Hi guys,

I'm modelling a steady state laminar flow in a circular pipe. This is gonna be used for an FSI simulation where the pipe will buckle due to an external pressure.

Anyway, I'm comparing my simulation with this paper and the thing that i'm having trouble in is this outlet boundary condition. It says,

A parallel, axially-traction-free outflow is fixed at the outlet

u1=u2=0, -p+2(du3/dz)=0 at outlet

*u1&u2 are the velocity in the x&y direction (radial) and u3 is velocity in the z direction (axial).

Can anyone please explain to me what this boundary means and how would I apply this in CFX?

Thanks in advance.

Regards, YY

Hi,

It says the exit velocity is purely axial, and has a pressure gradient defined as a function of the axial velocity gradient. This is strange, as for an incompressible flow with these constraints there will be no axial velocity gradient so I don't know what it is talking about - unless it is a compressible flow.

I would just use a normal outlet boundary for CFX.

Glenn Horrocks

Hi,

Thanks for your response. The flow is incompressible. Would you think its just a static pressure outflow boundary condition?

Regards, YY

Yes, I would think so.

Glenn Horrocks

Just specify the pressure at the outlet. The solver computes the rest for you.

Hi thanks for all your help. Really appreciate it.

I'm pretty sure now that it is a static pressure outlet but i'm still unsure about the magnitude of it. The equation -p+2(du3/dz)=0 doesn't seem right to me somehow as the units are different for both the terms. Could anyone please clarify this?

Regards, YY

It's like they want to constrain the axial velocity gradient as well for some reason. I don't know why.

Just set p=0 and let the solver come up with the answer.

Hi HekLer,

Thanks for your help. I have managed to contact the author for the thesis. You are right about setting p=0. The author had to apply a stress tensor because he was using FIDAP which needs BC's to be entered in terms of stress tensors.

Regards, YY