MSE 321

Engineering Thermodynamics and Heat Transfer

Heat Conduction in Transient Mode (Chapter 11)

Peyman Taheri

Introduction
The temperature of a body, in general, varies with time as well as position. In rectangular coordinates, this
variation is expressed as T = T ( x, y, z, t ) , where ( x, y, z) indicate variation in the x-, y-, and z-directions,
and t indicates variation with time. In the preceding chapter, we considered heat conduction under steady
conditions, for which the temperature of a body at any point does not change with time. This certainly
simplified the analysis, especially when the temperature varied in one direction only, and we were able to
obtain analytical solutions. In this chapter, we consider the variation of temperature with time as well as
position in one- and multidimensional systems.

Lumped System Analysis

Temperatures of some bodies are observed to remain uniform at all times during a heat transfer process.
The temperature of such bodies are only a function of time, T = T (t ) . The heat transfer analysis that uses
this case is called lumped system analysis.
Consider a body of arbitrary shape of mass m , volume V , surface area As , density and specific heat

cP initially at a uniform temperature Ti (Fig. 1).

Figure 1: The geometry and parameters involved in the lumped system analysis

At time t = 0 , the body is placed into a medium at temperature T (T > Ti ) with a convective heat
transfer coefficient h .We assume lumped system analysis to be applicable, so that the temperature
remains uniform within the body at all times and changes with time only, T = T (t ) .
An energy balance of the body for a time interval dt can be expressed as,

Heat transfer
The increase in the

=
into the body during dt energy of the body during dt

or in a mathematical form,

MSE 321 (Fall 2013)

Peyman Taheri

h As (T - T ) dt = m cP dT

(1)

Noting that m = V and dT = -d (T - T ) one can find,

d (T - T )
T - T

=-

h As
dt
V cP

(2)

Integrating from t = 0 at which T = Ti , to any time t at which T = T (t ) , and then taking the exponential
of both sides gives,
T (t ) - T
= - h As t
ln
T -T
V cP

i

h As
T (t ) - T
= exp [-bt ] with b =
Ti - T
V cP

1

s

(3)

The coefficient b is called time constant and is a positive value whose dimension is [1/s].

Figure 2: The temperature of a lumped system approaches the environment temperature as time gets larger.

Note:

Using Eq. (3), we can determine the temperature T (t ) of a body at time t, or alternatively, the
time t required for the temperature to reach a specified value T (t ) .

Note:

The temperature of a body approaches the ambient temperature T exponentially. A large value
of b indicates that the body will approach the environment temperature in a short time.

Note:

The constant b is proportional to the surface area, but inversely proportional to the mass and the
specific heat of the body.

The total amount of heat transfer in lumped systems

Once the temperature of the body at time t is calculated from Eq. (3), the total amount of heat transfer
between a body and its surroundings over a time interval t can be calculated from,
Q = m cP [T (t ) - Ti ]

MSE 321 (Fall 2013)

(4)

Peyman Taheri

where Ti is the initial temperature of the body at t = 0 .

Criterion for Lumped System Analysis

Lumped system approximation provides a great convenience in heat transfer analysis, but we should
know when it is appropriate to use. The first step is to define a characteristic length Lc ,
Lc =

V
A

(5)

and then define a dimensionless parameter, the Biot number Bi,

Bi =

h Lc
k

(6)

which also can be expressed as,

Bi =

h T
convection at the surface of the body
=
k T / Lc
conduction within the body

L /k
conduction resistance within the body
Bi = c
=
1/ h
convection resistance at the surface of the body

Note:

(7)

The Biot number is the ratio of the internal resistance (conduction) to the external resistance to
heat convection.

Lumped system analysis assumes a uniform temperature distribution throughout the body, which implies
that the conduction heat resistance is zero. Thus, the lumped system analysis is exact when Bi = 0 .
Note:

It is generally accepted that the lumped system analysis is applicable if,

Bi 0.1

(8)

Therefore, small bodies with high thermal conductivity are good candidates for lumped system
analysis.

Example
A thermocouple junction, which may be approximated by a sphere, is to be used for temperature
measurement in a gas stream. The convection heat transfer coefficient between the junction surface and
the gas is known to be h = 400 W/m2 K, and the junction thermophysical properties are k = 20 W/m K,

cP = 400 J/kg K, and = 8500 kg/m3. Determine the junction diameter needed for the thermocouple to
have a time constant of 1 s. If the junction is at 20C and is placed in a gas stream that is at 200C, how
long will it take for the junction to reach 199C?

MSE 321 (Fall 2013)

Peyman Taheri

Assumptions: (1) Temperature of the junction is uniform at any instant, i.e., lumped system analysis. (2)
Radiation is negligible. (3) Losses through the leads, by conduction, are negligible. (4) Constant
properties.

Solution:
To find the diameter of the junction D, we can use the time constant,
b=

h As
V cP

1=

400(4 r 2 )
(8500)(4 r 3 / 3)(400)

r = 0.353[mm]

Now, we can check the validity of the lumped system analysis. With Lc =

Bi =

h Lc (400)(0.000353 / 3)
=
= 2.3510-3 0.1
20
k

D = 0.706[mm]

V 4 r 3 / 3 r
=
= ,
A
4 r 2
3

Lumped analysis is applicable.

The time required for the junction to reach T = 199 C is,

T (t ) - T
Ti - T

= exp [-bt ]

where

b=

hAs
Vc P

Then,
199 - 200
= exp [-t ]
20 - 200

199 - 200

t = - ln
20 - 200

t = 5.2 s

Transient Conduction in Large Plane Walls, Long

Cylinders, and Spheres
As mentioned in the Introduction section, in transient mode, the temperature of a body essentially varies
with time as well as position, i.e., T = T ( x, y, z, t ) . In lumped system analysis we considered T = T (t ) ,
that means the temperature of the system is uniform (T does not depend on position). In this section, we
consider temperature to be dependent on position, as well as time, but we only consider one-dimensional

MSE 321 (Fall 2013)

Peyman Taheri

dependency of temperature to position, that is T = T ( x, t ) in Cartesian coordinate, and T = T (r, t ) in

cylindrical and spherical coordinates.
Consider a plane wall of thickness 2L , a long cylinder of radius r0 , and a sphere of radius r0 initially at a
uniform temperature Ti (Fig. 3).

Figure 3: Schematic of the simple geometries in which heat transfer is one-dimensional.

At time t = 0 , each geometry is placed in a large medium that is at a constant temperature T and kept in
that medium for t > 0 . Heat transfer takes place between these bodies and their environments by
convection with heat transfer coefficient h. Note that all three cases possess geometric and thermal
symmetry: the plane wall is symmetric about its center plane ( x = 0 ), the cylinder is symmetric about its
centerline ( r = 0 ), and the sphere is symmetric about its center point ( r = 0 ). We neglect radiation heat
transfer between these bodies and their surrounding surfaces.
The variation of the temperature profile with time in the plane wall is illustrated in Fig. 4. When the wall
is first exposed to the surrounding medium at T < Ti at t = 0 , the entire wall is at its initial temperature

Ti . But the wall temperature at and near the surfaces starts to drop as a result of heat transfer from the
wall to the surrounding medium. This creates a temperature gradient in the wall and initiates heat
conduction from the inner parts of the wall toward its outer surfaces. Note that the temperature at the
center of the wall remains at Ti until t = t2 , and that the temperature profile within the wall remains
symmetric at all times about the center plane. The temperature profile gets flatter and flatter as time
passes as a result of heat transfer, and eventually becomes uniform at T = T . That is, the wall reaches
thermal equilibrium with its surroundings. At that point, heat transfer stops since there is no longer a
temperature difference. Similar discussions can be given for the long cylinder or sphere.

MSE 321 (Fall 2013)

Peyman Taheri

Figure 4: Transient temperature profiles in a plane wall exposed to convection from its surfaces for Ti > T .

Governing equations for transient one-dimensional heat conduction

The governing equations that describe the above-mentioned one-dimensional transient temperature
distribution T ( x, t ) form a partial differential equation (PDE) system, which can be solved using
advanced mathematical methods. This PDE system for Cartesian geometry is,
2T
T
=
2
x
t

Equation:
Boundary condition 1:
Boundary condition 2:
Initial condition:

with

k
cP

T (0, t )
=0
x
T ( L , t )
-k
= h [T ( L, t ) - T ]
x
T ( x,0) = Ti
-k

(9)

where = k / cP is the thermal diffusivity coefficient.

It is more convenient to nondimensionalize the problem by defining dimensionless position X and
dimensionless temperature as,
X=

x
L

( x, t ) =

T ( x, t ) - T
Ti - T

(10)

Substituing x and T ( x, t ) from Eq. (10) into the PDE system in (9) gives,
Dimensionless equation:
Dimensionless boundary condition 1:
Dimensionless boundary condition 2:
Dimensionless initial condition:

MSE 321 (Fall 2013)

=
2
X

(0, )
=0
X
(1, )
= -Bi (1, )
X
( X ,0) = 1

(11)

Peyman Taheri

in which,

Bi =

hL
k

t
L2

= Fo

(12)

Note:

The dimensionless time is also called Fourier number.

Note:

Solution of the dimensional system [cf. Eq. (9)] becomes a function of several quantities,
T = T ( x, L, t, k , , h, Ti ) while the solution of dimensionless system [cf. Eq. (11)] becomes less
complicated, = ( X ,Bi,Fo) .

Exact solutions
The dimensionless solutions for one-dimensional transient conduction in a plane wall of thickness 2L , a
long cylinder of radius r0 , and a sphere of radius r0 subjected to convection from all surfaces are listed in
the following table.
Geometry

Solution

n s are the roots of

Plane wall

T ( x, t ) - T
4sin n
=
exp -n2 cos (n x / L)
Ti - T
n =1 2n + sin(2n )

n tan n = Bi

Cylinder

T (r , t ) - T
2
J1 (n )
=
exp -n2 J 0 (n r / r0 )
2
2
Ti - T
n =1 n J 0 (n ) + J 1 (n )

Sphere

4(sin n - n cos n )
sin (n r / r0 )
T (r , t ) - T
=
exp -n2

n r / r0
Ti - T
2
sin(2
)
n =1
n
n

1 - n cot n = Bi

Bi h L / k

J1 (n )
= Bi
J 0 (n )

or Bi h r0 / k

Fo t / L2 or Fo t / r02
J 0 and J1 are the Bessel functions of the first kind

Approximate solutions
The analytical solution listed in the above table involves infinite series which are difficult to evaluate.
Therefore, there is clear motivation to simplify the analytical solutions and to present the solutions in
tabular or graphical form using simple relations.
There are two approaches to approximate the solution:

MSE 321 (Fall 2013)

Peyman Taheri

1- Use the first term of the infinite series solution. This method is only valid for Fo > 0.2 .
2- Use the Heisler charts for each geometry as shown in Figs. 11-15, 11-16 and 11-17 of the
textbook.

Using the First Term Solution

The maximum error associated with method is less than 2%. For different geometries we have,

( x, t )wall =

T ( x, t ) - T
Ti - T

( x, t )cylinder =
( x, t )sphere =

= A1 exp -12 cos (1 x / L )

T ( r , t ) - T
Ti - T

T (r , t ) - T
Ti - T

= A1 exp -12 J 0 (1r / r0 )

= A1 exp -12

sin (1r / r0 )

(1r / r0 )

for

> 0.2

for

> 0.2

for

> 0.2

(13)

where A1 and 1 can be found from the follow table (Table 11-2 in the textbook).

MSE 321 (Fall 2013)

Peyman Taheri

Using Heisler Charts

There are three charts, Figs. 11-15 to 11-17, associated with different geometries. For a plate/wall
(Cartesian geometry) the Heisler chart is shown below.

The first chart is to determine the temperature at the center T0 at a given time.
By having the temperature at the center T0 at a given time, the second chart is to determine the
temperature at other locations at the same time in terms of T0 .
The third chart is to determine the total amount of heat transfer up to the time t.

MSE 321 (Fall 2013)

10