Conjugate Heat Transfer Simulation
AIM:
To simulate the flow of air and simultaneous heat transfer from the aluminum pipe.
OBJECTIVE:
1)To set up a case for conjugate heat transfer through the pipe
2)To simulate the case with different grid sizes and analyze grid dependence of the solution.
3)To use different super cycling stage intervals and compare results.
INTRODUCTION:
There are different modes of heat transfer predominant in different mediums. In solid the predominant mode of heat transfer is conduction and in fluid its convection. In CFD to capture the combined behavior of such heat transfers occurring together, multiple different equations have to be solved in a different medium. Such analysis is called conjugate heat transfer analysis.
Conjugate heat transfer is widely used in applications that include both fluid flow and heat transfer and combustion such as ic engine HVAC applications gas turbine applications.
In this simulation, we are simulating the flow of air through a pipe that is being heated from the outer surface.
GEOMETRY:
The geometry dimensions are as follows
Lx=0.04 m
Ly=0.0398294 m
Lz=0.2 m
Case Setup:
Materials:
Enable solid simulation and species.
Selecting Time Based Simulation With air as a predefined mixture.
1)Gas Simulation: Keeping default values
2)Solid simulation:
Include aluminum from predefined solids.
3)Global Transport Properties: Keeping default values
4)Species:
Include O2 and N2
3)Run parameters:
Using Transient density-based solver
4)Steady-state monitor
Turning off the steady-state monitor.
5)Simulation Time Parameter:
Running simulation for 0.5 sec and 1e-7 minimum time step.
6)Solver parameters:
Density-based solver
Region and Initialization:
Creating two regions
1)Solid Region: Pipe Body
Set stream id 1 and enable solid.
2)Fluid Region: Air flowing inside the pipe
Specify initial conditions of pressure and temperature.
Boundary Conditions:
1)Solid Outer Wall:
Region: Solid
BC: Heat flux of 10000 W/m2 with slip boundary condition.
2)Solid side thickness:
Region: Solid
BC: Wall
1)Inlet :
Region: Fluid
Inflow boundary condition with 7000 Reynolds number
Velocity at inlet: 3.71 m/s
2)Outlet:
Outflow Boundary condition with atm pressure of 101325 Pa if any backflow is there it is of air.
3) Interface:
Forward Side:Fluid region
Reverse Side:Solid Region
Turbulence Modeling:
Keeping it as RNG k-ε model.
Base Grid:
Case1)0.002 m for dx dy dz
Case2)0.003 m for dx dy dz
Case3)0.004 m for dx dy dz
Super cycling modeling:
For Grid Dependence test:
Storing data delay:0.05
Time length of each cycle:0.05
Output files:
Writing Output files for every 0.01 sec so that we can have 50 post files.
Results:
Case1:0.002 m
1.Mesh
2.Regions:
3.Velocity Contours:
4.Temperature Contours:
Case2:0.003 m
1.Mesh
2.Regions:
3.Velocity Contours:
4.Temperature Contours:
Case3:0.004 m
1.Mesh
2.Regions:
3.Velocity Contours:
4.Temperature Contours:
Grid Comparision:
Fluid Region:
Solid Region:
SuperCycling Comparision Grid Size 0.003m
Fluid Region:
Solid Region:
Conclusions:
1)The heat flows from the outer wall to the inner wall of the pipe and then air picks up that heat and air gets heated increasing temperature higher on outer side and graduating reducing towards the centre of the pipe.
2)Grid Dependence:
In solid region: as grid refines solver predicts lower mean temperature.
In fluid region: as grid refines solver predicts higher temperature.
3)The convergence of graphs of different supercycling stage intervals shows at the end of the simulation all 3 cases converge at single-temperature shows the super cycling affects only time taken by the simulation to run not accuracy of final results.