Mixing efficiency

Aim: To compare mixing effectiveness when hot inlet temperature is 36⁰ C and cold inlet is at 19⁰ C.

Introduction: The mixing tee simulation explains the common phenomenon by which heat is transferred from one fluid to another usually seen in Air Conditioners. The velocity and temperature of the respective fluids are important in calculating the effectiveness of mixing. We are using common turbulence models k-ϵ and k-ω SST model for comparison of fluid at different temperatures and velocities.

Theory: Mixing of 2 fluids at different temperatures works on the principle of conservation of energy. The decrease in temperature of one fluid subsequently increases the temperature of the other one when coming in to direct contact and then reaching an equilibrium temperature. The concept is used in HVAC applications, cooling tower applications, heat exchangers etc.

Simulation

Case 1: Short TEE with momentum ratio of 2, hot inlet (x) at 36⁰ C and cold inlet is at 19⁰ C.

1.1 Using k-ϵ model

• Geometry in space claim

• Defining boundaries

Inlet_X, Inlet_Y, Wall and Outlet are defined in Ansys Meshing

• Meshing

Element Size : 0.0015 m , 211671 elements. The quality of the mesh generated is acceptable and is shown below.

• Solver Settings

2d, pressure based, k-ϵ Realizable turbulence model. The boundary conditions for each named selection are specified.

Energy On

Material: Air

• Results

	Short TEE MR 2	Short TEE MR 4	Long TEE MR 2	Long TEE MR 4
No. of Elements	211671	211671	288343	288343
Avg Outlet Temp	30.31	27.55	30.43
No. of Iterations to converge	152	Truncated due to oscillating residuals

For the Short TEE MR 2 case, the standard deviation of the outlet temperature is found to be 2.3378^o C. For the Short TEE MR 4 case, the standard deviation of the outlet temperature is found to be 0.915^o C.

Short TEE MR 2	k-ϵ Simulation	K – ω Simulation	Theoretical
No. of Elements	211671	211671
Avg Outlet Temp	30.31	30.476	30.33
No. of Iterations to converge	152	230

We can see that k-ω solution requires more time to converge and is lesser accurate than k- ϵ Simulation for this particular case. Hence K-ϵ is preferred.

Conclusions

1. k-ϵ Simulation is more accurate than K – ω SST Simulation for this particular case.
2. The mixing is more effective at MR 4. This can be verified from the lower standard deviation and temperature at outlet.
3. The increase in length does not facilitate greater mixing.
4. Mesh refinement study is done to verify the results obtained for different mesh sizes.