Chapter 8

Temperature

Material from Theory and Design for Mechanical Measurements;

Figliola, Third Edition

Temperature Standards
‘ “Temperature can be loosely described as the
property of an object that describes its hotness or
coldness, concepts that are clearly relative. Our
experiences indicate that heat transfer tends to
equalize temperature; or more precisely, systems
that are in thermal communication will eventually
have equal temperatures. The zeroth law of
thermodynamics states that two systems in thermal
equilibrium with a third system are in thermal
equilibrium with each other. Although the zeroth
law of thermodynamics essentially provides the
definition of equality of temperature, it provides no
means for defining a temperature scale.”

Temperature Standards

‘ A temperature scale provides for three

essential aspects of temperature
measurement:
(1) the definition of the size of the scale
increment,
(2) fixed reference points for establishing known
temperatures
(3) a means for interpolating between these
fixed temperature points.

1
Temperature Standards

‘The calibration of a temperature

measurement device entails not only the
establishment of fixed temperature points,
but the indication of any temperature
between fixed points.
‘The operation of a mercury-in-glass
thermometer is based on the thermal
expansion of mercury contained in a glass
capillary, where the level of the mercury is
read as an indication of the temperature.

Temperature Standards

‘ Submerge the
thermometer in water at
the ice point, make a
mark on the glass at the
height of the column of
mercury, and label it
0oC.
‘ Submerge the
thermometer in boiling
water, and mark the
level of mercury, and
Figliola, 2000 label it 100oC.

Temperature Standards

‘ The process of establishing 50oC without a fixed-

point calibration is called interpolation.
‘ Theory of the behavior of the mercury in the
thermometer, or many fixed points for calibration
are necessary.
‘ Even by the late 18th century, there was no
standard for interpolating between fixed points
on the temperature scale; the result was that
different thermometers indicated different
temperatures away from fixed points, sometimes
with surprisingly large errors.

2
Temperature

‘ Temperature is one of the most commonly

measured engineering variables.
‘ Thermometry is based on thermal
expansion.
‘ Most materials exhibit a change in size as a
result of a change on temperature.
‘ Measure temperature variation through
thermal expansion of liquid in glass.
‘ Difference in thermal expansion between
liquid and glass results in a change in level.

Temperature

‘The modern engineering definition of the

temperature scale is provided by a standard
called the International Temperature Scale.
‘This standard establishes fixed points for
temperature and provides standard
procedures and devices for interpolating
between fixed points.

Temperature

‘ A liquid-in-glass thermometer
measures temperature by virtue
of the thermal expansion of
liquid.
‘ The liquid is contained in a
glass structure that consists of a
bulb and a stem.
‘ The bulb serves as a reservoir
and provides sufficient fluid for
the total volume change of the
fluid to cause a detectable rise
Figliola, 2000 of the liquid in the stem.

3
Temperature

‘ During calibration, the thermometer is

subject to one of three measuring
environments:
1. Complete immersion thermometer
2. Total immersion thermometer
3. Partial immersion thermometer

Bi-Metallic Thermometer

‘ Based on differential thermal expansion of two

metals
‘ Two different materials are bonded together
‘ At bonding temperature, the metals are straight

Figliola, 2000

Bi-Metallic Thermometer

‘ Rc ∝ d / [(C∝)A – (C∝)B](T2 - T1)

‘ Rc = radius of curvature
‘ C∝ = material thermal expansion coefficient
‘ T = temperature
‘ d = thickness
‘ INVar is common material where
• C∝ = 1.7 * 10-8 m/moC
‘ Where steel
• C ∝ = 2*10-5 to 20*10-5 m/moC

4
 Electrical Resistance
Thermometry
‘By physical nature of the conduction of
electricity, electrical resistance of a
conductor varies with temperature.
‘Two basic classes of sensor
– Resistance temperature detectors (RTD)
• Electrical conductors
– Thermistors
• Semiconductors

Electrical Resistance
Thermometry
‘The physical basis for the relationship
between resistance and temperature is the
temperature dependence of the resistivity
of the material, ρe.
‘The resistance of a conductor length l and
cross-sectional area Ac may be expressed
in terms of resistivity: R = Resistance
• R = ρel / Ac ρe = resistivity of material
(temp dependent)
l = length
Ac = cross-sectional area

Figliola, 2000

‘ The resistance temperature detector (RTD) is constructed by

mounting a metal wire on an insulating support structure to
eliminate mechanical strain and by encasing the wire to
prevent changes in resistance due to sensor’s environment,
such as corrosion.
‘ Mechanical strains change a conductor’s resistance and must
be eliminated for accurate temperature measurements are to
be made.

5
 Electrical Resistance
Thermometry
• Over small temperature ranges,
R = Ro [1 + ∝(T – To)]

• Ro= reference resistance at To ; ∝= coefficient

of resistivity.

• Helical coils provide strain relief as the

conductor expands and contracts. Platinum is
the most popular.

Resistance Measurements

‘ Platinum is the most

linear over the entire
region.
Figliola, 2000
‘ Platinum RTD’s are
used over a wide range
from cryogenic to ≈
o
650 C with uncertainty
o
at ± 0.005 C possible

Figliola, 2000

Resistance Measurements

‘ The choice of an appropriate resistance

measuring device must be made based on the
required level of uncertainty in the final
temperature measurement.
‘ Conventional ohmmeters cause a small current to
flow during resistance measurements, creating
self-heating in the RTD.
– An appreciable temperature change of the sensor may
be caused by this current, in effect a loading error.

6
Resistance Measurements

‘ Bridge circuits are used to measure the resistance

of RTDs, to minimize loading errors, and to
provide low uncertainties in measured resistance
values.
‘ Wheatstone bridge circuits are commonly used
for these measurements.
– However, the basic Wheatstone bridge circuit does not
compensate for the resistance of the leads in
measuring resistance of an RTD, which are a major
source of error in electrical resistance thermometers.

Bridge Circuits

Figliola, 2000

‘ Use a wheatstone bridge with conductor length

compensation.
‘ Fig. 8.6a is a 3-wire Callender-Griffith bridge circuit.
Leads 1, 2, 3 have resistance r1, r2, r3, where r2 does not
affect the circuit, since there is no current through the
galvometer at balance conditions.

Bridge Circuits

‘ At balanced condition, iG = 0
‘ R1/R2 = R3/RRTD Î R1/R2 = (R3 + r1)/ (RRTD + r3)
‘ If R1 = R2, RRTD = R3 + r1 – r3
‘ If r1 balances r3 Î RRTD = R3

‘ A 4-wire, shown in Fig. 8.6b is available for low-

error, the 3-wire is most common.
‘ RTD has a slow response time and may not be
suitable in transit conditions.

7
Thermister

‘ Thermally sensitive resistors

‘ They are ceramic-like semi-conductor devices.
Resistance decreases rapidly with decreasing
temperature β ranges from 3500 to 4600k
– R = Roe β (1/T - 1/To) depending on device material
and construction.

‘ This equation is valid over limited temperature

range, unless β(t) is given. Typically, β is
constant for set temperature range.

Thermistor

‘ Thermisters are generally used in applications

where high sensitivity, ruggedness, or fast
response times are necessary.
‘ High resistance of thermisters eliminates the lead
length compensation issues found in RTD.
However, thermisters are not directly
interchangeable, due to variation in β and
sensitivity to temperature change.

8
Thermoelectric Temperature
Measurement
‘ The most common temperature sensing and
control element is the thermocouple.
‘ Consists of two electrical conductors, made of
dissimilar materials joined together at a junction.
‘ The output is a voltage that is related to the
temperature at the junction.
‘ The thermoelectric phenomena is a result of
simultaneous flow of heat and electricity away
from the junction.

Thermoelectric Temperature
Measurement
‘If T1 ≠ T2, a finite electric potential emf1
exists whose magnitude depends on ∆T
and the difference in materials.

Thermoelectric Temperature
Measurement
‘ “In an electrical conductor that is subject to a
temperature gradient, there will be both a flow of
thermal energy and a flow of electricity. Both of
these phenomena are closely tied to the behavior of
the free electrons in a metal; it is no coincidence that
good electrical conductors are, in general, good
thermal conductors. The characteristic behavior of
these free electrons in an electrical circuit composed
of dissimilar metals results in a useful relationship
between temperature and emf.”
‘ Three effects can occur: (1) Seebeck effect, (2)
Peltier effect, (3)Thomson effect

9
Seebeck Effect

‘ Seebeck effect – generation of voltage potential

as a result of temperature difference between two
junctions in circuit. It is a fixed and reproducible
relationship between emf and junction temp T1
and T2
‘ Seebeck coefficient: ∝AB = [∂(emf) / ∂T]open circuit
‘ This is a measure of the rate of change of voltage
of the two materials, with respect to temperature,
and is the static sensitivity of the open circuit
thermocouple.

Seebeck Effect

‘Under ideal conditions, the measured emf

of thermocouple is due to Seebeck effect,
alone.
‘Should be measured with no current flow
through use of high impedance
measurement device.

Peltier Effect
‘ Peltier Effect: Joule heat = I2R
Figliola, 2000

‘In Peltier effect, heat must be removed at

I2R to maintain temperature of conductor.
At the junction, extra heat is generated
(Peltier heat),which is proportional to I.

10
 Peltier Effect

‘ The Peltier effect is due to the thermodynamically

reversible conversion of energy as a current flows
across the junction, in contrast to the irreversible
dissipation of energy associated with I2R losses.
‘ The proportionality constant is the Peltier
coefficient πAB, and the heat transfer required to
maintain a constant temperature is: Qπ = πABI
‘ If I ≈ 0, heat ≈ 0

Thompson Effect

‘ The conductor has

temperature gradient and
potential difference, there
is a flow of heat and
current.
‘ To maintain constant
temperature, an amount
of heat must be removed
different from I2R.
• Qσ = σI (T1-T2)
Figliola, 2000 • σ = Thompson coeff

Figliola, 2000

11
‘ “The basic thermocouple circuit can be used to measure
the difference between the two temperatures T1 and T2.
For practical temperature measurements, one of these
junctions becomes a reference junction and is maintained
at some known, constant reference temperature. The other
junction then becomes the measuring junction, and the
emf existing in the circuit for any temperature T1
provides a direct indication of the temperature of the
measuring junction.”

Reference Junctions

‘ Thermocouples can be
used to measure the
temperature difference
between two junctions of
dissimilar material. One
junction at reference
temperature, the other at
measured temperature.
– Ice bath of crushed ice and
water is common.
Figliola, 2000

Reference Junctions

‘ Law of intermediate materials ensures that the

extension wires on potentiometer do not effect
emf as long as they are at the same temperature.
‘ The reference junction must be at a known
temperature, be reproducible, and be stable. Ice
point 0oC is common, although it is only
relatively stable.
‘ Electronic reference junctions are available on
some measurement devices, which use a
thermister to measure local environment
temperature (internal compensation).

12
Thermocouple Designations

Figliola, 2000
‘ Thermocouples are made from many different
combinations of materials.
‘ Choice of thermocouple depends on temperature range,
application environment, and desired uncertainty level.

Voltage Output

‘ The slope of the curve

corresponds to the static
sensitivity of the
thermocouple circuit.
‘ The output emf voltage of
a thermocouple can be
interpreted either from
conversion tables or by
polynomial expressions
that are fitted to a
particular type of
thermocouple.
‘ See Table 8.6, 8.7
Figliola, 2000

Thermocouple Measurement
‘ The Seebeck voltage can be measured at no current
flow. If current is flowing in circuit, Thomson and
Peltier effects may be experienced.
‘ The ideal is to use either a potentiometer or a high
impedance digital voltmeter which results in minimal
current flow.
‘ In common applications, a “built-in electronic cold
junction compensation” method is used, which
employs the use of a thermister. Typical error is 0.5o
to 1.5o bias.
‘ Internal polynomial interpolation is often used to
convert from voltage to temperature, which
introduces a linearization error.

13
Thermocouple Measurement

‘Often gains of 100 to 500 are used to boost

the thermo-electric signal to a range
detectable by A/D board.
‘Typical use of differential-ended inputs
with twisted pair conductors and a 10kΩ to
100kΩ resistor between low input (-)
terminal and low-level ground.

Multiple-Junction Thermocouple
Circuits
‘More than two junctions can be employed
in a thermocouple circuit, and
thermocouple circuits can be devised to
measure temperature differences or
average temperature or to amplify the
output voltage of a thermocouple circuit.

Thermopiles

‘“Thermopile” is a term used to describe a

multiple-junction thermocouple circuit that
is designed to amplify the output of the
circuit.
‘Increasing the voltage output may be a key
element in reducing the uncertainty in the
temperature measurement.

14
‘ This figure shows a thermopile for providing an
amplified output signal; in this case, the output
voltage would be N times the single thermocouple
output, where N is the number of measuring
junctions in the circuit.
‘ The average output voltage corresponds to the
average temperature level sensed by the N
junctions.

Thermocouples

‘ When spatially
averaged temperature
is desired, multiple
thermocouple
junctions can be
arranged as shown:
N
emf = 1 / N ∑ (emf )i
i =1
N
T = 1 / N ∑ Ti
i =1

Data Acquisition Considerations

‘ Attention must be given to the cold junction

compensation method.
‘ The two connection points of the thermocouple to
the DAS board form two new thermocouple
connections.
‘ Use of external cold junction methods between
the thermocouple and the board will eliminate
this problem, but more frequently, the
thermocouple will be connected directly to the
board and use built-in electronic cold junction
compensation.

15
 Data Acquisition Considerations
‘ This is usually accomplished by using a separate
thermistor sensor, which measures the temperature
at the system connection point to determine the
cold junction error, and providing an appropriate
bias voltage correction either directly or through
software.
‘ These boards also may use internal polynomial
interpolation for converting measured voltage into
temperature.
‘ This introduces a “linearization” error, which is a
function of thermocouple material and temperature
range and typically specified with the DAS board.

Data Acquisition Considerations

‘ Thermocouple wire pairs should be twisted to

reduce noise.
‘ A differential-ended connection is preferred
between the thermocouple and the DAS board.
‘ The thermocouple becomes an isolated voltage
source, meaning there is no longer a direct
ground path keeping the input within its common
mode range.
‘ As a consequence, a common complaint is that
the measured signal may drift or suddenly jump
in the level of its output.

Data Acquisition Considerations

‘ This interference behavior is eliminated by

placing a 10-kΩ to 100-kΩ resistor between the
low terminal of the output and low-level ground.
‘ Since most DAS boards use A/D converters
having a ± 5V full scale, the signal must be
conditioned using an amplifier.
‘ Usually, a gain of 100 to 500 is significant.
‘ High gain, very low noise amplifiers are
important for accurate measurements.

16
Data Acquisition Considerations

‘ Thermocouples tend to have long time constant

relative to the typical sample rate capabilities of
general purpose DAS boards.
‘ If temperature measurements show greater then
expected fluctuations, high-frequency sampling
noise is a likely cause.
‘ Slowing down the sample rate or using a
smoothing filter are simple solutions.
‘ The period of averaging should be on the scale of
the time constant of the thermocouple.

17