[Lorddrinkalot]

Hey guys,

In my Simulation, I am observing backflow back into the domain.

I've attempted to resolve this using the inletOutlet boundary condition, which should prevent backflow. However, the backflow issue persists, and I'm puzzled as to why this is happening.
My outlet condition. Image of the backflow and my Domain.

    outlet
    {
        type            inletOutlet;
        inletValue      uniform (0 0 0);
        value           uniform (0 0 0);

I tried extending the outlet in many ways; however, I am observing backflow.

I also tried the pressureInletOutletVelocity with total pressure for the p_rgh.

Can someone help me out here?

I am using the Boussinesq equation of state and the buoyantPimpleMhdFoam solver, which is just a combination of the buoyantPimpleFoam and the Mhd one. I use Openfoam v2312

I have a GitHub link to my case if anyone wants to check it. https://github.com/Lorddrinkalot/Ope..._shape/L_shape

[Martin_Sz]

Backflow in OpenFOAM simulations can arise from various factors, even when using the inletOutlet boundary condition. Here are some potential causes and solutions for your specific scenario:
Inconsistent Pressure Field: Double-check that the pressure field within the domain behaves as expected. Unrealistic pressure gradients can induce backflow. You can visualize the pressure field to identify any anomalies.

* Insufficient Outlet Length: Extending the outlet further downstream can sometimes mitigate backflow, especially if the flow hasn't fully developed. This allows for a more natural pressure recovery.

* Incorrect Boundary Conditions
More at link below
https://howtooansys.blogspot.com/202...ckflow-in.html

[Lorddrinkalot]

Thanks for your answer.

I actually dislike some behaviour in the p_rgh field. (Image included.) It has a very strong gradient, but I thought that's the result of the backflow.


Mh ok. I will try to stabilise my simulation.

[Lorddrinkalot]

Hey, I have changed my boundary conditions, and it appears to work. I don't see any backflow.

However, when I want to change my timestep, backflow again occurs. How? One simulation with a fixed timestep has no backflow. When changing to an adjustable time step, the backflow is there again.