Flow over Channel

In this challenge, i will attempt to simulate the age old of problem in the CFD world using Converge CFD. I have used gas simulation considering the continous flow with the following input condition and demisions of the channel as shown in figure below

The dimension of the channel is as follows :

X= 0.1 m, Y= 0.01 m, Z= 0.01 m.

The boundry is analog is shown also in the figure i.e flow is coming from left and going towards the right. There were 3 different Mesh sizes were used in order to better simulate and understand the theorectical concept which we studied in the class. The cases were as follows:

1. The mesh size of 0.001 m i.e. (dx=0.001 m, dy= 0.001 m, dz=0.001 m).
2. The mesh size of 0.0015 m i.e. (dx=0.0015 m, dy= 0.0015 m, dz=0.0015 m).
3. The mesh size of 0.0002 m i.e. (dx=0.0002 m, dy= 0.0002 m, dz=0.0002 m).

As shown in Picture for different cases

Case 1

as shown in figure the geometry looks clean and mesh is generated using CFD Converge which is generated automatically
Case 2


Case 3
For this case since the mesh was so refine i choose to have more zoomed version, so here it as shown in image below

For the simulation purpose i have used the parallel processor with 4 cores for reducing the time. The time step for this simulation was consider 1 nanosecond and simulation run for 15000 cycles, since converge uses the dynamic time step i.e. it adjust the cfl number accordingly from the phyical inputs and codes. For solving purposes i am using the pressure based steady state solver with assumming the temperature as a constant and no slip in the wall.

Here are the results of the contours for different pressure and velocity:

Pressure

Case 1

Here the picture is not very clear because low resolution mesh. I have given input pressure as 1000001 Pa and output pressure as 100000 Pa. so this the reason we can see the red is region on the left and blue region on the right even though in the scale we cannot find the minute changes.

Case 2

Here is the better picture with more refine mesh.

Case 3

As seen in above the nature of the image remains the same. I conclude that the change in pressure does depend on the mesh size since the pressure remains the same on all over the places, I have aslo made the animation of it just to check it, Here is it Pressure Animation .Following are the plots for the total pressure for all the cases

Case 1

Case 2

Case 3

For the Pressure once the pressure reaches to the to it's stability over the channel there is the sharp rise on the total pressure at the inlet this sharpness width increases with the mesh refine which was expected for this kind of geometry. Just in case my animation goes wrong or the flow is flowing correctly i used the glyph function of the paraview just to check flow and image are follows:

Accoridng to mesh size i am execpting the velocity to be maximum in the middle of the plane so i have used the plot ove line as well as velocity contours for different mesh sizes using paraview as follows.
Case 1

Case 2

Case 3

As from the above picture the more intense the meshings are the better the curvature and you can see the curvature is maximum in the top were the point in the y direction is exaclty on the middle so this expected even though the magnitude of the velocitys are slightly different. Following are the velocity profiles contour

Case 1

Case 2

Case 3

As the mesh becomes more fine the velocity contor becomes more refine. The nature of the flow in all the cases remain the same i.e the flow is maximum in the middle of the channel for more clearity click on the animation for Initial flow in the channel and Full Simulation. Since i am also expecting the flow near the wall as near to zero because of the friction between the air and wall, i observed it clearly near the wall of the channel and it is behaving for all the case but only difference is it more refine for more refine mesh.

The Mass flow rate was pretty straight forward since i i have consider air was flowing in the channel and there was no reaction between the components but only change i have consider was surrounding fluid which had a reverse air concerntration i.e. o2 0.77% and N2 0.23% so here are the plots for the mass flow rate.

Case 1

Case2

Case 3

Here are the Total cell counts for all the cases

Case 1

Case 2

Case 3

The conditions and mechanisms for the simulation were remain unchanged i.e. it was used which were recommeded by the converge software, by while simulating the cases couple of things to remeber, just to make sure that the paraview install on the computer is updated to the latest one so that there are less bug in the software. In future we can use run the case on the turbulent flows using various models such as k epsilon or k - omega just to check the flow in z direction as well.