[wilhelms]

Good evening

I am creating mass in my model, and I get very high velocities in the air phase. Anybody that have an idea where the mistake/s could be?

Geometry data: Length of model = 10m, width=1m,
Inlet area and outlet area = 1.25m^2, height step =0,25m.

The geometry is set to be 70% filled with water.

Set-up:

- Pressure-based, velocity formulation absolute, transient, gravity=-9.81
- VOF, implicit, open channel flow, implicit body force,
- k-epsilon, relizable, standard wall functions
- Primary phase = air (constant density) and secondary phase = water
- Inlet boundary: Multiphase, free surface level -2.14,bottom level -3.015.
Air, mass flow rate = 0.045 kg/s
Water, mass flow rate = 87,5 kg/s
- Outlet boundary: Outflow, flow rate weighting = 1
- Reference values: v=0.1m/s
- Solution method: PISO

Any help would be appreciated

[laurentb]

Hi,
What's about your residuals ? Did you achieve convergence ?

[wilhelms]

Originally Posted by laurentb

Hi,
What's about your residuals ? Did you achieve convergence ?

After 800 iterations it has still not converged. The residual for the continuity is 10^-2.

Under relaxation factors was set up:
Pressure = 0.3
Density = 1
Body Forces = 1
Momentum = 0.5
Volume Fraction = 0.5
Turbulent Kinetic Energy = 0.8
Turbulent Viscosity =1

Thank you for trying to help me

[laurentb]

You must obtain convergence before checking mass imbalance.
Try to iterate more until you obtain convergence

[wilhelms]

Originally Posted by laurentb

You must obtain convergence before checking mass imbalance.
Try to iterate more until you obtain convergence

Thanks. I will do some more iteration to try to get convergence, but do you think the under relaxation factors seems reasonable? Is it corect to set this up as a transient problem, or should it have been steady state?

[laurentb]

The choice between steady and transient depends on what you want to do, what's the aim of your work ? don't matter about under relaxation factors before you are not sure of what you want to do exactly...

[wilhelms]

I want to see the effect the step has on the flow pattern.

[laurentb]

try to use steady state formulation

[delaneyluke]

If it's an open channel flow, the top wall of your model should have a pressure BC of atmosphere

[wilhelms]

Originally Posted by delaneyluke

If it's an open channel flow, the top wall of your model should have a pressure BC of atmosphere

Thank you. That sounds like a good advise