet, as stated earlier, will work but there will be flow reversals.

In essence, keep boundaries far away from vehicle and you can use either pressure outlet or free slip.

[johny_walker]

If we use free slip BC, we generally need to have a boundary layer on a wall with free slip. However, in my domain, I dont have a boundary layer resolved on top and side walls. Will this be an issue?



Best regards,

[vinerm]

On the contrary, free slip does not require any boundary layer mesh or fine mesh since due to free slip the boundary layer does not exist.

[johny_walker]

Dear all,



I have been talking with my professors about this condition. It was decided that having a pressure outlet condition will be more suitable to simulate no wall condition that having a symmetry because in pressure outlet condition, if we set the gauge pressure to be 0, it will take it same as atmospheric boundary. Also velocity gradients is 0, so the velocity will be same. However, in symmetry condition, there is nothing related to pressure and only normal velocity component is 0. So, pressure outlet would simulate no walls much better.



Furthermore, having a domain like this where one give pressure outlet on top would be more efficient because the car would generate some stream-line bending near the upper boundary, but these will be much smaller than the slope. So, the flow will be outlet everywhere.












Regards,

[vinerm]

Pressure outlet will certainly work, however, as mentioned earlier, it would have flow reversal. As long as boundary is far away from vehicle, this reversal would not affect the flow around the vehicle.

Symmetry implies symmetry for each field, including pressure. When symmetry is applied, mathematically, gradient of each field with respect to the plane of symmetry is set to 0. So, as far as velocity field is concerned, this is same as free-slip. But it imposes similar conditions on pressure, which is wrong. So, symmetry is not good, at least not for top boundary.

Do note that the pressure outlet works like pressure inlet for the flow that reverses and enters the domain. And in that case, the pressure applied is taken as total pressure. This causes static pressure to drop below operating pressure. But as mentioned, it will not have any effect if flow boundary is far away.