[LuckyTran]

The Pressure Outlet with target mass flow rate option is not working for me (maybe because it's using Bernoulli's).

I want to model a very long pipe using real gas equation of state, density is a function of pressure and temperature. I'm using the pressure based solver with coupled p-v scheme, steady, rke, etc. I am not having convergence issues (except for the wrong mass flow rate).

Suppose I know all the flow variables at the inlet (pressure, temperature, velocity, density, turbulence, mass flow rate everything). I'm using velocity inlet boundary condition which lets me specify everything at the inlet except for pressure and density. In order to get the correct mass flow rate, the solver needs to iterate and achieve the correct pressure and density. Ok no prob.

So on to the pressure outlet with the target mass flow rate option. This makes sense, the pressure at the outlet is whatever it needs to be to achieve my mass flow rate, so I should set the target mass flow to the desired mass flow (the known mass flow at the inlet). Here's the problem.

As far as I know (at least by default): Fluent uses Bernoulli's equation to adjust the pressure on the next iteration. So when the current mass flow is too low, Fluent decreases the outlet pressure to try to increase the mass flow on the next iteration. Because the flow is not incompressible, this is incorrect, since decreasing the pressure also decreases the density and decreases the mass flow at the exit. I end up in a runaway process where the massflow gets worse with each iteration. I need Fluent to do the opposite or to not use Bernoulli's to adjust the pressure at outlet.

Any ideas?

*update* my temporary fix is to set the under-relaxation factor in the target-mass-flow-rate-settings to a negative value.
*2nd update* I put in the wrong x,y coordinates on the inlet profile, which was interpolated, leading to incorrect inlet profile, which is a dumb mistake to make. Maybe this will resolve the issue.
*3rd update* Inlet profile didn't change the convergence behavior. Back to negative under-relaxation factors.

[Mazze[ITA]]

Hi LuckyTran,

did you fix the issue?