CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   Boundary Conditions for Simulation ofcontrol valve (http://www.cfd-online.com/Forums/fluent/49048-boundary-conditions-simulation-ofcontrol-valve.html)

 Ashraf Sharara August 20, 2008 03:15

Boundary Conditions for Simulation ofcontrol valve

Hi, I am trying to carry out a steady state simulation for turbulent water flow through a single seat globe control valve, with a flat faced plug. The experiments gave me the upstream and downstream pressure as well as the mass flow rate, at different valve openings. My question is: " What are the best boundary conditions to be used at inlet and outlet, in order to obtain good results for both the pressure and velocity fields?".

Notes: 1. I tried to use total pressure at inlet and gauge pressure at outlet, The pressure field was o.k. However, the computed mass flow rate is not ( A deviation of about 25% from the experimental value was noticed). 2. I tried to use velocity with direction vector at inlet and gauge pressure at outlet, The velocity field was o.k and the computed mass flow rate also. However,the pressure is not as the pressure values were noticed increased in all the computaional domain( Pressure at inlet is about 1,75 times of the experimental value).

Accordingly, iam looking for any solution of this problem.Please treat the problem as a matter of urgent due to time factor.

Thanks for your coorperation. B.Regards Ashraf Sharara

 Ravi Shankar August 20, 2008 07:54

Re: Boundary Conditions for Simulation ofcontrol v

Hello Ashraf,

CFD Problem setup is understood as follows: -------------------------------------------- 1. Flow through a control valve, for varying plug or seat positions, for varying velocities. 2. Flow through solid walls, i.e. can be defined as periodic flow, which transverses the flow direction, and has restriction and pressure drop across the valve. 3. The inlet boundary shall be best provided if its Velocity, and outlet as Guage pressure, which is a real measure. Ensure Guage pressure is set to Zero. 4. Outlet boundary shall be free stream flow. 5. Input the K and E terms with proper eddy dissipation values. A standard industrial constants KE model would be fine to track the vortices near the restriction plug or seating. 6. A SIMPLE solver with Second order pressure gradients shall be used for high resolution. 7. Start the simulation with steady state solution, this is a kind of unsteady solution but with respect to position but not time. A kind of time accelerator solution. 8. Validate the solution to correspoding experiments. 9. The 75% higher values of pressure, could be for reason improper KE values or Reference values. Check Drag reference values. 10. I believe the fluid mentioned is Liquid.