Computational Fluid Dynamics

Department of Mechanical Engineering

CoEP, Pune
 FVM for Diffusion Problems
𝝏
 𝝆. ∅ + 𝛁(𝛒. 𝑽. ∅) = ∇(Γ. ∇. ∅) + 𝑺
𝝏𝒕

- It discretize the integral form of the conservation equation directly

in the physical space.
- It discretize the integral form of the conservation equation directly
in the physical space.
- At the centroid of each of the control volumes the variable values
are calculated.
- Interpolation is used to express variable value at the CV surface in
terms of center values and suitable quadrature formulae are applied
to approximate the surface and volume integrals.
- An algebraic equation for each of the control volumes can be
obtained , in which a number of the neighboring nodal values
appear.
 FVM for Diffusion Problems
𝝏
 𝝆. ∅ + 𝛁(𝛒. 𝑽. ∅) = ∇(Γ. ∇. ∅) + 𝑺
𝝏𝒕

Advantages:
- FVM works with control volume, not at grid intersection points.
Therefore, has a capacity to accommodate any type of grid.
- Unstructured grid can be employed. Unstructured grid allows many
options for definition of the shape and location and the control volumes.
(unstructured mesh offers greater flexibility for handling complex
geometries.)
- FVM requires no transformation of the equation in terms of body fitted
coordinate system as is required in the finite difference method.
- FVM uses the control volume integration.
- In control volume, the bounding surface areas of the element are directly
linked to the discretization of the 1st and 2nd order derivatives for Ø ( the
generic flow field variable ).
 FVM for One Dimensional Steady State Diffusion
𝝏
 𝝆. ∅ + 𝛁(𝛒. 𝑽. ∅) = ∇(Γ. ∇. ∅) + 𝑺
𝝏𝒕

• Consider the steady state diffusion of a property φ in a one-dimensional

domain.

• The process is governed by:

𝒅 𝒅∅
∇ Γ. ∇. ∅ + 𝑺 = Γ. +𝑺=𝟎
𝒅𝒙 𝒅𝒙

 is the diffusion coefficient and S is the source term.

Boundary values of Ø at points A and B are known.
 FVM for One Dimensional Steady State Diffusion
𝒅 𝒅∅
∇ Γ. ∇. ∅ + 𝑺 = Γ. +𝑺=𝟎
𝒅𝒙 𝒅𝒙
Step1: Grid Generation
• Divide the domain into discrete control volumes.
• Let us place several nodal points in the space between A and B.
• Each node is surrounded by a control volume or cell.

• It is common practice to set up control volumes near the edge of the

domain in such a way that the physical boundaries coincide with the
control volume boundaries.
• It is essential to establish a common notation.
 FVM for One Dimensional Steady State Diffusion
Step2: Discretization
• Integrate the governing equation over the CV to yield a discretized
equation at its nodal point P.

𝒅 𝒅∅
∇ Γ. ∇. ∅ + 𝑺 = න Γ. . 𝒅𝑽 + න 𝑺. 𝒅𝑽 = 𝟎
∆𝑽 𝒅𝒙 𝒅𝒙 ∆𝑽

𝒅 𝒅∅ 𝒅∅ 𝒅∅
‫ 𝒙𝒅 𝑽∆׬‬Γ. 𝒅𝒙 . 𝒅𝑽 + ‫𝑺 𝑽∆׬‬. 𝒅𝑽 = Γ. 𝑨. 𝒅𝒙 − Γ. 𝑨. 𝒅𝒙 ഥ ∆𝑽 = 𝟎 -----(1)
+ 𝑆.
𝒆 𝒘

• A is the cross-sectional area of the control volume face, ΔV is the volume

ഥ is the average value of source S over the CV.
and 𝑺.
 FVM for One Dimensional Steady State Diffusion
Step2: Discretization
𝒅 𝒅∅ 𝒅∅ 𝒅∅
න Γ. . 𝒅𝑽 + න 𝑺. 𝒅𝑽 = Γ. 𝑨. − Γ. 𝑨. ഥ ∆𝑽 = 𝟎
+ 𝑆.
∆𝑽 𝒅𝒙 𝒅𝒙 ∆𝑽 𝒅𝒙 𝒆
𝒅𝒙 𝒘

• To derive useful forms of

the discretized equations,
the interface diffusion
coefficient Γ and the
gradient dØ/dx at east (e)
and west (w) are required.

• An approximate distribution of property Ø, is between nodal points

is required to calculate gradients at the control volume faces.
• Linear approximations is used for calculating interface values and
the gradients i.e. central differencing
 FVM for One Dimensional Steady State Diffusion
Step2: Discretization
𝒅 𝒅∅ 𝒅∅ 𝒅∅
න Γ. . 𝒅𝑽 + න 𝑺. 𝒅𝑽 = Γ. 𝑨. − Γ. 𝑨. ഥ ∆𝑽 = 𝟎 −− −(𝟏)
+ 𝑆.
∆𝑽 𝒅𝒙 𝒅𝒙 ∆𝑽 𝒅𝒙 𝒆
𝒅𝒙 𝒘
Γ𝑊 + Γ𝑃 Γ𝑃 + Γ𝐸
Γ𝑤 = Γ𝑒 =
2 2

• The term dØ/dx is:

𝒅∅ ∅𝐸 −∅𝑃
Γ. 𝑨. 𝒅𝒙 = Γ𝑒 . 𝐴𝑒
𝒆 𝛿𝑥𝑃𝐸
𝒅∅ ∅𝑃 −∅𝑊
Γ. 𝑨. 𝒅𝒙 = Γ𝑤 . 𝐴𝑤
𝒘 𝛿𝑥𝑊𝑃

• The source term S may be a function of the dependent variable.

• In such cases the FVM finite volume method approximates the source
term by means of a linear form: ഥ ∆𝑽 = 𝑺𝑼 + 𝑺𝑷 . ∅𝑷 −− −(𝟐)
𝑆.

• Equation 1 gives ∅𝑬 − ∅𝑷 ∅𝑃 − ∅𝑊
Γ𝒆 . 𝑨𝒆 − Γ𝑤 . 𝐴𝑤 + (𝑺𝑼 +𝑺𝑷 . ∅𝑷 ) = 𝟎
𝛿𝑥𝑃𝐸 𝛿𝑥𝑊𝑃
 FVM for One Dimensional Steady State Diffusion
Step2: Discretization
∅𝑬 − ∅𝑷 ∅𝑃 − ∅𝑊
Γ𝒆 . 𝑨𝒆 − Γ𝑤 . 𝐴𝑤 + (𝑺𝑼 +𝑺𝑷 . ∅𝑷 ) = 𝟎
𝛿𝑥𝑃𝐸 𝛿𝑥𝑊𝑃

Γ𝒆 Γ𝒘 Γ𝒘 Γ𝑬
. 𝑨 + 𝑨 − 𝑺𝑷 ∅𝑷 = 𝑨 ∅ + 𝑨 ∅ + 𝑺𝑼
𝛿𝑥𝑃𝐸 𝒆 𝛿𝑥𝑊𝑃 𝒘 𝛿𝑥𝑊𝑃 𝒘 𝑾 𝛿𝑥𝑃𝐸 𝑬 𝑬

• The above equation can be written as:

𝒂𝑬 + 𝒂𝑾 − 𝑺𝑷 ∅𝑷 = 𝒂𝑾 ∅𝑾 + 𝒂𝑬 ∅𝑬 + 𝑺𝑼

𝒂𝑷 ∅𝑷 = 𝒂𝑾 ∅𝑾 + 𝒂𝑬 ∅𝑬 + 𝑺𝑼 −− −(𝟑)
𝒂𝑷 = 𝒂𝑾 + 𝒂𝑬 − 𝑺𝑷
• The values of Su and Sp can be obtained from the equation 2
ഥ ∆𝑽 = 𝑺𝑼 + 𝑺𝑷 . ∅𝑷
𝑆.

• The equation 3 is discretized form of G.D.E.

 FVM for One Dimensional Steady State Diffusion
𝒂𝑷 ∅𝑷 = 𝒂𝑾 ∅𝑾 + 𝒂𝑬 ∅𝑬 + 𝑺𝑼 −− −(𝟑)

𝒂𝑷 = 𝒂𝑾 + 𝒂𝑬 − 𝑺𝑷
Step 3: Solution of Equations

• The discretized equations need to obtained at each of the nodal

points.
• A special treatment is necessary for nodal points which are located at
the boundary to incorporate the boundary condition imposed.
• The system of linear algebraic equations is then solved to obtain the
distribution of the property φ at nodal points.
 FVM for One Dimensional Steady State Diffusion
Consider the problem of source-free heat conduction in an insulated rod
whose ends are maintained at constant temperatures of 100°C and 500°C
respectively. The one-dimensional problem sketched in following Figure is
governed by

Calculate the steady state temperature distribution in the rod. Thermal

conductivity k equals 1000 W/m.K, cross-sectional area A is 10 × 10−3 m2.

Solution:
Let us divide the length of the rod into five equal control volumes,
x=0.1m
 FVM for One Dimensional Steady State Diffusion

Solution:

W P E
w e

Let us divide the length of the rod into five equal control volumes,
x=0.1m

For the CV at node 2

 FVM for One Dimensional Steady State Diffusion

Solution:

W P E
w e

For the CV at node 2

𝒂𝑬 + 𝒂𝑾 𝑻𝑷 = 𝒂𝑾 𝑻𝑾 + 𝒂𝑬 𝑻𝑬
𝒂𝑷 = 𝒂𝑬 + 𝒂𝑾
1000 × 10 × 10−3
𝑎𝐸 = 𝑎𝑊 = = 100 𝑎𝑃 = 𝑎𝐸 + 𝑎𝑊 = 100 + 100 = 200
0.1
𝒂𝑬 + 𝒂𝑾 − 𝑺𝑷 𝑻𝑷 = 𝒂𝑾 ∅𝑾 + 𝒂𝑬 ∅𝑬 + 𝑺𝑼
𝑺𝑷 = 𝑺𝑼 = 𝟎
 FVM for One Dimensional Steady State Diffusion

Solution:

W P E
w e

1000 × 10 × 10−3
𝑎𝐸 = 𝑎𝑊 = = 100 𝑎𝑃 = 𝑎𝐸 + 𝑎𝑊 = 100 + 100 = 200
0.1
For the CV at nodes 2,3,4 𝒂𝑬 + 𝒂𝑾 𝑻𝑷 = 𝒂𝑾 𝑻𝑾 + 𝒂𝑬 𝑻𝑬
Nodes 𝑎𝐸 𝑎𝑊 𝑎𝑃 𝑆𝑈 𝑆𝑃
2 100 100 200 0 0
3 100 100 200 0 0
4 100 100 200 0 0
𝟐𝟎𝟎 𝑻𝟐 = 𝟏𝟎𝟎 𝑻𝟏 + 𝟏𝟎𝟎 𝑻𝟑 −−− −(𝟐)
Algebraic Equations at
𝟐𝟎𝟎 𝑻𝟑 = 𝟏𝟎𝟎 𝑻𝟐 + 𝟏𝟎𝟎 𝑻𝟒 −−− −(𝟑)
Node at nodes 2,3,4 𝟐𝟎𝟎 𝑻𝟒 = 𝟏𝟎𝟎 𝑻𝟑 + 𝟏𝟎𝟎 𝑻𝟓 −−− −(𝟑)
 FVM for One Dimensional Steady State Diffusion

Solution:

P 𝒂𝑬 + 𝒂𝑾 𝑻𝑷 = 𝒂𝑾 𝑻𝑾 + 𝒂𝑬 𝑻𝑬
W E
w e

1000 × 10 × 10−3
𝑎𝐸 = 𝑎𝑊 = = 100 𝑎𝑃 = 𝑎𝐸 + 𝑎𝑊 = 100 + 100 = 200
0.1
Algebraic Equations at 𝟐𝟎𝟎 𝑻𝟐 = 𝟏𝟎𝟎 𝑻𝟏 + 𝟏𝟎𝟎 𝑻𝟑
Node at nodes 2,3,4 𝟐𝟎𝟎 𝑻𝟑 = 𝟏𝟎𝟎 𝑻𝟐 + 𝟏𝟎𝟎 𝑻𝟒
𝟐𝟎𝟎 𝑻𝟒 = 𝟏𝟎𝟎 𝑻𝟑 + 𝟏𝟎𝟎 𝑻𝟓

−𝟏𝟎𝟎𝑻𝟏 + 𝟐𝟎𝟎𝑻𝟐 − 𝟏𝟎𝟎𝑻𝟑 + 𝟎 + 𝟎 = 𝟎 −−− −(𝟐)

𝟎 − 𝟏𝟎𝟎𝑻𝟐 + 𝟐𝟎𝟎𝑻𝟑 − 𝟏𝟎𝟎𝑻𝟒 + 𝟎 = 𝟎 −−− −(𝟑)

𝟎 + 𝟎 − 𝟏𝟎𝟎𝑻𝟑 + 𝟐𝟎𝟎𝑻𝟒 − 𝟏𝟎𝟎𝑻𝟓 + 𝟎 = 𝟎 −−− −(𝟒)
 FVM for One Dimensional Steady State Diffusion

Solution:

w P E
e

For the CV at node 1: Special treatment is required for Boundary Node 1

𝒅𝑻 𝒅𝑻
𝑘. 𝑨. − 𝒌. 𝑨. =𝟎
𝒅𝒙 𝒆
𝒅𝒙 𝒘
𝑻𝑬 − 𝑻𝑷 𝑻 𝑷 − 𝑻𝑨
𝑘𝒆 . 𝑨𝒆 − 𝑘𝑨 . 𝑨𝑨 =𝟎
𝜹𝒙 𝜹𝒙ൗ
𝟐
𝑘 𝒆 . 𝐴𝒆 𝑘𝒆 . 𝐴𝒆 2𝑘𝑨 . 𝐴𝑨 𝟐𝑘𝑨 . 𝐴𝑨
𝑻𝑬 − 𝑻𝑷 − 𝑻𝑷 + 𝑻𝑨 =𝟎
𝜹𝒙 𝜹𝒙 𝜹𝒙 𝜹𝒙

𝒂𝑬 + 𝒂𝑾 − 𝑺𝑷 𝑻𝑷 = 𝒂𝑾 𝑻𝑾 + 𝒂𝑬 𝑻𝑬 + 𝑺𝑼
 FVM for One Dimensional Steady State Diffusion

Solution:

W P E
e

For the CV at node 1: Special treatment is required for Boundary Node 1

𝑘 𝒆 . 𝐴𝒆 𝑘 𝒆 . 𝐴𝒆 2𝑘𝑨 . 𝐴𝑨 𝟐𝑘𝑨 . 𝐴𝑨
𝑻 − 𝑻 − 𝑻𝑷 + 𝑻𝑨 =𝟎
𝜹𝒙 𝑬 𝜹𝒙 𝑷 𝜹𝒙 𝜹𝒙
𝑘 𝒆 . 𝐴𝒆 2𝑘𝑨 . 𝐴𝑨 𝑘 𝒆 . 𝐴𝒆 𝟐𝑘𝑨 . 𝐴𝑨 𝑎𝑊 = 0 𝑠𝑖𝑛𝑐𝑒 𝐵𝑜𝑢𝑛𝑑𝑎𝑟𝑦
𝑻 +𝟎+ 𝑻𝑷 = 𝟎+ 𝑻 + 𝑻𝑨
𝜹𝒙 𝑷 𝜹𝒙 𝜹𝒙 𝑬 𝜹𝒙
𝒂𝑬 + 𝒂𝑾 − 𝑺𝑷 𝑻𝑷 = 𝒂𝑾 𝑻𝑾 + 𝒂𝑬 𝑻𝑬 + 𝑺𝑼

𝑘 𝒆 . 𝐴𝒆 −2𝑘𝑨 . 𝐴𝑨 𝟐𝑘𝑨 . 𝐴𝑨
= 𝒂𝑬 , 𝒂𝑾 = 𝟎, = 𝑺𝑷 , 𝑻𝑨 = 𝑺𝒖
𝜹𝒙 𝜹𝒙 𝜹𝒙
 FVM for One Dimensional Steady State Diffusion

Solution:

W P E
e

1000 × 10 × 10−3
𝑎𝐸 = = 100
0.1
For the CV at node 1: Special treatment is required for Boundary Node 1
𝒂𝑬 + 𝒂𝑾 − 𝑺𝑷 𝑻𝑷 = 𝒂𝑾 𝑻𝑾 + 𝒂𝑬 𝑻𝑬 + 𝑺𝑼 −− −(𝑨)
𝑘 𝒆 . 𝐴𝒆 −2𝑘𝑨 . 𝐴𝑨 −2 × 1000 × 10 × 10−3
= 𝒂𝑬 , 𝒂𝑾 = 𝟎, = 𝑺𝑷 𝑺𝑷 = = −200
𝜹𝒙 𝜹𝒙 𝟎. 𝟏
𝟐 × 1000 × 10 × 10−3
𝑺𝒖 = 𝟏𝟎𝟎 = 𝟐𝟎𝟎𝟎𝟎
Equation A gives 𝟎. 𝟏
𝟏𝟎𝟎 + 𝟎 − (−𝟐𝟎𝟎) 𝑻𝟏 = 𝟎 𝑻𝑾 + 𝟏𝟎𝟎 𝑻𝟐 + 𝟐𝟎𝟎𝟎𝟎
𝟑𝟎𝟎) 𝑻𝟏 − 𝟏𝟎𝟎 𝑻𝟐 + 𝟎 + 𝟎 + 𝟎 = 𝟐𝟎𝟎𝟎𝟎 −− −(𝟏)
 FVM for One Dimensional Steady State Diffusion

Solution:

W P E W P E

e w

1000 × 10 × 10−3
𝑎𝐸 = 𝑎𝑊 = = 100
0.1
For the CV at node 5: Special treatment is required for Boundary Node 5
𝒅𝑻 𝒅𝑻
𝑘. 𝑨. − 𝒌. 𝑨. =𝟎
𝒅𝒙 𝑩
𝒅𝒙 𝒘
𝑻𝑩 − 𝑻𝑷 𝑻𝑷 − 𝑻𝑾
𝑘𝑩 . 𝑨𝑩 −𝑘𝒘 . 𝑨𝒘 =𝟎
𝜹𝒙ൗ 𝜹𝒙
𝟐
2𝑘𝑩 . 𝐴𝑩 2𝑘𝑩 . 𝐴𝑩 𝑘𝑾 . 𝐴𝑾 𝑘𝑾 . 𝐴𝑾
𝑻𝑩 − 𝑻𝑷 − 𝑻𝑷 + 𝑻𝑾 =𝟎
𝜹𝒙 𝜹𝒙 𝜹𝒙 𝜹𝒙

𝒂𝑬 + 𝒂𝑾 − 𝑺𝑷 𝑻𝑷 = 𝒂𝑾 𝑻𝑾 + 𝒂𝑬 𝑻𝑬 + 𝑺𝑼
 FVM for One Dimensional Steady State Diffusion

Solution:

W P E W P E

e w

1000 × 10 × 10−3
𝑎𝐸 = 𝑎𝑊 = = 100
0.1
For the CV at node 5: Special treatment is required for Boundary Node 5
2𝑘𝑩 . 𝐴𝑩 2𝑘𝑩 . 𝐴𝑩 𝑘𝑾 . 𝐴𝑾 𝑘𝑾 . 𝐴𝑾
𝑻𝑩 − 𝑻𝑷 − 𝑻𝑷 + 𝑻𝑾 =𝟎
𝜹𝒙 𝜹𝒙 𝜹𝒙 𝜹𝒙 𝑎𝐸 = 0 𝑠𝑖𝑛𝑐𝑒 𝐵𝑜𝑢𝑛𝑑𝑎𝑟𝑦
𝑘𝑾 . 𝐴𝑾 2𝑘𝑩 . 𝐴𝑩 𝑘𝑾 . 𝐴𝑾 2𝑘𝑩 . 𝐴𝑩
𝟎+ 𝑻𝑷 +𝟎 + 𝑻𝑷 = 𝑻𝑾 + 𝟎 𝑻𝑬 + 𝑻𝑩
𝜹𝒙 𝜹𝒙 𝜹𝒙 𝜹𝒙

𝒂𝑬 + 𝒂𝑾 − 𝑺𝑷 𝑻𝑷 = 𝒂𝑾 𝑻𝑾 + 𝒂𝑬 𝑻𝑬 + 𝑺𝑼 −− −(𝑩)

𝑘𝑾 . 𝐴𝑾 1000 × 10 × 10−3 2𝑘𝑩 . 𝐴𝑩 𝟐 × 1000 × 10 × 10−3

𝒂𝑾 = = = 𝟏𝟎𝟎 𝑺𝑷 = =− = −𝟐𝟎𝟎
𝜹𝒙 𝟎. 𝟏 𝜹𝒙 𝟎. 𝟏
 FVM for One Dimensional Steady State Diffusion

Solution:

W P E W P E

e w

For the CV at node 5: Special treatment is required for Boundary Node 5

𝒂𝑬 + 𝒂𝑾 − 𝑺𝑷 𝑻𝑷 = 𝒂𝑾 𝑻𝑾 + 𝒂𝑬 𝑻𝑬 + 𝑺𝑼 −− −(𝑩)
𝑘𝑾 . 𝐴𝑾 1000 × 10 × 10−3
𝒂𝑾 = = = 𝟏𝟎𝟎
𝜹𝒙 𝟎. 𝟏 2𝑘𝑩 . 𝐴𝑩 𝟐 × 1000 × 10 × 10−3
𝑺𝑷 = =− = −𝟐𝟎𝟎
𝜹𝒙 𝟎. 𝟏
2𝑘𝑩 . 𝐴𝑩 𝟐 × 1000 × 10 × 10−3
𝑺𝒖 = 𝑻𝑩 = × 𝟓𝟎𝟎 = 𝟏𝟎𝟎𝟎𝟎𝟎
𝜹𝒙 𝟎. 𝟏
Equation B gives
𝟎 + 𝟏𝟎𝟎 − (−𝟐𝟎𝟎) 𝑻𝟓 = 𝟏𝟎𝟎 𝑻𝟒 + 𝟎 𝑻𝑬 + 𝟏𝟎𝟎𝟎𝟎𝟎

𝟎 + 𝟎 + 𝟎 − 𝟏𝟎𝟎𝑻𝟒 + 𝟑𝟎𝟎𝑻𝟓 =𝟏𝟎𝟎𝟎𝟎𝟎 −−−− −(𝟓)

 FVM for One Dimensional Steady State Diffusion

Solution:

Algebraic Equations at Nodes 1 to 5 are:

𝟑𝟎𝟎) 𝑻𝟏 − 𝟏𝟎𝟎 𝑻𝟐 + 𝟎 + 𝟎 + 𝟎 = 𝟐𝟎𝟎𝟎𝟎 −− −(𝟏)

−𝟏𝟎𝟎𝑻𝟏 + 𝟐𝟎𝟎𝑻𝟐 − 𝟏𝟎𝟎𝑻𝟑 + 𝟎 + 𝟎 = 𝟎 −−− −(𝟐)

𝟎 − 𝟏𝟎𝟎𝑻𝟐 + 𝟐𝟎𝟎𝑻𝟑 − 𝟏𝟎𝟎𝑻𝟒 + 𝟎 = 𝟎 −−− −(𝟑)

𝟎 + 𝟎 − 𝟏𝟎𝟎𝑻𝟑 + 𝟐𝟎𝟎𝑻𝟒 − 𝟏𝟎𝟎𝑻𝟓 + 𝟎 = 𝟎 −−− −(𝟒)

𝟎 + 𝟎 + 𝟎 − 𝟏𝟎𝟎𝑻𝟒 + 𝟑𝟎𝟎𝑻𝟓 =𝟏𝟎𝟎𝟎𝟎𝟎 −−−− −(𝟓)

 FVM for One Dimensional Steady State Diffusion

Solution:

Algebraic Equations at Nodes 1 to 5 can be arranged in matrix form as:

𝟑𝟎𝟎) 𝑻𝟏 − 𝟏𝟎𝟎 𝑻𝟐 + 𝟎 + 𝟎 + 𝟎 = 𝟐𝟎𝟎𝟎𝟎 −− −(𝟏) −𝟏𝟎𝟎𝑻𝟏 + 𝟐𝟎𝟎𝑻𝟐 − 𝟏𝟎𝟎𝑻𝟑 + 𝟎 + 𝟎 = 𝟎 −−− −(𝟐)

𝟎 − 𝟏𝟎𝟎𝑻𝟐 + 𝟐𝟎𝟎𝑻𝟑 − 𝟏𝟎𝟎𝑻𝟒 + 𝟎 = 𝟎 −−− −(𝟑) 𝟎 + 𝟎 − 𝟏𝟎𝟎𝑻𝟑 + 𝟐𝟎𝟎𝑻𝟒 − 𝟏𝟎𝟎𝑻𝟓 + 𝟎 = 𝟎 −−− −(𝟒

𝟎 + 𝟎 + 𝟎 − 𝟏𝟎𝟎𝑻𝟒 + 𝟑𝟎𝟎𝑻𝟓 =𝟏𝟎𝟎𝟎𝟎𝟎 −−−− −(𝟓)

300 −100 0 0 0 𝑻𝟏 20000
−100 200 −100 0 0 𝑻𝟐 0
0 −100 200 −100 0 𝑻𝟑 = 0
0 0 −100 200 −100 𝑻𝟒 0
0 0 0 −100 300 𝑻𝟓 100000

Numerical Solution:
T1 = 140oC, T2 = 220oC, , T3 = 300oC, T4 = 380oC, T5 = 460oC
 FVM for One Dimensional Steady State Diffusion

Solution:

300 −100 0 0 0 𝑻𝟏 20000

−100 200 −100 0 0 𝑻𝟐 0
0 −100 200 −100 0 𝑻𝟑 = 0
0 0 −100 200 −100 𝑻𝟒 0
0 0 0 −100 300 𝑻𝟓 100000
𝒂𝑬 + 𝒂𝑾 − 𝑺𝑷 𝑻𝑷 = 𝒂𝑾 𝑻𝑾 + 𝒂𝑬 𝑻𝑬 + 𝑺𝑼
Node aW aE SU SP aP
1 0 100 200TA -200 300
2 100 100 0 0 200
3 100 100 0 0 200
4 100 100 0 0 200
5 100 0 200TB -200 300
 FVM for One Dimensional Steady State Diffusion

Solution:

Analytical Solution
T1 = 140oC, T2 = 220oC, , T3 = 300oC, , T4 = 380oC, T5 = 460oC

Numerical Solution:
T1 = 140oC, T2 = 220oC, T3 = 300oC, T4 = 380oC, T5 = 460oC
 FVM for One Dimensional Steady State Diffusion

Solution:

Analytical Solution
T1 = 140oC, T2 = 220oC, , T3 = 300oC, , T4 = 380oC, T5 = 460oC

Numerical Solution:
T1 = 140oC, T2 = 220oC, T3 = 300oC, T4 = 380oC, T5 = 460oC

End A Node 1 Node 2 Node 3 Node 4 Node 5 End B

x 0 0.05 0.15 0.25 0.35 0.45 0.5
T exact 100 139.9 219.9 299.9 379.9 459.9 500
T FVM 100 140.0 220.0 300.0 380.0 460.0 500
Error % - - - - - -
 FVM for One Dimensional Steady State Diffusion
The cooling of a circular fin by means of convective heat transfer along its length.
Convection gives rise to a temperature-dependent heat loss or sink term in the
governing equation. Extended surface is a cylindrical fin with uniform cross sectional
area A. The base is at a temperature of 100°C (TB) and the end is insulated. The fin
is exposed to an ambient temperature of 20°C. One dimensional heat transfer in this
situation is governed by

where h is the convective heat transfer coefficient, P the perimeter, k the

thermal conductivity of the material and T∞ the ambient temperature.
Calculate the temperature distribution along the fin and compare the results
with the analytical solution given by

where n2 = hP/(kA), L is the length of the fin and x the distance along the
fin. Data: L = 1 m, hP/(kA) = 25/m2 (note that k.A is constant).
 FVM for One Dimensional Steady State Diffusion

Solution:
Let us divide the length of the rod into five equal control volumes, x=0.1m

𝑑 𝑑𝑇
න 𝑑𝑉 − න 𝑛2 𝑇 − 𝑇∞ 𝑑𝑉 = 0
𝐶𝑉 𝑑𝑥 𝑑𝑥 𝐶𝑉
 FVM for One Dimensional Steady State Diffusion
Solution:
W P E
w e

𝑑 𝑑𝑇
න 𝑑𝑉 − න 𝑛2 𝑇 − 𝑇∞ 𝑑𝑉 = 0
𝐶𝑉 𝑑𝑥 𝑑𝑥 𝐶𝑉

𝑑𝑇 𝑑𝑇
𝐴 − 𝐴 − 𝑛2 𝑇𝑃 − 𝑇∞ . 𝐴. 𝛿𝑥 = 0
𝑑𝑥 𝑒
𝑑𝑥 𝑤
𝑇𝐸 − 𝑇𝑃 𝑇𝑃 − 𝑇𝑊
− − 𝑛2 𝑇𝑃 − 𝑇∞ . 𝛿𝑥 = 0
𝛿𝑥 𝛿𝑥
Rearranging the terms 1 1 1 1
+ + 𝑛2 . 𝛿𝑥 𝑇𝑃 = 𝑇𝑊 + 𝑇𝐸 + 𝑛2 𝑇∞ 𝛿𝑥
𝛿𝑥 𝛿𝑥 𝛿𝑥 𝛿𝑥
𝒂𝑬 + 𝒂𝑾 − 𝑺𝑷 𝑻𝑷 = 𝒂𝑾 𝑻𝑾 + 𝒂𝑬 𝑻𝑬 + 𝑺𝑼
1
𝒂𝑬 = =5 𝒂𝑷 𝑻𝑷 = 𝒂𝑾 𝑻𝑾 + 𝒂𝑬 𝑻𝑬 + 𝑺𝑼
𝛿𝑥
1 For interior Nodes
𝒂𝑾 = =5 𝑺𝑷 = −𝑛2 . 𝛿𝑥 = −5 𝑺𝑼 = 𝑛2 𝑇∞ 𝛿𝑥 = 15 2,4, and 4
𝛿𝑥
 FVM for One Dimensional Steady State Diffusion
Solution:
W P E
w e

𝒂𝑬 + 𝒂𝑾 − 𝑺𝑷 𝑻𝑷 = 𝒂𝑾 𝑻𝑾 + 𝒂𝑬 𝑻𝑬 + 𝑺𝑼
For interior Nodes 2,4, and 4
𝒂𝑷 𝑻𝑷 = 𝒂𝑾 𝑻𝑾 + 𝒂𝑬 𝑻𝑬 + 𝑺𝑼
1
𝒂𝑬 = =5 𝑺𝑼 = 𝑛2 𝑇∞ 𝛿𝑥 = 100
𝛿𝑥
1
𝒂𝑾 = =5 𝑺𝑷 = −𝑛2 . 𝛿𝑥 = −5
𝛿𝑥
𝟏𝟓 𝑻𝑷 = 𝟓 𝑻𝑾 + 𝟓 𝑻𝑬 + 𝟏𝟎𝟎

𝟏𝟓 𝑻𝟐 = 𝟓 𝑻𝟏 + 𝟓 𝑻𝟑 + 𝟏𝟎𝟎 −− −(𝟐)

𝟏𝟓 𝑻𝟑 = 𝟓 𝑻𝟐 + 𝟓 𝑻𝟒 + 𝟏𝟎𝟎 −− −(𝟑)

𝟏𝟓 𝑻𝟒 = 𝟓 𝑻𝟑 + 𝟓 𝑻𝟓 + 𝟏𝟎𝟎 −− −(𝟒)
 FVM for One Dimensional Steady State Diffusion
Solution:
P E
e
w

𝑇𝐸 − 𝑇𝑃 𝑇𝑃 − 𝑇𝑊
− − 𝑛2 𝑇𝑃 − 𝑇∞ . 𝛿𝑥 = 0
𝛿𝑥 𝛿𝑥
For Node 1
𝑇𝐸 − 𝑇𝑃 𝑇𝑃 − 𝑇𝐵
− − 𝑛2 𝑇𝑃 − 𝑇∞ . 𝛿𝑥 = 0
𝛿𝑥 𝛿𝑥/2
1 2 1 2
+ + 𝑛2 . 𝛿𝑥 𝑇𝑃 = 0 𝑇𝑊 + 𝑇𝐸 + 𝑇𝐵 + 𝑛2 𝑇∞ 𝛿𝑥
𝛿𝑥 𝛿𝑥 𝛿𝑥 𝛿𝑥
𝒂𝑬 + 𝒂𝑾 − 𝑺𝑷 𝑻𝑷 = 𝒂𝑾 𝑻𝑾 + 𝒂𝑬 𝑻𝑬 + 𝑺𝑼
𝒂𝑷 𝑻𝑷 = 𝒂𝑾 𝑻𝑾 + 𝒂𝑬 𝑻𝑬 + 𝑺𝑼 For Node 1
1
𝒂𝑬 = =5
𝛿𝑥 2
𝑺 = 𝑛 2
𝑇 𝛿𝑥 + 𝑇 = 1100 𝟐𝟎 𝑻𝑷 = 𝟎 𝑻𝑾 + 𝟓 𝑻𝑬 + 𝟏𝟏𝟎𝟎
𝒂𝑾 = 0 𝑼 ∞ 𝐵
𝛿𝑥
2 𝟐𝟎 𝑻𝟏 = 𝟎 𝑻𝑾 + 𝟓 𝑻𝟐 + 𝟏𝟏𝟎𝟎
𝑺𝑷 = − − 𝑛2 . 𝛿𝑥 = −𝟏𝟓
𝛿𝑥
 FVM for One Dimensional Steady State Diffusion
Solution:
P E P e
e w
w

𝑇𝐸 − 𝑇𝑃 𝑇𝑃 − 𝑇𝑊
− − 𝑛2 𝑇𝑃 − 𝑇∞ . 𝛿𝑥 = 0
𝛿𝑥 𝛿𝑥
For Node 5
𝑇𝑃 − 𝑇𝑊
0 − − 𝑛2 𝑇𝑃 − 𝑇∞ . 𝛿𝑥 = 0
𝛿𝑥
1 1
0+ + 𝑛2 . 𝛿𝑥 𝑇𝑃 = 𝑇𝑊 + 0 𝑇𝐸 + 𝑛2 𝑇∞ 𝛿𝑥
𝛿𝑥 𝛿𝑥
𝟎 + 𝒂𝑾 − 𝑺𝑷 𝑻𝑷 = 𝒂𝑾 𝑻𝑾 + 𝒂𝑬 𝑻𝑬 + 𝑺𝑼
𝒂𝑷 𝑻𝑷 = 𝒂𝑾 𝑻𝑾 + 𝒂𝑬 𝑻𝑬 + 𝑺𝑼 For Node 5
𝒂𝑬 = 0
1 𝑺𝑼 = 𝑛2 𝑇∞ 𝛿𝑥 = 100 𝟏𝟎 𝑻𝑷 = 𝟓 𝑻𝑾 + 𝟎 𝑻𝑬 + 𝟏𝟎𝟎
𝒂𝑾 = =5
𝛿𝑥
𝑺𝑷 = −𝑛2 . 𝛿𝑥 = −𝟓 𝟏𝟎 𝑻𝟓 = 𝟓 𝑻𝟒 + 𝟎 𝑻𝑬 + 𝟏𝟎𝟎
 FVM for One Dimensional Steady State Diffusion
P E P e
e w

Algebraic Equations at Nodes 1 to 5 can be arranged in matrix form as:

𝟏𝟓 𝑻𝟐 = 𝟓 𝑻𝟏 + 𝟓 𝑻𝟑 + 𝟏𝟎𝟎 −− −(𝟐) 𝟐𝟎 𝑻𝟏 = 𝟎 𝑻𝑾 + 𝟓 𝑻𝟐 + 𝟏𝟏𝟎𝟎 − −(𝟏)
𝟏𝟓 𝑻𝟑 = 𝟓 𝑻𝟐 + 𝟓 𝑻𝟒 + 𝟏𝟎𝟎 −− −(𝟑) 𝟏𝟎 𝑻𝟓 = 𝟓 𝑻𝟒 + 𝟎 𝑻𝑬 + 𝟏𝟎𝟎 −− −(𝟓)
𝟏𝟓 𝑻𝟒 = 𝟓 𝑻𝟑 + 𝟓 𝑻𝟓 + 𝟏𝟎𝟎 −− −(𝟒)
20 −5 0 0 0 𝑻𝟏 1100
−5 15 −5 0 0 𝑻𝟐 100
0 −5 15 −5 0 𝑻𝟑 = 100
0 0 −5 15 −5 𝑻𝟒 100
0 0 0 −5 10 𝑻𝟓 100
Numerical Solution:
T1 = 64.22oC, T2 = 36.91oC, , T3 =26.50oC, T4 = 22.60oC, T5 = 21.30oC
 FVM for One Dimensional Steady State Diffusion
P E P e
e w

w
20 −5 0 0 0 𝑻𝟏 1100
−5 15 −5 0 0 𝑻𝟐 100
0 −5 15 −5 0 𝑻𝟑 = 100
0 0 −5 15 −5 𝑻𝟒 100
0 0 0 −5 10 𝑻𝟓 100
𝒂𝑬 + 𝒂𝑾 − 𝑺𝑷 𝑻𝑷 = 𝒂𝑾 𝑻𝑾 + 𝒂𝑬 𝑻𝑬 + 𝑺𝑼

Node aW aE SU SP aP
1 0 5 100+10TB -15 20
2 5 5 100 -5 15
3 5 5 100 -5 15
4 5 5 100 -5 15
5 5 0 100 -5 10
 FVM for One Dimensional Steady State Diffusion

Solution:
Comparison between FVM and Exact Solution:

End A Node 1 Node 2 Node 3 Node 4 Node 5 End B

x 0 0.1 0.3 0.5 0.7 0.9 1
T exact 100 68.52 37.86 26.61 22.53 21.21 0
T FVM 100 64.22 36.91 26.50 22.60 21.30 0
Error % - 6.27% 2.51% 0.41% -0.31% -0.42% -
 FVM for Two-Dimensional Steady State Diffusion
• The methodology used in deriving discretized equations in the 1D case
can be extended to 2D.
 ∅  ∅
Γ. + Γ. +𝑺=𝟎
𝒙 𝒙 𝒚 𝒚

• Additional north (N) and

south (S) neighbours need to
be consider along with the
east (E) and west (W)
neighbours for node P
•Similar notation will be used
for designation of faces and
cell dimensions.
 FVM for Two-Dimensional Steady State Diffusion
 ∅  ∅
Γ. + Γ. +𝑺=𝟎
𝒙 𝒙 𝒚 𝒚
• The equation is integrated over the CV we obtain,
 ∅  ∅
න Γ. . 𝒅𝑽 + න Γ. . 𝒅𝑽 + න 𝑺. 𝒅𝑽 = 𝟎
∆𝑽 𝒙 𝒙 ∆𝑽 𝒚 𝒚 ∆𝑽

 ∅  ∅
න Γ. . 𝒅𝒙. 𝒅𝒚 + න Γ. . 𝒅𝒙. 𝒅𝒚 + න 𝑺. 𝒅𝑽 = 𝟎 = 𝟎
∆𝑽 𝒙 𝒙 ∆𝑽 𝒚 𝒚 ∆𝑽

• Now, Ae=Aw=y and As = An=x, above equation can be modified as:

∅ ∅ ∅ ∅
Γ𝒆 . 𝐴𝒆 − Γ𝒘 . 𝐴𝒘 + Γ𝒏 . 𝐴𝒏 − Γ𝒔 . 𝐴𝒔 +ഥ
𝑺. ∆𝑽 = 𝟎 −− −(𝟏)
𝒙 𝒆
𝒙 𝒘
𝒚 𝒏
𝒚 𝒔

∅𝑬 − ∅𝑷 ∅𝑵 − ∅𝑷
𝐹𝑙𝑢𝑥 𝑎𝑐𝑟𝑜𝑠𝑠 𝑡ℎ𝑒 𝑒𝑎𝑠𝑡 𝑓𝑎𝑐𝑒 = Γ𝒆 . 𝐴𝒆 𝐹𝑙𝑢𝑥 𝑎𝑐𝑟𝑜𝑠𝑠 𝑡ℎ𝑒 𝑛𝑜𝑟𝑡ℎ 𝑓𝑎𝑐𝑒 = Γ𝒏 . 𝐴𝒏
∆𝒙 ∆𝒚 𝒏
𝒆
∅𝑷 − ∅𝑾 ∅𝑷 − ∅𝑺
𝐹𝑙𝑢𝑥 𝑎𝑐𝑟𝑜𝑠𝑠 𝑡ℎ𝑒 𝑤𝑒𝑠𝑡 𝑓𝑎𝑐𝑒 = Γ𝒘 . 𝐴𝑤 𝐹𝑙𝑢𝑥 𝑎𝑐𝑟𝑜𝑠𝑠 𝑡ℎ𝑒 𝑠𝑜𝑢𝑡ℎ 𝑓𝑎𝑐𝑒 = Γ𝒔 . 𝐴𝒔
∆𝒙 𝒘 ∆𝒚 𝒔
 FVM for Two-Dimensional Steady State Diffusion
∅ ∅ ∅ ∅
Γ𝒆 . 𝐴𝒆 − Γ𝒘 . 𝐴𝒘 + Γ𝒏 . 𝐴𝒏 − Γ𝒔 . 𝐴𝒔 +ഥ
𝑺. ∆𝑽 = 𝟎 −− −(𝟏)
𝒙 𝒆
𝒙 𝒘
𝒚 𝒏
𝒚 𝒔

∅𝑬 − ∅𝑷 ∅𝑵 − ∅𝑷
𝐹𝑙𝑢𝑥 𝑎𝑐𝑟𝑜𝑠𝑠 𝑡ℎ𝑒 𝑒𝑎𝑠𝑡 𝑓𝑎𝑐𝑒 = Γ𝒆 . 𝐴𝒆 𝐹𝑙𝑢𝑥 𝑎𝑐𝑟𝑜𝑠𝑠 𝑡ℎ𝑒 𝑒𝑎𝑠𝑡 𝑓𝑎𝑐𝑒 = Γ𝒏 . 𝐴𝒏
∆𝒙 ∆𝒚 𝒏
𝒆

∅𝑷 − ∅𝑾 ∅𝑷 − ∅𝑺
𝐹𝑙𝑢𝑥 𝑎𝑐𝑟𝑜𝑠𝑠 𝑡ℎ𝑒 𝑤𝑒𝑠𝑡 𝑓𝑎𝑐𝑒 = Γ𝒘 . 𝐴𝑤 𝐹𝑙𝑢𝑥 𝑎𝑐𝑟𝑜𝑠𝑠 𝑡ℎ𝑒 𝑒𝑎𝑠𝑡 𝑓𝑎𝑐𝑒 = Γ𝒔 . 𝐴𝒔
∆𝒙 ∆𝒚 𝒔
𝒘

• Equation 1 can be modified as:

∅𝑬 − ∅𝑷 ∅𝑷 − ∅𝑾 ∅𝑵 − ∅𝑷 ∅𝑷 − ∅𝑺
Γ𝒆 . 𝐴𝒆 − Γ𝒘 . 𝐴𝑤 + Γ𝒏 . 𝐴𝒏 − Γ𝒔 . 𝐴𝒔 +ഥ
𝑺. ∆𝑽 = 𝟎
∆𝒙 𝒆
∆𝒙 𝒘
∆𝒚 𝒏
∆𝒚 𝒔

ഥ ∆𝑽 = 𝑺𝑼 + 𝑺𝑷 . ∅𝑷 −− −(𝟐)
𝑆.
• The above equation can be rearranged as:
Γ𝒘 . 𝐴𝑤 Γ𝒆 . 𝐴𝒆 + Γ𝒏 . 𝐴𝒏 Γ𝒔 . 𝐴𝒔 Γ𝒘 . 𝐴𝑤 Γ𝒆 . 𝐴𝑒 Γ𝒔 . 𝐴𝒔 Γ𝒘 . 𝐴𝑤
+ + + − 𝑆𝑃 . ∅𝑷 = ∅𝑾 + ∅𝑬 + ∅𝑵 + ∅𝑺 + 𝑺𝑼
∆𝒙 ∆𝒙 ∆𝒚 ∆𝒚 ∆𝒙 ∆𝒙 ∆𝒚 ∆𝒚

𝒂𝑷 . ∅𝑷 = 𝒂𝑾 ∅𝑾 + 𝒂𝑬 ∅𝑬 + 𝒂𝑵 ∅𝑵 + 𝒂𝑺 ∅𝑺 + 𝑺𝑼

𝒂𝑷 = 𝒂𝑾 + 𝒂𝑬 + 𝒂𝑵 + 𝒂𝑺 − 𝑺𝑷
 FVM for Two-Dimensional Steady State Diffusion
A two-dimensional plate of thickness 1 cm is shown. The thermal
conductivity of the plate material is k = 1000 W/m.K. The west boundary
receives a steady heat flux of 500 kW/m2 and the south and east
boundaries are insulated. If the north boundary is maintained at a
temperature of 100°C, use a uniform grid with Δx = Δy = 0.1 m to
calculate the steady state temperature distribution at all nodes.
• Governing Equation:
 𝑻  𝑻
𝑘. + 𝒌. +𝑺=𝟎
𝒙 𝒙 𝒚 𝒚
Governing Equation in its discretized form as:
𝒂𝑷 . ∅𝑷 = 𝒂𝑾 ∅𝑾 + 𝒂𝑬 ∅𝑬 + 𝒂𝑵 ∅𝑵 + 𝒂𝑺 ∅𝑺 + 𝑺𝑼
𝒌. 𝐴𝑤 𝒌. 𝐴𝒆 𝒌. 𝐴𝒏 𝒌. 𝐴𝒔
𝒂𝑾 = 𝒂𝑬 = 𝒂𝑵 = 𝒂𝑺 =
∆𝒙 ∆𝒙 ∆𝒚 ∆𝒚
𝒌. 𝑨 𝟏𝟎𝟎𝟎 × 𝟎. 𝟏 × 𝟎. 𝟎𝟏
𝒂𝑾 = 𝒂𝑬 = 𝒂𝑵 = 𝒂𝑺 = = = 𝟏𝟎
∆𝒙 𝟎. 𝟏

For Interior Nodes 6 and 7:

 FVM for Two-Dimensional Steady State Diffusion
A two-dimensional plate of thickness 1 cm is shown. The thermal
conductivity of the plate material is k = 1000 W/m.K. The west boundary
receives a steady heat flux of 500 kW/m2 and the south and east
boundaries are insulated. If the north boundary is maintained at a
temperature of 100°C, use a uniform grid with Δx = Δy = 0.1 m to
calculate the steady state temperature distribution at all nodes.
𝒂𝑷 . ∅𝑷 = 𝒂𝑾 ∅𝑾 + 𝒂𝑬 ∅𝑬 + 𝒂𝑵 ∅𝑵 + 𝒂𝑺 ∅𝑺 + 𝑺𝑼
𝒌. 𝑨 𝟏𝟎𝟎𝟎 × 𝟎. 𝟏 × 𝟎. 𝟎𝟏
𝒂𝑾 = 𝒂𝑬 = 𝒂𝑵 = 𝒂𝑺 = = = 𝟏𝟎
∆𝒙 𝟎. 𝟏
For Interior Nodes 6 and 7:
𝒂𝑷 . ∅𝑷 = 𝒂𝑾 ∅𝑾 + 𝒂𝑬 ∅𝑬 + 𝒂𝑵 ∅𝑵 + 𝒂𝑺 ∅𝑺 + 𝑺𝑼

𝟒𝟎 . ∅𝑷 = 𝟏𝟎 ∅𝑾 + 𝟏𝟎 ∅𝑬 + 𝟏𝟎 ∅𝑵 + 𝟏𝟎 ∅𝑺 + 0

𝟒𝟎 . 𝑻𝟔 = 𝟏𝟎 𝑻𝟐 + 𝟏𝟎 𝑻𝟏𝟎 + 𝟏𝟎 𝑻𝟓 + 𝟏𝟎 𝑻𝟕 − −(𝟔)

𝟒𝟎 . 𝑻𝟕 = 𝟏𝟎 𝑻𝟑 + 𝟏𝟎 𝑻𝟏𝟏 + 𝟏𝟎 𝑻𝟔 + 𝟏𝟎 𝑻𝟖 − −(𝟕)
 FVM for Two-Dimensional Steady State Diffusion
A two-dimensional plate of thickness 1 cm is shown. The thermal conductivity of the
plate material is k = 1000 W/m.K. The west boundary receives a steady heat flux of 500
kW/m2 and the south and east boundaries are insulated. If the north boundary is
maintained at a temperature of 100°C, use a uniform grid with Δx = Δy = 0.1 m to
calculate the steady state temperature distribution at all nodes.

𝒂𝑷 . ∅𝑷 = 𝒂𝑾 ∅𝑾 + 𝒂𝑬 ∅𝑬 + 𝒂𝑵 ∅𝑵 + 𝒂𝑺 ∅𝑺 + 𝑺𝑼
𝒌. 𝑨 𝟏𝟎𝟎𝟎 × 𝟎. 𝟏 × 𝟎. 𝟎𝟏
𝒂𝑾 = 𝒂𝑬 = 𝒂𝑵 = 𝒂𝑺 = = = 𝟏𝟎
∆𝒙 𝟎. 𝟏
For Boundary Node 1:
𝒂𝑷 . ∅𝑷 = 𝒂𝑾 ∅𝑾 + 𝒂𝑬 ∅𝑬 + 𝒂𝑵 ∅𝑵 + 𝒂𝑺 ∅𝑺 + 𝑺𝑼
Constant
𝒂𝑷 = 𝒂𝑾 + 𝒂𝑬 + 𝒂𝑵 + 𝒂𝑺 − 𝑺𝑷
Wall Temp.
Contribution to source term SU is due to
constant heat flux BC (bW) plus Constant
𝟐𝒌. 𝑨
Wall temp. BC. (bs) 𝑺𝑼 = 𝒃𝑾 + 𝒃𝑺 𝑺𝑷 =
𝜹𝒙
=𝟎
𝟐𝒌𝑨. 𝑻
𝑺𝑼 = 𝒒" × 𝑨 + = (𝟓𝟎𝟎 × 𝟏𝟎𝟑 × 𝟎. 𝟏 × 𝟎. 𝟎𝟏) + 𝟎 = 𝟓𝟎𝟎
∆𝒚
And aW=0 & 𝟐𝟎 . 𝑻𝑷 = 𝟎 𝑻𝑾 + 𝟏𝟎 𝑻𝑬 + 𝟏𝟎 𝑻𝑵 + 𝟎 𝑻𝑺 + 𝟓𝟎𝟎
aS=0,SP=0 𝟐𝟎 . 𝑻𝟏 = 𝟎 𝑻𝑾 + 𝟏𝟎 𝑻𝟓 + 𝟏𝟎 𝑻𝟐 + 𝟎 𝑻𝑺 + 𝟓𝟎𝟎 − −(𝟏)
 FVM for Two-Dimensional Steady State Diffusion
A two-dimensional plate of thickness 1 cm is shown. The thermal
conductivity of the plate material is k = 1000 W/m.K. The west boundary
receives a steady heat flux of 500 kW/m2 and the south and east
boundaries are insulated. If the north boundary is maintained at a
temperature of 100°C, use a uniform grid with Δx = Δy = 0.1 m to
calculate the steady state temperature distribution at all nodes.
𝒂𝑷 . ∅𝑷 = 𝒂𝑾 ∅𝑾 + 𝒂𝑬 ∅𝑬 + 𝒂𝑵 ∅𝑵 + 𝒂𝑺 ∅𝑺 + 𝑺𝑼
𝒌. 𝑨 𝟏𝟎𝟎𝟎 × 𝟎. 𝟏 × 𝟎. 𝟎𝟏
𝒂𝑾 = 𝒂𝑬 = 𝒂𝑵 = 𝒂𝑺 = = = 𝟏𝟎
∆𝒙 𝟎. 𝟏
For Boundary Node 4:
𝒂𝑷 . ∅𝑷 = 𝒂𝑾 ∅𝑾 + 𝒂𝑬 ∅𝑬 + 𝒂𝑵 ∅𝑵 + 𝒂𝑺 ∅𝑺 + 𝑺𝑼
Constant
𝒂𝑷 = 𝒂𝑾 + 𝒂𝑬 + 𝒂𝑵 + 𝒂𝑺 − 𝑺𝑷
Wall Temp.
Contribution to source term SU is due to
constant heat flux BC (bW) plus Constant
Wall temp. BC. (bs) 𝑺𝑼 = 𝒃𝑾 + 𝒃𝑺
𝟐𝒌𝑨. 𝑻 𝟐 × 𝟏𝟎𝟎𝟎 × 𝟎. 𝟏 × 𝟎. 𝟎𝟏 × 𝟏𝟎𝟎
𝑺𝑼 = 𝒒" × 𝑨 + = (𝟓𝟎𝟎 × 𝟏𝟎𝟑 × 𝟎. 𝟏 × 𝟎. 𝟎𝟏) + = 𝟐𝟓𝟎𝟎
∆𝒚 𝟎. 𝟏
And aW=0 & aS=0, 𝟒𝟎 . 𝑻 = 𝟎 𝑻 + 𝟏𝟎 𝑻 + 𝟎 𝑻 + 𝟏𝟎 𝑻 + 𝟓𝟎𝟎
−𝟐𝒌𝑨 𝑷 𝑾 𝑬 𝑵 𝑺
SP= ∆𝒚 = −𝟐𝟎 𝟒𝟎 . 𝑻𝟒 = 𝟎 𝑻𝑾 + 𝟏𝟎 𝑻𝟖 + 𝟎 𝑻𝑵 + 𝟏𝟎 𝑻𝟑 + 𝟐𝟓𝟎𝟎 − −(𝟒)
 FVM for Two-Dimensional Steady State Diffusion
𝒌. 𝑨 𝟏𝟎𝟎𝟎 × 𝟎. 𝟏 × 𝟎. 𝟎𝟏
𝒂𝑾 = 𝒂𝑬 = 𝒂𝑵 = 𝒂𝑺 = = = 𝟏𝟎
∆𝒙 𝟎. 𝟏
𝒂𝑷 . ∅𝑷 = 𝒂𝑾 ∅𝑾 + 𝒂𝑬 ∅𝑬 + 𝒂𝑵 ∅𝑵 + 𝒂𝑺 ∅𝑺 + 𝑺𝑼
𝒂𝑷 = 𝒂𝑾 + 𝒂𝑬 + 𝒂𝑵 + 𝒂𝑺 − 𝑺𝑷

Node aN aS aW aE SP aP SU
1 10 0 0 10 0 20 500
2 10 10 0 10 0 30 500
3 10 10 0 10 0 30 500
4 0 10 0 10 -20 40 2500 𝒂𝑷
= 𝒂𝑾 + 𝒂𝑬 + 𝒂𝑵 + 𝒂𝑺 − 𝑺𝑷
5 10 0 10 10 0 30 0
6 10 10 10 10 0 40 0 𝟐𝒌. 𝑨
𝑺𝑷 = . 𝑻𝑷 = 𝟎
𝜹𝒙
7 10 10 10 10 0 40 0 ∵ 𝑭𝒍𝒖𝒙 𝒊𝒔 𝟎 at nodes 1,6,9,12
8 0 10 10 10 -20 50 2000 𝟐𝒌𝑨. 𝑻
𝑺𝑼 = 𝒒" × 𝑨 +
9 10 0 10 0 0 20 0 ∆𝒚
10 10 10 10 0 0 30 0 𝑂𝑏𝑡𝑎𝑖𝑛 12 𝑎𝑙𝑔𝑒𝑏𝑟𝑎𝑖𝑐
11 10 10 10 0 0 30 0 𝑒𝑞𝑢𝑎𝑡𝑖𝑜𝑛𝑠 𝑎𝑛𝑑 𝑠𝑜𝑙𝑣𝑒
𝑓𝑜𝑟 𝑛𝑜𝑑𝑎𝑙 𝑣𝑎𝑙𝑢𝑒𝑠
12 0 10 10 0 -20 40 2000
 FVM for Two-Dimensional Steady State Diffusion
𝒌. 𝑨 𝟏𝟎𝟎𝟎 × 𝟎. 𝟏 × 𝟎. 𝟎𝟏
𝒂𝑾 = 𝒂𝑬 = 𝒂𝑵 = 𝒂𝑺 = = = 𝟏𝟎
∆𝒙 𝟎. 𝟏
𝒂𝑷 . ∅𝑷 = 𝒂𝑾 ∅𝑾 + 𝒂𝑬 ∅𝑬 + 𝒂𝑵 ∅𝑵 + 𝒂𝑺 ∅𝑺 + 𝑺𝑼
𝒂𝑷 = 𝒂𝑾 + 𝒂𝑬 + 𝒂𝑵 + 𝒂𝑺 − 𝑺𝑷

Node Temp
T1 260
T2 242
T3 205.6
T4 146.3
T5 222
T6 211
T7 178
T8 129
T9 212
T10 196
T11 166
T12 124
 Assignment 2
A two-dimensional plate of thickness 1 cm is shown. The thermal
conductivity of the plate material is k = 1000 W/m.K. The east boundary
receives a steady heat flux of 500 kW/m2 and the north and west
boundaries are insulated. If the south boundary is maintained at a
temperature of 100°C, use a uniform grid with Δx = Δy = 0.1 m to
calculate the steady state temperature distribution at all nodes.
Insulated North

Insulated West

East
Heat Flux 500 kW/m2

South
Temperature 100oC