[raf1111]

Hello,

I am working on a semester project dealing with a uniform flow past a rotating cylinder (Re=200, solved using icoFoam).
I have tried to use the rotatingWallVelocty boundary condition for U but when I check the flow field at the end of the simulations, the velocity at the cylinder's surface is not constant (and therefore not equal to the radius times the angular velocity that I specified in the boundary condition for U).

I'm new at openFoam and I've tried to find the solution to this problem on other threads but without success so I need your help ! Do you know if there are basic mistakes/problems that need to be avoided when using this BC?

Please let me know if you need more information about my problem!

Thank you in advance for your help,

Raphael

[ngj]

Hi Raphael,

I have just looked into the code of OF2.2.1. and it seems correct. The only thing I can thing of is the origin of your object. Are you sure that you have placed the origin in the centre of your object?

Kind regards

Niels

[raf1111]

Hello and thank you for your answer. I checked and the origin of the rotation does correspond to the center of the cylinder.
However, I did a slice on a plane normal to the axis of the cylinder, halfway through the z-direction of the domain, and it seems that I have the correct velocity around the cylinder. I think that my mistake comes from the boundary conditions applied on the side walls (connected to the cylinder's tips). I've tried using the slip BC and it looks like it somehow interferes with the velocity on the cylinder's surface right next to the side walls. This would actually explain why the BC on the cylinder is correct at the center of the domain but not on the side walls. I'll try to look in more details at the definition of the slip BC but otherwise I'll simply look at the flow at the center of the domain (which is probably the most relevant place to look actually).

Thanks again for your answer,

Raphael

[ngj]

Hi Raphael,

A couple of things come to my mind:

1. You could start with a cyclic/periodic condition in the z-direction, which will remove your dependency on end effects.

2. Have you visualised the velocities at the cylinder surface, i.e. turning on the cylinder boundary in paraView/paraFoam? These should be correct. If you look at the near-wall cell, then you already start to get effects from other boundaries into the result.

3. Also, if you are using interpolated and not cell centred values, the picture you obtain can be distorted from the actual computed picture.

4. Finally, I have also been confused in the beginning with the default perspective in paraView/paraFoam. I always use parallel perspective, as it is otherwise difficult to see the correct behaviour along one axis, due to a natural perspective representation. You can change this in Edit->View Settings->General.

Good luck,

Niels