CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   CFX (https://www.cfd-online.com/Forums/cfx/)
-   -   Failure in partitioning due to axial fluid-solid interfaces with frozen rotor model (https://www.cfd-online.com/Forums/cfx/200974-failure-partitioning-due-axial-fluid-solid-interfaces-frozen-rotor-model.html)

Fred_Erik April 18, 2018 08:46

Failure in partitioning due to axial fluid-solid interfaces with frozen rotor model
 
Dear all,


I have a problem with interfaces between fluid and solid domains. Some domains have rotational periodicity. So the connecting surfaces have a pitch change. For this I selected the frozen rotor interface model with specified pitch angles. The surfaces contain radial as well as axial surface parts. I separated them in one interface with pure axial faces and one interface with radial faces as it is recommended in ANSYS CFX Help. When I start the solving job the solver stops in partitioning part with failure in transformation of cylindrical coordinates of the interface with only these axial sections. The problematic axial surfaces are only on the side of the fluid domain. I have no explanation for this. Did I forget something in my settings? Can anybody help or has some advice? Thank you!

ghorrocks April 18, 2018 20:09

Can you attach an image of what you are modelling?

Fred_Erik April 19, 2018 04:59

1 Attachment(s)
Hello ghorrocks,

I attached an image which shows an extract of the problematic geometry. The pitch change is quite large, but this shouldn't be a problem, because it is full rotational periodic in fluid domain, isn't it? I tried it with pitch ratio of 1 in some other area, but it fails, too. The problem apperars if I choose the frozen rotor model.

What about meshing? The mesh in the fluid domain is fine with using prisms for the boundary layer. The mesh of the solid domain is coarser and has no prisms. Are there important recommandations concerning meshing?

Thanks for an answer!

ghorrocks April 19, 2018 07:52

Your fluid domain appears to be only 0.2 degree segment. An angle this small is going to cause problems as the mesh quality will be terrible and getting the GGI interfaces to match up will also be difficult. Try a larger segment angle, 2 to 5 degrees would be my recommendation.


All times are GMT -4. The time now is 09:07.