Temperature is a measure of a materials internal molecular activity.

As the level of
molecular activity rises, the temperature of a substance increases. Hot and cold are
subjective, qualitative descriptions of the rise in molecular activity. The need for more
consistent and accurate ways to describe temperature led to the invention of temperaturemeasuring devices, or sensors.
Temperature is a widely used measurement. Galileo is credited with inventing the first
thermometer in 1595. Over the years thermometer technology has evolved, and
measuring principles are continuously improved upon. Today, highly accurate and reliable
devices are available.
Sensors use standard, universally recognized temperature scales. Because these scales
rely on fixed points in nature (e.g., freezing point of water), they provide a way to describe
temperature that is both objective and quantitative.
Many types of sensors are available for applications both in the home and in industry. Two
commonly used types are discussed in this module: resistance temperature detectors
(RTDs) and thermocouples (T/Cs).

Why Measure Temperature?

PRODUCT QUALITY AND YIELD
Many chemical reactions can only create the desired products under specific conditions,
which often requires a specific temperature. Even a few degrees of variation from the
optimum temperature often results in very little production of the desired productinstead,
waste product is created. To ensure a quality product, accurate temperature measurement
and precise control are necessary. Effective separation of products in a distillation column
relies upon precise temperature control, as do catalytic cracking processes.
In a scientific laboratory, the objective of measurement may be accuracy, regardless of
cost and time. This might be the case, for example, when determining the melting point of
a new material.
In industry, however, the usual objective is to find a way to measure temperature that
yields the following:
The accuracy and speed of response required by the application
Acceptable initial investment
Low maintenance cost, which implies long life and stability of calibration

EFFICIENCY
Accurate temperature measurements increase process efficiency. For example, consider
an air conditioning supplier that services several large facilities. An accurate measurement
of energy expenditure determines customer billings. The supplier measures the amount of
cooling provided by comparing the temperature of the hot air coming into the air
conditioner to the temperature of the cooler air given off by the air conditioner. If one or
both of these temperatures is inaccurately measured, the supplier may expend more
energy than is necessary to cool the air.
Accurate temperature measurements prevent the supplier from overcooling the air, which
saves energy and increases efficiency.

SAFETY
Accurate temperature measurements help to ensure safety in process industry
environments. Temperatures can increase rapidly in exothermic (giving off heat) chemical
reactionsif the temperatures are not closely watched and controlled, explosions could
result.

CUSTODY TRANSFER
Often, the amount of material that is bought and sold (referred to as custody transfer) is
based on a measurement of the volumetric flow rate of gas. The amount of material
contained in a specific volume of gas decreases with rising temperatures and increases
with falling temperatures. Therefore, it is extremely important to know the exact
temperature of the gas when determining volumetric flow rate. Inaccurate temperature
measurements during custody transfer applications result in over- or undercharging of
customers. A natural gas custody transfer application is one example of when temperature
measurements are required.

Temperature Measurement Scales

The three temperature measurement scales in use today are:
Fahrenheit
Celsius (also called Centigrade)
Absolute (Kelvin and Rankine)
Three features are common to the development of these scales, even though the scales
were developed at different times and by different people:

Two or more fixed points are chosen and assigned to a specific temperature on the
scale. For example, the freezing point of water is used as a fixed point for both
Celsius (0 C) and Fahrenheit (32 F) scales.
Each fixed point is associated with an observable or measurable phenomenon
usually a physical property that changes predictably at the fixed point (e.g., the
points at which water boils or freezes).
A standard method is used to establish numerical values for measurements that fall
between the fixed points.

Figure 4.1 shows where the boiling and freezing points of water (fixed points) fall on each
of the four temperature scales:

FAHRENHEIT SCALE
Gabriel Fahrenheit is credited with creating the first universally recognized temperature
scale. The scale originally used three fixed points:

Temperature of a mixture of ice water and salt (assigned 0)

Temperature of ice water (assigned 30)
Body temperature of a healthy person (assigned 96)

On this scale, Fahrenheit measured the boiling point of water to be 212. He later adjusted
the freezing point of water to 32 so that the interval between the boiling and freezing
points of water could be represented by the more rational number 180.
Temperatures measured on this scale are given in degrees Fahrenheit (F).
To convert a temperature from Fahrenheit to Celsius, use the following equation:

CELSIUS SCALE
The Celsius scale, also known as the Centigrade scale, was devised by Anders Celsius.
On the original scale, the freezing point of water was 100 and the boiling point of water
was 0.
Several years later, the two points were reversed. Today, 100 designates the boiling point
of water, and 0 designates the freezing point of water. Temperatures measured on this
scale are given in degrees Celsius (C).
To convert a temperature from Celsius to Fahrenheit, use the following equation:

ABSOLUTE SCALES
The search for more consistency and accuracy than the Fahrenheit and Celsius scales
provided led to the development of the Kelvin and Rankine scales, or absolute scales.
Absolute scales are scales that measure temperature relative to absolute zero.
Absolute zero is the lowest possible temperature of matter, or the temperature at which
molecular activity ceases. Absolute scales are not often used in process industries.
Kelvin Scale
Absolute zero on the Kelvin scale is equal to 0 K. Increments on the Kelvin scale are
equal to increments on the Celsius scalethat is, 1 K = 1 C.
Rankine Scale
Absolute zero on the Rankine scale is equal to 0 R. Increments on the Rankine scale are
equal to increments on the Fahrenheit scalethat is, 1 R = 1 F.

Temperature Measurement
Physical properties that change with temperature are used to measure temperature. For
example, the property of material expansion when heated is used in liquid-in-glass,
bimetallics, and filled-system measurement. The electromotive force (emf) principle is
used in thermocouples, and electrical resistance changes are used in resistance
temperature detectors (RTDs). Other means of temperature measurement include
temperature-sensitive paint and crayons, and optical devices.
In the traditional medical thermometer, liquid-in-glass measurement takes the form of
mercury enclosed in glass. Obviously, the delicate nature of the glass and the toxicity of
mercury limit the usefulness of this type of thermometer in industrial applications. An
improvement on the liquid-in-glass thermometer is the filled system.
Temperature-sensitive paint and crayons can be applied to a surface to determine its
temperature. Some of them are reversible, such as desktop thermometers, while others
are irreversible. They are available in a wide range of temperatures. Crayons have a
calibrated melting point, and the crayon mark melts when the rated temperature is
reached. Paints, which are suspended in a medium, function similarly. Both crayons and
paints have ranges varying from 120 to 2000F (50 to 1090C).
All temperature measurements are indirect. That is to say, the measurement is the
measurement of volumetric expansion (liquid-filled thermometer), dimensional change
(bimetallic thermometer), electromotive force (thermocouple), resistance (resistance
temperature detector, or RTD), radiated energy (radiation thermometer), or some other
characteristic of a material that varies predictably and reproducibly with temperature.
However, in industrial process measurement and control, the concept of direct and indirect
temperature measurement has a different meaning. A direct measurement is a
measurement of the temperature of the product itself. An indirect measurement is a
measurement of some other temperature from which one can infer the product
temperature.

Thermometers Temperature indicators

Filled System
Principle of Measurement
A filled system (see figure 4.1.1) is a metallic assembly that consists of a bulb, smalldiameter tubing (known as capillary), and a Bourdon spring. An indicator linked to the
Bourdon tube indicates temperature. Sometimes bellows and diaphragms are used
instead of a Bourdon.

Fig. 4.1.1: Filled system

The system is filled with a liquid or gas that expands and contracts as the temperature
sensed at the bulb increases and decreases. This expansion/contraction is translated into
a mechanical motion. Liquid causes volume changes, and gas causes pressure changes.
Application Notes
The filled-system type of measurement is generally used for local indication or for
temperature sensing in self-actuated temperature control valves. Its use has decreased
over the years, but there are still some applications for it. This device is an improvement
over the liquid-in-glass thermometer. It needs no power to function and is simple, rugged,
self-contained, and accurate over narrow temperature spans.
However, the units bulb may be too large to fit existing applications, and if the filled
system fails, the whole system must be replaced, which is expensive. In addition, the
capillary tubing is generally limited to a distance of 250 ft (80 m), and the filled system as a
whole is slow to respond and relatively expensive. Moreover, it is susceptible to ambient
temperature changes around the capillary, and ambient temperature compensation is
often necessary.
Filled-system measuring devices must be free of leaks in order to maintain their accuracy.
Therefore, plants must occasionally check and test them, and continuously support and
protect the capillary tubing against damage. In addition, the capillarys material of
construction should be compatible with the surrounding environment. Finally, the bulb
must be sufficiently immersed to ensure that the actual temperature is being measured.

Bimetallic
Principle of Measurement
In a bimetallic device (see figure 4.1.2), a spiral made of two metals with different
coefficients of expansion expands as the temperature increases. The movement
generated by the expansion drives an indicator on a scale. Industrial bimetallics use a
helical coil to fit inside a stem. Most temperature switches operate on this principle, except
that the pointer is replaced with a microswitch. Precision-made bearings and guides
provide minimum acceptable friction for the moving components.

Fig. 4.1.2: Bimetallic thermometer

Application Notes
The bimetallic method of measurement is generally used in local temperature gages and
switches. To facilitate the reading of process temperatures, plants usually select all-angle
gages with a 5 in. (120 mm) diameter dial. A capillary type is sometimes used for operating
visibility. If vibration exists, the plant may have to fill the thermometer with a dampening
fluid that is compatible with the process fluid, in case of leakage.
The bimetallic has a simple construction and few moving parts and requires little
maintenance. Its cost is the lowest of all temperature-measuring devices. However, its
accuracy is low, and it provides no remote indication. Calibrating bimetallics requires
immersing them in a bath of known temperature.

Temperature Sensors
In industrial environments, high process temperatures, pressures, and vibration make it
necessary to have a robust temperature sensor. Fast response time, accuracy, and
stability are also needed. While several types of temperature sensors are available, the
two most commonly used in the process measurement industry are resistance
temperature detectors (RTDs) and thermocouples (T/Cs).
RTDs and T/Cs have several construction characteristics in common (Figure 4.2):
Sensing element
Sensor sheath
Lead wires
Threaded adaptor (optional)

SENSING ELEMENT
The sensing element (RTD element or T/C hot junction) is located at the tip of the
temperature sensor, on the end that is exposed to the process temperature. The sensing
element, which is constructed of metal, responds to the process temperature by
generating a measurable resistance (in RTDs) or voltage (in T/Cs) signal.

SENSOR SHEATH
The sensor sheath, or cable housing, is constructed of metal and holds most of the
component parts of the temperature sensor. The sensor packing, typically magnesium
oxide (MgO), surrounds the sensing element and is contained within the sensor sheath.
The sensor packing decreases the impact of process vibration on the sensing element and
thus ensures a more accurate measurement. The end of the sensor sheath is sealed with
a fill (e.g., epoxy) that keeps moisture out of the sheath and away from the sensing
element.

LEAD WIRES
Lead wires are attached to the sensing element and then passed through the sensor
sheath. The lead wires are insulated before exiting the sensor sheath. Lead wires connect
the sensing element either directly to a distributed control system (DCS) or to a
temperature transmitter connected to a DCS.
Internal lead wires are the part of the lead wires inside the sensor sheath. External lead
wires are the part of the lead wires outside the sensor sheath.

THREADED ADAPTOR
The threaded adaptor is welded ovr the rear housing of the sensor sheath. One end of the
threaded adaptor threads into a thermowell or directly into the process piping. The other
end is typically threaded into a temperature transmitter, connection head, or conduit
connection (Figure 4.3):

Resistance Temperature Detectors

Pure metals will produce an increase in resistance with an increase in temperature. In
resistance temperature detectors (RTD), the electrical resistance of a metal increases as
temperature increases, a phenomenon known as thermoresistivity. A temperature
measurement can be inferred by measuring the resistance of the RTD element.

RTD SENSING ELEMENT MATERIALS

Thermoresistive characteristics of RTD sensing elements vary depending on the metal or
alloy from which they are made. To be useful in an RTD element, the metal or alloy must
have the following characteristics:

A predictable relationship between temperature and resistance

Relatively high resistance that can be measured easily
Physical strength
Stability (will not melt or freeze) over the measured temperature range
A large, easily measurable change in resistance for a given change in temperature

The three types of resistance metals most commonly used to construct RTDs are
platinum, copper, and nickel.
Platinum
Platinum RTD elements are the most common type of RTD element used in process
industries. Platinum elements have high accuracy, high repeatability, and a high resistance
change per degree of temperature change. In addition, platinum RTD elements are highly
linear throughout their temperature range.
Copper
Copper RTD elements are highly linear throughout their temperature range, but have
limited accuracy and a narrower temperature range than platinum elements. Copper
elements are most often used for measuring temperature in bearings and motor
windingsapplications in which accuracy is not critical.
The most common RTD element is platinum; nickel is generally the second choice.
Nickel
Nickel RTD elements have a high resistance change per degree of temperature change,
but have poor linearity, limited accuracy, and a relatively narrow temperature range. Nickel
elements are most commonly used in applications in which accuracy is not critical.
Figure 4.4 compares thermoresistive characteristics for platinum, nickel, and copper RTD
elements. (Note the platinum elements wider temperature range and more linear curve).

Table 4.1 compares resistance temperature characteristics for platinum, nickel, and
copper RTD elements.

Resistance temperature detectors are usually protected from the environment by a sheath
made of stainless steel or any other temperature- and corrosion-resistant material. The
element fits snugly inside the sheath to produce a high rate of heat transfer. A fine powder
is used to eliminate air pockets. Ceramic insulators are typically used to isolate the internal
lead wires. At the end of the tube a hermetic seal protects the element. The assembly may
be terminated with the lead wires or may be supplied with an appropriate terminal block.

Fig. 4.4b: 3-wire RTD

RTD elements are available in several designs. The two most common designs are:
Wire wound
Thin film
Wire-Wound RTD Sensing Elements
Wire-wound RTD sensing elements are constructed by coiling a platinum (or other
resistance metal) wire in or around a ceramic mandrel (spindle). Figure 4.5 shows an
externally wound sensing element.
A dual-element RTD can be created by coiling a second set of wires inside (internally
wound) or outside the ceramic mandrel. If connected to a second transmitter, a transmitter
with dual-sensor capabilities, or to another DCS card, a dual-element sensor increases the
reliability of the temperature measurement.

Wire-wound RTD elements are very sturdy and reliable. Compared to thin-film RTD
elements, their accuracy tends to be higher, but their time response (how quickly the
electronic output reflects temperature change) tends to be slightly slower.
Wire-wound RTD elements work well for a wide variety of applications, although they may
fail in high-vibration areas.

Thin-Film RTD Sensing Elements

Thin-film RTD sensing elements (Figure 4.6) are constructed by depositing a thin film of
resistance metal onto a ceramic substrate (base piece) and trimming the metal to
specifications.
Sensing elements of thin-film construction are typically less expensive than those of wirewound construction because less resistance metal is required for construction. Thin-film
RTD elements have a response rate that is 12 seconds faster than wire-wound RTD
elements.

To get an accurate temperature reading from an RTD, the resistance of the RTD sensing
element must be measured.
Each copper lead wire that connects the RTD sensing element to the resistance
measuring device adds a small amount of resistance into the measurement. If this added
resistance is ignored, an error is introduced and an inaccurate temperature measurement
results. The error is referred to as a lead wire effect. The longer the wire run, the greater
the error, or lead wire effect, reflected in the temperature measurement.
To compensate for lead wire effect, three-wire and four-wire RTDs are used instead of
two-wire RTDs. Three-wire RTDs are created by soldering one additional copper wire onto
one of the lead wires. Four-wire RTDs are created by soldering one additional copper lead
wire to each of the existing lead wires (Figure 4.7).

Two-Wire RTDs
Two-wire RTDs do not compensate for lead wire effects. Resistance of the two lead wires
is added into the resistance of the RTD element, which results in an inaccurate
temperature reading.
Three-Wire RTDs
Three-wire RTDs allow for compensation of lead wire effects. There are several methods
for performing lead wire compensation and finding the resistance of the RTD element.
With a three-wire element, the impedance in the wires will cancel because the wires are in
opposite legs of the bridge. In other words, the three-wire method compensates for the
effect of lead resistance. This is the most practical and commonly used RTD method
One common method uses Ohms law (Voltage [V] = Current [I] Resistance [R]):
1. A small, known current is sent through LW 1 and LW2 (Figure 4.8), and the voltage
drop across LW 1 and LW2 is measured. The resulting value is used later in the
calculation.

2. Next, the voltage drop across LW 1 and LW3 is measured. The resulting value
includes the effects of the lead wires:

From Ohms law, the resistance of the RTD element plus the lead wires is equal to
the measured voltage drop divided by the known current (R = V/I).

3. To find the voltage of the RTD element alone, the lead wire voltage of LW 1 and
LW 3 must be subtracted from the measured voltage across LW 1 and LW 3. The
voltage of LW1 and LW 2 found in Step 1 can be used as a close approximation to
the voltage of LW 1 and LW3.

4. Calculate the resistance of the RTD element using Ohms law. (Some error may
still be present since Vlead wires 1&3 will never exactly equal Vlead wires 1&2).

Three-wire RTDs are less accurate than four-wire RTDs because of the assumption that
the resistance of LW2 and LW 3 is equal.
Four-Wire RTDs
Four-wire RTDs provide the best lead wire compensation. A four-wire element requires an
extra wire but provides additional accuracy. It is generally used only rarely where very high
degrees of accuracy are required. The four-wire element is immune to lead resistance. Its
current is sourced on one set of leads and the voltage is sensed on another set of leads.
The lead resistance is not part of the measurement and the output voltage is directly
proportional to the RTD resistance.
The following steps are used to find the resistance of a four-wire RTD element:
1. A small, known current is sent through LW 1 and LW4 (Figure 4.9), and the voltage
drop is measured across LW 2 and LW 3.

2. Because no current runs along LW2 and LW 3, the measured voltage drop is due
only to the RTD element:

CONVERTING RESISTANCE TO TEMPERATURE

Once the resistance measurement is determined, it is converted to a temperature
measurement. One of two conversion methods is used by the transmitter:
RTD standard (e.g., IEC 751 standard)
Callendar-Van Dusen equation

The IEC 751 Standard

The IEC 751 standard describes an ideal relationship between the resistance of a platinum
RTD and the temperature to which the RTD is subjected. For example, at 212 F (100 C),
the IEC 751 standard shows that an ideal platinum RTD (one that exactly matches the IEC
751 standard) would have a resistance value of 138.5 (Figure 4.10).
When a transmitter or control system accepts a resistance signal from a platinum RTD
subjected to 212 F (100 C), the IEC 751 standard curve is used to translate that
resistance signal into a temperature reading. However, since actual RTDs are never ideal,
they do not match the resistance versus temperature relationship as described in the IEC
751 standard. The difference between the actual RTD curve and the ideal RTD curve
results in a measurement error, which is referred to as a sensor interchangeability error
(Figure 4.10).

The maximum allowable sensor interchangeability error at a given process temperature is

defined by two IEC 751 standard classificationsClass A and Class B. These
classifications are used to identify platinum RTDs. Figure 4.11 compares these two
classes to the IEC 751 standard ideal.

Table 4.2 shows the sensor accuracy requirements for Class A and Class B RTDs as
defined by the IEC 751 standard.

Class A RTDs
Class A RTDs are defined as being accurate to the IEC 751 standard ideal 0.15 C +
(0.002 T) C (where T is the process temperature in C). Class A RTDs are more
expensive than Class B RTDs.
Class B RTDs
Class B RTDs are defined as being accurate to the IEC 751 standard ideal 0.3 C +
(0.005 T) C (where T is the process temperature in C). Class B RTDs, while less
accurate than Class A RTDs, are sufficient for most industry applications.

Using the Callendar-Van Dusen Equation

The Callendar-Van Dusen equation offers an alternative to the IEC 751 standard. The
equation, used in transmitter-sensor matching, can be used to create a curve that
approximates an RTDs resistance-temperature relationship.
The Callendar-Van Dusen equation is:

Where:
t = Temperature in C
Rt = Resistance of the RTD at t
Ro = Resistance of the RTD at t = 0 C (a Callendar-Van Dusen constant)
, , and = Callendar-Van Dusen constants
The Callendar-Van Dusen equation can be programmed into a transmitter so that the
transmitter can use the actual RTD curve, rather than an ideal curve (e.g., IEC 751
standard), to translate the sensors resistance signal into a temperature value.
The Callendar-Van Dusen equation provides a significant improvement in measurement
accuracy, even when compared to Class A RTDs.

The Four Callendar-Van Dusen Constants

Each RTD is associated with four Callendar-Van Dusen constants specific to that
particular RTD. The process of generating the specific constants for each RTD is referred
to as calibration or characterization. During calibration, an RTDs resistance is measured
at several temperatures in different temperature baths. The data collected are fit to a
fourth-order curve, from which the four Callendar-Van Dusen constants can be
determined:

Ro (ice point resistance)Resistance output of the RTD at 0 C (32 F)

: Constant that gives the linear approximation to the R versus t curve; common
values are 0.00385 and 0.00392 / C
Ro: Average slope between 0 C and 100 C; common values are 0.385 /C and
0.392 /C.
: Constant that adjusts for non ideal behavior at <0 C temperatures
: Constant that adjusts for non ideal behavior at >0 C temperatures

A resistance versus temperature curve can be generated for any RTD by plugging the
RTDs four constants into the Callendar-Van Dusen equation. (Figure 4.12).

Application Notes
As a rough rule of thumb, RTDs are used where the temperature is less than 250F
(120C), whereas T/Cs are used where the temperature is greater than 930F (500C).
Since the accuracy of RTDs varies with temperature, the user must assess the processs
operating temperature and deemed whether an RTD is acceptable.
Of all temperature-measuring devices, RTDs are, at moderate temperatures, the most
stable and the most accurate. Their output is stronger than that of a T/C, they are less
susceptible to electrical noise, and they operate on a higher level of electrical signals.
Moreover, they are more sensitive and more linear than a T/C (output versus
temperature), use copper extension wire (not special extension wire), require no reference
junction, and are easy to interchange.
However, RTDs are relatively expensive compared to thermocouples, have a slow
response, and require a current source. They are susceptible to small resistance changes,
and self-heating appears as a measurement error (the main source of RTD error). In
addition, RTDs have a limited temperature range, are susceptible to strain and vibration,
generate some nonlinearity, and require three (or four) extension wires. Their resistance
curves vary from manufacturer to manufacturer, and their accuracy and service life are
limited at high temperatures.

Thermocouples
T/Cs, like RTDs, use electricity to measure temperature. Instead of using electrical
resistance, however, T/Cs determine temperature by measuring voltage.
In 1821, T. J. Seebeck discovered that when two dissimilar metals are joined together, an
electromotive force (emf) is generated between the hot and cold (reference) junctions
(e.g., 4 mV for 100C between the two junctions). An increase in temperature produces an
increase in voltage output. Originally, the cold junction was actually immersed in an ice
bath to maintain a constant reference temperature. Modern electronics have replaced this
ice bath.
Theoretically, any two dissimilar metals will form a thermocouple (T/C) (see figure 7-4).

Fig. 7-4. Thermocouple

However, only a few are used because of their superior response to temperature changes
(i.e., sensitivity) and performance in general. There are many types of thermocouples,
each with its advantages and disadvantagesrefer to table 7-1 (p. 158/470, Battikha).

Thermocouple wires are manufactured to close tolerances and tend to be expensive. Their
use is thus limited to the probe itself. Thermocouple extension wires, which are compatible
with the T/C wires, are used as the link between the T/C and the measuring device or
transducer.
The evolution of modern electronics has created transducers that are small enough to fit
inside the T/C box (or head). The major advantage of this arrangement is that it avoids
long-distance transmission of a very low T/C voltage signal, which is prone to electrical
noise (as opposed to the 4-20 mA signal).

A T/C consists of two wires of dissimilar metals (e.g., iron and constantan) that are joined
at one end to form a hot junction (or sensing element). The temperature measurement is
made at the hot junction of the T/C, which is exposed to the process. The other end of the
T/C lead wires, when attached to a transmitter or volt meter, form a cold or reference
junction (Figure 4.13).

Several types of T/Cs are available, each differing by the metals used to construct the
element. The T/C categories are:

Type EChromel and constantan

Type JIron and constantan
Type KChromel and alumel
Types R and SPlatinum (Pt) and rhodium (Rh) (differing in the % of platinum)

Type TCopper and constantan

Figure 4.14 compares five types of T/Cs.

Type E
Type E T/Cs (chromel-constanan) are gaining popularity because they have the highest
output (most sensitive) and can be used over a wide range of temperatures.
Type J
Type J T/Cs (iron-constanan) are among the first thermocouples to be widely used. The
materials are rugged, but iron wire is susceptible to oxidation, especially at high
temperatures. The corrosive effects of oxidation (poisoning) reduce the accuracy of the
T/C. Type J T/Cs remain one of the most common T/C types in use today.
Type K
Type K T/Cs (chromel-alumel) are commonly used for high-temperature applications.
Because they are relatively inexpensive and offer a wide measurement range. Type K
T/Cs are the most popular T/C type in industrial environments.
Types R and S
Types R (Pt/13%Rh, Rh) and S (Pt/10%Rh, Rh) T/Cs are useful for extremely high
temperature measurements. Because they are constructed from a platinum alloy, types R
and S T/Cs are expensive and are used only when the process temperatures are too high
for other T/Cs to function properly.
Type T
Type T T/Cs (copper-constanan) are generally used in very low temperature (cryogenic)
applications. Type T T/C elements are robust in high-humidity environments.

CREATING A HOT JUNCTION

Several different methods are used to join the T/C lead wires at the hot junction, such as
twisting, clamping, soldering, brazing, and various types of welding (e.g., bead and butt
welding) (Figure 4.15).
For good performance, the hot junction must be constructed to be mechanically sound,
electrically continuous, and not poisoned by the chemical ingredients of welding or brazing
materials.

Hot Junction Configurations

Junctions can be grounded or ungrounded to the sensor sheath. With dual-element T/Cs
(two T/Cs in one sheath), the elements can be isolated or connected (unisolated) (Figure
4.16). Each configuration offers benefits and limitations:

GroundedGrounding creates improved thermal conductivity, which in turn gives

the quickest response time. However, grounding also makes T/C circuits more
susceptible to electrical noise (which can corrupt the T/C voltage signal) and may
cause more susceptibility to poisoning over time.

UngroundedUngrounded junctions have a slightly slower response time than

grounded junctions, but are not susceptible to electrical noise.

UnisolatedUnisolated junctions are at the same temperature, but both junctions

will typically fail at the same time.

IsolatedIsolated junctions may or may not be at the same temperature. Reliability

of each junction is increased, however, because failure of one junction does not
cause a failure in the second junction.

VOLTAGE MEASUREMENT AND THE SEEBECK EFFECT

T/Cs use a phenomenon known as the Seebeck effect to determine process temperature.
According to the Seebeck effect, a voltage measured at the cold junction of a T/C is
proportional to the difference in temperature between the hot junction and the cold
junction.
The voltage measured at the cold junction is commonly referred to as the Seebeck
voltage, the thermoelectric voltage, or the thermoelectric EMF. As the temperature of the
hot junction, or process fluid, increases, the observed voltage at the cold junction also
increases by an amount nearly linear to the temperature increase.
If the hot junction temperature is held constant, an increasing cold junction temperature
will produce a decreasing voltage, because the temperature difference between the hot
and cold junction is decreasing. When the cold and hot junctions reach the same
temperature, the observed voltage will be 0 V.
The magnitude of the voltage signal produced at the cold junction also depends on the
type of metals used to form the T/C. Different metals, or different type of T/Cs, have
different thermoelectric voltages at the same temperatures.
EMF Calculations
For each type of thermocouple there is a corresponding reference table that converts mV
reading (i.e., thermocouple output) into a temperature. These tables are available from
many sources and vendors, including National Bureau of Standards Monograph 125, ISA
Recommended Practices, or ISA Standards (ISA-MC96.1-1982). As an example, a portion
of a K type reference table is shown in table 7-2. From that table, it can be seen that if a Ktype thermocouple has an output of 4.012 mV and the temperature of the reference
junction is at 0C, then the measured temperature is of 98C.

Note that sometimes the mV values must be interpolated from the table to obtain the
corresponding temperature. Linear interpolation is acceptable. (p. 162/470, Battikha).

Partial type K thermocouple temperature emf table.

Cold Junction Compensation

Like RTDs, each type of T/C has a standard curve. The standard curve describes a T/Cs
voltage versus temperature relationship when the cold junction temperature is
32 F (0 C).
Because the voltage measured at the cold junction is proportional to the difference in
temperature between the hot and cold junctions, the cold junction temperature must be
known before the voltage signal can be translated into a temperature reading. The process
of factoring in the actual cold junction temperature (rather than assuming it is at 32 F
[0C]) is referred to as cold junction compensation (CJC).
The general steps for performing CJC are as follows:
1. Determine the temperature of the cold junction (room temperature) and find the
corresponding voltage (CJV) in a reference book.
2. With the hot junction exposed to the process temperature, read the voltage output.
3. Add the measured process temperature voltage to the cold junction voltage (CJV)
and look up the result in the standard curve table to find the true process
temperature.

EXTENSION WIRE
T/C extension wires are often used to wire the T/C back to a control system or to a remote
transmitter, which may be anywhere from 22,000 ft away.
T/C extension wire, with a few rare exceptions, must be of the same type of metal as the
T/C lead wires. If the metals do not match, the cold junction will be created prematurely.
For example, in Figure 4.17, copper wires are used to connect a T/C to the measuring
instrument. The temperature measurement is made with the assumption that the cold
junction is at Junction 1, rather than at Junction 2. The temperature measurement will be
off by a factor of 25 F.

The measured voltage from the T/C with copper extension wires does not equal the
measured voltage of the T/C with correct extension wires. In fact, if copper extension wires
are used, it is nearly impossible to find the process temperature from the measured
voltage.
Copper extension wires can be used in rare instances. If the two additional hot junctions (1
and 2 in Figure 4.17) are kept at the same temperature, no temperature gradient exists,
and the net voltage drop is zero. Thus, the temperature measurement is not affected.
However, keeping the hot junctions at identical temperatures is generally too difficult to be
practical.
T/C wire is relatively costly, particularly for platinum T/C Types R and S. Often, it is not
economically feasible to make continuous runs of T/C wire from the hot to the cold
junctions (often hundreds of feet long). To compensate for this problem, special lead wire
is used that closely approximates the thermoelectric properties of T/C wire. The special
wire allows the user to minimize cost without sacrificing performance.

WHY USE AN RTD RATHER THAN A T/C?

The main reasons for selecting RTDs rather than T/Cs are as follows:
RTDs have better accuracy and repeatability.
RTD signals are less susceptible to noisethey have a higher signal-to-noise ratio.
RTDs have better linearity over temperature ranges.
RTDs can use the Callendar-Van Dusen equation to eliminate sensor
interchangeability error.
CJC and related errors are not associated with RTDs.
RTD drift is predictable, while T/C drift is erratic and unpredictable. In addition, T/C
drift errors can be large as a result of element poisoning and element oxidation at
high temperatures (Figure 4.18).

The changes that affect the output of an RTD or T/C occur over time due to
mechanical shock, poisoning, and temperature cycling. These changes can be
eliminated by performing another RTD characterization, an option not available for
a T/C.
RTDs do not need special extension wire.

WHY USE A T/C RATHER THAN AN RTD?

The main reasons for selecting T/Cs rather than RTDs are as follows:
T/Cs function at higher temperatures (above 1,100 F [593 C]) than RTDs.
T/Cs are approximately 20% less expensive than RTDs.
T/Cs have a faster response time than RTDs.

Thermowells
A thermowell is a closed-end, metal tube that encapsulates and protects a sensor from
process flow, pressure, vibration, and corrosion. Thermowells also allow for sensor
removal without process shutdown. Thermowells are mounted in various ways onto the
process pipe or tank. The temperatura sensor is then threaded into the thermowell.
Thermowells are available in several different materials, mounting methods, and stem
types. The variety of design features renders thermowells suitable for various applications
and environmental conditions (Figure 4.19).

MATERIALS
Thermowells are constructed of one of several materials, some of which are:
316 and 304 stainless steel
Carbon steel
Monel
Inconel 600
Hastelloy
Three factors affect the choice of material:
Type of corrosive environment to which the thermowell will be exposed
Temperature and pressure limits of the material
Compatibility with the process piping material to ensure solid, noncorroding welds
and junctions

MOUNTING METHODS
Thermowells can be mounted in one of three methods, which results in three different
construction designs (Figure 4.20). Thermowells can be threaded, welded, or bolted
(flanged style) onto the process pipe.

Threaded
Thermowells are most commonly threaded onto the process piping or tank, which enables
them to be easily installed and removed. Threaded thermowells are the weakest type of
thermowell and may leak.
Welded
Welded thermowells are permanently welded onto the process pipe or tank. Thus, removal
is very difficult and requires cutting the thermowell out of the system. Welded thermowells
are the strongest type of thermowell and are used with fluids of high velocity, high
temperature, or high pressures. Welded thermowells are necessary for applications that
require a leak-proof seal.

Flanged
Flanged thermowells are bolted onto a pipe or tank and can be easily removed or installed.
Flanged thermowells are used in corrosive environments as well as in high-velocity, hightemperature, or high-pressure applications. Flanged thermowells are the most expensive
type of thermowell.

STEM DESIGNS
The stem of a thermowell is the part that is inserted into the process piping. Stems can be
tapered, straight, or stepped.
The performance of a thermowell varies with its stem design. Figure 4.21 examines the
following performance factors in relation to tapered, straight, and stepped stem designs:
Process pressure
Time response
Wake frequencyFrequency of alternating side-to-side movement of a fluid,
generated by properties of the fluid
Drag forceResistance of motion of a solid shape through a body of fluid
Price

All stem designs perform equally well at equal process pressures, as long as the
thicknesses of the thermowell walls are identical.
Thermowells with straight stems have the slowest time response because they possess
the most material at the tip (largest diameter). Thermowells with stepped stems have the
fastest time response because they possess the least material at the tip (smallest
diameter). A small diameter also results in the least amount of drag force. Thermowells
with stepped stems also cause the highest wake frequency and have the lowest natural
frequency (oscillation rate determined by the properties of the thermowell itself).
Generally, thermowells with tapered stems are slightly more expensive as a result of a
more complicated manufacturing process.

Thermowell Performance
THERMOWELL FAILURE
Thermowells can fail for many reasons, such as high drag forces, static pressure, and
corrosion. The von Karman effect can also cause thermowell failure.
Fluid flowing around a thermowell forms a turbulent wake called the von Karman trail
(Figure 4.22). The wake alternates from side to side at a specific frequency, called the
wake frequency. The wake frequency is dependent on many process variables. If the wake
frequency coincides with the thermowells natural frequency, the thermowell can break.
This phenomenon is known as the von Karman effect.

Wake frequency is a determining factor in selecting a thermowell for a high-velocity

application. To ensure safe operation, the wake frequency divided by the natural frequency
of the thermowell must be <0.8. If it is >0.8, the thermowell may fail and a different
thermowell should be chosen. (Software is available to aid in determining wake
frequency.)
Preventing Thermowell Failure
Thermowell failure can be prevented by selecting the appropriate thermowell for the
process conditions. To ensure safe and successful operation, the following information
should be known about the thermowell:
Material
Mounting method
Stem design
Dimensions (e.g., diameter, length)
In addition, the following information should be known about the product being measured:
Velocity
Pressure
Temperature
Fluid density
Fluid type (e.g., steam, natural gas)

TEMPERATURE POINT RESPONSE TIME

Thermowells have a significant effect on a temperature sensors time response.
Thermowell time constants (mathematical quantities used to determine time response) are
so large that the temperature sensors time constants become only a small factor. The
thermowell time constant becomes the major contributing factor to the overall time
constant for the entire thermowell/sensor/transmitter system.
Several factors affect the overall time response of the temperature measurement when
thermowells are used, including:
Thermowell design
Tip thickness and diameter
Sensor sheath fill
Thermowell Design
Stepped thermowells are the fastest to respond to changes in temperature, with a time
constant of about 2023 seconds. Tapered thermowells have time constants of about 26
seconds.
Straight thermowells are the slowest to respond to temperature fluctuations, with time
constants approaching 30 seconds.
Tip Thickness and Diameter
The time constant of a thermowell is also affected by the thickness at the tip. Less material
at the stem tip results in smaller time constants. Stepped thermowells have the smallest tip
diameter and thus have a faster response time.
Sensor Sheath Fill
The distance and material between the sensor sheath and the thermowell also affect time
response. The smaller the distance (x and y in Figure 4.23), the quicker the time response.
Spring-loaded sensors can be used to ensure contact between the tip of the temperature
sensor and the bottom of the thermowell (x would equal 0 in Figure 4.23), which provides a
faster time response due to increased thermal transfer.

To increase the time response even further, industry practice suggests using a thermally
conductive fill between the sensor sheath and the inside of the thermowell. Rather than
being filled with air, which is a poor thermal conductor, y in Figure 4.23 is filled with a
thermally conductive fill, which can significantly reduce the overall time constant from 20
30 seconds to 1011 seconds.

Transmitting the Temperature Sensor Signal

To be useful for control, safety, or monitoring purposes, a temperature-measurement
signal must be relayed from the point of measurement to the DCS of the process. The
signal is relayed back to the DCS in one of many ways. The two most common ways,
which will be discussed in this section, are:
TransmittersThe sensor is wired a short distance to the transmitter, where its
signal is converted to a digital or 420 mA signal. The transmitter is wired back and
the converted signal is relayed to the DCS.
Wiring directThe sensors lead wires are wired the entire distance to the DCS.
No signal conversion takes place along the route.

WHAT DOES A TRANSMITTER DO?

A transmitter converts a temperature sensors signal from resistance or voltage into a
common digital or analog 420 mA signal. Typically, the analog signal is linear with the
process-temperature measurement. As an example, consider a transmitter with a range of
32212 F (0100 C). With this transmitter, a 4 mA output corresponds to a 32 F (0 C)
process temperature, 12 mA corresponds to 122 F (50 C), and 20 mA to 212 F (100 C)
(Figure 4.24).
Temperature transmitters can accept sensor inputs from a wide variety of temperature
sensor types (including many types of RTDs and T/Cs).

SMART TRANSMITTERS
A technology that has emerged and gained popularity within the last decade is smart
communications. Smart transmitters have the ability to transmit digital information
simultaneously with the standard 420 mA analog signal. Smart transmitters offer many
benefits over transmitters that only communicate a 420 mA signal, such as reduced plant
inventory, improved measurement accuracy and stability, reduced maintenance, and
remote diagnostics.

TRANSMITTER MOUNTING STYLES

There are four common styles of transmitters. Transmitter styles are named according to
their mounting method:
Field mount (most common)
Head mount
Rail mount
Rack mount (least common)
Field-Mount Transmitters
Field-mount transmitters are the most rugged of all transmitter styles. Their robust
housings protect against corrosion and humidity. Some field-mount transmitters house the
electronics in dual-compartment housings, which completely isolates them from the effects
of humidity. Dual-compartment transmitters are the best design for use in hazardous
locations where explosion-proof standards are being used.
Field-mount transmitters can be integrally or remotely mounted (Figure 4.25):
Integral mountThe transmitter is threaded onto the sensor directly (mounts
directly to U.S. style [1/2 NPT] sensor).
Remote mountThe transmitter is mounted on a pipe stand or other support near
the sensor. Remote mount is preferred when the measurement point is
inaccessible or when the process environment is too harsh for the transmitter to be
installed directly on top of the sensor.

Head-Mount Transmitters
Head-mount transmitters are small, puck-shaped transmitters. Head-mount transmitters
are typically housed in a protective enclosurea connection head for direct mounting or a
junction box for remote mounting.
Head-mount transmitters are very common in Europe and are gaining popularity in the
U.S. as well.
Head-mount transmitters can be mounted in one of three ways (Figure 4.26):
Integral mountThe transmitter is housed in a connection head that is threaded
directly onto the sensor.
Remote mountThe transmitter is housed in a junction box that is mounted on a
pipe stand or other support near the sensor.
Rail mountThe transmitter is used in lieu of a rail-mount style and is mounted to
a G-rail or top-hat rail.

Rail-Mount Transmitters
Rail-mount transmitters are designed to be attached to a DIN-rail (G-rail or top-hat rail) or
directly screwed onto a wall. Rail-mount transmitters are also designed for compact
mounting, which allows for a number of transmitters to be mounted very closely together
(Figure 4.27).

Rail-mount transmitters are usually mounted in a control room or other mild or controlled
environment, which makes them both convenient for maintenance and less susceptible to
harsh environmental conditions.
A limitation of rail-mount transmitters is their susceptibility to noise. When rail-mount
transmitters are installed near the DCS, the sensors lead wires have to run the entire
distance to the DCS. In addition, there is no standard accessory that would allow a railmount transmitter to be used in an explosion-proof area.
Rack-Mount Transmitters
Rack-mount transmitters are very similar to rail-mount transmitters, except that rack-mount
transmitters use a mounting chassis. The common backplane of the mounting chassis,
rather than the loop wires, often provides power to the transmitter (Figure 4.27).

WHY USE A TEMPERATURE TRANSMITTER?

Three benefits of using a temperature transmitter over wiring direct are:
A more robust signal is deliveredThe 420 mA signal output from the transmitter
is much more robust than a sensors signal. Noise interference has less impact on
420 mA signals.
Measurement accuracy is optimalTransmitters offer improved measurement
accuracy over wiring direct. For example, sensors can be matched to transmitters
(transmitter sensor matching), which improves the accuracy of the temperature
measurement.
Time and money are savedTransmitter installation is often less expensive than
wiring a sensor direct because of savings from cabling costs and installation
(sensor wire, especially T/C wire, is relatively expensive). Also, a robust signal and
accurate measurements produce time and money savings through increased
functionality and diagnostic capabilities of the transmitter.

WIRING DIRECT
Wiring direct refers to wiring the sensors lead wires back to the DCS (Figure 4.28).
Because the sensors lead wire (and original signal) is traveling the entire distance from
the point of measurement to the DCS, care must be taken to avoid two key problems:
NoiseT/Cs are especially sensitive to noise interference and must be routed
around such sources as turbines and motors.
Heat sourcesA large change in the ambient temperature can effect the sensors
signal as it travels to the control system.

Although transmitters offer many benefits over wiring direct, many temperature
measurement points are still wired directly to the DCS.

INPUT/OUTPUT CARDS
Input cards, usually located near the DCS or control room, collect measurement signals
(inputs) from sensors or transmitters in the field. An input card typically accepts 8 or 16
inputs, depending on its design. Output cards send the signal back to the field for process
control.
Input/output (I/O) cards can be either digital or analog. Analog input cards can accept
inputs of only voltage (mV), resistance (W), or current (mA). Each type of signal requires a
specific type of input card. For example, transmitters are wired into input cards that accept
420 mA inputs. Type K T/Cs are wired into Type K T/C input cards, and Type J T/Cs are
wired into Type J T/C input cards. RTDs also have their own input card types.
Figure 4.29 shows sensors connected to I/O cards via transmitter and via wiring direct.

TRANSMITTERS AND INPUT CARDS

Transmitters allow for a standard, 420 mA analog input card to be used. If a transmitter is
used, the 420 mA analog input cards accept any type of measurement (temperature,
flow, pressure, level, etc.). If smart transmitters are used, digital cards may be used.
Standardizing the type of input card used reduces costs and ensures full use of each card.
Changing the sensor type does not necessitate switching the input card. The transmitter is
simply reconfigured for the new sensor type.

WIRING DIRECT AND I/O CARDS

When a sensor is wired directly to the DCS, specific types of input cards are used that
correspond to the sensor type. Each type of T/C must be used with a specific type of mV
analog input card. RTDs must be used with special analog input cards that measure
resistance signals.
One of the limitations of wiring direct is the need for specific types of input cards for
specific sensor types.