CONJUGATE HEAT TRANSFER SIMULATION

CONJUGATE HEAT TRANSFER SIMULATION

Introduction:

• The term conjugate heat transfer is used to describe processes which involve variations of temperature within solids and fluids, due to thermal interaction between the solids and fluids.
• The exchange of thermal energy between the two physical bodies is called study of Heat Transfer, the rate of transferred heat is directly proportional to the temperature difference between the bodies.
• CHT occurs only when heat transfer takes place between fluid & solid region. And also when we have multiple modes of heat transfer.

Objective:

• The objective of this project is to understand the concept of super-cycling and grid dependency by simulating the flow through the solid pipe and post processing the result.

Case Setup:

It is created using Converge CFD

Application: General flow

Materials: Solid & Gas simulation

Solid simulation : Aluminum

Species:

1. Gas: Air
2. Solid: Aluminum

Simulation Parameters:

Solver: Transient

Start time: 0s

End time: 0.5s

Boundary Conditions:

Calculation:

Inlet Reynolds number= 7000

Dynamic viscoscity= 18.45E-6 Pa s

Rho= 1.225

D= 0.03

Re = (rho*V*D)/(Dynamic viscosity)

Therefore, V= 3.7m/s

Regions & Initialization:

• Here the stream ID for the fluid region is set as zero and for solid region it is set as one.

Physical Models:

Turbulence modelling - K-Epsilon

Super cycling:

Supercycling stage intervals

1. 0.01
2. 0.02
3. 0.03

Base grid:

Case-1:

dx=dy=dz= 0.002

Super-cycling intervals= 0.01, 0.02 & 0.03

Case-2:

dx=dy=dz= 0.003

Super-cycling intervals= 0.01, 0.02 & 0.03

Case-3:

dx=dy=dz= 0.005

Super-cycling intervals= 0.01, 0.02 & 0.03

Simulation Results:

Surface with edge contour:

Case-1:

Case-2:

Case-3:

Explanation:

• By reducing the base grid size we will get the more refined mesh and so the results will be accurate.

Pressure Contours:

Case-1:

Case-2:

Explanation:

Case-3:

Explanation:

• The Pressure at inlet is set to zero normal gradient(NE), Therefore the pressure at inlet will depend on the initial condition.
• Initially the pressure is high in both inlet and outlet, After few cycles it converges and reaches the steady-state.

Temperature Contour:

Case-1:

Case-2:

Case-3:

Explanation:

• Here the wall(solid) is heated by using the heat flux boundary and as we are using super-cycling modelling the solid region performs the steady state simulation as a result it reaches the high temperature in few cycle and attains the steady state.
• And by visualizing the contour we can see that the fluid region also heated up slightly that is because we are using the negative heat flux, So the wall heat is transferred into the system.

Velocity Contour:

Case-1:

Case-2:

Case-3:

Explanation:

• The velocity value at inlet is calculated and applied, And the velocity value at outlet is set to zero normal gradient(NE), So the outlet velocity will depend on initial or inlet condition.
• Then the velocity converges and reaches the steady state condition, As we can see the velocity at outlet is high that is because the pressure at the outlet is decreased and thus velocity is increased.

Y+ Contour:

Case-1:

Case-2:

Case-3:

Explanation:

• The Y+ value is based on local cell fluid velocity and is a non-dimensional distance from wall to first mesh node. To use a wall function approach for a particular turbulence model with confidence, We need to ensure that our y+ values are within certain range.
• The Range of Y+ for RNG K-Epsilon turbulence modelling is 0.5-600. For the mesh we generated has the range of 9-12, So the wall function can be applied for all three mesh sizes.

Grid Dependency Test:

Explanation:

• The grid dependency test means that the analysis begins with a coarse mesh and it is refined to finer mesh for each run until the results does not change significantly.
• The above two plots are mean temperatures of fluid and solid regions of grid sizes(2mm,3mm & 5mm).
• By visualizing the plot we can understand that from refining the mesh from coarse to finer there is no significant changes, All three grids sizes converges at same time but by refining the mesh we get the more accurate results.

Effect of Supercycle Stage Interval:

Base Grid Size

dx=dy=dz= 0.005

Supercycling Stage Interval = 0.01

Supercycling Stage Interval = 0.02

Supercycling Stage Interval = 0.03

Explanation:

• By using the super-cycling, We can run steady-state solver in transient solver. We know that the solid takes much time to heat than the fluid, as the specific heat capacity is higher for solids. So if we use transient solver for solid region it will be more time consuming, And so the super-cycling is used.
• The super-cycling runs the solid region in the steady-state solver and captures the temperature change for the given intervals
• Here we are setting the supercyle stage interval to 0.01, 0.02 and 0.03s, And the time period value to save data at begining is same as supercycle interval value.
• By visualizing the plot we can understand that for every 0.01 or 0.02 or 0.03s the temperature is captured and stored. And it begins to store data at 0.01, 0.02, 0.03 respectively.

Simulation time for the different stage intervals:

1. 0.01 - 139.52s
2. 0.02 - 138.97s
3. 0.03 - 145.31s

• By increasing the stage intervals the simulation time is also increased. So in order to reduce the simulation time the lower stage interval values should be used.

Conclusion:

• The flow is simulated through a solid pipe and the results is post-processed, The grid dependency test is carried out and also the effect of supercycling is explained.