Learning Forum

[angie07]

Hi everyone,

I am working on a dissertation simulation and I require some assistance. Air enters a stepped tee pipe through the inlet and exits from the slits at the top and bottom of the horizontal pipe. My goal is to monitor the output velocity field.

I have attached some pictures to show you the geometry. They are section views of the top and right plane respectively.

I would like first to carry out the simulation in 2D to establish the boundary conditions and then proceed with the 3D simulation.

My questions are as follows:

1. How can I set the slits as the outlet in 2D and 3D? I am not sure which edges, faces, etc to pick for the named selection.
2. To carry out the 2D simulation, what is the best method to define the domain since I am interested in internal flow? I tried carrying out the simulation on internal geometry (i.e. the shape of the gap from the pictures attached) and also subtracting that from a large rectangular domain.

I would appreciate any help you can provide on those questions, as well as any general advice.

Thank you.

[Rob]

Given the change in surface area between the inlet and outlet I'd question the benefit of trying this in 2d. Given the apparent symmetry I'd split the model.
The steps you should take now are fairly similar to an experiment, once you've clicked on Help and done some tutorials:
On paper (A3 is good for this or A4 if your hand writing hasn't yet degenerated!)
Define the problem
Sketch out the geometry and add ALL boundary conditions, material properties etc. Include everything you don't know but want to.
Sketch where you think the flow is going to go.

Post the image of your sheet of paper.

Open the geometry tool (you have CAD so you can use that) and create the fluid region.
Mesh the model
Set up the solver & run the case
Post process

[angie07]

Hi Rob,
Thanks for your reply. I will think on whether the 2D step is necessary then.
Here is an annotated diagram of my simulation. Air enters at the inlet (in blue) and exits through the slit outlet (in red). The turbulence model is Transition SST. The inlet velocity is known and the goal of the simulation is to monitor the velocity field at the outlet
My primary trouble with setting up the simulation is defining the outlet. When setting up the named selections in the meshing step, i don't know how to define the slit gap as the outlet. Selecting the surrounding edges did not work. Could you shed some light on how a gap could be defined as an outlet?
Thanks again.

[Rob]

I think you've fallen into the classic hole that all new users of CFD manage to do. You've drawn the solid part of the nozzle: we're interested in the bit where the solid isn't. So you need to subtract the solid from a fluid region. As you're interested in the outlet flow you'll need to add an external domain too. If I just did a volume fill on the above I'd set a boundary on the outlet surface and that might (or might not) modify the result.
Have a look at https://ansyshelp.ansys.com/account/secured?returnurl=/Views/Secured/corp/v221/en/flu_tg/flu_tg_exhaust_manifold_cht.html and think how it's similar to your model.