COMPUTA

C ATIONAL FLUID DY
YNAMICS:: FVM: Applications
A s to Scalar Transport Problems

Lectu
ure 19
AP
PPLICA
ATION
NS OF FINITE
F VOLU
UME ME
ETHOD D TO
SCA
ALAR TRANS
T SPORT PROBL LEM

19.1 A
APPLICA
ATION OF
O FVM
M TO 1-D
D DIFFUSION PR
ROBLEM
M
Consideer the steadyy state diffuusion of a prroperty  inn a one-dim
mensional doomain definned in
Figure 19.1. The prrocess is gooverned by thet differenntial equatioon

d  d 
  S  0 (19.1)
dx  dx 

m. Boundaryy values off  at

where  is the diiffusion coeefficient and S is the source term
points A and B aree given.

Figure 199.1 Schemaatic of the 1--D diffusionn problem.

19.2 G
GRID GE
ENERAT
TION
The firrst step in the
t finite volume
v metthod is to divide the domain innto discrete control
volumees. The scheematic of grrid generatioon is shownn in Fig. 19
9.1. The usu ual notationn for the
CV is sshown in Figg. 19.2. A general
g nodal point is identified
i byy P and its neighbors
n inn a one-
dimensiional geomeetry, the noddes to the west
w and easst, are identiified by W anda E respeectively.

2 Notationss used for the

Fiigure 19.2 t 1-D con
ntrol volum
me

D K M Sin
Dr ngh, Indian
n Institute oof Technoloogy Roorkeee NPTE
EL 19.1
COMPUTATIONAL FLUID DYNAMICS: FVM: Applications to Scalar Transport Problems

19.3 DISCRETIZATION
The key step of the FVM is the integration of the governing equation(s) over a control
volume to yield a discretized equation at its nodal point P. For the control volume defined
above this gives

d  d   d   d 
 dx   dx dV   SdV   A dx 
V V e
  A   SdV  0
 dx  w
(19.2)

Here A is the cross-sectional area of the control volume face, dV is the volume and S is the
average value of source S over the control volume.

In order to derive useful forms of the discretized equations, the interface diffusion
coefficient  and the gradient d  / dx at east face (e) and west face (w) are required. To
calculate gradients (and hence fluxes) at the control volume faces, an approximate
distribution of properties between nodal points is used. Using simple averaging, one can
write

W   P
w 
2 (19.3)
P  E
e 
2
Using central difference method, the diffusive fluxes can be written as

 d   E  P 
 A    e Ae  
 dx e   xPE 
(19.4)
 d   P  W 
 A    w Aw  
 dx  w   xWP 

The source term S may also depend on the dependent variable. In such cases the finite volume
method approximates the source term by means of a linear form expressed as

SdV  Su  SPP (19.5)

where Su and SP are independent of ϕ. Use of the preceding approximations leads to

   P   P  W 
 e Ae  E    w Aw     Su  S PP   0
  xPE    xWP 

Rearrangement of the preceding equation yields

 e        
 Ae  w Aw  S P  P   w Aw  W   e Ae  E  Su (19.6)
  xPE  xWP    xWP    xPE 

The above equation can be written in a short form as

aPP  aW W  aEE  Su (19.7)

Dr K M Singh, Indian Institute of Technology Roorkee NPTEL 19.2

COMPUTATIONAL FLUID DYNAMICS: FVM: Applications to Scalar Transport Problems

where
aW aE aP
W e
A A aW  aP  SP
 xWP W  xPE e

19.4 SOLUTION OF EQUATIONS

Discretized equations for each of the control volume are formulated and suitable
modifications are incorporated for the boundary control volumes. The resulting set of linear
equations is solved using suitable linear solver (e.g. TDMA would be the best solver for the
present example).

Example 19.1
Consider the steady state heat conduction in a slab of width l = 0.5 m with heat generation.
The left end of the slab (x = 0) is maintained at T = 373 K. The right end of the slab (x = 0.5
m) is being heated by a heater for which the heat flux is 1 kW/m2. The heat generation in the
slab is temperature dependent and is given by Q = (1273 – T) W/m3. Thermal conductivity is
constant at k = 1 W/(m-K).Write down the governing equation and boundary conditions for
the problem. Use the finite difference method (central difference scheme) to obtain an
approximate numerical solution of the problem. For the first order derivative, use forward or
backward difference approximation of first order. Choose Δx = 0.1, and use the TDMA. (We
have chosen the same as the one solved previously in Example 15.1 using FDM to illustrate
comparison between FVM and FDM.)

Solution
Let us recall that governing equation for the steady state heat conduction with constant heat
generation is the slab is
d 2T
k 2 Q 0 (i)
dx
Given: Q = 1273 – T. Thus, Eq. (i) becomes
d 2T
k 2  (1273  T )  0 (ii)
dx

Left end of the slab is maintained at constant temperature; hence boundary condition at this
end is given by
T (0)  373 (iii)
At the right end, heat influx is specified. Thus, boundary condition at this end is
dT
k  1000 (iv)
dx
For discretization, let us use a face centered grid. Computational nodes are 1 (x=0), 2 (x=0.1),
3(x=0.2) , 4(x=0.3), 5(x=0.4) and 6 (x=0.5). Each interior finite volume is of width Δx = 0.1,
whereas finite volumes around boundary nodes 1 and 6 are of width 0.05.

Dr K M Singh, Indian Institute of Technology Roorkee NPTEL 19.3

COMPUTATIONAL FLUID DYNAMICS: FVM: Applications to Scalar Transport Problems

1 2 3 4 ∆x 5 6

W P E ∆x/2
w e

Let us assume that standard discrete equation can be written as

AWi Ti 1  APi Ti  AEi Ti 1  bi (v)
Temperature is specified at node 1, hence, we have
T1  373, i.e. AW1  0, AP1  1, AE1  0, b1  373 (vi)

For a finite volume around computational node P, finite volume integral equation yields
d  dT   dT   dT 
V dx  k dx dV  V (1273  T )dV   kA dx e   kA dx w  (1273  TP )V  0 (vii)

where we have used mid-point rule for evaluation of the second volume integral. For an
interior finite volume, using central difference scheme for the derivative, and assuming unit
area of CV surfaces and substituting k = 1, the preceding equation can be re-written as
TE  TP TP  TW
  (1273  TP )x  0 (viii)
x x
Using Δx = 0.1, the preceding equation simplifies to

Ti 1  2.01Ti  Ti 1  12.73, i  2,3,4,5 (ix)

where we have used indices i for P, (i-1) for W and (i+1) for node E. Hence, the coefficients
in the standard equation AWi Ti 1  APi Ti  AEi Ti 1  bi are:
AWi  1, APi  2.01, AEi  1, bi  12.73, i  2,3, 4,5 (x)

For the last FV around boundary node 6, terms in finite volume equation (vii) are
 dT   dT  T6  T5
 kA   1000,  kA  
 dx e  dx  w x
(xi)
 (1273  T )dV  (1273  T
V
m ) x / 2

Using linear interpolation, temperature at the centroid of this FV can be expressed as

(T  (T6  T5 ) / 2)
Tm  (T6  Tw ) / 2  6 (xii)
2
From Eq. (vii), (xi) and (xii), discrete equation for the boundary finite volume becomes
T T  (3T  T )  x
1000  6 5  1273  6 5  0 (xiii)
x  4  2
Using Δx = 0.1, the preceding equation simplifies to
1.00375T6  0.99875T5  106.365 (xiv)
Thus, AW6  0.99875, AP6  1.00375, AE6  0, b6  106.365 .
Therefore, the discrete system obtained from finite volume discretization is

Dr K M Singh, Indian Institute of Technology Roorkee NPTEL 19.4

COMPUTATIONAL FLUID DYNAMICS: FVM: Applications to Scalar Transport Problems

1 0 0 0 0 0  T1   373 
 1 2.01 1
 0 0 0  T2   12.73 
0 1 2.01 1 0 0  T3   12.73 
     
0 0 0 2.01 1 0  T4   12.73  (xv)
0 0 0 1 2.01 1  T5   12.73 
    
 0 0 0 0 0.99875 1.00375 T6  106.365
The preceding system is very similar to the discrete system (x) obtained using finite
difference discretization in Example 15.1 (only coefficients in last row differ). Numerical
calculations using TDMA are given in the following table:

AWi AEi 1 * AWi bi* bi*  AEi Ti 1 Tex

AWi AEi APi  APi  b  b  T 
i APi bi APi 1
i i
APi 1
i
APi

1 0 1.00 0 373 1 373 373.000 373.00

2 -1 2.01 -1 12.73 2.01 385.73 498.931 498.95
3 -1 2.01 -1 12.73 1.512487 204.6355 617.121 617.18
4 -1 2.01 -1 12.73 1.34883754 148.02734 728.753 728.85
5 -1 2.01 -1 12.73 1.26862087 122.47438 834.942 835.08
6 W6 1.00 0 100.36 0.21174243 196.54136 936.751 936.92

where W6=-0.99875. Comparison with results obtained in Example 15.1 using finite
difference method, we can clearly observe that the FV results using identical grid spacing are
more accurate. This is simply due to the fact that all interpolation or integration formulae used
in finite volume formulation were second order accurate. On the other hand, first order
backward order method was used in finite difference solution for incorporation of flux
boundary condition.

REFERENCES/FURTHER READING
Chung, T. J. (2010). Computational Fluid Dynamics. 2nd Ed., Cambridge University Press,
Cambridge, UK.
Ferziger, J. H. And Perić, M. (2003). Computational Methods for Fluid Dynamics. Springer.

Versteeg, H. K. and Malalasekera, W. M. G. (2007). Introduction to Computational Fluid

Dynamics: The Finite Volume Method. Second Edition (Indian Reprint) Pearson Education.

Dr K M Singh, Indian Institute of Technology Roorkee NPTEL 19.5