CFD Online Discussion Forums

CFD Online Discussion Forums (
-   CFX (
-   -   Flow Reversal at the inlet (

salihovic July 15, 2010 03:34

Flow Reversal at the inlet
I have analyzed a flow field over a cylinder in pulsating flow (unsteady flow). At the inlet, we have periodic forcing with high amplitudes.

Velocity inlet: u(t)=u0+2*u0*sin(w*t);

At some time the flow becomes reversal at the inlet because the time dependent amplitude is twice the steady state amplitude (u0). We have implied a velocity inlet for the inlet boundary condition.

The solver gives a warning about the flow reversal.

My question is about:

How does the CFX solver face this problem?
Thanks in advance,

ghorrocks July 15, 2010 06:47

When you specify a velocity boundary with reverse flow does the solver construct walls to stop it? I am not sure - if it does not then you are fine and just proceed.

salihovic July 15, 2010 07:03


The solver gives me a warning. Something like ''A wall is constructed.....flow reversal...'' But, I am happy with my results.

How can I interpret the construction of the wall? What does it mena physically? Is this not anymore a BC that I am imposing (velocity inlet)?

At the outlet I a have pressure outlet BC (in this case I may also use an opening type BC)

Is there any documentation of CFX about that issue?

Thanks a lot,

mvoss July 15, 2010 10:50

cfx is building artificial walls, since it is an "inlet" the inwards face normal has to be same like the vel.components.

Try an opening with specified vel.comp. and set Expert Parameter-->Discret.-->Misc.--> build art.walls =false

this should do it.


All times are GMT -4. The time now is 07:10.