## Fluids

#### Creating an internal inlet within a geometry or using outlet velocity turbulence profile for inlet

• kwl8rs
Subscriber

I am modeling additional variable (AV) flow in a turbulent flume but have run into the problem of significant dissipation of the AV before it reaches the developed turbulent section of the flume. The velocity profile can be seen in the attached image and analyzing the AV concentration as a result of advection is the primary goal, so this poses a problem. I was hoping someone could give me advice on creating an internal inlet to release the additional variable inside the geometry rather than at the face inlet the water enters from. This would allow the additional variable to immediately interact with the turbulent environment rather than only encountering a developed boundary layer after 5 seconds of dissipiation. I've tried doing this by creating a plane inside the geometry, sketching a circular inlet, splitting it from the plane and making it a named selection, but the face split is superimposed onto the outer face of the flume where the initial water inlet is. Do you know of any way to create an internal inlet within a geometry?

Another idea someone told me is that water from the outlet can be used for the inlet, which would allow for a turbulent environment to be present in the entire plume through recirculation. Again I'm not sure if this can be done or what steps to take. Any advice would be appreciated.

• Vishal Ganore
Ansys Employee

Hi, could you please share some snapshots of geometry after creating circular inlet just to understand what is happening exactly?

• kwl8rs
Subscriber

I attached a few images of the geometry, additional variable concentration, and velocity field to get a feel for the project.

entire geometry with inlet on upper left and outlet at bottom right

geometry showing inlet and roughness blocks to create realistic turbulence

Additional Variable in XY plane along centerline

velocity field: you can see turbulence takes time to develop, so inserting the additional variable in the middle of the plume would create the most realistic model. I'm increasing the size of the domain to prevent the additional variable hitting the roof as well.

Thanks a lot for your help!

• Holger Grotjans
Ansys Employee

Hi, did you consider to create a local source in the domain? So instead of an inlet you can create the AV at the required location by a source term.

• kwl8rs
Subscriber

Do you mean a sub-domain, point source, or source term? I'm using CFX so I'm not sure if I can create a source term.