FINITE VOLUME

METHOD
1 DIMENSIONAL HEAT CONDUCTION
M. Hilman Gumelar Syafei
Let consider, an only diffusif transpost
 
process of a general property
Consider, one dimensional
steady state transport

*The approximation technique that are needed to obtain the so-called discretised equation will be
intorduced in terms of finite volume technique.

3 main basic steps in finite volume method: Grid Generation, discretisation,

solution of equation
#1 Grid
Generation
 #2 DISCRETISATION
Integration of the governing equation over a control volume to yield a
discretised equation at its nodal point P as follows:

 • The equation above stated that the diffusive flux of leaving the east face (e) minus the diffusive flux of entering at the
west face (w) is equal to the generation of .
• The values of and the diffusion coefficient are defined and calulcated an nodal point.

Using Central Differencing, that is Linear approximations to calculate interface values and the gradient. The value of the
diffusion coefficient at the interface can be approximated by:

The diffusive flux

terms
 #2 DISCRETISATION
It can be rearranged into :
The source Terms S, is defined by the linear form as follows:

(Eq. 1)

Then, the discretised conservation equation form will be:

Where,
 #3 SOLVING EQUATION
In order to get a better understanding, let’s see the example 4.1 in Versteeg & Malalasekera
Solution 4.1

Let’s consider nodes 2,3,4 first. From Eq. 1 (in slide 5),
we obtain that:

The source terms, Sp and Su are zero since there is No.

source term in these nodes
Solution 4.1 (CON’T)
Node 1 and 5 are the boundary nodes, therefore it
needs a special attention.

In node 1, we obtain that:

 
The fixed temperature boundary condition The dicretised equation for the boundary 1:
enters the calculation as a source term (Su
+ SPTp) with Su =( and Sp .

The value of is consider as zero, because

there is no west side at point 1. It is only
boundary nodes.
Solution 4.1 (CON’T)
With the same procedure in obtaining the discretised
equation in node 1, In node 5, we obtain that:

The dicretised equation for the boundary 5:

 Solution 4.1 (CON’T)
 
Hence, after we have determine the descretised equation at each node, and calculate that , we can get the value of the
equation’s contant at each node as follows:
 Solution 4.1 (CON’T)
 
Hence, after we have determine the descretised equation at each node, and calculate that , we can get the value of the
equation’s contant at each node as follows:
Solution 4.1 (CON’T)

using any method for solving the matrix, such as gauss elimination, or by using appropriate SOFTWARE such as Modelica we
can obtain:

Compare to the analytical solutions:

# EXAMPLES 4.2
Solution 4.2

The equation 4.28 can re arrarenged as:

The second term integral in the equation above is the

source terms of the equation

This equation above is valid for control

volumes at nodal 2,3,4
Solution 4.2 (CON’T)
Node 1 and 5 we apply the linear approximation for
tempertures betwwen a boundary point and the
adjacent nodal point.

In node 1, we obtain that:

Solution 4.2 (CON’T)
With the same procedure in obtaining the discretised
equation in node 1, In node 5, we obtain that:
 Solution 4.2 (CON’T)
Hence, after we have determine the descretised equation at each node, we can get the value of the equation’s contant at
each node as follows:
 Solution 4.2 (CON’T)
using any method for solving the matrix, such as gauss elimination, or by using appropriate SOFTWARE such as Modelica we
can obtain:

Compare to the analytical solutions:

# EXAMPLES 4.3
Solution 4.3

From equation 4.43 it is obtained that:

We divided the left side and the right side by A,

hence For nodal 2, 3, 4 it is obtained:

The second term integral due to the source term in

the equation is evaluated by assuming that the
integrand is locally constant within each control
volume
Solution 4.3 (CON’T)
Solution 4.3 (CON’T)
At node 1, the west control volume boundary is kept
at a specified temperature. Hence, it can be
developed as:
Solution 4.3 (CON’T)
At node 5, the flux across the east boundary is Zero
flux boundary condition. The coefficeis at boundary
node 5 are given by:
 Solution 4.3 (CON’T)
Hence, after we have determine the descretised equation at each node, we can get the value of the equation’s contant at
each node as follows:
 Solution 4.3 (CON’T)
using any method for solving the matrix, such as gauss elimination, or by using appropriate SOFTWARE such as Modelica we
can obtain:

Compare with the fine grid, where the number of node N is higher, let say 10,
Compare to the analytical solutions: it can be obtained that the result is better
 Solution 4.3 (CON’T)
using any method for solving the matrix, such as gauss elimination, or by using appropriate SOFTWARE such as Modelica we
can obtain:

Compare with the fine grid, where the number of node N is higher, let say 10,
Compare to the analytical solutions: it can be obtained that the result is better
Solution 4.3 (CON’T)