CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   turbulence quires (http://www.cfd-online.com/Forums/main/2221-turbulence-quires.html)

ram June 9, 2000 02:39

turbulence quires
 
hi .. i have some quires on turbulence flow.. i hope u can help. 1.we use continuity equation and momentum equations to solve for variables u,v,w,p ie 4 equations and four unknowns. thus the problem is mathematically closed and we use SIMPLER algorithm. similarly, if we take k-e model, we have 3 equations and three unknowns ie turbulent viscosity, e,k. Can we obtain an algorithm to SIMPLER to solve for k,e,turbulent viscosity with or without SIMPLER to u,v,w,p? if so how can it be done?

2.we have generally used time averaged NS equations for turbulent flow analysis. can we use space averaged NS equations b'cos turbulent flow fluctuations have 3D spatial character. why i am asking u this is that if unsteady flow for eg.,is taking place over a flat plate, with coordinate system in the fluid space; the probelm can be modified to a steady one if the coordinate is shifted to the flat plate. can similar procedure be done for turbulent flow to obtain space averaged NS equations?

thanking u ram

John C. Chien June 9, 2000 23:45

Re: turbulence quires
 
(1). These equations are all coupled. (2). In the momentum equations, you needs the eddy viscosity, which must be derived from the eddy viscosity definition in terms of the k and epsilon. And k and epsilon are solutions of k- and epsilon-equations, which include the convection terms etc... So, it requires the knowledge of the velocity field from the momentum equations. (3). It is nice to have new ideas, so, why not pick a simple flat plate and apply your concept, reasonable?


All times are GMT -4. The time now is 08:07.