[gopisandy]

Hi all,

I am trying to simulate a flow between two annular regions which are in contact. Please see images.

The flow enter the inner ring and exits through the outer. In real life they would be separated by a very thin foil and hence to avoid meshing issues they are in contact.

In this case I tried having two fluid domains(same fluid), but initially I thought I could have just one domain having a surface as a wall. Which is the right way? Can I define a contact region or surface as a wall?

When I look at the velocity contour its not clear if the wall is being detected.

Any help is greatly appreciated.

Thanks!!

[mome]

Hi,

yes it looks like one single fluid domain, not separated at least.. Did you share topology and mesh the two bodies continuously? Check that first.. For the interface of the two bodies, what condition is applied in fluent, or doesn't it even appear? If not, you should be able to fix that easily by just defining a BC on the interface.

Mo

[gopisandy]

Hi Mortem,

Thanks for the reply. I did not do anything special in meshing and by the look of things, it isn't continuous(see image).
But anyway, after setting the proper interface BC, it looks good (image).

In this case I had two fluid domains, I was wondering if I had only one domain with a surface at the current interface, could I set the surface as a wall and get the same result? I can't do it for some reason.

That brings me to my general question, if I have a surface body in a solid body, can I define the interface between them? If so how? I cant seem to do it. FYI, all of this is to avoid having a thin membrane and getting rid of meshing issues.

Thanks again!

[mome]

Good that it works already.
About your additional question: doing the same with only one fluid zone is also possible in principle - at least I wouldn't know why not. The question is how you can obtain that fluid zone and still have a well-defined interface. For example, in Ansys Workbench you can simply produce a single part with different bodies and upon meshing define a named selection on the interface between them. Fluent will interpret the single part as your only fluid domain and your named selection as a BC which you can then set to your internal (two-sided) wall or what you need.

That brings me to my general question, if I have a surface body in a solid body, can I define the interface between them?

I don't ever use surface bodies in fairness, but in general, when a volume body sits within another one, you can define interfaces decently when subtracting the inner one from the outer one and then use the resulting surfaces as interfaces or similar approaches.. but that may be a question for the respective forum

Regards