Mixing efficiency using ANSYS Fluent
Objective:
To run three different cases of mixing - t problem to compare mixing efficiency following are the three cases -
1) Short mixing tee with a hot inlet velocity of 3m/s and Momentum ratio of 2.
2) Long mixing tee with a hot inlet velocity of 3m/s and Momentum ratio of 2.
3) Short mixing with a hot inlet velocity of 3 m/s and Momentum ration of 4.
- Momentum ratio = velocity at cold inlet / velocity at hot inlet
----------------------------------------------------------------------------------------------------------------------------------------------
Mixing tee problem:
Our main objective is to get the desired temperature of air from the outlet by mixing the cold and hot air in the mixing-tee having two inlets one introduces the chilled air and other hot air. The question is what is the best way to get the desired temperature of the air at the outlet along with an effective mixing. To answer this question the CFD simulation is being performed in ANSYS Fluent for the above three mentioned cases.
Following is the geometry of mixing-tee:
Above image represents the solid volume of the mixing-tee geometry.
The Above image is the wet volume of the mixing-tee on which the CFD simulation has been performed.
For geometry manipulations, SpaceClaim is used.
After this, we move to meshing to create the mesh and perform boundary flagging.
First, we: define the different boundaries in the geometry which are as follows -
CASE - I:
Now we define mesh under the mesh option and generate Mesh. We are using tetrahedral mesh for this case with an element size of 1.021e-2 m.
Following image shows the element quality of the mesh -
The above image shows the cut-plane section of the geometry which was created to get a better view and understanding of the meshing in the domain.
Next, we move to ANSYS Fluet to set up a boundary and initial conditions.
For this case, we have a Short mixing tee with a hot inlet velocity of 3m/s and Momentum ratio of 2.
We are using velocity inlet and pressure outlet boundary condition for this problem. The following table shows the boundary settings for this case:
Material is defined as air and we are running a pressure-based steady-state solver for this problem.
Results:
Followings are the output graphs for this case:
The above graph shows the residuals for different parameters calculated as the function of the number iterations.
Above plot shows the mean temperature of the air at the outlet.
Above plot shows the standard deviation of the temperature at the outlet which indicates how efficient was the mixing in this case.
Post-processing view -
CASE - II:
For this case, we have a Long mixing tee with a hot inlet velocity of 3m/s and Momentum ratio of 2.
Now we define mesh under the mesh option and generate Mesh. We are using tetrahedral mesh for this case with an element size of 1.385e-2 m.
Following image shows the element quality of the mesh -
The above image shows the cut-plane section of the geometry which was created to get a better view and understanding of the meshing in the domain.
Next, we move to ANSYS Fluet to set up a boundary and initial conditions.
We are using velocity inlet and pressure outlet boundary condition for this problem. The following table shows the boundary settings for this case:
Material is defined as air and we are running a pressure-based steady-state solver for this problem.
Results:
Followings are the output graphs for this case:
The above graph shows the residuals for different parameters calculated as the function of the number iterations.
Above plot shows the mean temperature of the air at the outlet.
Above plot shows the standard deviation of the temperature at the outlet which indicates how efficient was the mixing in this case.
Post-processing view:
CASE - III:
For this case, we have a Short mixing tee with a hot inlet velocity of 3m/s and Momentum ratio of 4.
Now we define mesh under the mesh option and generate Mesh. We are using tetrahedral mesh for this case with an element size of 1.021e-2 m.
Following image shows the element quality of the mesh -
The above image shows the cut-plane section of the geometry which was created to get a better view and understanding of the meshing in the domain.
Next, we move to ANSYS Fluet to set up a boundary and initial conditions.
We are using velocity inlet and pressure outlet boundary condition for this problem. The following table shows the boundary settings for this case:
Material is defined as air and we are running a pressure-based steady-state solver for this problem.
Results:
Followings are the output graphs for this case:
The above graph shows the residuals for different parameters calculated as the function of the number iterations.
Above plot shows the mean temperature of the air at the outlet.
Above plot shows the standard deviation of the temperature at the outlet which indicates how efficient was the mixing in this case.
Post-processing view -
Following table represents the number of cells in each case:
Convergence:
For the CASE -I with Short mixing tee with a hot inlet velocity of 3m/s and Momentum ratio of 2 if we look at the residual plot for this case we see that the residual does not change after 550 iterations hence we can say that for this case the convergence is achieved in 550 iterations.
For CASE -II with Long mixing tee with a hot inlet velocity of 3m/s and Momentum ratio of 2 if we look at its residual plot we can see the similar phenomenon starts occurring at 350 iterations hence we can say that the convergence is achieved in 350 iterations.
For CASE -III with Short mixing with a hot inlet velocity of 3 m/s and Momentum ration of 4 similarly for this case the convergence is achieved in 500 iterations.
Conclusion:
- Among the three cases that we have simulated if we look at the mixing efficiency, the long duct is the one which has better mixing of the cold and hot air compared to the other two. This is evident by looking at the standard deviation graph for all the three cases the long pipe case has the lowest standard deviation which indicates that temperature variations at the outlet are less for the long pipe.
- If we have a desire of achieving a relatively lower temperature at the outlet then the setup with a short pipe with the increased velocity of cold air is better than the other two cases. Which is evident from the area-weighted average plot for the temperature at the outlet, where the short pipe with the increased velocity of cold air has the lowest outlet temperature.
- In conclusion, if we increase the length of the pipe without changing the other simulating parameters then we see a decrease in Standard deviation which is an indication of the increase in mixing efficiency.
- But, if we want a lower temperature output then increasing the length of the pipe will not create any significant change in the outlet temperature but increasing the velocity of the cold air at the inlet will be able to fulfill the desired result.