[kooki_13]

Hi,


I simulated turbulent jet flow.
For the outlet and lateral sides, I have chosen "pressureInletOutletVelocity," inspired by the case setup of the previous simulation.
Now, I want to interpret it for myself, but it is confusing.
It has been mentioned in .H file that:

31 This velocity inlet/outlet boundary condition is applied to pressure
32 boundaries where the pressure is specified. A zero-gradient condition is
33 applied for outflow (as defined by the flux); for inflow, the velocity is
34 obtained from the patch-face normal component of the internal-cell value.

So,
1- Outflow is OK, but what about inflow?
2- In other words, does inflow mean a back-flow at the outlet boundary? or it is inflow at the inlet of the domain?
3- What does patch-face normal refer to?


Thanks in advance
Koukab

[LuckyTran]

It is a pressure boundary condition template that you use for the U field. Other fields have their own equivalent InletOutlet types. For flow going out of the domain it applies the zero gradient condition. If there is inflow, it uses a specified flux phi that you specify.

Inflow means backflow. But you have to keep in mind that mathematically and programmatically, there are no "inlets' and "outlets" there are just mathematical BC's. So inflow means stuff flowing into the domain regardless of whether you label a boundary an inlet, an outlet, an elephant, or a pineapple. Hence, backflow terminology is needlessly ambiguous

Surfaces are called patches in OpenFOAM. And patches contain collections of faces. patch-normal is your surface normal or boundary normals. Normal also means orthonormal, or 90 degrees if that wasn't clear enough.

[kooki_13]

Originally Posted by LuckyTran

It is a pressure boundary condition template that you use for the p field. Other fields have their own equivalent InletOutlet types. For flow going out of the domain it applies the zero gradient condition. If there is inflow, it uses a specified flux phi that you specify.

Inflow means backflow. But you have to keep in mind that mathematically and programmatically, there are no "inlets' and "outlets" there are just mathematical BC's. So inflow means stuff flowing into the domain regardless of whether you label a boundary an inlet, an outlet, an elephant, or a pineapple. Hence, backflow terminology is needlessly ambiguous

Surfaces are called patches in OpenFOAM. And patches contain collections of faces. patch-normal is your surface normal or boundary normals. Normal also means orthonormal, or 90 degrees if that wasn't clear enough.

Thank you, Lucky.

For boundary condition of velocity I am speaking, i.e., U file:
I guess that, when one applies "pressureInletOutletVelocity" on the outlet of the domain, this B.C. implements zero normal gradients for outflow fluxes and calculates the magnitude of u from normal patches for inflow fluxes for each cell.

Am I right?

[LuckyTran]

The InletOutlet BC applies zero gradient for outgoing flow and fixedValue for incoming flow. If you pair InletOutlet on U with a fixedValue for p, the problem becomes ill-posed for many flow scenarios when fixedValue for U and fixedValue for p are applied simultaneously.

pressureintletOutletVelocity is meant to be used together with a pressure BC. So instead of using fixedValue for U, you specify the mass flux and calculate the normal velocity component U from the specified mass flux.