[Engineer1]

Hi everyone,

I am new in Fluent. I am doing a simulation of a 3D nozzle which has pressure inlet and pressure outlet BCs.

The geometry consist only in the fluid volume internal in the nozzle (from the inlet to the outlet and stop).

When I start the simulation, I get the message "reverse flow in XXX faces on pressure outlet" for every iteration.

Can anyone help me?

Thanks in advance

[edoan]

you can wait, let the sim run and the reverse flow # of cells might start to shrink and eventually go away. If not then you can extend your domain further with a straight section. in the case of a recirculation region near the outlet this method will fix that.

[Engineer1]

Thanks for your reply.
The number of cell does not shrink so I think I have to extend my domain.

Sorry for my ignorance, but how do I do that?
I'm working with Workbench, should I open the DesignModeler and extrude a new portion of domain from the outlet of the nozzle? Or i must create a bigger volume, also broader than the nozzle outlet?
And then, should I do anything special with the mesh or is it sufficient to merge the volumes and remesh again? And what about the BCs for the new volume? Where should I apply them?

Thanks for your patience and sorry for all these question.

[sircorp]

Originally Posted by Engineer1

Hi everyone,

I am new in Fluent. I am doing a simulation of a 3D nozzle which has pressure inlet and pressure outlet BCs.

The geometry consist only in the fluid volume internal in the nozzle (from the inlet to the outlet and stop).

When I start the simulation, I get the message "reverse flow in XXX faces on pressure outlet" for every iteration.

Can anyone help me?

Thanks in advance

I have same problem at the outlet(Reverse Flow). Tried few trick but non worked out. However, that does not stop system running(if reverse flow is confined to few cells only) My problem is confined to only to top two cell at the outlet. Trying to analyse it how to stop.

Do one thing : Try to setup Negative Pressure at the outlet. It may work.

Shane

[sircorp]

Originally Posted by edoan

you can wait, let the sim run and the reverse flow # of cells might start to shrink and eventually go away. If not then you can extend your domain further with a straight section. in the case of a recirculation region near the outlet this method will fix that.

I extended the outlet domain further. But now Reverse flow still happens at Top few cells. Extending may or may not help. (I am not an Expert though)

[Engineer1]

@Sircorp

I read in other posts that a reverse flow may happen in a few cells. For exemple in my case, moving the outlet 1 meter beyond the "old" outlet reduced the cells from about 500 to about 10-15 at the top cells. A such so low number is probably due to physiological recirculation, not due to numerical problems, so it is not a problem.

[edoan]

Right and since you have extended the outlet away from your area of interest then the small number of cells where the backflow properties are being imposed will not have an effect on the info you are trying to gather from the sim. glad it is resolved

[sircorp]

Originally Posted by Engineer1

@Sircorp

I read in other posts that a reverse flow may happen in a few cells. For exemple in my case, moving the outlet 1 meter beyond the "old" outlet reduced the cells from about 500 to about 10-15 at the top cells. A such so low number is probably due to physiological recirculation, not due to numerical problems, so it is not a problem.

I had a terrible experience with Adaption(Refining the Mesh) Earlier I was getting back flow problem just in few cells(only at top layer cells of the outlet), now back flow problem happens in 100s of cells and eventually solution collapsed.

However I got a idea from another gentleman post (Thanks to CFD Online).

Setting pressure slightly -negative ( relative to outlet) in the operating conditions.

My system operates at high flow rate (Several MPa pressure) so a tiny negative outlet pressure setting will not affect the solution but may solve CFD problem.