In Fluent, when pressure outlet boundary condition is mentioned, we are usually mentioning only the value of the pressure at that boundary, but it is also necessary that even velocity should be mentioned at that boundary due to the very basic elliptic nature of NS equations. Or in other words boundary of the domain should be completely closed for the variables to be found. So how does fluent calculate the other variables at that location internally? (For steady state case)
Actually, the magnitude of velocity at the boundary can be calculated with a known pressure from the momentum equation.
The following picture is provided by Prof. Tao Wen-Quan's teaching courseware.
You can learn more from his personal website(http://nht.xjtu.edu.cn/nht.htm).
Dear abhey
salaam
To be honest I couldn't digest your question but I think your answer is in SIMPLE,SIMPLEC,piso and etc. algorithms in which we couple continuity to find pressure in a staggered mesh and solve another equation.I hope this link initiate to solving your question.
http://en.wikipedia.org/wiki/SIMPLE_algorithm
regards
The velocity is calculate based on the pressure drop at outlet. The velocity at outlet is part of the unknown variables that are solved in the linear system,. e.g. the SIMPLE method aforementioned. The boundary volumes have balances including the boundary pressure information, and then velocity field is unknown. After the solver iteraction (with any of P-v coupling methods available), a velocity field is found trough domain, including the outlet. This field needs respect the mass balance of course.
To impose a pressure at outlet assurance that the pressure field will have one solution, once to NS its gradient is the only information that the system needs. Without a defined pressure on boundaries, the pressure field is undetermined, and it has infinite solutions with the same gradient. The gauge or reference pressure, is summed to the pressure field to recovery the absolute pressure.
- Badges
- Science topic