Boundary conditions
 

Dear cfd people,

A few questions: 1. When is it justified to prescribe p=0 at outflow when the inflow velocity is prescribed? Does this boundary condition permit the fluid structures to pass the boundary or it reflects pressure waves? And can it be used for unsteady problems as well? * 2. You mention about the dissipative nature of diffusion. What to you mean by that? How is diffusion dissipative and for what? * 3. Equation df/dt+U*df/dn (f being a quantity like velocity) is a more realistic outflow boundary condition than the ones usualy used. However how do you calculate the U velocity to maintain overall conservation of mass flux? In addition is this condition for both velocities u,v (for an incompresible problem)? Is this condition implemented in any of your codes? * 4.In Peric's (very good by the way) i read that "Notable exceptions occur when very fine grids are used near solid surfaces because there are important structures of small sizes in these regions. Here, instability may arise from the viscous terms involving derivatives normal to the wall and these derivatives may need to be treated implicitly". What exactly happens when very fine grids ocuur near solid walls and explicit methods become unstable. Is it because the stability criterions are violated? Could you clarify it please? * Best regards Achilleas Tsompanos, Eng.

Re: Boundary conditions
 

(1). You are allowed to ask one question a time. (2). Please describe your problem background so that a person on the other end of the line can figure out your problem and question. (3). Otherwise, you will have to go back to school to take some cfd and turbulence courses.