Simulation of Mixing efficiency in a T-joint using Ansys Fluent

Title:-Steady-state simulations to compare the mixing effectiveness in a T-shape geometry when hot inlet temperature is 36⁰C & the Cold inlet is at 19⁰C using Ansys Fluent.

Objective:-

1. To Set up Steady-State Simulations and to compare the mixing effectiveness.

2. To use the k-epsilon and k-omega SST model for the first case and to use a more suitable model for further cases and to give a reason for selecting a suitable model.

3. To vary the momentum ratio for each case.

• Momentum ratio = velocity at cold inlet / velocity at hot inlet.

Expected Results:-

1. Velocity and temperature contour plots on the cut planes along and across the pipe.

2. Velocity and temperature line plots along and across the length of the pipe.

3. Perform Mesh independent study for any one case.

4. Discuss the effect of length and momentum ratio on results.

Given Data:-

1. The hot inlet temperature is 36⁰C.

2. The cold inlet temperature is 19⁰C.

3. Hot inlet velocity of 3m/s.

Procedure:-

• Open Workbench and double click Fluid Flow Fluent in the toolbox/drag and drop into the project schematic window.
• Now in the project schematic window double click on the geometry to open SpaceClaim.
• Now open the geometry part from the location where u saved it.
• Now for the current project as we are working only on the fluid without the effect of walls we are gonna eliminate the walls
• Go to prepare and select volume extract and select all the three edges of the part and click ok.
• Now hide the solid part and to be fully aware that the wall part is not loaded for the further process we use suppress for physics option.
• Now close the SpaceClaim and Open the Meshing Module by double-clicking on it.
• Give Names to every part by using the face selection option and by pressing shortcut key N.
• Now edit the element size to the desired value and Generate Mesh.
• Keep the geometry in Y-direction now select the section option, click and drag on the geometry to cut the geometry for sectional view.
• Now go to Mesh Quality, select Mesh metric and select element quality to see the quality of the mesh.
• Double click setup and select Double Precision and Select Serial Processing Option or if u know the number of processors in the system select parallel processing option and enter the number and click ok.
• In the console, we get important information regarding Geometry/Simulation, etc. In Graphics we are gonna see our simulation.
• First, check the Mesh Quality by check mesh option and u can see the process in the console when it says check mesh done proceed further.
• Now go to setting up physics, set time to steady, solver type to the pressure-based solver and velocity formulation to absolute.
• As we are interested in solving for Temperature we use Energy Equation.
• As the flow can be turbulent we select the viscous model and we are using K-epsilon and K-omega SST models.
• Click on Materials and check that air is the working fluid.
• Go the cell zones, select volume go to edit and check if the material name is air, if we want to change the material name to water or any other we have to change it here or else we are not gonna see any difference.
• select Boundaries and assign the cold and hot inlet velocity and temperature values.
• Go to Solving option as we want to monitor some physical properties at the outlet like the temperature etc. Go to definitions, new, surface report and select Area weighted Average, Temperature to static temperature at the outlet and check on report file and report plot and as we want it to be done for every iteration we set the frequency to 1 and click ok.
• Similarly, do the same for the Standard deviation report.
• Now the Initialize the process by Hybrid Initialization so after 10 iterations the Initialization is done.
• Now we are ready to run the solver, run the calculation by entering the desired no of iterations and press calculate.
• Now we Post process the Results, double click on Results, select walls, Render and apply some transparency
• Select insert, location, plane and click ok.
• As we want the plane to be in the XY axis ie Z value is 0, click on apply and it will create a plane in the middle.
• Now select plane, color, model, variable, temperature and click on apply and we can see the temperature contour and similarly we can follow the same procedure for velocity contour.
• For Range go to user-defined and enter the min and max values and apply and we can also change the precision and units in the default legend view
• Finally, plot the charts of temperature and velocity by line plot.

comparing k-epsilon and k-omega SST models for hot inlet velocity of 3 m/s and momentum ratio of 2 for short tee geometry

Mixingtee (short) before volume extract:-

Mixingtee (short) after volume extract:-

Mixingtee (short) after using the suppress for physics option:-

The same geometry had been used to compare the viscous models(K-epsilon and K-omega).

Mesh:-

element size -0.0011(chosen this value because of the academic version as the no of cells/elements limited to 512000).

No of elements -458358

Mesh developed is the same for both types of models

Mesh Metrics:-

from the above fig, we can observe that most of the elements fall into 90% of mesh quality which in turn provides that the developed mesh is a good quality mesh.

Sectional view of the Mesh:-

Setup details:-

• Hot inlet velocity =3 m/s
• Cold inlet velocity =6 m/s
• Hot inlet Temperature =36° C.
• Cold inlet Temperature =19° C.

Residuals:-

So when we are calculating we discretize the PDE and the approximate values are called Residuals. As the number of Iterations increases the Residuals will be close to zero.

K-epsilon

Here from the above fig, we can see that for the k-epsilon model the solution got converged at 137 iterations.

K-omega

Here from the above fig, we can see that for the k-omega model the solution got converged at 274 iterations.

The standard deviation of outlet temperature:-

k-epsilon

Here from the above graph, the standard deviation of temperature for the k-epsilon model is approximately 1.8

k-omega

Here from the above graph, the standard deviation of temperature for the k-omega model is approximately 2.3

Area-Weighted average of outlet temperature:-

k-epsilon

Here from the above graph, we can observe that the Area-Weighted Average at the outlet temperature for the k-epsilon model is about 30.45

k-omega

Here from the above graph, we can observe that the Area-Weighted Average at the outlet temperature for the k-omega model is about 30.5

Post-processing:-

Temperature contour along the pipe:-

k-epsilon

k-omega

From the above contours, it is observed that the mixing of cold and hot fluids is more effective in case of k-epsilon model.

Velocity contour along the pipe:-

k-epsilon

k-omega

Temperature plot along the length of the pipe:-

k-epsilon

k-omega

Velocity plot along the length of the pipe:-

k-epsilon

k-omega

Choosing the suitable model:-

• k-epsilon model has less S.D which indicates that the values tend to be close to the mean of the set which clearly implies it is accurate for the given project(work).
• k-epsilon has taken fewer no of iterations to converge where as k-omega iterations are comparatively high which saves the computation time.
• Based on these two all the further cases will be solved by k-epsilon and the same mesh will be considered for better results.

Case 1(b):-

setup details:-

• Hot inlet velocity =3 m/s
• Cold inlet velocity =12 m/s
• Hot inlet Temperature =36° C.
• Cold inlet Temperature =19° C.

Residual:-

At 184 Iterations the solution is converged.

The standard deviation of outlet temperature:-

Standard deviation of temperature is about 1.2

Area Weighted of temperature:-

From the above graph Area -Weighted Average of temperature is 27.5

Post-processing:-

Temperature contour along the length of the pipe:-

Velocity contour along the length of the pipe:-

Temperature plot along the length of the pipe:-

Here the sudden drop of temperature line is at the cold inlet point on going further the temperature line becomes constant

Velocity plot along the length of the pipe:-

Case 2(a):-(hot inlet velocity of 3 m/s and momentum ratio of 2 for long tee geometry)

Mixingtee (long) part:-

Mesh:-

element size -0.0012(chosen this value because of the academic version as the no of cells/elements limited to 512000).

No of elements -506453

Mesh Metrics:-

from the above fig, we can observe that most of the elements fall into 90% of mesh quality which in turn provides that the developed mesh is a good quality mesh.

Sectional view of the Mesh:-

Setup details:-

• Hot inlet velocity =3 m/s
• Cold inlet velocity =6 m/s
• Hot inlet Temperature =36° C.
• Cold inlet Temperature =19° C.

Residuals:-

Here at 142 iterations, the solution got converged.

The standard deviation of outlet temperature:-

The standard deviation for this case is about 1.75

Area-Weighted average of outlet temperature:-

Here the Area-Weighted average of outlet temperature is 30.5

Post-processing:-

Temperature contour along the length of the pipe:-

Velocity contour along the pipe:-

Temperature plot along the length of the pipe:-

Velocity plot along the length of the pipe:-

case 2(b):-(hot inlet velocity of 3 m/s and momentum ratio of 4 for long tee geometry)

Geometry and Mesh is same as that of case 2(a)

Setup details:-

• Hot inlet velocity =3 m/s
• Cold inlet velocity =12 m/s
• Hot inlet Temperature =36° C.
• Cold inlet Temperature =19° C.

Residuals:-

Here the solution did not converge for 300 iterations but got a fair results

The standard deviation of outlet temperature:-

Here the standard deviation of outlet temperature is 0.9

Area-Weighted average of outlet temperature:-

Here the Area-Weighted average of outlet temperature is 27.5

Post-processing:-

Temperature contour along the length of the pipe:-

Velocity contour along the pipe:-

Temperature plot along the length of the pipe:-

Velocity plot along the length of the pipe:-

Conclusion:-

Table:-

• Increment in length showed a slight increment in temperature when compared with the short pipe and the standard deviation got reduced. So as we can see there are no huge difference between the two types of flow it is preferred to use the short pipe.
• As from the above table, we can observe that both types of flows have the same outlet temperature so it is best preferred to use the short pipe.

Mesh independent study:-

• Two types of mesh(i.e one is course mesh and other is fine mesh) have been considered to perform the mesh independent study.
• short tee geometry is considered for both the type

Mesh details:-

case 1

1. Element size =0.04
2. No of elements=12602

case 2

1. Element size =0.02
2. No of elements=106222

case 1

case 2

Element Quality

case 1

The mesh quality is good because most of the elements fall into 90% of mesh quality which in turn provides that the developed mesh is a good quality mesh ie in case of course mesh

case 2

In case of fine mesh also most of the elements fall into 90% of mesh quality which in turn provides that the developed mesh is a good quality mesh.

Setup details:-

• Hot inlet velocity =3 m/s
• Cold inlet velocity =6 m/s
• Hot inlet Temperature =36° C.
• Cold inlet Temperature

Results Comparison:-

Residuals:-

case 1

Here the solution got converged in 152 iterations.

case 2

Here the solution got converged in 176 iterations.

Standard deviation at outlet temperature:-

case 1

Here in case of the course mesh the standard deviation it almost 1.75

case 2

Here the standard deviation 1.8 which is almost same as that of the course mesh

Area-Weighted Average at outlet temperature:-

case 1

The area-weighted average at the outlet temperature is about 30.25

case 2

In case of fine mesh, it is almost 30.5

Temperature contours:-

case 1

As the meshing is not fine the results obtained are not that good

case 2

Here meshing is good compared to that of course which in turn gives good results

Velocity contour:-

case 1

Similarly the velocity contour also the contour is not good

case 2

It is somewhat good compared to that of course mesh as more number of elements are added while meshing.

Temperature plot along the length of the pipe:-

case 1

case 2

Velocity variation along the length of the pipe:-

case 1

case 2

Conclusion for Mesh Independent test:-

To get accurate results, it is always advisable to use fine mesh and the meshing can be changed at different parts based on the necessity.