Hi, I encountered backflow in my simulation. Once there's backflow, there is problem with convergence. Any suggestion to minimise backflow through the pressure outlet? Does an increase of total backflow temperature (default=300k) help?

Best regards, May

hallo, backflow doesn't necessary mean problem. It might be natural to the problem you are solving but you might consider extending computational domain at the outlet side so that you get more accurate result in the area close to the "backflow". The backflow might by because of some flow separation at wall or in the area where some kind of recirculation might be real. If the outlet BC is close to area of interest you shoud anlarge computational domain.

If it is the case of fluid flow in adverse pressure gradient (like compressor stage) you may start with low backpressure level and increase it succesively till you reaches some operation point that is of interest.

Another possible solution might be to set backflow direction from adjacent field and set reasonable value of backflow teperature (especially in compressible flow). a.

Hey May,

If you are modeling a flow in a region and extending the domain is a choice, please do so. The flow in this case will be uniform by the time it reaches the outlet BC. No one can tell you how far you should go; so use your imagination! Don't forget that "Pressure-outlet" BCs are more stable that "Outflow" BCs, since pressure is perdefined beforehand there. If you are using Pressure-outlet, you can specify the "backflow total temperature". Change it to the average temperature at the outlet of your domain. (in case backflow exists, the cells adjacent to the outlet will be assigned this temperature, it helps!). How many outlets do you have in your domain? More than one? If so, never use both the pressure-outlet and the outflow together (they hate each other!).

Kind Regards