Rotating Flow test case

[kanuk]

Hello,

I'm wondering whether the rotating flow test case is working properly. TEST CASE 1 solves very well, creating the desired rotating flow. However, TEST CASE 2, which I believe is intended to produce similar results, only produces a zero velocity field (or at least zero down to machine precision). Is this correct? Or am I missing something?

Thanks for your help.

[Juha]

Hi,

i think that in the second case the flow field is relative to the rotation.
The effect of rotation only shows in the pressure field in this case.

Juha

[kanuk]

Thank you for your response. I see now how it works...

I have one more question I hope you can answer. Let's say I would like to model a rotating cylinder in a cross flow using this method. Since I would need to use a rotating boundary condition, is there any way I could ensure just a zero pressure condition on the outflow? In other words, I am currently using the following:

Code: Select all

 	
Velocity 1 = Variable Time
  Real MATC "VEL*cos(OMEGA*(tx))"
Velocity 2 = Variable Time
  Real MATC "-VEL*sin(OMEGA*(tx))"

as my boundary condition on the circular far-field fluid.

However, this is sometimes fraught with numerical instability and I would like something a little more stable at the outlet. Usually for a nonmoving frame of reference I would set the velocity parallel to flow to be zero at the outlet, and set a pressure there. Is there anything that can be done in my case? Or, does a new solver need to be added to account for rotating meshes? I am afraid this may be the case...

Thank you so much for all your help - here and elsewhere on the forums.