CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   Turbine Model (http://www.cfd-online.com/Forums/cfx/93790-turbine-model.html)

 fergal October 26, 2011 13:24

Turbine Model

Hi,

I am currently modelling (inviscid model) a wind/tidal turbine blade using a single rotating reference frame. It is a two-bladed rotor, so I applied rotational periodicity (180 degrees). Ansys recommends a velocity inlet and pressure outlet. The turbine blade is contained within a cylindrical domain.

What boundary conditions are the most suitable for the cylindrical shell which contains the fluid. My opinion is to apply an opening boundary condition but what boundary details are suitable. At the outlet I applied an opening which allows for flow in both directions and again in my opinion is the most realistic and the most suitable choice but what boundary details are applicable?

Your help in this matter is much appreciated.

F

 ghorrocks October 26, 2011 16:58

An opening is physically good, but can cause convergence problems. A wall is less physically realistic but often converges easier.

Regardless of which boundary you use for the outside you should do a sensitivity analysis to check it is far enough away to be a good approximation of a far-field boundary.

 fergal October 26, 2011 17:35

Hi Glenn,

Thanks again for your speedy response. Your advice is good, what is your opinion on selecting a symmetry boundary for the fluid domain boundary?

When selecting an opening boundary for the outlet, what is the best boundary details option (if entrainment is selected is a value of 0 pascals adequate for relative pressure).

Thanks,

F

 ghorrocks October 26, 2011 17:42

I should point out that both the opening and the wall approach is accurate providing it is far enough away from the region of interest. Just that the wall approach will probably have to be further away for accuracy then the opening.

You can use a symmetry condition, it is effectively a slip wall condition.

 fergal October 27, 2011 13:16

Thanks again for the response. Have you any thoughts on the outlet boundary condition, a pressure outlet is the most robust but is specifying a 0 relative pressure good practice?

F

 ghorrocks October 27, 2011 16:52

Yes, a pressure boundary at 0 pressure is the usual way to proceed. Put any pressure offset in as a reference pressure. This minimises round off errors.

 fergal October 27, 2011 17:58

Hi Glenn,

Thanks again, I was using these boundary conditions I wanted to make sure I was applying the correct conditions to accurately predict the flow characteristics. I was a little worried about using a free slip wall for the cylindrical domain but it seems to be accurate!! I will do a sensitivity analysis as you discussed to model the effect of the location with respect to the turbine blade.