CFD Online Discussion Forums - Help!simulation of free water jet

Hi, everyone:

I am trying to use CFD and am inexperienced with it. Can anyone with experience in modelling free water jets help me? I need to calculate a axisymmetry jet spray to air atmosphere, and I am using Fluent to know its velocity and pressure distribution in the air.

The jet got high velocity(950m/s) and pressure(350 MPa) after it coming through a very fine nozzle(1 mm). The Re number is 312000.

I use a rectangle computational domain(100mm*10mm). i.e.I have got a length of X=100mm downstream of the jet nozzle and the distance from central line of nozzle is y=10 mm. Along the X direction, there are 800 points and along the Y direction, there are approximately 60 points.

My settings of model are:

Space: Axisymmetric

Time: Steady

Viscous:RNG k-epsilon turbulence model

The settings of boundary condition are:

velocity_inlet.1: velocity-inlet

pressure_outlet.2: pressure-outlet

pressure_outlet.3: pressure-outlet

pressure_outlet.4: pressure-outlet

axis.6 : axis

default-interior : interior

and multiphase is used in my simulation.

The result which I got is very strange. The velocity along the downstream of jet going down when x reach about 20mm, after that, the velocity going up, or even higher than the initial velocity. I am sure there must be something wrong.

My questions are:

1. how to decide free boundary conditions of computational domain. I am not quite sure about the boundary condition what I used.

2. how to define the size of computational domain. It seems to me that Fluent is very sensitive to the size of domain.

3. could anybody give me some reference about the simulation of free jet.

Thanks in advance!