Week 1- Mixing Tee

AIM – The main goal of this challenge to create two versions of the mixing tee. One of them is longer than the other. Your job is to set up steady-state simulations to compare the mixing effectiveness when hot inlet temperature is 36⁰C & the Cold inlet is at 19⁰C.  Use the k-epsilon and k-omega SST model for the first case and based on your judgment use the more suitable model for further cases. Giving the reason for choosing a suitable model is compulsory. 

OBJECTIVE-

• To clean the geometry and mesh it in pre-processing.
• To solve the model with best solver in Ansys solver.
• To generate contour and line plots and analyse it in postprocessor.

THOERY –

Computational fluid dynamics is a branch of fluid mechanics, which uses mathematics, physics and computational software to visualize how fluid flows and affects its medium and its neighbours. In CFD mostly computers are used to predict the fluid flow phenomena based on some governing equations like conservation of momentum – Navier stokes equation, conservation of energy etc. In this study we are going to simulating effect of fluid using Ansys fluent software. Ansys fluent is a very important tool in CFD domain, which is allows you to perform all the three major steps of any simulation pre-processing, solving and post-processing. Ansys fluent facilitates for modelling flow, turbulence, heat transfer and chemical reactions. With Ansys we can accomplice more in less time and with less training.

Mixing tee is a device which uses specifically engineered internal geometry to mix two fluid streams into one steam, efficiently. In this study we are going to find effectiveness of mixing tee. Hot air and cold air are going to be mixed in mixing tee. They will have different velocities. We will use two different mixing tee distinguished with length and analyse the effect of length on mixing. In industrial process engineering, mixing is a unit operation that involves the manipulation of a heterogeneous physical system with the intent to make it more homogeneous. Mixing is performed to allow heat or mass transfer to occur between one or more streams, components, or phases. Modern industrial processing almost always involves some form of mixing. Mixing Tess is a type of mixing mechanism. Mixing Tee utilizes a specifically engineered internal geometry to efficiently mix two fluid streams into one combined stream. Mixing tees are widely used in the petrochemical, HVAC industry, etc, in which two fluids streams with different physical or chemical properties mix together.

Hot air flowing in the main pipe is mixed with cold air flowing through a tee. The standard deviation of temperature is computed to quantify the degree of mixing. The velocity and temperature fields are also computed. The effects of the mesh size, turbulence model on results were examined. For this, we have created two versions of the mixing tee. One of them is longer than the other.

For this, we have created two versions of the mixing tee. One of them is longer than the other.

It is set up as steady-state simulations to compare the mixing effectiveness when hot inlet temperature is `36^0 C` & the cold inlet is at `19^0C`.

      1. Case 1

                Short mixing tee with a hot inlet velocity of 3m/s.

                Momentum ratio of 2,4.

      2. Case 2

                 Long mixing tee with a hot inlet velocity of 3m/s.

                 Momentum ratio of 2,4.

The momentum ratio is defined as the ratio of velocity at the cold inlet to the velocity at the hot inlet.

PROCEDURE - 

There are three major steps to perform any type of simulation as follows pre-processor – it includes geometry clean up and generate best quality mesh. Solving includes applying boundary conditions, initialize and convergence. Last step is post processor which includes generate plots, contours, vectors to analyse results. In this study we are going to follow all three steps to find out effectiveness of mixing tee. The following steps to be followed –

1. Open new project, by dragging Fluid flow (Fluent) in Ansys workbench.

2. Now double click on geometry to import mixing tee model. It will redirect to Spaceclaim, Ansys’s geometry model manipulator.

3. We are to analyse the mixing of two fluid, so there is role of parts other than two pipes. To supress that we need to extract volume of tee. Go to prepare menu and click on volume extract. Switch to edge selection and select three outer edges and click on green check button.

4. Hide solid by unclicking it from structure menu to see volume, which we have extracted. Supress solid by clicking on option supress for physics.

5. Exit from Spaceclaim and click on mesh to open mesh menu.

6. Select outer face and press right click of mouse, it will open a menu. Click on option create named selection and write suitable name for face. Repeat for other faces.

7. Now click on mesh option and choose generate mesh. It will generate mesh according to some predefined criteria. Element size can be selected but it is going to affect simulation time. Section plane can be created to see inner portion of geometry.

8. Select the option element quality to see, how many elements are defying quality. Quality is measured on the scale of 0 to 1. In our graph most of the element are near to 1 so we can say it is good quality.

9. Exit mesh and open setup. First, Check mesh, it is throwing no error then only we can proceed further. We are going to do pressure based steady state simulation, so choose these options on task page.

10. Go to physics menu and select viscous model. It will open a menu, choose realizable k-epsilon model and close it.

11. Click on material and choose air as working fluid. Check working fluid by clicking on cell zone conditions option, if it is air then proceed.

12. Now go to boundaries option to set boundary conditions. Select inlet_x and click on option edit. It will open a menu, where all the parameter can be set whether it is thermal, radiation, momentum etc. Put the value of velocity as 3m/s and go to thermal parameter to set temperature as 36C, for hot fluid.

13. Repeat the same procedure for cold fluid coming, velocity –6 m/s, temperature – 19C.

14. We want to generate temperature report of outlet and we want standard deviation of temperature also. Go to solutions tab and select definition and select area weighted average. In the opened menu select temperature as field variable. Similarly select standard deviation in definitions menu.

15. Now click on initialization to calculate approximate values of velocity and pressure.

16. Change the number of iterations to 250 and click on calculate. If solution is not converged, then iterations can be increased.

17. Exit setup and double click on result to open CFD post. Check boxes to see walls and other parts. Click on walls and go to render set transparency of 0.4 to inside of wall. Colour can also be changed to white for clear visibility.

18. Now insert a XY plane in wall by clicking on plane in location option of insert.

19. Go to colour of plane and change it by variable choose option temperature. Similarly, velocity contours can be created by just changing variable as velocity.

20. Right click on plane which we created and select streamline. Go to appearance and select temperature to see plot line on plane. Similar procedure can be repeated for velocity plot line.

RESULTS –

Case 1 – In this short mixing tee is used, which was mixing two fluid with temperature 6C and 19C. Velocity of hot fluid was 3m/s and cold fluid was 6m/s. we used two different turbulent solvers for it first was realized k-epsilon and second was k-omega SST.

In this case k-epsilon is giving better results because solution was converged after few iterations, whereas after using k-omega model solution was not converged. K-omega is preferred in studying fluid behaviour near wall, but here it is not the case. So, k-epsilon model is recommended to study of fluid far from wall i.e. flow region, in further cases.

Plot and contours from k-epsilon model.

A mesh element metric that identifies the quality of the mesh is employed and it can be seen that the elements of the lowest quality employed are minimal. Most of the elements have quality in the range of 0.7 to 1. Hence the mesh is of acceptable quality.

                                                                       Fig. Residuals

                                                                        Fig.Area-Weighted Average of Temperature vs Iterations

                                                                    Fig. Standard Deviation of Temperature vs Iteration

                                                                      Fig. Temeprature Contour

                                                                         Fig. Velocity Contour

 Plot and contours from k-omega model

                                                                           Fig. Residuals

                                                                              Fig.Area-Weighted Average of Temperature vs Iterations

                                                                             Fig. Standard Deviation of Temperature vs Iteration

                                                                      Fig. Temeprature Contour

                                                                         Fig. Velocity Contour

Case 2 - In this case long mixing tee is used, which was mixing two fluid with temperature 6C and 19C. Velocity of hot fluid was 3m/s and cold fluid was 12m/s.

We used realized k-epsilon model and solution was converged after 156 iterations.

A mesh element metric that identifies the quality of the mesh is employed and it can be seen that the elements of the lowest quality employed are minimal. Most of the elements have quality in the range of 0.7 to 1. Hence the mesh is of acceptable quality.

                                                                           Fig. Residuals

                                                                            Fig.Area-Weighted Average of Temperature vs Iterations

                                                                             Fig. Standard Deviation of  Temperature vs Iteration

                                                                      Fig. Temeprature Contour

                                                                         Fig. Velocity Contour

COMPARISON TABLE –

Case	Cell Count	Area weighted Average	Number of iterations
Case 1	12812	30.2C	212
Case 2	15588	28C	158

MESH INDEPENDENT STUDY –

Mesh independent test allows mesh refinement without affecting results. After some tests we can be ensured by solution that mesh refinement won’t affect the result. The following graphs are showing the results after mesh refinement on short mixing tee, earlier element size is used as 0.0195 and now it got 0.001.

                                                                           Fig. Residuals

                                                                            Fig.Area-Weighted Average of Temperature vs Iterations

                                                                             Fig. Standard Deviation of  Temperature vs Iteration

CONCLUSION –

Ansys fluent is high performance CFD software recognized for accuracy, robustness etc. it is not confined to only aerospace and mechanical domain, it has various application like to simulation of calculation of air flow on air craft wing, combustion in furnace, bubble columns in oil refineries, combustion in IC engine etc. In this study we performed simulation on simple geometry of mixing tee and got insights of various feature of fluent. we can conclude by analysing graph that If momentum ratio is increased, outlet temperature will decrease and there is no effect of velocity even though length is increased.

1. As seen, the temperature at the outlet drops down significantly due to the high velocity of cold air from the inlet. This is possible due to turbulent mixing at the tee joint between hot air and cold air because of the high velocity of cold air, which is not possible in the case of low-velocity cold air.

2.The temperature contour at the outlet for short tee cases shows that some of the high-temperature flow sips through without properly mixing with the flow. Whereas, in the temperature contour of long tee cases, the high temperatures flow thoroughly mixes with the rest of the flow.

3. Therefore, we can recommend using the long tee and high-velocity inlet for application. This is because we have a low standard derivation in the case of long tee compared to the short tee for both momentum ratios.

4. K-epsilon model predicts well far from boundaries (wall) and K- omega model predicts well near the wall. Even though it depends on Y+. Therefore, the K-epsilon model is majority used for free flow away from the wall and as in this project, there is no dealing with the wall of the mixing tee for change in temperature along the length of it, so the K-epsilon model was chosen for further simulations.