Parallel Tutorial
Parallel Tutorial
Abstract—In this exercise, heat conduction in parallel is modeled. The solid is represented in two dimensions. The
thickness of the two solids can be specified as an input. Coarse, medium, and fine meshes are available. The thermal
conductivity of the two solids can be specified within limits. The temperature or heat flux at the left wall, and the
temperature at the right wall can be specified. Temperature gradient and heat flux are reported. A plot of temperature
along the interface is available. Contours of temperature can be displayed.
1 Introduction
Thermal conduction is an important mode of heat transfer. Fourier’s law of heat conduction relates the heat
transfer rate to the temperature gradient, where thermal conductivity is represented by a proportionality
constant. The temperature gradient adjusts so that the heat flux across the right wall is constant.
2 Modeling Details
The solid is represented in two dimensions by a rectangle. The procedure for solving the problem is:
FlowLab creates the geometry and mesh, and exports the mesh to FLUENT. The boundary conditions and
flow properties are set through parameterized case files. FLUENT continues to solve the problem until the
convergence limit is satisfied or the specified number of iterations is achieved.
2.1 Geometry
The geometry is created from a set of four vertices. Edges are created over the vertices and are stitched
with faces to facilitate meshing. The geometry consists of four walls and an interface (Figure 2.1).
You have to specify the length of the two solids (L) and the thickness of two walls (H1 and H2).
1
2.2 Mesh
Coarse, medium, and fine mesh types are available. The discretization scheme used is based on the following
logic:
The numerical values for the two discretization parameters are given in Table 2.1.
Mesh Type $H $L
Coarse Mesh 2 10
Medium Mesh 3 20
Fine Mesh 4 30
The face is meshed using the map scheme after the edges are discretized into intervals (Figure 2.2).
The energy equation is solved in FLUENT for the solid domain. Since there is no fluid flow, mass and
momentum equations are not solved.
The default solid materials are Aluminum and Copper. The thermal conductivity of the solids can be
specified within limits. Other materials such as Steel, Wood and a User Defined solid can also be selected.
* Based on the thermal condition (heat flux or temperature specification at the left wall).
2
c Fluent Inc. [FlowLab 1.2], April 16, 2007
The following boundary conditions are assigned in FLUENT:
2.6 Solution
The mesh is exported to FLUENT along with the physical properties and the initial conditions specified.
The material properties and the initial conditions are read through the case file. Instructions for the solver
are provided through a journal file. Once the solution is converged or the number of specified iterations
is met, FLUENT exports data to a neutral file and to .xy plot files. GAMBIT reads the neutral file for
postprocessing activities.
The maximum temperature allowed by FLUENT (hence FlowLab as well) is 5000 K. If the exercise temper-
ature exceeds this limit, the temperature will be artificially restricted to 5000 K. Results obtained for cases
where the temperature exceeds this limit may not be correct.
Difficulty in obtaining convergence or poor accuracy may result if input values are used outside the upper
and lower limits suggested in the problem overview.
4 Exercise Results
4.1 Reports
• Heat flux
• Average temperature gradient
The average temperature gradient is calculated by taking the difference between the average temperature
at left and right walls and dividing it by the total length of the domain.
4.2 XY Plots
• Residuals
• Temperature distribution
Figure 4.1 presents temperature across the interface between Solid 1 and Solid 2.
c Fluent Inc. [FlowLab 1.2], April 16, 2007 3
Figure 4.1: Temperature Distribution along Two Solids
5 Verification of Results
1 1 1 kA AA kB AB
P = + = + (5-1)
R RA RB L L
∆T (T1 − T2 )
qx = P = P (5-2)
R R
4
c Fluent Inc. [FlowLab 1.2], April 16, 2007
The net heat flux is determined by dividing the net heat rate by the total area.
The following default settings are used to verify the CFD model:
L1 = 1m
AA = 0.1 m2
AB = 0.1 m2
kA = 202.4 W/(m − K)
kB = 387.6 W/(m − K)
T2 = 300 K
Table 5.1 compares the heat flux reported by FlowLab to theoretical values determined using the parameters
summarized above. The FlowLab results were generated using the fine mesh option and a convergence
criterion of 1e-12.
6 Sample Problem
1. Run the cases using the default exercise settings and all three mesh options. It will be observed that
the temperature gradient is slightly influenced by mesh density. Compare the results obtained using
FlowLab with theoretical results.
2. Set the left wall boundary condition to the wall temperature type and run additional cases using the
default material properties and all three mesh options. Compare the results obtained using FlowLab
with theoretical results.
7 Reference
[1] Incropera, F. P., and DeWitt, D. P., “Fundamentals of Heat and Mass Transfer”, 4th Ed., Ch. 3.
c Fluent Inc. [FlowLab 1.2], April 16, 2007 5