Simulation of flow over Backward Facing Step
SIMULATION OF FLOW OVER BACKWARD FACING STEP
1.)AIM:- The objectives of the current project are: -
1.) To simulate the air flow over a Backward Facing step using Converge Studio.
2.) To study the turbulent flow characteistics downstream to BFS.
3.)To compare the flow over BFS for different mesh grading of 2e-3 , 1.5e-3 , 1.0e-3.
2.)INTRODUCTION:-The study of flow over a Backward facing step (BFS) is a rudimentary problem. Although it is very basic problem, It has many scientific important applications. Like Turbulent flow, flow separation are very common in flow over the airfoil. Such flow can be seen in automobile and aircraft industries, fluid-solid interaction etc. There are many unwanted effects obtained as the results of turbulent flow like the increase in the pressure drag force on airfoil, increase in the instabilities, noise etc. The current solution to reduce the undesirable effect by altering the turbulent flow properties and to do this we need to study the nature of turbulent flow, flow separation and their impacts. The study of turbulent flow is usually preferred on various geometries like rib, fence, bluff body with the splitter plane, forward/Backward facing step etc. The study of turbulent flow and separation of flow using the BFS is very easy and convenient because there is one flow separation point i.e. the top point of a step of the BFS, the turbulent flow is obtained from the step wall to reattachment zone downstream to the step. Additionally, there are many research papers available on the flow through BFS so it is easier to validate any new numerical schemes. These features of (flow through) BFS maintain its utilitarian.
Flow characteristic downstream to the Step of BFS
The general characteristics of a BFS flow begins with an upstream boundary layer separating at the step edge due to the adverse pressure gradient that develops into a thin shear layer. The size of the shear layer increases in the downstream. It consists of many vortices and further vortices start merging together to form larger vortices. The shear layer becomes the shear zone when the size of shear layer increases due to vortices, coherent structures. Finally, the separated flow attaches with the bottom of the BFS when the favourable pressure gradient is obtained. The point of reattachment is known as the reattachment point. The distance between the wall of the step and the reattachment point is known as the reattachment length. The location of the reattachment is unsteady due to inherent oscillatory motion of the shear layer. Hence, the reattachment zone is formed in which reattachment point can present. The turbulent structures in the shear layer entrains irrotational fluid outside the shear zone. This creates low velocity circulation between step wall and the shear zone. This leads to the formation of the recirculation zone. The recirculation zone is mainly comprised of a primary vortex in the centre and a secondary vortex adjacent to the corner of the step.To study the flow over the BFS i have used the geometry as shown in figure 02.I have used fluid as air and assumed that it consists of oxygen and nitrogen.
Figure 02 Geometry of BFS
I have simulated 2- dimensional flow but it is compulsory to provide third dimension for geometric purpose in converge studio so i have provided 0.025m along z-direction. The dimension along this direcion is assumed to have no effects on the flow.The flow is assumed to be compressible,2-Dimensional,turbulent,non-reactive but genral flow.
Figure 03 complete 3-Dimensional geometery of BFS
In Figure 03, The complete geometery is made up of triangles because in converge studio triangle is the basic structural element.
3.)PROCEDURE:-The following steps have been performed in converge studio for the simulation of flow over the BFS:-
1.) Forming the geometry of BFS.
2.) Assigning Boundary names to the surfaces of the geometry.
3.) Setting the case setup files.
4.) Post-Processing
I have used the same procedure for three case studies. In the project i have shown three cases for three differnent mesh sizes:-
1.) Case 1-> dx=dy=dz=2e-3
2.) Case 2->dx=dy=dz=1.5e-3
3.) Case 3->dx=dy=dz = 1.0e-3.
Also,i have used paraview for post-processing along with converge_Studio.
3.1) Forming the geometry of BFS:- The dimension of the geometery is shown in figure 02 and along z-direction 0.025m has been given.The whole geometry has been made by triangles.
3.2)Assigning Boundary names: - In this process, different surfaces of the boundary have been assigned different names. It is important to do because boundary conditions and initial conditions are applied by using names of different boundaries/surfaces.
Figure 04:- Flagging Boundary Names to surfaces
3.3)Setting case setup files:-Table number 01 shows all the files that have been filled for complete setting of the case setup file for the simulation: -
Table 01 :- Complete Case Setup Table
Figure 04: - Time Parameters used for the simulation
Table No 02 Boundary Conditions
Figure 05 :- Meshing for all three cases
Figure 06 :- Complete case setup
3.4) Post-Processing: - I have done post-processing of all the three cases using Paraview and Converge_Studio and I have used Converge_Studio for comparing the results of all the three cases.Post_processing is actually to obtain the desired results from the simulation.I have provided graphs for velocity, pressure, mass flow, animation of the development of velocity profile and pressure profile and velcity contours, pressure contours for all the three cases as the results of the simulation.
4.)RESULTS: -
4.1)Profiles: -4.1.1)Velocity profiles for all the three cases: - The velocity profile obtained for all the three cases are very similar (fig. 07 (A),(B),(C)) but the velocity profile for case 3 (fig 07(C):-for the finest meshing) is much more clear than the first case (fig 07(A)). The velocity has not changed after second case (fig. 07(B)).
Figure 07 (A) :- Velocity profile for case 01
Figure 07 (B): - Velocity profile for case 02
Figure 07 (C): - Velocity profile for case 03
4.1.2) Pressure Profile for all the three cases: - The pressure just after the step has changed a lot from the figure 08(A) to figure 08(C). The Pressure profile has not changed after figure 08(B).Initially the presssure was higher at the inlet but final pressure profile demonstrates the reverse.
Figure 08 (A): - Pressure profile for case 01
Figure 08 (B): - Pressure profile for case 02
Figure 08 (C): - Pressure profile for case 03
4.2)Contours: -
4.2.1)Velocity Contours :- The velocity contours shows the flow is turbulent just downstream to the step as it was expected.It can be seen in fig 09 (A),(B),(C). The flow is separated at the top of the step and the shear zone can be seen also.The recirculation zone between the step and the shear zone can also be seen in the figure 09(A),(B),(C).
Figure 09 (A): - Velocity contour for case 01
Figure 09 (B): - Velocity contour for case 02
Figure 09 (C): - Velocity contour for case 03
4.2.2)Pressure Contours: -
Figure 10 (A):- Pressure conotur for case 01
Figure 10 (B):- Pressure conotur for case 02
Figure 10 (C):- Pressure conotur for case 03
4.3)Graphs :-
4.3.1)velocity profiles at x= 0.13m from the inlet (just before the step) for the three cases :-Figure 11 (A),(B),(C) all are symmetric about x=0.005m i.e. the midpoint of the thinner section of the geometery.The nature of all the three graphs(Fig 11(A),(B),(C)) is very different. The velocity has been increased upto 0.002m and then decreased from 0.002m to 0.008m and thenincresed upto 0.009m in figure 11(A) while in figure 11 (B) and 11(C) the velocity keeps increasing irregularly step wise upto the midpoint and then is decreased in the same manner it has increased.The pattern of increasing and decreasing is different for case 02(figure 11(B)) and case 03(figure 11(C)) but The highest value is 150m/s from 0.004m to 0.005m for both the cases.
Figure 11 (A): - Velocity variation with y-point at x=0.13m from the inlet for case 01.
Figure 11 (B): - Velocity variation with y-point at x=0.13m from the inlet for case 02.
Figure 11 (C): - Velocity variation with y-point at x=0.13m from the inlet for case 03.
4.3.2)pressure variation at x=0.13m from the inlet (just before the step) for all the three cases: -Pressure is symmetric about the midpoint only in figure 12(C).All the graphs are different from each other.
Figure 12 (A): - pressure variation with y-point at x=0.13m from the inlet for case 01.
Figure 12 (B): - pressure variation with y-point at x=0.13m from the inlet for case 02.
Figure 12 (C): - pressure variation with y-point at x=0.13m from the inlet for case 03.
4.3.3)velocity profiles at x=0.14m from the inlet (just after the step) for the three cases: -All the three graphs (figure 13 (A),(B),(C))are very different from each other in many aspects like The value of velocity at -0.01m is negative for first case( figure 13(A)) , zero for case 2(figure 13(B)) and approximately 8m/s for the third case( figure 13(C)),The highest value is 160m/s for first case (figure 13(A)) and 150m/s for case 02 (figure 13(B)) and case 03 (figure 13(C)) but the range for which it is attained is different. The nature of graph is same for all the three cases at x=0m where there is tremendous increase in velocity of approximately 100m/s.This is due to the nature of flow below x=0m is turbulent and it is very different from the flow above x=0m.
Figure 13 (A): - velocity variation with y-point at x=0.14m from the inlet for case 01.
Figure 13 (B): - Velocity variation with y-point at x=0.14m from the inlet for case 02.
Figure 13 (C): - Velocity variation with y-point at x=0.14m from the inlet for case 03.
4.3.4)pressure profiles at x=0.14m from the inlet (just after the step) for the three cases: -The pressure varition is also maximum at x=0m for all the three cases (Figure 14 (A),(B),(C)) because of the turbulent flow, recirculation zone below x=0m which is not above x=0m.
Figure 14 (A): - pressure variation with y-point at x=0.14m from the inlet for case 01.
Figure 14 (B): - pressure variation with y-point at x=0.14m from the inlet for case 02.
Figure 14 (C): - pressure variation with y-point at x=0.14m from the inlet for case 03.
4.4) Graphs using Converge_Studio
4.4.1) Graph for the average mass flow rate through both inlet and outlet for all the three cases: -The mass flow rate for all the three cases are same after 6000 cycles approximately.The outflow and inflow mass rate is equal to 1.6m/s approximately.
4.4.2) Graph for average velocity at Outlet for all the three cases: -The average outlet velocity is same for all the three cases after 6000 iterations and it is 68m/s.
4.4.3) Graph for average velocity at inlet for all the three cases: - The average velocity is approximately 135m/s for all the three cases after 6000 iterations. The value of the mass rate is same for both inlet and outlet but the average velocity for the inlet is greater than that of oulet because the cross sectional area of inlet is smaller than that of outlet.
4.4.4)Graph for cell counts for all the three cases: -The cell numbers for the third, second and third case are given below.The cell number for the third case is maximum because the size of the cells is the smallest.
4.5) Images showing different zones of turbulence: -
4.6) Animation: -
4.6.1) Animation for velocity development for case 03.
4.6.2) Animation for pressure development for case 03: -
5.)CONCLUSIONS:- The results obtained in all the cases are similar but i find the case 03 results are the most appropriate. 6000 iterations are enough to obtain the same average results for all the three cases.I find the nature of graphs are not varying much after mesh size 0.0015m. The turbulent zone has been shown clearly and it matches with figure 01. The detailing is missing that may be acieved with either by futher resolution or more number of iterations.