[falsc233]

Hello,
I am trying to model laminar flow inside of a duct with square cross sections. I am using cyclic bundary conditions and I am starting with simplefoam to solve for a steady-state solution. I set the inlet and outlet boundary conditions to be "cyclic" and specified Ubar to be 0.04m/s. The axial velocity profile looks alright and is quite close to the analytical solution calculated using the pressure gradient printed out after each iteration.

However, my simulation has two major problems:
1. The residual from the pressure solver is very high and fluctuates at around 0.28, and the residuals from the velocity solvers are also much higher than the case where the boundary conditions are not cyclic;
2. The pressure distribution shown in paraview does not match the pressure gradient provided by the solver from the terminal. I was expecting the pressure to be droping linearly from the inlet to the outlet, but instead it is close to zero throughout the channel and oscillates at an amplitude of 1e-10.

I have included some of my codes in the attachment. I'd wish to know how to model a laminar pipe flow using cyclic boundary conditions and what are the causes of the problems I've encountered. Thank you very much!
blockMeshDict.txt

controlDict.txt

fvSchemes.txt

fvSolution.txt

fvOptions.txt

[mostanad]

Hi,
Can you please just tell me whether you found any solution for this problem?
Thank you for your help in advance,
Mohammad

[falsc233]

Originally Posted by mostanad

Hi,
Can you please just tell me whether you found any solution for this problem?
Thank you for your help in advance,
Mohammad

Hi, unfortunately I did not find a solution to this problem, and eventually I gave up using cyclic BCs. I hope you have better luck!

[mostanad]

Hi,
Thank you for your great reply. I am still struggling with that. What method did you use instead of that?

[falsc233]

Originally Posted by mostanad

Hi,
Thank you for your great reply. I am still struggling with that. What method did you use instead of that?

Hi, I just made my computational domain longer instead.
Good luck!