Conduction and Radiation

MED 546

INSTRUCTOR
Dr. Ajay Bhandari
Assistant Professor
Department of Mechanical Engineering
Indian Institute of Technology (ISM), Dhanbad
826004 (Jharkhand) INDIA
 Conduction and radiation (MED 546)
Credits: L-T-P: 3-0-0 (Total: 39 lectures (3 Lectures per week))
Days and Time: Monday (02:00 to 02:50 pm), Tuesday (06:00 to 06:50 pm) and Thursday (02:00 to
02:50 pm)
Intended for: Masters (PG) and Ph.D. students
Prerequisite: Heat and Mass transfer at UG level
Aim: The aim of the course is to provide fundamentals concepts and their application in conduction
and radiation heat transfer. Students will be learning different solution methods (analytical) to
handle the complex problem in conduction and radiation
Grading Pattern: Mid-sem and End-sem exam + May be some Assignments
Books to be Referred:
• Latif M. Jiji, Heat Conduction, 3/e, Springer, 2009.
• M.F. Modest, Radiative Heat Transfer, Academic Press, 3/e 2013.
• E. R. G. Eckert and Robert M. Drake, Analysis of Heat and Mass Transfer, McGraw-Hill, 1st Edition, 1987.
• Vedat S. Arpaci, Conduction Heat Transfer, Addison-Wesley series, 1st Revised Edition, 1966.
• F. Incropera, D. J. Dewitt, T. Bergman and A. Lavine, Fundamentals of heat and mass transfer, Wiley & Sons Inc., 7th Edition,
2011.
 Conduction and Radiation (MED 546)
The entire course will be divided in 6 modules.

Conduction and
Radiation

3
1. Review of Basic
2. Multi
Dimensional 3. Transient Heat 4. Conduction With
5. Perturbation
Solution and Heat 6. Radiation Heat
Concepts (6L) Steady State Conduction (7L) Phase Change (5L) Conduction in Transfer (12L)
Conduction (4L) Porous Media (5L)
 Module 1 (Review of basic concepts)
 Introduction to heat transfer
 Modes of heat transfer
 Brief introduction to conduction
 Differential formulation of heat conduction equation
➢ Different types of boundary conditions
 One dimensional steady state conduction
➢ Variable thermal conductivity
➢ Heat Generation
 Extended surfaces
➢ Variable area fins
➢ Bessel differential equations and functions
 Introduction to heat transfer
 From the study of thermodynamics, we know that energy can be transferred by interactions of a system
with its surroundings in the form of work and heat.
 Thermodynamics provides no information concerning the nature of the interaction or the time rate at
which it occurs.
 The objective of studying heat transfer to extend thermodynamic analysis through the study of the modes
of heat transfer and through the development of relations to calculate heat transfer rates.
In this slide we will address following questions
Q1. What is heat transfer?
A1. Heat transfer (or heat) is thermal energy in transit due to a spatial temperature difference. Whenever a
temperature difference exists in a medium or between media, heat transfer must occur. The energy transfer
by heat can not be measured directly but it can be measured with the help of the measurable quantity
temperature.
Q2. How is heat transferred?
A2. For answering this question we have to look into different type of heat transfer process technically known
as modes of heat transfer.
1. Conduction
2. Convection
3. Radiation
Application areas of heat transfer
 Modes of heat transfer
CONDUCTION
 Conduction is a mode of heat transfer in which energy exchange takes place from the region of high
temperature to that of low temperature by the kinetic motion or direct impact of molecules (molecular motion).
 Rate of equation for conduction is given by Fourier’s law. For heat conduction in x direction normal to area A
𝝏𝑻
assuming material to be isotropic and homogeneous, rate of heat flow (q) = -𝐤𝐀 𝝏𝒙
CONVECTION
 The convection heat transfer mode is comprised of two mechanisms. In addition to energy transfer due to
random molecular motion (diffusion), energy is also transferred by the bulk, or macroscopic, motion of the fluid
 Regardless of the nature of the convection heat transfer process, the appropriate rate equation is of the form
𝒒 = 𝒉𝑨 𝑻𝑺 − 𝐓∞ h=Heat transfer coefficient. (newton’s law of cooling)
RADIATION
 All bodies continuously emit energy because of their temperature and energy thus emitted is called thermal
radiation .
 The energy of the radiation field is transported by electromagnetic waves (or alternatively, photons).
 The transfer of energy by conduction or convection requires the presence of a material medium, radiation
does not. In fact, radiation transfer occurs most efficiently in a vacuum.
 Rate at which energy is released per unit area (W/m2 ) is termed the surface emissive power, E. There is an
upper limit to the emissive power, which is prescribed by the Stefan–Boltzmann law 𝑬𝒃 =𝝈𝑻𝟒
Eb = Emissive power of black body
𝝈 = Stefan –Boltzmann constant T= Absolute temperature in kelvin
 Differential formulation of heat conduction equation
Rectangular/Cartesian Coordinates ሶ
𝑄𝑧+∆𝑧 ሶ
𝑄𝑦+∆𝑦
Let’s consider a differential control volume (a small rectangular
element) of length dx, width dy and height dz. We assume 𝝆 and c
to be density and specific heat capacity of the body. Let’s apply
the principle of conservation of energy 𝑄𝑥ሶ

Rate of energy added + Rate of energy generated - Rate of energy

removed = Rate of energy change within element.
ሶ
𝑄𝑥+∆𝑥
ሶ +𝑬𝒈ሶ -𝑬𝒐𝒖𝒕
𝑬𝒊𝒏 ሶ = 𝑬ሶ (1)
𝑄𝑦ሶ
• In this analysis we assume that energy is being exchanged
within the element by conduction only (neglecting mass
motion)any sort of energy interaction in the form viscous 𝑄𝑧ሶ
dissipation and radiation is also negligible .
∆𝑬
𝑸ሶ𝒙 +𝑸𝒚ሶ + 𝑸ሶ𝒛 -𝑸𝒙+∆𝒙
ሶ ሶ − 𝑸𝒛+∆𝒛
− 𝑸𝒚+∆𝒚 ሶ + 𝑬𝒈𝒆𝒏,𝒆𝒍𝒆𝒎𝒆𝒏𝒕
ሶ = 𝒆𝒍𝒆𝒎𝒆𝒏𝒕 (2)
∆𝒕

ሶ
ሶ = 𝑸ሶ𝒙 +𝝏(𝑸𝒙 ) 𝒅𝒙 Similarly we can write 𝑸𝒚+∆𝒚
Now, 𝑸𝒙+∆𝒙 ሶ and 𝑸𝒛+∆𝒛
ሶ
𝝏𝒙

𝑬𝒈ሶ within the element can be written as 𝒒′′′ 𝑽𝒆𝒍𝒆𝒎𝒆𝒏𝒕 = 𝒒′′′dxdydz Substituting all these in (2) and dividing
dxdydz we get
 Differential formulation of heat conduction equation
𝟏 𝝏(𝑸𝒙ሶ ) 𝟏 𝝏(𝑸𝒚ሶ ) 𝟏 𝝏(𝑸ሶ𝒛 ) 𝑻𝒕+𝒅𝒕 −𝑻𝒕
− 𝒅𝒚𝒅𝒛 − 𝒅𝒙𝒅𝒛 − 𝒅𝒙𝒅𝒚 + 𝒒′′′ = 𝝆c( ) (3)
𝝏𝒙 𝝏𝒚 𝝏𝒛 𝒅𝒕

𝝏𝑻 𝝏𝑻 𝝏𝑻
Now, according to Fourier’s law 𝑸ሶ𝒙 = -𝐤𝑨𝒙 𝝏𝒙 ; 𝑸𝒚ሶ = -𝐤𝑨𝒚 𝝏𝒚 ; 𝑸ሶ𝒛 = -𝐤𝑨𝒛 𝝏𝒛 Substituting all these in (3) we get

𝝏𝟐 𝑻 𝝏𝟐 𝑻 𝝏𝟐 𝑻 q’” 𝝆𝒄𝒑 𝝏𝑻
(𝝏𝒙𝟐 +𝝏𝒚𝟐 + 𝝏𝒛𝟐 )+ 𝒌 = 𝒌 ( 𝝏𝒕 ) (4)

Cylindrical Coordinates

Substitute x= rcos∅; y = rsin∅ and z= z. Substitute this in eqn. (4)

𝟏 𝝏 𝝏𝑻 𝟏 𝝏 𝝏𝑻 𝝏 𝝏𝑻 q’” 𝝆𝒄𝒑 𝝏𝑻
(𝒓 )+ + 𝝏𝒛 + = 𝒌 ( 𝝏𝒕 ) (5)
𝒓 𝝏𝒓 𝝏𝒓 𝒓𝟐 𝝏∅ 𝝏∅ 𝝏𝒛 𝒌

Spherical Coordinates

Substitute x= rcos∅sin𝜽; y = rsin∅sin𝜽 and z= rcos𝜽. Substitute this in eqn. (4)

𝟏 𝝏 𝟐 𝝏𝑻)+ 𝟏 𝝏 𝝏𝑻 𝟏 𝝏 𝝏𝑻 q’” 𝝆𝒄𝒑 𝝏𝑻

(𝒓 + 𝒓𝟐 𝒔𝒊𝒏𝜽 𝝏𝜽 𝒔𝒊𝒏𝜽 𝝏𝜽 + = ( 𝝏𝒕 ) (6)
𝒓𝟐 𝝏𝒓 𝝏𝒓 𝒓𝟐 𝒔𝒊𝒏𝟐 𝜽 𝝏∅ 𝝏∅ 𝒌 𝒌
 Different types of boundary conditions
Specified Temperature Boundary Conditions: When the quantitative value of the temperature is specified at
the boundary. Let’s say T = 100 °𝑪

Specified Heat flux Boundary Conditions: Whenever there is sufficient information about energy interaction at
𝝏𝑻
a surface it is then possible to determine the rate of heat flux 𝒒′′ . Let’s say 𝒒′′ = −𝐤 𝝏𝒙

Insulated Boundary Conditions: Heat transfer through a perfectly insulated surface can be taken to be zero. A
well insulated surface can be modeled at a surface with specified heat flux = 0. Let’s say 𝒒′′ = 0

Thermal Symmetry: Some heat transfer problems pose thermal symmetry as a result of symmetry in imposed
thermal conditions.

Convection Boundary Conditions: When heat transfer surfaces are exposed to environment at specified
𝝏𝑻
temperature −𝐤 𝝏𝒙 = h (𝑻 𝑳, 𝒕 − 𝑻∞ )

Radiation Boundary Conditions: When heat transfer surfaces are surrounded by an evacuate surface and
there is no convection HT between surface and surrounding medium. In those cases radiation HT between
𝝏𝑻
surface and surroundings becomes significant −𝐤 𝝏𝒙 = 𝝐𝝈 (𝑻(𝑳, 𝒕)𝟒 − 𝑻𝟒𝒔𝒖𝒓𝒓 )

Interface Boundary Conditions: Two bodies in contact must have same temperature at the area of contact
and an interface cannot store any energy and thus heat flux on two sides of the interface must be same.
Identification of boundary conditions: Numerical
 ONE-DIMENSIONAL STEADY STATE CONDUCTION
Q: Consider a steam pipe of length L = 20 m; inner radius r1 = 6 cm and outer radius r2 = 8 cm, k = 20 W/mK.
Inner and outer surfaces of pipe are T1= 150 °𝑪 and T2 = 60 °𝑪 . Obtain a general solution for temperature
distribution inside the pipe and rate of heat loss from steam through pipe. Ignore the internal heat generation
and take constant thermal conductivity.
A: Procedure: Formulate the problem by obtaining applicable differential equation and specifying the
boundary conditions

Assumptions: HT is steady, 1D and there is thermal symmetry about centreline and no

variation in axial direction T=T(r), k is constant and no heat generation.

𝒅 𝒅𝑻
𝒓 =𝟎 T1
𝒅𝒓 𝒅𝒓

𝒅𝑻 T2
Integrating 𝒓 𝒅𝒓 = C1, which implies T(r) = C1lnr+C2

Find C1 and C2
L
𝒓
𝐥𝐧 𝒓𝟏
T(r) = 𝒓𝟐 𝑻 𝟐 − 𝑻𝟏 + 𝑻𝟏
𝐥𝐧 𝒓𝟏

𝒅𝑻 𝑪𝟏
ሶ
𝑸𝒄𝒚𝒍𝒊𝒏𝒅𝒆𝒓 = -kA 𝒅𝒓 = -k(2𝝅rL) 𝒓 = 786KW
 ONE-DIMENSIONAL STEADY STATE CONDUCTION
Q: Consider a spherical shell container of inner radius r1 = 8 cm and outer radius r2 = 10cm, k = 45W/mK.
Inner and outer surfaces of pipe are T1= 200 °𝑪 and T2 = 80 °𝑪 . Obtain a general solution for temperature
distribution inside the shell under steady conditions and rate of heat loss from container. Ignore the internal
heat generation and take constant thermal conductivity.
A: Procedure: Formulate the problem by obtaining applicable differential equation and specifying the
boundary conditions

Assumptions: HT is steady, 1D as there is thermal symmetry about midpoint T = T(r) T=T(r), k

is constant and no heat generation. 𝑇2
𝑟2

𝒅 𝟐
𝒅𝑻
𝒓 =𝟎
𝒅𝒓 𝒅𝒓 𝑟1
𝑇1
𝒅𝑻 C
Integrating 𝒓𝟐 = C1, which implies T(r) = - r1 + C2
𝒅𝒓

𝒓 𝒓 𝒓𝟐 𝑻𝟐−𝒓𝟏 𝑻𝟏
Find C1 = -𝒓 𝟏−𝒓𝟐 𝑻𝟏 − 𝑻𝟐 𝒂𝒏𝒅 𝑪𝟐 =
𝟏 𝟐 𝒓𝟐−𝒓𝟏

𝒓 𝒓 𝒓𝟐 𝑻𝟐−𝒓𝟏 𝑻𝟏
T(r) = 𝒓(𝒓 𝟏−𝒓𝟐 𝑻𝟏 − 𝑻𝟐 +
𝟏 𝟐) 𝒓𝟐−𝒓𝟏

𝒅𝑻 𝑪
ሶ
𝑸𝒄𝒚𝒍𝒊𝒏𝒅𝒆𝒓 = -kA 𝒅𝒓 = -k(4𝝅𝒓𝟐 ) 𝒓𝟐𝟏 = 27.1KW
 ONE-DIMENSIONAL STEADY STATE CONDUCTION
Q: A large homogeneous resistance wire of radius r0 = 0.5 cm and thermal conductivity k = 13.5 W/m0C is being
used to boil water at atmospheric pressure by the passage of electric current as shown in Figure. Heat is generated
in the wire as a result of resistance heating at the rate of q’” = 4.3×107 W/m3. If the outer surface temperature of the
wire is measured to be Ts = 1080C, obtain a relation for the temperature distribution and determine the temperature
at the centerline of the wire when steady operating conditions are reached.
A: Procedure: Formulate the problem by obtaining applicable differential equation and specifying the boundary
conditions
Assumptions: HT is steady, 1D and there is thermal symmetry about
centreline and no variation in axial direction T=T(r), k is constant,
heat generation in the wire is uniform.
𝟏𝟎𝟖𝟎𝑪
𝟏 𝒅 𝒅𝑻 q’”
𝒓 + =𝟎
𝒓 𝒅𝒓 𝒅𝒓 𝒌
𝒓𝟎
𝒅𝑻 q’” 𝒓𝟐 𝒅𝑻(𝟎)
Integrating 𝒓 𝒅𝒓 = − + 𝑪𝟏, Applying BC = 𝟎; C1 = 0
𝒌 𝟐 𝒅𝒓 𝟎 r
q’”
T(r) =− 𝒓 𝟐 + 𝑪 , Applying BC 𝑻 𝒓
𝟐 𝟎 = 𝑻𝒔 = 𝟏𝟎𝟖𝟎𝑪;
𝟒𝒌
q’”
C2 = Ts + 𝟒𝒌 𝒓𝟐

q’”
𝑻(𝒓) = Ts + (𝒓𝟐𝟎 − 𝒓𝟐 ) = 1280C
𝟒𝒌
 ONE-DIMENSIONAL STEADY STATE CONDUCTION (Variable thermal conductivity)
Q: Consider a plate with internal heat generation 𝒒′′′ and a variable thermal conductivity k given by 𝒌 =
𝒌𝟎 (𝟏 − 𝜸𝑻) where 𝒌𝟎 and 𝜸 are real constants and T is the temperature. Both surfaces are maintained at 0°𝑪.
Determine the temperature distribution inside the plate?
A: Assumptions: (1)HT is steady and1D; (2) uniform heat generation. (3) Stationary material
𝒅 𝒅𝑻
Governing Equations: 𝒅𝒙 𝒌 𝒅𝒙 + 𝒒′′′ = 𝟎 Put the value of 𝒌 = 𝒌𝟎 (𝟏 − 𝜸𝑻)
𝒅 𝒅𝑻 𝒒′′′
(𝟏 − 𝜸𝑻) +𝒌 =𝟎 (1)
𝒅𝒙 𝒅𝒙 𝟎

Boundary Conditions: T(0) = 0 and T(L) = 0 (2)

𝒅𝑻 𝒒′′′
Integrating (1) (𝟏 − 𝜸𝑻) + x = C1, (3)
𝒅𝒙 𝒌𝟎
𝜸𝑻𝟐 𝒒′′′
Integrating (3) 𝐓− + 𝒙𝟐 = C1x+C2 (4) Incorporating boundary conditions (2) and
𝟐 𝟐𝒌𝟎
finding C1 and C2 we get

𝟏 𝟏 𝒒′′′ 𝑳𝒙 𝒙
T= ± (𝜸𝟐 − 𝜸𝑲 (1− 𝑳 )) (5)
𝜸 𝟎

To satisfy BC T(0) = 0 and T(L) = 0 minus sign in eqn.(5) should be considered. Therefore, final temperature
𝟏 𝟏 𝒒′′′ 𝑳𝒙 𝒙
distribution is T = − (𝜸𝟐 − 𝜸𝑲 (1− 𝑳 )) (6)
𝜸 𝟎
 Radial conduction in composite cylinder with interface friction
Q: A shaft of radius Rs rotates inside a sleeve of inner radius Rs and outer radius Ro. Frictional heat is generated at the interface at a
flux 𝒒′′
𝒊 . The outside surface of the sleeve is cooled by convection with an ambient fluid 𝑻∞ . The heat transfer coefficient is h.
Consider 1D steady state conduction in the radial direction, determine the temperature distribution in the shaft and sleeve.
A: Assumptions: (1)1D radial conduction; (2) Steady state; (3) Constant k1 and k2 (4) No energy generation and uniform frictional
energy flux at the interface; (5) Stationary material (No motion in the radial direction)
𝒅 𝒅𝑻𝟏 𝒅 𝒅𝑻𝟐
Governing Equations: 𝒓 = 𝟎 (1); 𝒅𝒓 𝒓 = 𝟎 (2)
𝒅𝒓 𝒅𝒓 𝒅𝒓

Boundary Conditions:

𝒅𝑻𝟏 (𝑹𝟎 )
(i) Convection BC: −𝑘1 = h (𝑻𝟏 (𝑹𝟎 ) − 𝑻∞ ) (3)
𝒅𝒓

𝒅𝑻𝟐 (𝑹𝑺 ) 𝒅𝑻𝟏 (𝑹𝑺 )

(ii) Interface BC: T2 (Rs) = T1 (Rs) (4); −𝑘2 + 𝒒′′ = −𝑘1 (5)
𝒅𝒓 𝒅𝒓

𝒅𝑻𝟐
(iii) Symmetry BC: = 0 (6)
𝒅𝒓
 Extended surfaces: Fins
Need : One of the main applications of heat transfer study is to increase
the rate of heat transfer from a heated surface to a cool fluid.
According to Newton’s Law of cooling 𝑸ሶ = hA( Ts – 𝑻∞ )
To increase the surface heat transfer rate:
(1) Increase the heat transfer coefficient h by changing the fluid and/or
manipulating its motion.
(2) Lower the ambient temperature
(3) Increase surface area This option is exercised in many engineering
applications in which the heat transfer surface is "extended" by adding
fins.
Practical examples: Rods attached to the condenser present at back of
the refrigerator, honeycomb surface of a car radiator, the corrugated
surface of a motorcycle engine, and the disks attached to a baseboard
radiator. Straight fin of Straight fin of Annular fin Pin fin
uniform cross non-uniform
section cross section

Types of fins:
(i) Constant area straight fins: Cross section remains same if one moves
from base of the wall to the tip.
(ii) Variable C/S area fins: When C/S area changes if one moves from
base of the wall towards tip
(iii) Spine/pin fin: Distinguished by circular C/S
(iv) Annular/Cylindrical fins: Disk mounted on a tube.
 Heat transfer and temperature distribution in variable C/S area fins
Let’s consider a general case of a variable area fin with volumetric energy generation under transient conditions. The fin
exchanges heat with an ambient fluid by convection. The heat transfer coefficient is h and the ambient temperature is
𝑻∞ . We assume that the Biot number is small compared to unity and thus invoke the fin approximation that the
temperature does not vary within a cross section. Since the temperature depends on a single spatial variable x, our
starting point should be the selection of an appropriate element. We select a fin slice at location x of infinitesimal
thickness dx, which encompasses the entire cross section.
Applying conservation of energy principle to the element we get:
ሶ +𝑬𝒈ሶ -𝑬𝒐𝒖𝒕
𝑬𝒊𝒏 ሶ = 𝑬ሶ (1)
𝒅𝒒
Energy enters the element by conduction at a rate 𝒒𝒙 and leaves at a rate 𝒒𝒙 + ( 𝒙)dx. Energy also leaves by convection
𝒅𝒙
at a rate dqc. However, this rate is infinitesimal because the surface area of the element is infinitesimal.
ሶ = 𝒒𝒙 + (𝝏𝒒𝒙)dx+ dqc (2)
ሶ = 𝒒𝒙 and 𝑬𝒐𝒖𝒕
𝑬𝒊𝒏
𝝏𝒙
𝝏𝒒𝒙
Substituting (2) in (1) we get 𝑬𝒈ሶ - ( )dx- dqc = 𝑬ሶ (3)
𝝏𝒙

Introducing Fourier’s law and Newton’s law to eliminate 𝒒𝒙 and dqc

𝝏𝑻
𝒒𝒙 = −𝐤𝑨𝑪 and dqc = hdAs( T – 𝑻∞ ) (4)
𝝏𝒙
𝝏𝑻
𝑬𝒈ሶ = 𝒒′′ ′Ac(x)dx and 𝑬ሶ = 𝝆𝒄𝒑 Ac(x) 𝒅𝒙 (5)
𝝏𝒕

Substituting (4) and (5) into (3) we get

𝝏 𝝏𝑻 𝝏𝑻
𝒌𝑨𝒄 𝒙 𝒅𝒙 − 𝒉 T – 𝑻∞ 𝒅𝑨𝒔 + 𝒒′′ ′Ac(x)dx = 𝝆𝒄𝒑 Ac(x) 𝒅𝒙 (6)
𝝏𝒙 𝝏𝒙 𝝏𝒕
 Heat transfer and temperature distribution in variable C/S area fins
𝝏𝟐 𝑻 𝟏 𝒅𝑨𝑪 𝝏𝑻 𝒉 𝒅𝑨 𝒒′′′ 𝟏 𝝏𝑻
+( ) − 𝒌𝑨 (𝒙) T– 𝑻∞ 𝒅𝒙𝑺 + = 𝜶 𝝏𝒕 (7)
𝝏𝒙𝟐 𝑨𝑪 (𝒙) 𝒅𝒙 𝝏𝒙 𝑪 𝒌

𝒅𝑨𝒔 𝒅𝒚𝒔 𝟐
Now dAs = C(x)ds and ds = 𝒅𝒙𝟐 + 𝒅𝒚𝒔 𝟐 and = 𝑪(𝒙) 𝟏 + ( )
𝒅𝒙 𝒅𝒙

Once the temperature is known Heat transfer rate can be determined by two methods:
𝝏𝑻(𝟎)
(i) Conduction at the fin base: 𝒒𝒇 = 𝒒 𝟎 = −𝐤𝑨𝑪
𝝏𝒙

(ii) Convection from surface: 𝒒𝒇 = 𝒒𝒇 = ‫ 𝒙 𝑻 𝒉 ׬‬− 𝑻∞ 𝒅𝑨𝒔

Let’s discuss some special case: Fins of Uniform Cross-Sectional Area

𝒅𝑨𝑪 𝒅𝑨𝒔 𝒅𝒚
(i) Straight Rectangular Fin: So for constant c/s area fin (Ac) = 0 and Further = 𝑪(𝒙) 𝟏 + ( )𝟐 = C where C =
𝒅𝒙 𝒅𝒙 𝒅𝒙
𝒅𝟐 𝑻 𝒉𝑪
Circumference. So equation (7) becomes − (𝑻 − 𝑻∞ ) = 0 (8).
𝒅𝒙𝟐 𝑲𝑨𝑪

𝒅𝟐 𝜽
Let’s assume (𝑻(𝒙) − 𝑻∞ ) = 𝜽(x) So equation (8) can be written as - 𝒎𝟐 𝜽 = 0 (9)
𝒅𝒙𝟐

General solution for the this second order differential equation is

𝜽(x) = 𝑪𝟏 𝒆𝒎𝒙 +𝑪𝟐 𝒆−𝒎𝒙 (10)
 Heat transfer and temperature distribution in variable C/S area fins
Case 1: Infinitely long fin: At x=0 T(0)=T0 Therefore, at x = 0 𝜽(𝟎) = T0 - 𝑻∞ =𝜽0
And at x = ∞ T = T∞ Therefore, at x = ∞ 𝜽 = T0 - 𝑻∞ =𝟎 Applying these boundary conditions to eqn (10) and finding C1 and
C2 we get
𝜽(𝒙) = 𝜽𝟎 𝒆−𝒎𝒙 (11)
𝒅𝜽
𝒒𝒇 = -kA( )x = 0 = 𝒉𝑪𝒌𝑨 𝜽𝟎 (12)
𝒅𝒙

Case 2: Fin of finite length having insulated tip: At x=0 T(0)=T0 Therefore, at x = 0 𝜽(𝟎) = T0 - 𝑻∞ =𝜽0
𝒅𝜽(𝑳)
And at x = L; = 0 Applying these boundary conditions to eqn. (10) and finding C 1 and C2 we get
𝒅𝒙

(13)

𝒅𝜽
𝒒𝒇 = -kA( )x=L = (14)
𝒅𝒙

Case 3: Fin of finite length having convecting tip: At x=0 T(0)=T0 Therefore, at x = 0 𝜽(𝟎) = T0 - 𝑻∞ =𝜽0
𝒅𝑻(𝑳) 𝒅𝜽(𝑳)
And at x = L; -k = 𝒉 𝑻 𝑳 − 𝑻∞ . Therefore, at x = L; -k = 𝒉𝜽(𝑳). Applying these boundary conditions to eqn. (10)
𝒅𝒙 𝒅𝒙
and finding C1 and C2 we get
(15)

𝒅𝜽 (16)
𝒒𝒇 = -kA( )x=L =
𝒅𝒙
 Fin efficiency and effectiveness
Fin efficiency : This is defined as the ratio of total heat transfer by fin to the maximum possible heat transfer from the fin
Under ideal condition the maximum heat transfer is possible if the entire fin surface area will be at base temperature
𝒒𝒎𝒂𝒙 = hAs(T0 – T∞)

𝒒𝒇
𝜼=
𝒒𝒎𝒂𝒙
Fin effectiveness : This is defined as the ratio actual heat transfer to heat that would be transferred from same base area A 0
without the fin with base temperature T0 remaining constant.

𝒒𝒇𝒊𝒏
𝜺=
𝒒𝒏𝒐 𝒇𝒊𝒏
Corrected Length (Lc): We see that fins with insulated tip have a simpler temperature profile and HT solutions than fin with
convection at the tip. Therefore, if the error is negligible we can use simpler solutions.
Tip area is small as compared to fin surface area and (T(x) – T∞) is very small at the tip. Therefore, HT by convection from
tip is very less than HT from surface of the fin. This little error is introduced by assuming tip to be insulated. To compensate
for ignoring heat loss from tip using insulated model fin length is increased by ∆𝑳𝒄.
Therefore corrected length Lc = L+ ∆𝑳𝒄

It depends upon geometry of the fin.

 Fin efficiency and effectiveness
Q: A 15cm X 20 cm integrated circuit board is to be cooled by attaching a 4 cm long aluminium (k = 237W/mK) fins on
one side of it. Each fin has a 2mm X 2mm square cross section. The surrounding ambient temperature is 25 0C and the
convection heat transfer coefficient on each fin surface is 20W/m2K. To prevent the circuit board from overheating the
upper surface of the circuit board needs to be at 850C. Design a finned surface having the appropriate number of fins with
an overall effectiveness of 3 that can keep the circuit board from overheating.
A: Assumptions: HT is steady, 1D, HT from the fin tips is negligible, The h is constant and uniform over the entire fin surface.
The thermal properties of the fins are constant.
Analysis: C/S area of the fins are constant the efficiency of the square C/S fins that are assumed to have adiabatic tips
can be determined to be
𝒉𝑷
𝒎= = 12.99 m-1
𝒌𝑨𝒄

𝒕𝒂𝒏𝒉𝒎𝑳
Efficiency of the fin 𝜂 = = 𝟎. 𝟗𝟏𝟗
𝒎𝑳

Afin = nfin*4*(0.002m)(0.04m) = 0.00032nfin 𝒎𝟐

Aunfinned = (0.15m)(0.2m)-nfin (0.002m)(0.002m) = 0.03-0.000004nfin 𝒎𝟐
ሶ
𝑸𝒇𝒊𝒏𝒏𝒆𝒅 = 𝜂𝒇𝒊𝒏 𝑸𝒇𝒊𝒏,𝒎𝒂𝒙
ሶ = 𝜂𝒇𝒊𝒏 𝒉𝑨𝒇𝒊𝒏(𝑻𝟎 − 𝑻∞ ) = 0.3529nfinW

ሶ
𝑸𝒖𝒏𝒇𝒊𝒏𝒏𝒆𝒅 = 𝒉𝑨𝒖𝒏𝒇𝒊𝒏𝒏𝒆𝒅(𝑻𝟎 − 𝑻∞ ) = 36-0.0048nfinW

Total heat transfer from the finned surface (circuit board) 𝑸𝒕𝒐𝒕𝒂𝒍,𝒇𝒊𝒏
ሶ ሶ + 𝑸𝒖𝒏𝒇𝒊𝒏𝒏𝒆𝒅
= 𝑸𝒇𝒊𝒏𝒏𝒆𝒅 ሶ = 0.3529nfin+36-0.0048nfin
 Fin efficiency and effectiveness
The rate of heat transfer if there were no fins attached to the plate would be
Ano fin = (0.15m)(0.2m) = 0.03m2
ሶ
𝑸𝒏𝒐,𝒇𝒊𝒏 = 𝒉𝑨𝒏𝒐 𝒇𝒊𝒏 𝑻𝒃 −ሶ 𝑻∞ = 𝟑𝟔𝑾

ሶ
𝑸𝒇𝒊𝒏 0.3529nfin+36−0.0048nfin
𝝐=𝟑= =
ሶ
𝑸𝒏𝒐,𝒇𝒊𝒏 𝟑𝟔

Therefore, number of fins required are = 207

Q: A constant area fin of length 2L and cross-sectional area Ac generates heat at a volumetric rate 𝒒′′ ′. Half the fin is
insulated while the other half exchanges heat by convection. The heat transfer coefficient is h and the ambient
temperature is 𝑻∞ . The base and tip are insulated. Determine the steady state temperature at the mid-section.
A: Observations. (i) This is a constant area stationary fin. (ii) Surface conditions are different for each half. Thus two
equations are needed. (iii) Both ends are insulated.
Origin and Coordinates. The origin is selected at one end of the fin and the coordinate x is directed as shown.
Assumptions. (1) One-dimensional, (2) steady state, (3) constant cross-sectional area, (4) uniform energy generation, (5)
Bi <<1, (6) constant conductivity, (7) no radiation and (8) uniform h and 𝑻∞ .
Governing Equations. Since half the fin is insulated while the other half exchanges heat by convection, two fin equations
are needed. Let
T1 (x) = temperature distribution in the first half, 0 ≤ x ≤ L
T2 (x) = temperature distribution in the second half, L ≤ x ≤ 2L
 Fin efficiency and effectiveness
𝝏𝟐 𝑻 𝟏 𝒅𝑨𝑪 𝝏𝑻 𝒉 𝒅𝑨𝑺 𝒒′′′ 𝟏 𝝏𝑻
The fin equation: + ( ) − T– 𝑻∞ + =
𝝏𝒙𝟐 𝑨𝑪 (𝒙) 𝒅𝒙 𝝏𝒙 𝒌𝑨𝑪 (𝒙) 𝒅𝒙 𝒌 𝜶 𝝏𝒕
𝒅𝑨𝑺
Note that Ac is constant and = C, where C is the circumference. Above equation becomes
𝒅𝒙
𝒅𝟐 𝑻 𝒉𝑪 𝒒′′′
− T – 𝑻∞ + = 0 (1)
𝒅𝒙𝟐 𝒌𝑨𝑪 𝒌

In the first half the fin is insulated along its surface so convection term vanishes therefore, eqn. (1) becomes
𝒅𝟐 𝑻𝟏 𝒒′′′
+ = 0 (2)
𝒅𝒙𝟐 𝒌

For second half

𝒅𝟐 𝑻𝟐 𝒒′′′
− 𝒎 T 2 – 𝑻∞ +
𝟐
= 0 (3)
𝒅𝒙𝟐 𝒌

Boundary conditions:
𝒅𝑻𝟏(𝟎)
(1): Insulated surface at x = 0 = 𝟎 (4)
𝒅𝒙

(2): Equality of temperature at x = L T1(L) = T2(L) (5)

𝒅𝑻𝟏(𝑳) 𝒅𝑻𝟐(𝑳)
(3): Equality of flux at x= L = (6)
𝒅𝒙 𝒅𝒙
𝒅𝑻𝟐(𝟐𝑳)
(4): Insulated surface at x = 2L = 0 (7)
𝒅𝒙
 Fin efficiency and effectiveness
−𝒒′′′ 𝟐
𝑻𝟏 𝒙 = 𝒙 + 𝑨𝒙 + 𝑩 (8)
𝟐𝒌
𝒒′′′
𝑻𝟐 𝒙 = 𝑻∞ + + 𝑪𝒔𝒊𝒏𝒉𝒎𝒙 + 𝑫𝒄𝒐𝒔𝒉𝒎𝒙 (9)
𝒌𝒎𝟐

Application of boundary conditions (4)-(7) will give A,B,C and D

After substitution of these constants we will get the temperature to be

 Moving Fins
There are applications where a material exchanges heat with the surroundings while moving through a
furnace or a channel.
Examples: The extrusion of plastics, drawing of wires and sheets from rollers. Such problems can be modeled
as moving fins as long as the criterion for fin approximation is satisfied.

Figure shows a sheet being drawn with velocity U through rollers. The sheet exchanges heat with the
surroundings by radiation. It also exchanges heat with an ambient fluid by
convection. Thus its temperature varies with distance from the rollers

Consider a constant area fin with constant velocity U. For steady state the conservation of energy for the
element dx gives:

ሶ ሶ ෡ሶ
𝒅𝒉
෡ ෡
𝒒𝒙 + 𝒎𝒉 = 𝒒𝒙+𝒅𝒙 + 𝒎𝒉 + 𝒎 𝒅𝒙 𝒅𝒙 + dqr + dqc (1)

𝝏𝑻
𝒒𝒙 = −𝐤𝑨𝑪 𝝏𝒙 (2) and dqc = h ( T – 𝑻∞ )dAs (3)

dqr = 𝜺𝝈 (𝑻𝟒 – 𝑻𝟒𝒔𝒖𝒓𝒓 ) dAs (4)

 Moving Fins
Q: A thin plastic sheet of thickness t and width W is heated in a furnace to temperature T0 . The sheet moves
on a conveyor belt traveling with velocity U. It is cooled by convection outside the furnace by an ambient
fluid at T∞. The heat transfer coefficient is h. Assume steady state, Bi<0.1, negligible radiation and no heat
transfer from the sheet to the conveyor belt. Determine the temperature distribution in the sheet.

A: Observations. (i) This is a constant area moving fin problem. Temperature distribution can be assumed
one-dimensional. (ii) Heat is exchanged with the surroundings by convection. (iii) The temperature is
specified at the outlet of the furnace.

Origin and Coordinates. The origin is selected at one end of the fin and the coordinate x is directed as
shown.

Assumptions. (1) One-dimensional, (2) steady state, (3) constant conveyer speed, (4) constant h, (5) no
radiation and (8) Bi<<1.

Governing Equations.
 Moving Fins

Boundary Conditions: T(0) = T0 and T(∞) = finite

Solution:
 Temperature distribution in a triangular fin
Consider a triangular fin with length of the fin to be L and its
thickness at the base is b. The fin extends a distance W in the Base at
h, 𝑻∞ T = T0
direction normal to the paper. The triangular end surfaces are
ys
assumed insulated. Consider an element of length dx as
shown. The conduction area Ac is x b

𝑨𝒄 = 𝟐𝑾𝒚𝒔 𝒙 (𝟏) h, 𝑻∞

dx
𝒚𝒔 (𝒙) is half the element height and is given by
L

𝒙 𝒃 x=0
𝒚𝒔 = ( )( ) (2)
𝑳 𝟐

𝑾𝒃 𝒅𝑨𝒄 𝑾𝒃
Therefore, 𝑨𝒄 = 𝒙 (𝟑); = (𝟒)
𝑳 𝒅𝒙 𝑳

𝒅𝑨𝒔 𝒅𝒚
= 𝑪(𝒙) 𝟏 + (𝒅𝒙)𝟐 = 2W[𝟏 + (𝒃/𝟐𝑳)𝟐 ]𝟏/𝟐 (5)
𝒅𝒙

Substituting (3), (4) and (5) in the variable area fin equation
we solved earlier we get
𝟏
𝒅𝟐 𝑻 𝟏 𝝏𝑻 𝟐𝒉𝑳 𝒃 𝟐 𝟐𝟏
+ − 𝟏+ T – 𝑻∞ = 𝟎 (6)
𝒅𝒙𝟐 𝒙 𝝏𝒙 𝒌𝒃 𝟐𝑳 𝒙
 Temperature distribution in a triangular fin
Since b<<L. Therefore eqn. (6) becomes
𝒅𝟐 𝑻 𝟏 𝝏𝑻 𝟐𝒉𝑳 𝟏
+ 𝒙 𝝏𝒙 − T – 𝑻∞ = 𝟎 (7)
𝒅𝒙𝟐 𝒌𝒃 𝒙

Using 𝜽 = T - 𝑻∞ and multiplying by 𝒙𝟐 throughout eqn (7) becomes

𝟐
𝟐 𝒅 𝜽 𝒅𝜽 𝟐𝒉𝑳
𝒙 𝒅𝒙𝟐 + 𝒙 𝒅𝒙 − 𝒙𝜽 = 𝟎 (8)
𝒌𝒃

Now the solution to equation (8) can be provided through Bessel functions.

General form of Bessel differential equations

A class of linear ordinary differential equations with variable coefficients is known as Bessel differential
equations. Such equations are encountered in certain variable area fins and in multi-dimensional
conduction problems.

(9)
 Bessel Differential equation and solution
Salient features of Bessel differential equations

(1) It is a linear, second order ordinary differential equation with variable coefficients. That is, the coefficients
of the dependent variable y and its first and second derivatives are functions of the independent
variable x.

(2) A, B, C, D, and n are constants. Their values vary depending on the equation under consideration. Thus,
eq. (9) represents a class of many Bessel differential equations.

(3) n is called the order of the Bessel function.

(4) D can be real or imaginary.  Hildebrand, F.B., Advanced Calculus for Applications, 2nd
edition, Prentice-Hall, Englewood Cliffs, New Jersey, 1976.
 McLachlan, N.W., Bessel Functions for Engineers, 2nd
Solution of Bessel differential equation: Bessel Functions edition, Clarendon Press, London, 1961

The general solution to Bessel differential equation can be constructed in the form of infinite power series.
Since the equation is second order, two linearly independent solutions are needed. The form of the solution
depends on the constants n and D. There are four possible combinations:

(i) n is zero or positive integer, D is real: The solution is

(10)
Bessel function of first kind Bessel function of second kind
 Bessel Differential equation and solution
(ii) n is neither zero nor a positive integer, D is real: The solution is

(11)

(iii) n is zero or positive integer, D is imaginary: The solution is

(12)

Where p = D/i Where i is imaginary, i = −𝟏

(iv) n is either zero nor a positive integer, D is imaginary: The solution is

(13)

Where Jn and J-n are Bessel functions of order n of first kind Yn are Bessel functions of order n of second kind
and In and I-n are modified Bessel functions of order n of first kind Kn are modified Bessel functions of order n
of second kind and they are the symbols representing different infinite power series. For example Jn(mx)
represents the following infinite series
(14)
 Bessel Differential equation and solution
Salient relations of Bessel differential equations
𝑱−𝒏 𝒙 = −𝟏 𝒏 𝑱𝒏 (𝒙) 𝒀−𝒏 𝒙 = −𝟏 𝒏 𝒀𝒏 (𝒙) 𝑰−𝒏 𝒙 = 𝑰𝒏 (𝒙) 𝑲−𝒏 𝒙 = 𝑲𝒏 (𝒙)
Bessel Differential equation and solution
Derivatives and Integrals of Bessel Functions
 Temperature distribution in a triangular fin
To make eqn. (8) similar to eqn. (9) we define a new independent variable 𝜂 = 𝒙. Thus eqn (8) get
transformed to
𝒅𝟐 𝜽 𝒅𝜽
𝜼𝟐 +𝜼 𝒅𝜼 − 𝟒𝒎𝟐 𝜼𝟐 𝜽 = 𝟎 (15)
𝒅𝜼𝟐

Comparing (15) with (9) we get A = 0, B = 0, n = 0, C = 1, D = −𝟒𝒎𝟐 = 2im. Now n = 0 and D is imaginary
therefore solution of Bessel differential equation (15) is

𝜽 𝜼 = 𝜼𝑨 𝒆𝒙𝒑 𝑩𝜼 𝑪𝟏 𝑰𝒏 𝒑𝜼𝑪 + 𝑪𝟐 𝑲𝒏 𝒑𝜼𝑪

𝜽 𝒙 = 𝑪𝟏 𝑰𝟎 (𝟐𝒎 𝒙) + 𝑪𝟐 𝑲𝟎 (𝟐𝒎 𝒙) (16)

Applying BC 1: at x = 0; T = finite or 𝜽 = 𝜽𝒎 = 𝑪𝟏 𝑰𝟎 (𝟎) + 𝑪𝟐 𝑲𝟎 (𝟎). Now 𝑲𝟎 (𝟎) = ∞ and 𝜽(0) is finite. Therefore C2
=0
Applying BC 2: at x = L; T = 𝑻𝒃 (𝒃𝒂𝒔𝒆 𝒕𝒆𝒎𝒑𝒆𝒓𝒂𝒕𝒖𝒓𝒆) or 𝜽 = 𝜽𝒃 = 𝑪𝟏 𝑰𝟎 (𝟐𝒎 𝑳).

𝜽𝒃
𝜽 𝒙 = 𝑰 (𝟐𝒎 𝒙) (17)
𝑰𝟎 (𝟐𝒎 𝑳) 𝟎
 Variable area fin: Numerical
Q: For the steady state 1D heat transfer from an annular fin obtain an expression for the temperature
distribution in the fin. Assume the tip of the fin to be insulated.
A: Heat is conducted in the radial direction and convected from the upper and
lower surfaces of the fin. The heat transfer coefficient is h and the
ambient temperature is T∞. To determine we select a circular
element (ring) of radius r, thickness dr and height t. The area through
which heat is conducted is normal to r, given by
𝑨𝒄 (𝒓) = 𝟐𝝅𝒓𝒕 (𝟏)

𝒅𝑨𝒔 𝒅𝒚
= 𝑪(𝒓) 𝟏 + ( 𝒅𝒓𝒔)𝟐 = 2(2𝝅𝒓) (2)
𝒅𝒓
Governing equations: Fin heat equation with variable cross section area is given by
𝒅𝟐 𝑻 𝟏 𝒅𝑻 𝟐𝒉
+( ) − T – 𝑻∞ = 𝟎 (3)
𝒅𝒓𝟐 𝒓 𝒅𝒓 𝒌𝒕
𝒅 𝟐𝜽 𝒅𝜽
In non-dimensional form we can write it as 𝒓𝟐 𝒅𝒓𝟐 + 𝒓 𝒅𝒓 − 𝒎𝟐 𝒓𝟐 𝜽 = 𝟎 (4)

At BC 1: at r = r1; 𝜽 = 𝜽𝟎 = T0-T∞
𝒅𝜽
At BC 2: at r = r2; 𝒅𝒓 = 𝟎
Therefore general solution is: 𝜽 𝐫 = 𝑪𝟏 𝑰𝟎 (𝒎𝒓) + 𝑪𝟐 𝑲𝟎 (𝒎𝒓)
Equation (4) is a Bessel equation. Comparing it with
𝜽 𝒓 𝑰𝟎 𝒎𝒓 𝑲𝟏 𝒎𝒓𝟐 +𝑲𝟎 𝒎𝒓 𝑰𝟏 𝒎𝒓𝟐
Final solution: = general form we get A = 0, B = 0, n = 0, C = 1, D = m𝐢,
𝜽𝟎 𝑰𝟎 𝒎𝒓𝟏 𝑲𝟏 𝒎𝒓𝟐 +𝑲𝟎 𝒎𝒓𝟏 𝑰𝟏 𝒎𝒓𝟐 p = D/i = 𝒎. Now n = 0 and D is imaginary therefore
solution of Bessel differential equation is
 Equidimensional (Euler) Equation
Consider the following second order differential equation with variable coefficients

𝒅𝟐 𝒚 𝒅𝒚
𝒙𝟐 + 𝒂𝟏𝒙 + 𝒂𝟎𝒚 = 𝟎
𝒅𝒙𝟐 𝒅𝒙

The solution to the above equation depends on the root of the following equation

There are three possibilities:

(1) If the roots are distinct then the solution is

(2) If the roots are imaginary then the solution is

(3) If there is only one root then the solution is

 Variable area fin: Numerical
Q: The profile of a straight fin is given by
𝒙
𝒚𝒔 = 𝜹(𝑳)𝟐 (1)
The fin exchanges heat with the ambient by convection. The ambient temp is T∞ and heat transfer coefficient
is h. The base temperature is T0. Assume (𝜹Τ𝑳)<<1, determine the temperature distribution in the fin. Fin
extends a distance W normal to the paper.
Observations: (i) Temperature distribution in the fin and Fourier’s law of conduction give the heat transfer rate.
(ii) This is a variable area straight fin with specified temperature at the base. (iii) Temperature distribution can
be assumed one-dimensional.
Assumptions: (1) One-dimensional, (2) steady state, (3) no energy generation, (4) Bi <<1, (5) constant
properties, (6) uniform ambient temperature and h (7) negligible radiation, (8) Fin tips to be insulated
Governing equations: Fin heat equation with variable cross section area is given by

𝒅𝟐 𝑻 𝟏 𝒅𝑨𝑪 𝒅𝑻 𝒉 𝒅𝑨𝑺
𝟐 +( ) − 𝒌𝑨 T – 𝑻∞ = 𝟎 (2)
𝒅𝒙 𝑨𝑪 (𝒙) 𝒅𝒙 𝒅𝒙 𝑪 (𝒙) 𝒅𝒙

The cross sectional area Ac is 𝑨𝒄 = 𝟐𝑾𝒚𝒔 𝒙 (𝟑)

𝒙
𝒚𝒔 (𝒙) is half the element height and is given by 𝒚𝒔 = 𝜹( )𝟐 (4)
𝑳

𝑾𝒃 𝒅𝑨𝒄 𝑾𝒃
Therefore, 𝑨𝒄 = 𝒙 (𝟑); = (𝟓)
𝑳 𝒅𝒙 𝑳
𝒅𝑨𝒔 𝒅𝒚𝒔 𝟐
= 𝑪(𝒙) 𝟏 + ( ) = 2W[𝟏 + (𝟐𝜹𝒙/𝑳𝟐 )𝟐 ]𝟏/𝟐 = 2W (6)
𝒅𝒙 𝒅𝒙
 Variable area fin: Numerical
𝟐 𝒉𝑳𝟐
𝟐𝒅 𝜽 𝒅𝜽
𝒙 𝒅𝒙𝟐 + 2x 𝒅𝒙 − 𝛃𝜽 = 𝟎 (7) Where 𝛃 = 𝒌𝜹

Boundary Conditions: : at x = 0; T(0) = finite or 𝜽(0) = T(0)-T∞ = finite

At x = L T(L) = T0 or or 𝜽(L) = T(L)-T∞ = T0 - T∞

Equation (7) is a second order differential equation with variable coefficient. Examination of the coefficients
shows that it is an Euler equation.
 Variable area fin: Numerical
Therefore the solution becomes:

In the non-dimensional form: