Simple rectangular open-channel flow
 

I'm using OF 5.x for multiphase flows.
The geometry is simple such as rectangular open channel (not duct flows).
The channel is divided to air part and water part
so the boundries I have to set are inlet_water, inlet_air, outlet_water, outlet_air, walls (sidewall+bottomwall), atmosphere

inlet_water and outlet_water are set by cyclic conditions
and I used the momentumSource (constant/fvOptions) for the driving force (below), without inclination of channel or correction of gravity acceleration.

momentumSource
{
type patchMeanVelocityForce;
selectionMode all;
patch inlet_water;
fields (U);
Ubar (0.1924 0 0);
}

However, the simulation results are somewhat strange. the velocity of air part become faster than the water part. So, the air part make the water part accelerate.... I've been trying many BCs for air_inlet & air_outlet, but I'm still very confused..

The details are attached.

https://drive.google.com/open?id=1Cx...5GQcKivKNxsmaX

Please let me knwo what is wrong.

Why did you use cyclic condition?

I hope to get a developed flow by using cyclic condition in the rather short domain. But i realized that using the cyclic condition at the interFoam simulation make nonsense in the physical view and cause instability at the free surface interface..

Did you manage to solve the problem? Could you upload the files, because I have the same case to solve, but I never used multiphase method.