In Palabos there is a Neumann boundary condition where the gradient is zero. Is there a flux boundary condition where the gradient has a specified value that can be user defined or can change during the simulation?

/Maka.

In Palabos there is a Neumann boundary condition where the gradient is zero. Is there a flux boundary condition where the gradient has a specified value that can be user defined or can change during the simulation?

/Maka.

Hi,

Currently, there is no implementation of a boundary condition for fixed velocity-gradients in Palabos, except for zero-gradients.

But, a flux boundary condition, isn’t this for boundaries on which you fix the value of the momentum? In a (nearly) compressible fluid, this is equivalent to imposing the value of the velocity. In this case, you should be looking for a Dirichlet condition on the velocity.

I should mention that there is a risk for confusion with the names for boundary conditions. In Palabos, we say “Dirichlet boundary” whenever we impose the value of something (velocity or pressure), and “Neumann boundary” when we impose the gradient of some quantity (velocity-gradient or pressure-gradient). This terminology is however not unique. In Sukop’s book for example, in Section 4.4.3, the term “Von Neumann (flux) boundary” is used for a boundary on which the velocity is fixed.

So, I could imagine that you read this book or some other source, made the conclusion that flux=“Von Neumann”, saw that there is a “Neumann” thing in Palabos and tried to go for it. However, as explained above, there’s a mix-up with the names. “Dirichlet velocity boundary” is the term you need to look for in Palabos to get this flux boundary condition.