Hi all,

I am puzzled by the vorticity or angular velocity of fluid. Their definitions can be found in the following link:

http://en.wikipedia.org/wiki/Vorticity

I found that in my CFD solution, the vorticity in the first cells close to wall will be much bigger than others as the fluid velocity at walls is equal to zero. This becomes more obvious when my mesh size near wall is increased. This sounds unreasonable. I wonder how it would be in in practice or experiments. In order to acquire the correct value of vorticity, I have to use fine density of grids near wall, right? Is there any other method to evaluate it during CFD simulations. I need this information for modelling others things?

I would appreciate your help.

With best regards,

David

Why does it sound unreasonable? The vorticity is directly connected to the velocity gradient, meaning regions of large velocity gradient are typically going to have high levels of vorticity. That's the case at the wall, where the velocity gradient is large. Resolution of the gradient controls the accuracy of the vorticity calculation, and high resolution is going to require a fine mesh or a high-order (4th order or higher) scheme with minimal numerical dissipation.

The problem is that what is the criterion for high solution. For some reason, I can not use high density of grid. Is there another way to evaluate vorticity?

Are you using a turbulence model? If so, you might be able to use wall functions to reduce the resolution you need at the wall. Otherwise, if you have a laminar flow, you might be able to determine the velocity gradient (and therefore vorticity) at the wall from boundary layer theory.

In my case laminar flows is considered. However, the high concentration of particles makes me feel difficult to evaluate the velocity gradient theoretically.

How are you evaluating the velocity gradient in your solver? It is still unclear why this is an issue. Why can you not increase the grid resolution? If the velocity gradient is not being resolved well in your solver then you have bigger problems than just not getting the vorticity correct.