Guidelines for Temperature Measurement
can be measured via a diverse array of sensors. All of them infer
temperature by sensing some change in a physical characteristic.
Six types with which the engineer is likely to come into contact
are: thermocouples, resistive temperature devices (RTDs and
thermistors), infrared radiators, bimetallic devices, liquid expansion
devices, and change-of-state devices. It is well to begin
with a brief review of each.
consist essentially of two strips or wires made of different metals
and joined at one end. As discussed later, changes in the temperature
at that juncture induce a change in electromotive force (emf)
between the other ends. As temperature goes up, this output emf
of the thermocouple rises, though not necessarily linearly.
temperature devices capitalize on the fact that the electrical
resistance of a material changes as its temperature changes. Two
key types are the metallic devices (commonly referred to as RTDs),
and thermistors. As their name indicates, RTDs rely on resistance
change in a metal, with the resistance rising more or less linearly
with temperature. Thermistors are based on resistance change in
a ceramic semiconductor; the resistance drops nonlinearly with
sensors are noncontacting devices. As discussed later, they infer
temperature by measuring the thermal radiation emitted by a material.
devices take advantage of the difference in rate of thermal expansion
between different metals. Strips of two metals are bonded together.
When heated, one side will expand more than the other, and the
resulting bending is translated into a temperature reading by
mechanical linkage to a pointer. These devices are portable and
they do not require a power supply, but they are usually not as
accurate as thermocouples or RTDs and they do not readily lend
themselves to temperature recording.
devices, typified by the household thermometer, generally come
in two main classifications: the mercury type and the organic-liquid
type. Versions employing gas instead of liquid are also available.
Mercury is considered an environmental hazard, so there are regulations
governing the shipment of devices that contain it. Fluid-expansion
sensors do not require electric power, do not pose explosion hazards,
and are stable even after repeated cycling. On the other hand,
they do not generate data that are easily recorded or transmitted,
and they cannot make spot or point measurements.
Change-of-state temperature sensors consist of labels, pellets,
crayons, lacquers or liquid crystals whose appearance changes
once a certain temperature is reached. They are used, for instance,
with steam traps - when a trap exceeds a certain temperature,
a white dot on a sensor label attached to the trap will turn black.
Response time typically takes minutes, so these devices often
do not respond to transient temperature changes. And accuracy
is lower than with other types of sensors. Furthermore, the change
in state is irreversible, except in the case of liquid-crystal
displays. Even so, change-of-state sensors can be handy when one
needs confirmation that the temperature of a piece of equipment
or a material has not exceeded a certain level, for instance for
technical or legal reasons during product shipment.
the chemical process industries, the most commonly used temperature
sensors are thermocouples, resistive devices and infrared devices.
There is widespread misunderstanding as to how these devices work
and how they should be used.
Consider first the thermocouple, probably the most-often-used
and least-understood of the three. Essentially, a thermocouple
consists of two alloys joined together at one end and open at
the other. The emf at the output end (the open end; V1 in Figure
1a) is a function of the temperature T1 at the closed end. As
the temperature rises, the emf goes up.
the thermocouple is located inside a metal or ceramic shield
that protects it from a variety of environments. Metal-sheathed
thermocouples are also available with many types of outer
coatings, such as polytetrafluoroethylene, for trouble-free
use in corrosive solutions.
open-end emf is a function of not only the closed-end temperature
(i.e., the temperature at the point of measurement) but
also the temperature at the open end (T2 in Figure 1a).
Only by holding T2 at a standard temperature can the measured
emf be considered a direct function of the change in T1.
The industrially accepted standard for T2 is 0°C; therefore,
most tables and charts make the assumption that T2 is at
that level. In industrial instrumentation, the difference
between the actual temperature at T2 and 0°C is usually
corrected for electronically, within the instrumentation.
This emf adjustment is referred to as the cold-junction,
or CJ, correction.
changes in the wiring between the input and output ends
do not affect the output voltage, provided that the wiring
is of thermocouple alloy or a thermoelectric equivalent
(Figure 1a). For example, if a thermocouple is measuring
temperature in a furnace and the instrument that shows the
reading is some distance away, the wiring between the two
could pass near another furnace and not be affected by its
temperature, unless it becomes hot enough to melt the wire
or permanently change its electrothermal behavior.
The composition of the junction itself does not affect the
thermocouple action in any way, so long as the temperature,
T1, is kept constant throughout the junction and the junction
material is electrically conductive (Figure 1b). Similarly,
the reading is not affected by insertion of non-thermocouple
alloys in either or both leads, provided that the temperature
at the ends of the "spurious" material is the same (Figure
ability of the thermocouple to work with a spurious metal in the
transmission path enables the use of a number of specialized devices,
such as thermocouple switches. Whereas the transmission wiring
itself is normally the thermoelectrical equivalent of the thermocouple
alloy, properly operating thermocouple switches must be made of
gold-plated or silver-plated copper alloy elements with appropriate
steel springs to ensure good contact. So long as the temperature
at the input and output junctions of the switch are equal, this
change in composition makes no difference.
It is important to be aware of what might be called the Law of
Successive Thermocouples. Of the two elements that are shown in
the upper portion of Figure 1d, one thermocouple has T1 at the
hot end and T2 at the open end. The second thermocouple has its
hot end at T2 and its open end at T3. The emf level for the thermocouple
that is measuring T1 is V1; that for the other thermocouple is
V2. The sum of the two emfs, V1 plus V2, equals the emf V3 that
would be generated by the combined thermocouple operating between
T1 and T3. By virtue of this law, a thermocouple designated for
one open-end reference temperature can be used with a different
A typical RTD consists of a fine platinum wire wrapped around
a mandrel and covered with a protective coating. Usually, the
mandrel and coating are glass or ceramic.
mean slope of the resistance vs temperature plot for the RTD is
often referred to as the alpha value (Figure 2),alpha standing
for the temperature coefficient. The slope of the curve for a
given sensor depends somewhat on purity of the platinum in it.
most commonly used standard slope, pertaining to platinum of a
particular purity and composition, has a value of 0.00385 (assuming
that the resistance is measured in ohms and the temperature in
degrees Celsius). A resistance vs temperature curve drawn with
this slope is a so-called European curve, because RTDs of this
composition were first used extensively on that continent. Complicating
the picture, there is also another standard slope, pertaining
to a slightly different platinum composition. Having a slightly
higher alpha value of 0.00392, it follows what is known as the
the alpha value for a given RTD is not specified, it is usually
0.00385. However, it is prudent to make sure of this, especially
if the temperatures to be measured are high. This point is brought
out in Figure 2, which shows both the European and American curves
for the most widely used RTD, namely one that exhibits 100 ohms
resistance at 0°C.
The resistance-temperature relationship of a thermistor is negative
and highly nonlinear. This poses a serious problem for engineers
who must design their own circuitry. However, the difficulty can
be eased by using thermistors in matched pairs, in such a way
that the nonlinearities offset each other. Furthermore, vendors
offer panel meters and controllers that compensate internally
for thermistors' lack of linearity.
are usually designated in accordance with their resistance at
25°C. The most common of these ratings is 2252 ohms; among the
others are 5,000 and 10,000 ohms. If not specified to the contrary,
most instruments will accept the 2252 type of thermistor.
sensors: These measure the amount of radiation emitted by
a surface. Electromagnetic energy radiates from all matter regardless
of its temperature. In many process situations, the energy is
in the infrared region. As the temperature goes up, the amount
of infrared radiation and its average frequency go up.
materials radiate at different levels of efficiency. This efficiency
is quantified as emissivity, a decimal number or percentage ranging
between 0 and 1 or 0% and 100%. Most organic materials, including
skin, are very efficient, frequently exhibiting emissivities of
0.95. Most polished metals, on the other hand, tend to be inefficient
radiators at room temperature, with emissivity or efficiency often
20% or less.
function properly, an infrared measurement device must take into
account the emissivity of the surface being measured. This can
often be looked up in a reference table. However, bear in mind
that tables cannot account for localized conditions such as oxidation
and surface roughness. A sometimes practical way to measure temperature
with infrared when the emissivity level is not known is to "force"
the emissivity to a known level, by covering the surface with
masking tape (emissivity of 95%) or a highly emissive paint.
Some of the sensor input may well consist of energy that is not
emitted by the equipment or material whose surface is being targeted,
but instead is being reflected by that surface from other equipment
or material. Emissivity pertains to energy radiating from a surface
whereas reflection pertains to energy reflected from another source.
Emissivity of an opaque material is an inverse indicator of its
reflectivity Ð substances that are good emitters do not reflect
much incident energy, and thus do not pose much of a problem to
the sensor in determining surface temperatures. Conversely, when
one measures a target surface with only, say, 20% emissivity,
much of the energy reaching the sensor might be due to reflection
from, e.g., a nearby furnace at some other temperature. In short,
be wary of hot, spurious reflected targets.
infrared device is like a camera, and thus covers a certain field
of view. It might, for instance, be able to “see” a 1-deg visual
cone or a 100-deg cone. When measuring a surface, be sure that
the surface completely fills the field of view. If the target
surface does not at first fill the field of view, move closer,
or use an instrument with a more narrow field of view. Or, simply
take the background temperature into account (i.e., to adjust
for it) when reading the instrument.
are more stable than thermocouples. On the other hand, as a class,
their temperature range is not as broad: RTDs operate from about
-250 to 850°C whereas thermocouples range from about -270 to 2,300°C.
Thermistors have a more restrictive span, being commonly used
between -40 and 150°C, but offer high accuracy in that range.
and RTDs share a very important limitation. They are resistive
devices, and accordingly they function by passing a current through
a sensor. Even though only a very small current is generally employed,
it creates a certain amount of heat and thus can throw off the
temperature reading. This self-heating in resistive sensors can
be significant when dealing with a still fluid (i.e. neither flowing
nor agitated), because there is less carry-off of the heat generated.
This problem does not arise with thermocouples, essentially zero-current
sensors, though relatively expensive, are appropriate when the
temperatures are extremely high. They are available for up to
3,000°C (5,400°F), far exceeding the range of thermocouples or
other contact devices. The infrared approach is also attractive
when one does not wish to make contact with the surface whose
temperature is to be measured. Thus, fragile or wet surfaces,
such as painted surfaces coming out of a drying oven, can be monitored
in this way. Substances that are chemically reactive or electrically
noisy are ideal candidates for infrared measurement. The approach
is likewise advantageous in measuring temperature of very large
surfaces, such as walls that would require a large array of thermocouples
or RTDs for measurement.