turbomachinery rotating shaft surface velocity

[mattkoch]

I am trying out the turbomachinery module in the CS GUI. I still do not understand the turbomachinery setup completely, but one thing that I think needs to be present is a no-slip velocity boundary condition at the shaft wall. This cannot simply be [u,v,w] = [0,0,0], but must represent the circumferential velocity of the shaft. Something like [u,vw] = [-omega*R*sin(phi), omega*R*cos(phi),0] perhaps, where omega is the rotation velocity and phi is the rotation angle, and the rotation axis points in the z-direction and the shaft has radius R. The CS GUI however only allows for straight numbers in units [m/s] for u,v,w to be put in. How do I enter the circumferential velocity equations above instead? Thanks.

[Yvan Fournier]

Hello,

I am not sure whether this is already accounted for or not, so checking on a unit case can be good (I'll try to check on my side).

In any case, with the GUI, you can use MEI expressions to express more complex velocity values.

Regards,

Yvan

[mattkoch]

Thank you kindly, Yvan. When I try to enter anything else in the U, V or W fields shown in the attached screen shot, then the GUI does not accept the input. It seems I can only enter numbers here. I also recall that wherever I can enter equations, the GUI tends to have a button that I can push and then the MEI editor comes up. That is not the case on this Boundary Conditions screen in the GUI, though. What am I missing? Was my copy of Code_Saturne (version 5.0) not compiled correctly?

Also, if I had access to the MEI editor in this screen, I would expect to find [x, y ,z] values representing any point on the boundary. I would then use these to calculate the phi from my question above, and then compute [u, v, w] based on that?

Sorry for the continued bother, but this turbomachinery module is a tough nut I have been trying to crack for a number of years now - on and off - and I would really like to see some success. I think I am close, it is just these boundary conditions don't seem to cooperate. I could not find anything, but is there a complete example, with files, available anywhere?

[Yvan Fournier]

Hello,

Sorry, I might have been a bit optimistic. It seems the sliding wall option has more limited options thant those for inlet, et no MEI options. Which would mean you need to handle this in user subroutines (cs_user_boundary_conditions.f90).

But I believe the non-slip condition is already handled correctly without any additional setting. I'll check, but you can also test this with a simple cylinder, checking whether the flow is driven by the wall rotation or not.

Best regards,

Yvan