[liviaadi]

Dear all,

I am simulating in OpenFoam an open channel flow in laminar condition with the solver interFoam.
The purpose is to explore and try to understand the physical meaning of the boundary conditions. I am considering a large time step for the analysis in order to reach te steady state and I am running in laminar condition.
I have played a little bit with it and now I would like to verify my numerical results.
I was wondering if somebody has already made something like this and how was validated.

I used the same approach for a flow in a pipe (Poiseuille flow), but in that case was simpler because it was not a multiphase problem and the analytical solution was very easy to find.

I hope that someone can help, I am using these it for the first time.

Thanks !

[FMDenaro]

You are simulating in practice two laminar boundary layers evolving spatially on the upper and lower walls until they merge in a fully developed laminar profile (Poiseuille). Therefore, you can use the BL exact solution to check your numerical results. Of course, avoid to check the solution in the singularity point x=0.

[liviaadi]

Hi FMDenaro,

thanks a lot for your answer.
Probably my post was not clear.
I simulated and validated analytically the case of the laminar flow between two walls (Poiseuille flow). I found this in many tutorials so I did not have problems and it was a good training.

Now I am treating a multiphase problem, water/air. So it is an open channel.
I have set the bottom as a wall, the left side as inlet, the right side as outlet and the top as atmosphere.

I need to compare the free surface elevation provided by OF and the one calculated theoretically. The problem is that in open flow problems, the uniform conditions that I want to simulate is solved using the "empirical" Chezy equations, which is for turbulent flow and not for laminar. In addition the most used form is the Manning equation, which requires the definition of the Manning coefficient and I do not know hoe to introduce this in the turbulence model.

Sorry for the long post and thanks for the help.

[RobertHB]

To verify your fluid flow you can compare the vel. profile from paraview with the theoretical profile of a laminar flow (Stigler, 2014). The theoretical law is for a tube flow (Link) but you can use e.g. Matlab to ensure it is representing your plate flow. Here is what i did for my simulation:

y = -0.5:0.01:0; %distance from the wall. note that my array goes from -0.5 to 0
R = 0.5; %height of my channel
v=max(U0)*(1-(y/R).^2); %theoretical law. see paper and link

% shift data to reseble the plate bounded flow
yabs = abs(y)
yadd = y+0.5;

% make axial value dimensionless
v=v/max(U0);
U=U0/max(U0);
Points2=Points2/0.5;

Stigler ANALYTICAL VELOCITY PROFILE IN TUBE FOR LAMINAR AND TURBULENT FLOW Engineering MECHANICS, Vol. 21, 2014, No. 6, p. 371–379

[liviaadi]

Dear Robert,
many thanks for your explanations. But is this valid for an open channel as well, or just for the circular pipe ?
I have to compare the results from interFoam solver, which uses the VOF method with a LAMINAR OPEN FLOW. It is a different situation because I have the interface water/air. I think that the comparison is the free surface height in this case.

Thank you anyway!