[medaouarwalid]

Hello every one

I have been carrying out a numerical simulation of a free air jet discharged through 7 nozzles (nozzles's walls are inside the domaine) , i use Ansys software . i set the boundaries conditions like the photo. after the mesh, and when the calculations begins, it displays a warning message after each iteration "reversed flow in XXX faces on pressure-outlet" or ( "reversed flow in XXX faces on outflow" when i switch the boundries from pressure outlet to outflow). could any one explain this to me, and does it have an influence on the convergence ?

[venturi]

This means that flow is entering the domain at your outlet boundary condition (in some faces, not all of them). And yes, it can damage the convergence.

To get rid of the problem you should extend your geometry, so the flow has more space to fully develop.

[medaouarwalid]

Originally Posted by venturi

This means that flow is entering the domain at your outlet boundary condition (in some faces, not all of them). And yes, it can damage the convergence.

To get rid of the problem you should extend your geometry, so the flow has more space to fully develop.

Thank you for replying. I set the dimensions of the domain like the real domain , is it accepteable if i extend it more thane the reality ?

[venturi]

Well, that depends on your problem.

What is there past your geometry? The flow goes out to atmosphere? Or to another chamber?

[venturi]

Oh, I just saw that your side surfaces are also pressure outlet, right? So, it is like an injector of air in the atmosphere?

If so, maybe you should look into a better boundary condition. I don't have experience with this kind of cases, sorry.

[medaouarwalid]

Originally Posted by venturi

Well, that depends on your problem.

What is there past your geometry? The flow goes out to atmosphere? Or to another chamber?

It is a study for ventilation in a room, my domain is a room i just chose the cylindre shape and not the parallelepipede shape to gain a little time of calcul in the corners . So yes, air goes out to atmosphere

[LuckyTran]

The domain seems to me plenty large compared to the tiny jets. However, we cannot rule out such obvious concerns such as mesh quality.

Which surface does the reversed flow occur on? Is it pressureoutlet1 or 2 or both? There should be some flow entering in pressureoutlet1 due to entrainment by the jet. If there is a small number of reversed flow on 2, then the warning might be ok and the solution not adversely affected.

[medaouarwalid]

Originally Posted by LuckyTran

The domain seems to me plenty large compared to the tiny jets. However, we cannot rule out such obvious concerns such as mesh quality.

Which surface does the reversed flow occur on? Is it pressureoutlet1 or 2 or both? There should be some flow entering in pressureoutlet1 due to entrainment by the jet. If there is a small number of reversed flow on 2, then the warning might be ok and the solution not adversely affected.

Bouth surfaces 1 and 2 does the reversed flow, and the dimensions are close to the real dimensions of the room