22 February 2024 Mechanical Engineering Lab [1]

Experiment #01
Name of the Experiment: To study the characteristics of different sensors:
Thermocouple, Thermistor, and Temperature sensor IC.
Objectives:
The purpose of this experiment is-
i) To describe the construction and operating principles of three different types of
temperature sensors: thermocouple, thermistor, and temperature sensor IC.
ii) To determine the characteristics of different temperature sensors.
Theory:
Temperature sensors are among the most commonly used sensors. All types of
equipment use temperature sensors, including computers, cars, kitchen appliances,
air conditioners, and (of course) home thermostats. The five most common types of
temperature sensors include:
1) Thermistor
2) Thermocouple
3) RTDs (resistive temperature detector)
4) Analog thermometer ICs
5) Digital thermometer ICs.
Thermocouple:
A Thermocouple is a sensor commonly used for measuring higher temperatures and
larger temperature ranges.
Any conductor subjected to a thermal gradient will generate a small voltage. This
phenomenon is known as the Seebeck effect. The Seebeck effect describes how a
temperature difference (ΔT) between the two ends of a metal wire can produce a
voltage difference (ΔV) across the length of the wire. This effect is characterized by the
equation
∆
𝑆 (1.1)
∆

where S denotes the Seebeck effect of the material. The magnitude of the generated
voltage is dependent upon the type of metal.
Practical applications of the Seebeck effect involve two dissimilar metals that are
joined at one end and separated at the other end.

Fig. 1.1: Structure of a thermocouple temperature sensor.

Thermocouples consist of two wire legs made from dissimilar metals. The wires legs
are welded together at one end, creating a junction. The junction is called the
measurement (or the hot) junction. The opposite end where the sensor connects to the
measurement system is called the reference (or the cold) junction. An isothermal

Prof. Sazzad M.S. Imran, PhD; Dept. of Electrical and Electronic Engineering
University of Dhaka; Dhaka-1000, Bangladesh; url- [Link]
 22 February 2024 Mechanical Engineering Lab [2]

block, which is made of a heat conducting material, is commonly required to keep the
leads of the thermocouple at the same temperature. When the hot junction experiences
a change in temperature, a voltage is created. The voltage can then be interpreted using
thermocouple reference tables to calculate temperature.
There are various types of thermocouples. Certain combinations of alloys have become
popular, and the desired combination is driven by variables including cost, availability,
chemical properties, and stability. Different types are best suited for different
applications, and they are commonly chosen based on the required temperature range
and sensitivity. Some common thermocouple types are
i) T (Copper-Constantan) – S = 40 μV/K at 298K (= 25 0C)
ii) J (Iron-Constantan) – S = 51 μV/K at 298K
iii) K (Chromel-Alumel) – S = 40 μV/K at 298K
iv) S (Platinum (10% Rhodium)-Platinum) – S = 11 μV/K at 298K
Fig. 1.2 shows a graph of characteristics for different thermocouple types.

Fig. 1.2: Characteristics for different thermocouple types.

Thermistor:
The thermistor (i.e., thermal resistor) is a temperature-sensing device whose
resistance strongly depends on temperature. A thermistor is a temperature-sensing
element composed of sintered semiconductor material and sometimes mixture of
metallic oxides such as Mn, Ni, Co, Cu and Fe, which exhibits a large change in
resistance proportional to a small change in temperature. In some materials the
resistance of thermistor at room temperature may change as much as 6% for 1°C
change in temperature. This high sensitivity to temperature change make the
thermistor extremely well suited to precision temperature measurement, control and
compensation. Therefore, especially in lower temperatures range of -100 0C to 300
0C, or to detect very small changes in temperature which cannot be observed with an

RTD or a thermocouple. Their resistance range from 0.5 Ω to 75 MΩ and they are
available in wide variety of shapes and sizes.
Pure metals have positive temperature coefficient of resistance, alloys have nearly
equal zero temperature coefficient of resistance and semiconductors have negative
temperature coefficient of resistance. Thermistors can be classified into two types.

Prof. Sazzad M.S. Imran, PhD; Dept. of Electrical and Electronic Engineering
University of Dhaka; Dhaka-1000, Bangladesh; url- [Link]
 22 February 2024 Mechanical Engineering Lab [3]

1) Positive temperature coefficient (PTC) thermistor: resistance increases with

increase in temperature.
2) Negative temperature coefficient (NTC) thermistor: resistance decreases with
increase in temperature, and this type seems to be the most commonly used
thermistor.
The thermistor exhibits a highly non-linear characteristic of resistance vs.
temperature.

Fig. 1.3: Characteristic graphs of resistance vs. temperature for PTC and NTC type
thermistors.
We can linearize a thermistor's behavior by placing a normal resistor in parallel with
the thermistor. If the value of the resistor is the same as that of the thermistor at room
temperature, the linearization region will be symmetrical around room temperature,
as shown in Fig. 1.4.

Fig. 1.4: Resistive mode linearization of thermistor’s behavior.

The standard equation for an NTC thermistor's resistance as a function of temperature
is given by
𝑅 𝑅 exp 𝛽 (1.2)

where, R is the resistance of thermistor at the temperature T (in K), R0 is the resistance
at given temperature T0 (in K), and β is the material specific-constant.

Prof. Sazzad M.S. Imran, PhD; Dept. of Electrical and Electronic Engineering
University of Dhaka; Dhaka-1000, Bangladesh; url- [Link]
 22 February 2024 Mechanical Engineering Lab [4]

The material specific-constant of an NTC thermistor is a measure of its resistance at

one temperature compared to its resistance at a different temperature. Its value may
be calculated by the formula
𝛽 (1.3)

and is expressed in degrees Kelvin (0K).

A graph plotted with logR–logR0 in Y axis and 1/T–1/T0 in X axis for NTC is shown in
Fig. 1.5. The slope of graph gives the value of β.

Fig. 1.5: logR–logR0 vs. 1/T–1/T0 graph for NTC thermistor.

Temperature coefficient of resistance is then found from the equation
𝛼 (1.4)
Temperature Sensor IC:
The thermistor and thermocouples have some significant limitations, e.g.
thermocouples have a low output signal which varies nonlinearly with temperature.
Thermistor has high sensitivity but they exhibit highly nonlinear resistance
temperature characteristics. For each of these transducers, electronic compensation
circuits have to be used in order to overcome their shortcomings. Also additional
circuitry may be needed to increase their voltage or current output.
Temperature sensor IC is a two terminal integrated circuit temperature transducer
that produces an output voltage or current proportional to absolute temperature. The
sensor package is small with a low thermal mass and a fast response time.

Fig. 1.6: Pin configuration of the IC LM35.

LM35 combines temperature sensing element (type K thermocouple) with signal
conditioning electronics to produce an IC containing 16 transistors, 9 resistors and 2
capacitors contained in a transistor type package. In contrast to a thermistor, this

Prof. Sazzad M.S. Imran, PhD; Dept. of Electrical and Electronic Engineering
University of Dhaka; Dhaka-1000, Bangladesh; url- [Link]
 22 February 2024 Mechanical Engineering Lab [5]

analog IC provides an almost linear output voltage; the slope is 10 mV/0C over a
temperature range of -55 to +150 0C with an accuracy of ±0.5 0C. Measurements of
output voltage therefore indicate the ambient temperature directly in degrees
centigrade e.g. at a temperature of 32 0C the output voltage will be 320 mV.
Apparatus:
1) Thermocouple (K type)
2) Thermistor (Rod type)
3) LM35 temperature sensor IC
4) Breadboard
5) Multimeter
6) Thermometer
7) Table lamp
8) +5V DC power supply
Working Diagrams:

Fig. 1.7: Experimental setup to study characteristics of thermocouple.

???
Fig. 1.8: Experimental setup to study characteristics of thermistor.
???
Fig. 1.9: Experimental setup to study characteristics of temperature sensor IC.
Data:
Table 1.1: Data get from thermocouple experiment.
Time Voltage Temperature Temperature Temperature
(minutes) Difference Hot Junction Reference Junction Difference (H-R)
(V) H (0C) R (0C) (0C)

Prof. Sazzad M.S. Imran, PhD; Dept. of Electrical and Electronic Engineering
University of Dhaka; Dhaka-1000, Bangladesh; url- [Link]
 22 February 2024 Mechanical Engineering Lab [6]

(a)

(b)
Fig. 1.10: Graphs for thermocouple. (a) Temperature difference vs. time and (b)
voltage vs. temperature difference.
Table 1.2: Data get from thermistor experiment.
Temperature Resistance Reference Reference 1 1 logR–logR0
T (oC) R (Ω) Temperature Resistance 𝑇 𝑇
T0 (oC) R0 (Ω)

Prof. Sazzad M.S. Imran, PhD; Dept. of Electrical and Electronic Engineering
University of Dhaka; Dhaka-1000, Bangladesh; url- [Link]
 22 February 2024 Mechanical Engineering Lab [7]

(a)

(b)
Fig. 1.11: Graph for NTC thermistor. (a) logR–logR0 vs. 1/T–1/T0 and (b) Resistance
vs. temperature.
Table 1.3: Data get from temperature sensor IC experiment.
Time Voltage Temperature Time Voltage Temperature
(minutes) (mV) (0C) (minutes) (mV) (0C)

(a)

Prof. Sazzad M.S. Imran, PhD; Dept. of Electrical and Electronic Engineering
University of Dhaka; Dhaka-1000, Bangladesh; url- [Link]
 22 February 2024 Mechanical Engineering Lab [8]

(b)
Fig. 1.10: Graphs for temperature sensor IC. (a) Temperature vs. time and (b) voltage
vs. temperature.
Calculation:
For thermocouple:
∆
Measured Seebeck coefficient 𝑆 =?
∆
Theoretical Seebeck coefficient of K type thermocouple, S = 40 μV/K at 298K
% of error = 100 = ?%
For thermistor:
Measured material constant of thermistor, 𝛽 =?

Measured temperature coefficient of thermistor, 𝛼 = ? (K-1)

Theoretical temperature coefficient of thermistor, α = ? (K-1)
% of error = 100 = ?%
For temperature sensor IC:
∆
Measured slope of the temperature sensor IC, = ? mV/0C
∆

Actual slope of the temperature sensor IC = mV/0C

% of error = 100 = ?%
Discussion:
Regular thermometers cannot withstand very low or very high temperatures. They
typically measure between -30 and 200 0C. Although thermometers that measure
significantly beyond this range do exist, they are usually quite expensive.
Thermocouples are inexpensive devices that allow us to measure system temperatures
in extreme conditions. For example, a K-type thermocouple probe can measure from
-100 to 1250 0C. The most likely source of error is from the two wires of thermocouples
touching another electrically conductive object. During use, thermocouples are
positively charged in the hot region and negatively charged in the cold region. And the
voltage is measured across the two ends of thermocouple to allow us to determine the

Prof. Sazzad M.S. Imran, PhD; Dept. of Electrical and Electronic Engineering
University of Dhaka; Dhaka-1000, Bangladesh; url- [Link]
 22 February 2024 Mechanical Engineering Lab [9]

temperature in test environment. If any of the thermocouple wire came into contact
with another conductive material, electrons would escape from the thermocouple and
give us inaccurate voltage reading. Therefore, it is crucial to make sure that
thermocouples do not come in contact with another electrically conductive material.
It is clearly observed that the time response of the temperature sensor IC shows a
linear behavior which confirm its linear sensitivity. Voltage increases directly
proportionately with increasing temperature. Characteristics of LM35 transducer
gives an idea of time scale required for the unit to reach stable conditions. Enough
space should be allowed around the temperature controller to ensure proper heat
dissipation. We should wire properly with correct polarity of terminals to ensure the
correct result.
References:
[1]
[2]

Prof. Sazzad M.S. Imran, PhD; Dept. of Electrical and Electronic Engineering
University of Dhaka; Dhaka-1000, Bangladesh; url- [Link]
22 February 2024 Mechanical Engineering Lab [10]

Experiment #02
Name of the Experiment: To measure the unknown length of a given aluminum
rod by determining the heat flux.
Objectives:
The purpose of this experiment is-
i) To determine the heat transfer rate of the metal (aluminum).
ii) To determine the length of an aluminum rod with an unknown length using heat
flux through an aluminum rod.
Theory:
Conduction is a process of heat transfer through solids. When a temperature gradient
exists in a body, there is a transfer of heat from the high temperature region to the low
temperature region. The heat transfer rate per unit area is proportional to the
temperature gradient given by
∆
∝ (2.1)
∆

where q is the heat transfer rate (watts), A is the area of heat transfer (m2), ΔT/Δx is
the temperature gradient in the direction of heat flow (0C/m). When the
proportionality constant is inserted, we get
∆
𝑄 𝑘 (2.2)
∆

Here, Q is the heat flux in Watts per meter squared. Q is described as the amount of
energy (in joules) which passes through a given surface over some unit of time. Using
the heat flux of an aluminum rod of known length, a prediction can be made of the
length of another aluminum rod of unknown length.
The positive constant k is called the coefficient of thermal conductivity of the material.
The negative sign indicates that heat transfer takes place in the direction of decreasing
temperature. Coefficient of thermal conductivity has the units of watts/m0C. The
numerical value of the coefficient of thermal conductivity indicates how fast heat will
flow in a given material. Thermal conductivity coefficient is a physical property of the
material. Although it is fairly constant in a narrow temperature range, it varies over a
wide temperature range.
Metals which are good conductors of heat have high values of coefficient of thermal
conductivity; for example, 385 watts/m0C for copper. Insulating materials have low
values of coefficient of thermal conductivity, for example 0.048 watts/m0C for fiber
insulating board. In any conduction heat transfer problem, it is essential to have the
knowledge of coefficient of thermal conductivity of the material involved in the heat
transfer process.
Apparatus:
1) 2 Aluminum rods of same diameter, one with unknown length and one with
known length (= ? cm)
2) 2 Styrofoam insulation blocks
3) Hot plate with heat source
4) Thermometer (digital and alcohol)
5) Thermocouple
6) Beaker (500 ml)

Prof. Sazzad M.S. Imran, PhD; Dept. of Electrical and Electronic Engineering
University of Dhaka; Dhaka-1000, Bangladesh; url- [Link]
 22 February 2024 Mechanical Engineering Lab [11]

7) Lab stand
8) Stopwatch
9) Ruler and slide caliper
10) Water
Working Diagrams:

Fig. 2.1: Experimental setup to measure coefficient of thermal conductivity of an

aluminum rod of known length.
Procedure:
Phase-1:
1. Pour enough water to fill three quarters of your beaker.
2. Measure the diameter and length of the aluminum rod and pierce a hole in the
Styrofoam block ensuring that the aluminum rod fits tightly within it.
3. Ensure 2-3mm of aluminum rod is exposed on one end of the Styrofoam block.
4. Attach the thermocouple to the free end of the aluminum rod.
5. Place the aluminum rod and Styrofoam block into the hot water bath and record
the temperature of the hot water bath submerging the exposed rod and 2-3mm of
the Styrofoam block.
6. Ensure the block is clamped to the lab stand in order to avoid falling of moving.
7. Heat the water bath for 5 minutes and record the temperature.
8. Using the temperature acquired calculate dT = T1 – T2, where T1 is the initial
temperature and T2 is the final temperature.
9. Continue steps 7 and 8 and calculate ~5 different dT values.
10. Calculate Q using the thermal conductivity constant (k), average temperature
difference (T1 – T2) and the length of the aluminum rod (Δx).
Phase-2:
1. Repeat Phase-1 using the second rod of unknown length (ensure the same area of
aluminum rod is exposed to the hot water bath)
2. Acquire ~5 new T1 – T2 and calculate their average dT.

Prof. Sazzad M.S. Imran, PhD; Dept. of Electrical and Electronic Engineering
University of Dhaka; Dhaka-1000, Bangladesh; url- [Link]
22 February 2024 Mechanical Engineering Lab [12]

3. Using the previously calculated heat flux, thermal conductivity constant and new
T1 – T2 calculate the length of the aluminum rod with the heat flux equation
rearranged for Δx.
4. Measure the actual length of the aluminum rod using a ruler and compare the
results.
Data:
Table 2.1: Data for aluminum rod of known length.
Length of the rod: ? m
t (min) T1 (oC) T2 (oC) dT = T1 – T2 (oC)
0
5
10
15
20
25
Table 2.2: Data for aluminum rod of unknown length.
t (min) T1 (oC) T2 (oC) dT = T1 – T2 (oC)
0
5
10
15
20
25
Calculation:
From Table-1 (rod with known length):
∑
Average ? 0C = ? K

Area of the rod 𝜋𝑑 ? m2

Coefficient of thermal conductivity k = 237 W.m-1.K-1
∆
Heat transfer rate 𝑘𝐴 ?W
∆

From Table-2 (rod with unknown length):

∑
Average ? 0C = ? K

Area of the rod 𝜋𝑑 ? m2

Coefficient of thermal conductivity k = 237 W.m-1.K-1

Prof. Sazzad M.S. Imran, PhD; Dept. of Electrical and Electronic Engineering
University of Dhaka; Dhaka-1000, Bangladesh; url- [Link]
 22 February 2024 Mechanical Engineering Lab [13]

∆
Length of the rod ∆𝑥 𝑘𝐴 ? m (= calculated length)

Measured length of the rod using ruler = ? m

% of error = 100 = ?%
Discussion:
For the heat flux to be considered the same for both rods, the exact same procedure
must be followed. Leaving the rod in the hot water bath for longer than the other rod
will decrease the heat flux via an increased T2. The same is also true for the opposite,
should the one rod be left in the hot water bath for a shorter period, the T2 will be
smaller and thus the heat flux greater. An insulator was used to prevent heat loss from
the aluminum rod.
The equation employed in this experiment work under the assumption that the system
is defined as steady state. Therefore, on small scale (rods smaller than ~30 cm and
differing less than ~10 cm) the calculations will hold. More precise instrumentation
should be used if larger scale experiments were to be performed.
Heat flux is a measure of energy passing through an area over a given time regardless
of the length of that material. Should the length of a rod increase in length, the
temperature at the end of that material will decrease and vice versa if exposed to the
same amount of heat.
References:
[1]
[2]

Prof. Sazzad M.S. Imran, PhD; Dept. of Electrical and Electronic Engineering
University of Dhaka; Dhaka-1000, Bangladesh; url- [Link]
 22 February 2024 Mechanical Engineering Lab [14]

Experiment #03
Name of the Experiment: To determine the mean emissivity of a copper/
aluminum plate at different temperatures.
Objectives:
The purpose of this experiment is-
i) To measure the property of emissivity of the test plate surface (copper/aluminum)
at various temperature.
Theory:
All substances at all temperature emit thermal radiation. Thermal radiation is an
electromagnetic wave and does not require any material medium for propagation. All
bodies can emit radiation and have also the capacity to absorb all or a part of the
radiation coming from the surrounding towards it.
When thermal radiation strikes a body, part of it is reflected, part of it is absorbed and
part of it is transmitted through body. The fraction of incident energy reflected by the
surface is called reflectivity (ρ). The fraction of incident energy absorbed by surface is
called absorptivity (α) and the fraction of energy transmitted is called transmissivity
(τ).
An ideal black surface is one which absorbs all the incident radiation with reflectivity
and transmissivity equal to zero. The radiant energy per unit time per unit area from
the surface of the body is called as the emissive power and is usually denoted by e. The
emissivity of the surface is the ratio of the emissive power of the surface to the emissive
power of a black surface at the same temperature. It is denoted by ε. Thus,
𝜀 (3.1)

For black body, absorptivity α = 1 and by the knowledge of Kirchhoff’s Law (at thermal
equilibrium, the power radiated by an object must be equal to the power absorbed),
emissivity of the black body becomes unity.
It is obvious from the Stefan Boltzmann’s Law (the radiant heat energy emitted from
a unit area of the black body in one second (ε) is directly proportional to the fourth
power of its absolute temperature, or ε = σ × T4) that the prediction of emissive power
of a surface requires knowledge about the values of its emissivity and therefore we aim
to measure the values of emissivity as function of surface temperature.
Any hot body maintained by a constant heat source, loses heat to surroundings by
conduction, convection and radiation. If two bodies made of same geometry are heated
under identical conditions, the heat loss by conduction and convection can be assumed
same for both the bodies, when the difference in temperatures between these two
bodies is not high. In such a case, when one body is black and the other body is grey
from the values of different surface temperatures of the two bodies maintained by a
constant power source emissivity can be calculated. The heat loss by radiation depends
on
a) Characteristic of the material
b) Geometry of the surface and
c) Temperature of the surface

Prof. Sazzad M.S. Imran, PhD; Dept. of Electrical and Electronic Engineering
University of Dhaka; Dhaka-1000, Bangladesh; url- [Link]
 22 February 2024 Mechanical Engineering Lab [15]

The heat loss by radiation when one body is completely enclosed by the other body is
given by

𝑄 3.2

If a body is losing heat to the surrounding atmosphere, then

area of atmosphere A2 >> area of body A1 (3.3)
Thus if anybody is losing heat by radiation to the surrounding atmosphere Eqn. (3.2)
takes the form
𝑄 𝜎𝐴 𝑇 𝑇 3.4
where σ = Stefan Boltzmann constant = 5.6697×10-8 W/m²K4
A1 = Surface area in m²
ε = Emissivity
T1 = surface temperature of the body in K and
T2 = surrounding atmospheric temperature in K
Let us consider a black body and a grey body with identical geometry being heated
under identical conditions, assuming conduction and convection heat loss to remain
the same.
Let 𝑄𝑏 and 𝑄𝑔 be the heat supplied to black and grey bodies, respectively. If heat input
to both the bodies are same, 𝑄𝑏 = 𝑄𝑔.
Assuming, heat loss by conduction and convection from both bodies to remain same.
Heat loss by radiation by the black body = Heat loss by radiation by the grey body
𝜎𝐴 𝜀 𝑇 𝑇 𝜎𝐴 𝜀 𝑇 𝑇 (3.5)
As geometry of two bodies are identical 𝐴 = 𝐴𝑔 = 𝐴𝑏 and 𝜀𝑏 = 1 for black body.
Therefore,

𝜀 (3.6)

where Suffix ‘b’ stands for black body,

Suffix ‘g’ stands for grey body,
Suffix ‘c’ stands for chamber.
Experimental Setup:
The experimental set up consists of two aluminum/copper plates of identical physical
dimensions. One of the plates is made black by applying a thick layer of lamp black
while the other plate whose emissivity is to be measured is a grey body. Heating coils
are provided at the bottom of the plates. The plates are mounted on brackets and are
kept in an enclosure so as to provide undisturbed natural convection surroundings.
The temperatures of the plates are measured by thermocouples. Separate wires are
connected to diametrically opposite points to get the average surface temperature of
the plates. Another thermocouple is kept in the enclosure to read the ambient
temperature of enclosure. The heat input can be varied with the help of variac for both
the plates that can be measured using digital voltmeter and ammeter.

Prof. Sazzad M.S. Imran, PhD; Dept. of Electrical and Electronic Engineering
University of Dhaka; Dhaka-1000, Bangladesh; url- [Link]
 22 February 2024 Mechanical Engineering Lab [16]

Apparatus:
1) 2 Specimen: 1 black plate square and 1 test plate square
2) Specimen size: ? cm × ? cm
3) 2 Digital voltmeter: 0-300 volts AC
4) 2 Digital ammeter: 0-3 amps AC
5) 2 Dimmer stats: 0-240 volts, 2 amps
6) 3 Temperature indicator: Digital type, 0-300 0C
7) 3 Thermocouples: K type
8) 2 Heaters: Sandwiched type Nichrome heater, 400 W
9) Enclosure size: 60 cm × 60 cm × ? cm
Working Diagrams:

Fig. 3.1: Experimental setup to measure emissivity of a test plate.

Procedure:
1) Gradually increase the input to the heater to black plate and adjust it to some value
viz. 30, 50, 75 W.
2) Adjust the heater input to test plate slightly less than the black plate viz. 27, 35, 55
W etc.
3) Check the temperature of the two plates with small time intervals and adjust the
input of test plate only, by the dimmer stat so that the two plates will be maintained
at the same temperature.
4) This will require some trial and error, and one has to wait sufficiently to obtain the
steady state condition.
5) After attaining the steady state condition record the temperature, voltmeter and
ammeter readings for both the plates.
6) The same procedure is repeated for various surface temperature in increasing
order.
Data:
Table 3.1: Temperature, voltmeter and ammeter readings for various surface
temperature.

Prof. Sazzad M.S. Imran, PhD; Dept. of Electrical and Electronic Engineering
University of Dhaka; Dhaka-1000, Bangladesh; url- [Link]
 22 February 2024 Mechanical Engineering Lab [17]

Sr. Black Plate Test Plate Enclosure Temp

No. Vb (V) Ib (A) Tb (0C) Vt (V) It (A) Tt (0C) Te (0C)

1
2
3

Calculation:
From the Stefan Boltzmann’s Law
𝑊 𝑊 𝜀 𝜀 𝜎𝐴 𝑇 𝑇
1 𝜀 𝜎𝐴 𝑇 𝑇 (3.7)
where σ = Stefan Boltzmann constant = 5.6697×10-8 W/m²K4
A = Area of disc/plate = ? m²
at ? 0C
Wb = Heat input to body coated with lamp black = Vb Ib = ? Watts
Wt = Heat input to test body = Vt It = ? Watts
Tt = plate temperature = ? 0C + 273.15 = ? K
Te = enclosure temperature = ? 0C + 273.15 = ? K
𝜀 1 ?

at ? 0C
Wb = Vb Ib = ? Watts
Wt = Vt It = ? Watts
Tt = ? 0C + 273.15 = ? K
Te = ? 0C + 273.15 = ? K
𝜀 1 ?

at ? 0C
Wb = Vb Ib = ? Watts
Wt = Vt It = ? Watts
Tt = ? 0C + 273.15 = ? K
Te = ? 0C + 273.15 = ? K
𝜀 1 ?
∑
Mean emissivity = 𝜀mean
Discussion:
We should preferably use stabilized AC single phase supply. We need to use proper
voltage range on Voltmeter and gradually increase the heater inputs. The black plate

Prof. Sazzad M.S. Imran, PhD; Dept. of Electrical and Electronic Engineering
University of Dhaka; Dhaka-1000, Bangladesh; url- [Link]
 22 February 2024 Mechanical Engineering Lab [18]

should have a layer of lamp black uniformly. We should never put our hands or papers
at the enclosure. We need to keep at least 200 mm distance between the back side of
unit and the wall.
There is a possibility of getting absurd results if the supply voltage is fluctuating or if
the input is not adjusted till the satisfactory steady state condition reached.
References:
[1]
[2]

Prof. Sazzad M.S. Imran, PhD; Dept. of Electrical and Electronic Engineering
University of Dhaka; Dhaka-1000, Bangladesh; url- [Link]
 22 February 2024 Mechanical Engineering Lab [19]

Experiment #05
Name of the Experiment: To determine the linear expansion of
steel/copper/aluminum/brass as a function of temperature using a dilatometer.
Objectives:
The purpose of this experiment is-
i) To determine and analyze the coefficients of thermal expansion α of different
materials (brass, steel, copper, and aluminum) using dilatometer.
Theory:
Thermal expansion is the tendency of matter to change in volume in response to a
change in temperature through transfer of heat. When a substance is heated, the
kinetic energy of its atoms increases. Consequently, the atoms start to vibrate and
move to a greater average separation. In other words, when temperature increases, the
average interatomic separation between atoms increases, and the material expands.
Thermal expansion is defined as a material property that is indicative of the extent to
which a material expands upon heating, and has units of reciprocal temperature.

Fig. 5.1: Potential energy versus interatomic separation for materials (a) with and (b)
without asymmetry in the potential well.
The origin of thermal expansion is the irregularity of the interatomic potential energy
(U) vs. interatomic separation (r) curve as shown in Fig. 5.1(a). The equilibrium
interatomic separation at 0 K occurs near to the minimum of the U vs. r curve. When
heat is transferred to the material, interatomic vibrations above the ground state (E1)
are excited. Because the asymmetry of the interatomic potential energy curve is more
distinct for larger interatomic distances, the average interatomic distance increases for
excited energy levels. However, a symmetrical potential energy curve would result in
no thermal expansion due to the fact that there is no change in the average interatomic
separation as temperature (energy) increases, as demonstrated in Fig. 5.1(b).
This behavior can be defined numerically as the coefficient of thermal expansion, α
(oC-1 or K-1), which is the stress-free strain per unit temperature change. The coefficient
of thermal expansion is defined as the degree of expansion divided by the change in
temperature and describes how the size of an object changes with a change in
temperature. The change in length of a material is thus given by
Δl/lo = αΔT (5.1)
where lo is the initial length, Δl is the change in length and ΔT is the change in
temperature.

Prof. Sazzad M.S. Imran, PhD; Dept. of Electrical and Electronic Engineering
University of Dhaka; Dhaka-1000, Bangladesh; url- [Link]
22 February 2024 Mechanical Engineering Lab [20]

Since thermal expansion of a material depends upon its interatomic bonding, the
coefficient of thermal expansion depends on the class of material. Linear and branched
polymers exhibit the largest α ≈ 50-400×10-6 0C-1, because they have weak secondary
bonds. Most metals have intermediate bond strengths, giving α ≈ 5-25×10-6 0C-1. In
general, ceramics exhibit strong interatomic bonding, and thus have relatively low α ≈
0.5-15×10-6 0C-1.
Thermal expansion can be measured by three main techniques.
i) optical interference,
ii) thermomechanical analysis, and
iii) mechanical dilatometer.
Although optical interference has the best precision and accuracy, it is limited in its
temperature range, which is less than that of the mechanical dilatometer (-150 to
2,000 0C). Thermomechanical analysis is to some extent inconsistent compared to the
other techniques. With an expansion dilatometer, the sample length is measured as a
function of temperature.
Apparatus:
1) Dilatometer
2) Dial Vernier caliper
3) Metallic tubes: copper, aluminum, steel and brass
4) Water bath system
5) Clamps and stands for holding tubes
6) Immersion thermostat
7) External hot water circulator set
8) Thermometer
Working Diagrams:

Fig. 5.1: Experimental setup for measuring thermal expansion.

Procedure:
1) Measure initial length of each sample using a dial Vernier caliper.
2) Clump the first tube over its whole length (~600mm) in the dilatometer and
connect it to the circulating pump of the thermostat via rubber tubing.
3) Set the temperature with the accuracy of the thermostatic control.

Prof. Sazzad M.S. Imran, PhD; Dept. of Electrical and Electronic Engineering
University of Dhaka; Dhaka-1000, Bangladesh; url- [Link]
22 February 2024 Mechanical Engineering Lab [21]

4) After the initial temperature (~25 0C) has been reached set the length measuring
meter (measuring gauge) to zero.
5) Then raise the temperature in ten steps, each of 0.5 0C to 1 0C.
6) Record the respective changes in length and corresponding temperature.
7) Repeat this measurement series with each of the remaining four materials.
Data:
Table 5.1: Length expansion data at different temperature for different metals.
Aluminum Brass Copper Steel
l0 = ? mm l0 = ? mm l0 = ? mm l0 = ? mm
T (0C) Δl (mm) T (0C) Δl (mm) T (0C) Δl (mm) T (0C) Δl (mm)

Fig. 5.2. The linear expansion as a function of temperature for aluminum, brass,
copper and steel [this is just an example, draw the figure according to the data in
Table 5.1].
Calculation:
From Eqn. (5.1), the coefficient of thermal expansion

Prof. Sazzad M.S. Imran, PhD; Dept. of Electrical and Electronic Engineering
University of Dhaka; Dhaka-1000, Bangladesh; url- [Link]
22 February 2024 Mechanical Engineering Lab [22]

∆
𝛼 (5.2)
∆

For aluminum,
∆
= slope of the curve = ?
∆

𝛼aluminum ? ? 0C-1
?
?
% of error = 100 100 = ?%
For brass,
∆
= slope of the curve = ?
∆

𝛼brass ? ? 0C-1
?
?
% of error = 100 100 = ?%
For copper,
∆
= slope of the curve = ?
∆

𝛼copper ? ? 0C-1
?
?
% of error = 100 100 = ?%
For steel,
∆
= slope of the curve = ?
∆

𝛼steel ? ? 0C-1
?
?
% of error = 100 100 = ?%
Discussion:
The experiment suggested that harder materials such as steel when compared to pure
metals such as aluminum are more likely to have lower thermal expansion.
Furthermore, changes in material’s length or its dimension can be sometimes noticed
but most of the time cannot be seen directly by the naked eye due to lower value of the
coefficient of thermal expansion. For that reason, knowledge in thermal expansion
must be learned by engineers and other industries to avoid its harmful effects. That is
why, knowledge in determining the coefficient of linear expansion is necessary. This
area is most applicable in engineering practices especially in testing the materials used
in the construction of buildings, bridges, highways and even railroads.
Coefficient of linear expansion need not to be of very large value since alteration in the
dimension of the object cannot be directly observed. Hence, when computing for the
coefficient of linear expansion, it is expected to have a lower value. The experimental
value of the coefficient of linear expansion of the four rods, namely, aluminum, copper,
steel and brass metal tubes are acceptable as these results mostly agree with the
reported values found in literature.
Sources of the inaccuracy (percentage of error in measuring the coefficient of linear
expansion) in the calculation might depend on several things. Aside from the

Prof. Sazzad M.S. Imran, PhD; Dept. of Electrical and Electronic Engineering
University of Dhaka; Dhaka-1000, Bangladesh; url- [Link]
22 February 2024 Mechanical Engineering Lab [23]

coefficient of linear expansion there are other factors that affect the change in length
of a material in thermal expansion and these are the initial length of the body and the
change in the temperature.
A little discrepancy in measurements directly affects the whole experiment. The
problem might exist while analyzing the rotation of the dial gauge is somewhat
confusing. To achieve a more consistent result, it is better to perform this experiment
in a room where the temperature does not alter. Since temperature plays an important
role in this activity, an inaccurate reading can control the overall manipulation of the
data and information collected.
References:
[1]
[2]

Prof. Sazzad M.S. Imran, PhD; Dept. of Electrical and Electronic Engineering
University of Dhaka; Dhaka-1000, Bangladesh; url- [Link]