I had a propeller geometry for which i prepared one rotating Domain but this propeller is situated in a bend. I defined fr rot interfaces between the bend and rot domain and defined ggi between bend boundaries with the tank domain,all other boundaries are no slip adiabatic walls.

I tried with mass flow at one of the Interface ,is this a right approach?or sholuld I have a Free surface at top? Or is there any other way to deal with?

Please clarify.......... Do you have a drawing....

Gert-Jan

Thanks for a reply. Suppose a tank as a pool of liquid in which a propeller is situated at one end upside down to create currents and to guide this flow from vertical direction to horizontal direction one 90 degree bend is there.So the propeller will have its suction facing towards bottom. Hence the propeller is in a pipe with bend which is in a pool. This could clarify the situation.

Are you interested in the flow in the bath of in the flow around the blades of your pump. I guess the first.

Then do not model the propellor as such with a rotating domain and stationary domain etc. Just use a subdomain with a momentum source equal to the momentum the pump provides. Much easier to model, and it provides the same flow in your pool.....

Model the free surface as a wall with free slip, as long as the pump is well below the free surface. You can check the disturbance of the free surface by postprocessing the pressure and compare it with rho*g*h.

Gert-Jan

www.bunova.nl

Dear Gert-Jan, Thanks for the reply. I have done the same thing. Removed propeller from the model and defined its outlet as pools inlet and its inlet as pools outlet,which will give a pressure drop indirectly the head required by the propeller. Now I have defined mass flow at inlet and avg static pressure as 0 pa at outlet of a pool with ref pre=0 pa.

Will this work?

I think this will do. As long a you only study the water. When you include the free surface again, p=0 at the outlet will give troubles.....

Gert-Jan