[ttu004]

Dear all

I have been trying to simulate aerofoil under pitch oscillation using Transient stater-rotor method. When modelling it at high Reynolds number e.g. Re:0.75E6, 25.2735 m/s, the results have shown great consistency with the experiments. However, when I try to model it at Low Reynolds number e.g. Re:1.1E5, 3.67 m/s, velocity vector changes direction when flow across the domains interface.(this does not happen to the High Reynolds number cases...) Figures attached are the velocity vector plots for high and low Reynolds numbers, clearly velocity has changed its direction once it cross the interface in the low Reynolds number cases.

I am very confused about what is happening and trying to find out the cause of problem.
Here are more details of the setup

1. Pitching frequency is 1.28505 Hz and it is the same used in the High Reynolds number cases. The purpose is to compare flow behavior under two Reynolds number for the same pitching frequency. .

Could the pitching frequency (1.28505 Hz and so 0.778s for 1 pitching period...) be so high relative to the free-stream velocity (3.67 m/s) so that the flow somehow mismanaged as it cross the interface into the rotational domain...... If so any idea to resolve this??

2. This problem raised whenever the "rotor" methods are introduced (frozen rotor, transient rotor stator...) to the interface between rotational and stationary domains.
3. Rotational movement of the domain are done by assigning an angular velocity to the domain motion. Alternate rotation model is also used.
4. Region of motion specified is used in mesh deformation with the displacement relative to "previous mesh". Mesh stiffness has a model exponent of 10 and increase near the small volumes.
5. General connection and Transient rotor stator are used on the interface. Pitch Change are set to automatic.
6. The mesh motion: Parallel to boundary is used on the interface of the rotational domain.

Very appreciated if anyone could give me my idea and suggestion

Thanks

[ttu004]

The free-stream velocity is set at 8 degrees so the angle of velocity vectors across the interface should be the same. Velocity vectors at two sides of interface should be aligned not discontinuous and changed in angle....

[ghorrocks]

First read this FAQ: http://www.cfd-online.com/Wiki/Ansys...f_reference.3F

Also note that the rotating domain of reference model does not include terms from the angular acceleration of the frame of reference. The centripetal and coriolis forces are included, but angular acceleration terms are not. So if the angular acceleration is significant then you cannot use a rotating frame of reference. In this case you will need to use a moving mesh.

[ttu004]

Thanks Glenn for your reply

You are right~ as the angular acceleration is significant relative to the free-stream flow, my model could not handle it and this is the source of problem.
So instead of having the same pitching frequency, I made the reduced frequency the same so now I got the slower pitching rate...

After all, having high pitching rate under slow wind speed is not realistic for a wind turbine....

Thanks so much