04 Temperature Sensors

MSE 355
Sensors, Measurements and Data

Acquisition System
Assoc. Prof. Dr. Mohamed Atef Ismail Kamel
moatef@msa.edu.eg
MSA University - Faculty of Engineering

Mechatronics Systems Engineering (MSE) Program
Fall Semester, 2023

Assoc. Prof. Dr. M. A. Kamel MSE 355 Year 3 – Fall Semester 1 / 33
Temperature Sensors

Outlines
1 Introduction
2 Thermoelectric Sensors
3 Thermoresistive Sensors
4 Q&A and Discussion
5 Recommended Readings and Practice Problems

Introduction
Outlines
1 Introduction

Introduction
Historical Background
Temperature sensors are one of the oldest sensors in use.

Early thermometers were introduced in the early 1600s.
Around the middle of 1600s, Robert Boyle mentioned the need for temperature
standards.
By 1742, most of temperature scale including the Celsius scale were established.
By 1848, Lord Kelvin proposed the absolute scale bearing his name.
Temperature scales were further developed and improved until the establishment of the
International Practical Temperature Scale in 1927.
The first temperature sensor is the classical thermometer.
Introduction
In 1821, the German scientist Thomas Johann Seebeck discovered a phenomenon

that named after him: the Seebeck Effect.
Seebeck Effect: when two wires made from dissimilar metals are joined at two ends
to form a loop, and if the two junctions are maintained at different temperatures, a
compass needle will move.
https://3-e-learning.blogspot.com/2013/04/thomas-johann-seebeck-discoverer-of.html
Introduction
Based on the Seebeck Effect, in 1826, the French

scientist Antoine Cesar Becquerel developed
the first temperature sensor called the thermo-
couple.
Idea: when two wires made from dissimilar met-

als are joined at two ends to form a loop, and if
the two junctions are maintained at different tem-
peratures, a voltage develops in the circuit.

Introduction
In 1834, the french scientist Charles Athanase

Peltier discovered a phenomenon that named af-
ter him: the Peltier Effect.
Peltier Effect: It is the reverse phenomenon

of the Seebeck effect. The electrical current
flowing through the junction connecting two ma-
terials will emit or absorb heat per unit time at
the junction.
This principle more often used for cooling or heat-

ing as well as for thermoelectric power generation.
Introduction
In 1871, the German–British scientist William
Siemens invented a method for temperature mea-
surement based on the resistance–temperature re-
lation in platinum.
This invention has become the basis of the ther-
moresistive sensors, on which resistance tem-
perature detectors (RTDs) are based.

Introduction
Types of Temperature Sensors

Thermoelectric Sensors
Outlines
1 Introduction

Introduction
They are among the oldest sensors (have been used
about 150 years ago).
They are

Introduction
They are among the oldest sensors (have been used
about 150 years ago).
They are passive sensors.
Basically, thermocouple is mode of two different
metals connected at the measurement end.
If many thermocouples are connected in series,
they are called a thermopile.

Basic Thermocouple
It is based on the Seebeck Effect.
Seebeck Effect: When two wires made from dis-

similar metals are joined at two ends to form a
loop, and if the two junctions are maintained at
different temperatures, a voltage develops in the
circuit.
This voltage can be used to measure temperature.
Note that, this voltage is too small. So, it has to

be amplified before interfacing.
Basic Analysis of a Basic Thermocouple
Assume two junctions at different temperature T1 and T2 .

The Seebeck emf across each conductor:
emfa = αa (T2 − T1 ) emfb = αb (T2 − T1 )
αa and αb are the Seebeck coefficients (µV/◦ C).

The thermometric emf generated by the thermocouple:
emfT = emfa − emfb = (αa − αb )(T2 − T1 ) = αab (T2 − T1 )
αab is the relative Seebeck coefficient of the material combination a and b.

Examples of Seebeck Coefficients

Types of Thermocouples
There are many types of thermocouples, the most common are:
Type-K (Chromel – Alumel): The most common general purpose thermocouple.
It ranges from -200◦ C to 1260◦ C with sensitivity of 41µV/◦ C.
Type-J (Iron – Constantan): Ranges from -40◦ C to 750◦ C with sensitivity of
55µV/◦ C.
Type-T (Copper – Constantan): A very stable thermocouple at low temperature
ranges. It ranges from -200◦ C to 350◦ C with sensitivity of 43µV/◦ C.
Type-E (Chromel – Constantan): It gives the highest measurement sensitivity of
68µV/◦ C, with range from -200◦ C to 900◦ C.
Type-N (Nicrosil – Nisil): It is the developed version of Type-K, with longer sta-
bility and life span.
Applications of Thermocouples
Industrial Process Control: Temperature control in heat treatment of metals, chem-
ical reactions, and food production.
HVAC Systems: Flame sensor in a gas water heater.
Automotive Industry: Monitoring the exhaust gas temperatures.
Medical Devices: Temperature measurement of MRI machines.
Oil and Gas Industry: Temperature measurement during drilling operations.

Advantages and Limitations of Thermocouples
Advantages: Disadvantages:
Quick response time. Low accuracy.
Durable. not perfectly linear.
Self-powered (Passive sensor). susceptible to drift over time (sensitive to
Simple and inexpensive. noise).
Wide range.

Thermoresistive Sensors
Outlines
1 Introduction

Introduction
Principle: A change in temperature causes electrical resistance of a material to change.

They are divided into two basic types: resistance temperature detectors (RTDs)
and thermistors.
RTDs: Thermoresistive sensors based on solid conductors, usually in the form of metal
wires or films. They have a positive temperature coefficient (PTC) of resistance.
Thermistors: Semiconductor-based devices that usually have a negative temperature
coefficient (NTC) of resistance.
PTC: Materials that experience an increase in electrical resistance when their tem-
perature is raised.
NTC: Materials that experience a decrease in electrical resistance when their tem-
perature is raised.
Resistance Temperature Detectors (RTDs)
All RTDs are based on the change in resistance due to

the temperature coefficient of resistance of the metal
being used.
The resistance of a conductor of length L with constant
cross-sectional area S and conductivity σ is:
L
R=
σS
The conductivity of the material itself is temperature dependent, and given as:
σ0
σ=
1 + α [T − T0 ]
α is the temperature coefficient of resistance of the con-

ductor.
T is the temperature.
σis the conductivity of the conductor at the reference
temperature T0 .
T0 is usually given at 20◦ C.
The resistance can be reformulated as:
L
R(T ) = (1 + α [T − T0 ]) = R0 (1 + α [T − T0 ])
σ0 S

Conductivity and temperature coefficients of resistance for selected materials

(at 20◦ C unless otherwise indicated)
Example: A temperature sensor made of copper wire with a length of 5 m and a diameter
0.2 mm. The proposed range is between -45◦ C and 10◦ C. A milliammeter is used to display
the temperature by connecting the sensor directly to a 1.5 V battery and measuring the
current through it.
Calculate the resistance of the sensor and the corresponding currents at the minimum
and maximum temperatures.
Calculate the maximum power the sensor dissipates.

Example
Step 1: Given:
L l = 5 m.
R(T ) = (1 + α [T − T0 ])
σ0 S d = 0.2 mm
Material: Copper.
Range: −45◦ C to 10◦ C.
V = 1.5 Volts.

Example
Step 1: Given:
L l = 5 m.
R(T ) = (1 + α [T − T0 ])
σ0 S d = 0.2 mm
Step 2: Get the characteristics of copper: Material: Copper.
σ0 = 5.8 × 107 S/m at T0 = 20◦ C. Range: −45◦ C to 10◦ C.
α = 0.0039/◦ C. V = 1.5 Volts.

Example
Step 1: Given:
L l = 5 m.
R(T ) = (1 + α [T − T0 ])
σ0 S d = 0.2 mm
Step 2: Get the characteristics of copper: Material: Copper.
σ0 = 5.8 × 107 S/m at T0 = 20◦ C. Range: −45◦ C to 10◦ C.
α = 0.0039/◦ C. V = 1.5 Volts.
Step 3: Get the resistance at the minimum and
maximum temperatures:
5
R−45 = (1 + 0.0039 [−45 − 20]) = 2.04 Ω.
5.8 × 107 × π × (0.1 × 10−3 )2
5
R10 = (1 + 0.0039 [10 − 20]) = 2.63 Ω.
5.8 × 107 × π × (0.1 × 10−3 )2
Example
Step 4: Get the currents: Given:
l = 5 m.
1.5 1.5
I−45 = = = 0.732A d = 0.2 mm
R(−45) 2.04
Material: Copper.
1.5 1.5 Range: −45◦ C to 10◦ C.
I10 = = = 0.569A
R(10) 2.63 V = 1.5 Volts.
R(−45) = 2.04 Ω.
R(10) = 2.63 Ω.

Example
Step 4: Get the currents: Given:
l = 5 m.
1.5 1.5
I−45 = = = 0.732A d = 0.2 mm
R(−45) 2.04
Material: Copper.
1.5 1.5 Range: −45◦ C to 10◦ C.
I10 = = = 0.569A
R(10) 2.63 V = 1.5 Volts.
Step 5: Get the dissipated power: R(−45) = 2.04 Ω.
R(10) = 2.63 Ω.
P−45 = V × I(−45) = 1.098W
P10 = V × I(10) = 0.853W

Self-Heat of RTDs
RTDs are very much subject to errors due to increases in their own temperature pro-
duced by the heat generated in them by the current used to measure their resistance.
The rise in temperature may be understood from the fact that the higher the current
in the sensor, the larger the output signal available.
On the other hand, power dissipated in the conductor is proportional to the square of
the current (P = I 2 R). This power can raise the temperature of the sensor, introducing
an error.
Typically, as part of the sensor specifications, the temperature increase per unit power
(◦ C/mW in most cases) is given by the manufacturer, allowing the designer to com-
pensate for these errors in the reading of the sensor.
Example
Example: Consider the self-heat of an RTD operating in the range −200◦ C to 850◦ C that
has a nominal resistance of 100 Ω at 0◦ C and a temperature coefficient of resistance of
0.00385/◦ C. Its self-heat is provided in its data sheet as 0.08◦ C/mW in air (typically this
value is given at a low airspeed of 1 m/s). Calculate the maximum error expected due to
self-heat if:
The resistance is measured by applying a constant voltage of 0.1 V across the sensor.
The resistance is measured by applying a constant current of 1 mA through the sensor.

Example
Step 1: At constant voltage source: Given:
α = 0.00385/◦ C.
R(T ) = R0 (1 + α [T − T0 ]) R0 = 100 Ω at 0◦ C.
Range: −200◦ C to 850◦ C.

Example
Step 1: At constant voltage source: Given:
α = 0.00385/◦ C.
R(T ) = R0 (1 + α [T − T0 ]) R0 = 100 Ω at 0◦ C.
Range: −200◦ C to 850◦ C.
Step 2: Get the resistance at the minimum and
maximum temperatures:
R−200 = 100 × (1 + 0.00385 [−200 − 0]) = 23 Ω.
R850 = 100 × (1 + 0.00385 [850 − 0]) = 427.25 Ω.

Example
Step 3: Get the dissipated power for a constant Given:
voltage source (0.1 Volts): α = 0.00385/◦ C.
R0 = 100 Ω at 0◦ C.
V 2
P−200 = = 0.435 mW Range: −200◦ C to 850◦ C.
R−200
V2
P850 = = 0.0234 mW
R850

Example
voltage source (0.1 Volts): α = 0.00385/◦ C.
R0 = 100 Ω at 0◦ C.
V 2
P−200 = = 0.435 mW Range: −200◦ C to 850◦ C.
R−200
V2
P850 = = 0.0234 mW
R850
Step 4: Get the error:
0.435
Error at − 200◦ = = 5.435◦ C
0.08
0.0234
Error at 850◦ = = 0.292◦ C
0.08
Example
current source (I = 1 mA): α = 0.00385/◦ C
R0 = 100 Ω at 0◦ C
P−200 = I 2 R−200 = 0.023 mW Range: −200◦ C to 850◦ C.
V = 0.1 Volts
P850 = I 2 R850 = 0.427 mW

Example
current source (I = 1 mA): α = 0.00385/◦ C
R0 = 100 Ω at 0◦ C
P−200 = I 2 R−200 = 0.023 mW Range: −200◦ C to 850◦ C.
V = 0.1 Volts
P850 = I 2 R850 = 0.427 mW
Step 6: Get the error:
0.023
Error at − 200◦ = = 0.2875◦ C
0.08
0.427
Error at 850◦ = = 5.341◦ C
0.08
Configuration of the RTD Sensing element
Wire-Wound RTD:
A fine wire which is wound around a ceramic or glass core,
and a protective sheath (enclosure).
The wire used in wire-wound RTDs is of high purity to en-
sure accurate and stable temperature measurements.
The wire should be very thin (less than 0.1 mm).
Thin-Film RTD
A thin layer of temperature-sensitive material deposited
onto a ceramic substrate, and mounted inside a protective
sheath or casing.
Etched to form a long strip.

Wire-Wound RTD Vs. Thin-Film RTD
Wire-Wound RTD
Accurate.
Expensive.
Sensitive to shocks and vibrations
Thin-Film RTD
Faster response.
Rugged.
Low cost.
Less sensitive to shocks and vibrations.

Practical Considerations
Tension or strain on the wires affect resistance. (Tension may case change in length
and cross-sectional area).
Wire RTDs has relatively low resistance. High resistances would require very long
wires or excessively thin wires.
Cost: High resistance RTDs require more material, and since most RTDs are based
on platinum, material costs can be significant.
When high demand applications: platinum is the first choice, because of its ex-
cellent mechanical and thermal properties. platinum also is chemically stable even at
elevated temperatures (resisting corrosion and oxidation) even at high temperatures.
For less demanding applications: nickel and copper offer less expensive alternatives
at reduced performance.
Q&A and Discussion
Outlines
1 Introduction

Q&A and Discussion
Q&A and Discussion

Recommended Readings and Practice Problems
Outlines
1 Introduction

Recommended Readings:
Ida, N., Sensors, Actuators and Their Interfaces: A multidisciplinary introduction.
Chapter 3 – Sections 3.2 and 3.3.
Practice Problems:
Ida, N., Sensors, Actuators and Their Interfaces: A multidisciplinary introduction.
Chapter 3 – Section 3.6: Problems 3.3 - 3.17 - 3.18

04 Temperature Sensors

Uploaded by

Copyright:

Available Formats

04 Temperature Sensors

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

04 Temperature Sensors

Uploaded by

Copyright:

Available Formats

MSE 355

Sensors, Measurements and Data

MSA University - Faculty of Engineering

Fall Semester, 2023

Assoc. Prof. Dr. M. A. Kamel MSE 355 Year 3 – Fall Semester 2 / 33

4 Q&A and Discussion

5 Recommended Readings and Practice Problems

Assoc. Prof. Dr. M. A. Kamel MSE 355 Year 3 – Fall Semester 3 / 33

4 Q&A and Discussion

5 Recommended Readings and Practice Problems

Assoc. Prof. Dr. M. A. Kamel MSE 355 Year 3 – Fall Semester 3 / 33

Temperature sensors are one of the oldest sensors in use.

In 1821, the German scientist Thomas Johann Seebeck discovered a phenomenon

Based on the Seebeck Effect, in 1826, the French

Idea: when two wires made from dissimilar met-

Assoc. Prof. Dr. M. A. Kamel MSE 355 Year 3 – Fall Semester 6 / 33

In 1834, the french scientist Charles Athanase

Peltier Effect: It is the reverse phenomenon

This principle more often used for cooling or heat-

In 1871, the German–British scientist William

Siemens invented a method for temperature mea-

surement based on the resistance–temperature re-

This invention has become the basis of the ther-

moresistive sensors, on which resistance tem-

perature detectors (RTDs) are based.

Assoc. Prof. Dr. M. A. Kamel MSE 355 Year 3 – Fall Semester 8 / 33

Types of Temperature Sensors

Assoc. Prof. Dr. M. A. Kamel MSE 355 Year 3 – Fall Semester 9 / 33

4 Q&A and Discussion

5 Recommended Readings and Practice Problems

Assoc. Prof. Dr. M. A. Kamel MSE 355 Year 3 – Fall Semester 9 / 33

They are among the oldest sensors (have been used

about 150 years ago).

Assoc. Prof. Dr. M. A. Kamel MSE 355 Year 3 – Fall Semester 10 / 33

They are among the oldest sensors (have been used

about 150 years ago).

They are passive sensors.

Basically, thermocouple is mode of two different

metals connected at the measurement end.

If many thermocouples are connected in series,

they are called a thermopile.

Assoc. Prof. Dr. M. A. Kamel MSE 355 Year 3 – Fall Semester 10 / 33

It is based on the Seebeck Effect.

Seebeck Effect: When two wires made from dis-

This voltage can be used to measure temperature.

Note that, this voltage is too small. So, it has to

Basic Analysis of a Basic Thermocouple

Assume two junctions at different temperature T1 and T2 .

emfa = αa (T2 − T1 ) emfb = αb (T2 − T1 )

αa and αb are the Seebeck coefficients (µV/◦ C).

emfT = emfa − emfb = (αa − αb )(T2 − T1 ) = αab (T2 − T1 )

αab is the relative Seebeck coefficient of the material combination a and b.

Assoc. Prof. Dr. M. A. Kamel MSE 355 Year 3 – Fall Semester 12 / 33

Examples of Seebeck Coefficients

Assoc. Prof. Dr. M. A. Kamel MSE 355 Year 3 – Fall Semester 13 / 33

Industrial Process Control: Temperature control in heat treatment of metals, chem-

ical reactions, and food production.

HVAC Systems: Flame sensor in a gas water heater.

Automotive Industry: Monitoring the exhaust gas temperatures.

Medical Devices: Temperature measurement of MRI machines.