[silent2608]

Hey,

This is actually a transient multiphase vof case but that fact doesn't change anything about the problem, so I'll post it here.
I have a case in which I have a fluid domain with one pressure outlet and one pressure inlet, otherwise walls.

I'm using a time dependent gravity as momentum source term in cell zone conditions and a time dependent pressure profile for my inlet boundary.

At t=0, pressure inlet = pressure outlet = 0 gauge pressure.
Eventually pressure inlet will have a value higher than zero --> therefore it should be pressure inlet.

However, due to gravity(t) I will eventually get flow in the "wrong" direction across pressure inlet and outlet. What this means is that fluent will tell me that I have a lot of backflow across pressure inlet and pressure outlet.


Will this be a problem for fluent, or can I ignore these messages? Is there a logical error in this setup? I must have a pressure difference between pressure inlet/outlet and I will get flow in +y axis after a while due to gravity and through this backflow. I do not know about mass flow or velocites (these are subject of the simulation).

edit: In fact due to phase oscillation I may always have backflow. I'm actually pretty certain that it doesn't affect fluent in terms of getting a bad simulation (other than false velocity vectors of backflow volume fraction) but I'm worried that it will affect convergence bahaviour and therefore will add to computing cost. Is this true? I could potentially move to a closed system setup and live with its drawbacks.