11 - Simulation of Flow Through a Pipe Part 2 2

This document represents the second part of the project related to the simulation of an incompressible laminar flow through a pipe. In the previous part of the project, a 4 edge degrees simulation was carried out using icoFoam and wedge boundary conditions. The purpose of this part is to employ the symmetry boundary conditions, applied at a geometry with 3 different angles (10,25 and 45) and compare both results. As the transient part is no longer of interest, and for further comparison, the solver simpleFoam was used in this part.

First, all the new meshes were generated with the Matlab program of the first part of the project, and the boundary conditions were changed accordingly. In the following picture, the mesh for the 45 degrees case (symmetry BC) is shown as an example. The results of the computational simulations will be discussed next.

RESULTS

Comparison of the velocity profiles (Fully developed vs Analytical)

Comparison of the Pressure Profile through the Pipe

Average Velocity [m/s]:

Hagen-Poiseuille	4 Degrees	10 Degrees	25 Degrees	45 Degrees
0.10731	0.1110	0.1096	0.1082	0.1076

Maximum Velocity [m/s]:

Hagen-Poiseuille	4 Degrees	10 Degrees	25 Degrees	45 Degrees
0.2146	0.2210	0.2188	0.2166	0.2148

Pressure Drop [Pa]:

Hagen-Poiseuille	4 Degrees	10 Degrees	25 Degrees	45 Degrees
43.4	37.45	37.52	37.62	38.85

DISCUSSION:

The results above show that there is not an important difference between the wedge boundary conditions and the symmetry boundary conditions. The small differences are most likely due to the angle simulated. In all of them, profile is fully developed, as the velocity profile is pretty similar to the analytic solution. The first thing to notice is that, the higher the angle simulated, the higher the fidelity of the results (higher values of mean and maximum velocity, closer to the analytical). Also notice that, as expected, the biggest differences appear near the wall and near the axis.

In term of the pressure drop, all the simulations approximate to the analytical solution, but none of them reaches the analytical value close enough. The tendency shows that also the pressure drop increases when we increase the angle simulated. This means that the simulation cannot predict with exactitud the pressure drop in the tube, with the simple model used here. Further prove with other solvers, maybe icoFoam, should be carried out in order to choose the correct one.