10/5/23, 11:12 AM MSL Tech Guide 22 - Calibration Infrared Thermometer | PDF | Infrared | Thermometer

Search Read free for 30 days

 100% (2) ·
Download
2K views ·
now 8 pages 8 of 8  Search document 

MSL Tech Guide 22 - Calibration Infrared Thermometer

Uploaded by acanis1016 AI-enhanced title

Calibration of Infrared Thermometer Full description

     
Save 100% 0% Embed Share Print

Download now 8 of 8  Search document 

HVAC eGuide for professionals Download

Are you in control of your HVAC system? Vaisala

https://www.scribd.com/document/358553314/MSL-Tech-Guide-22-calibration-Infrared-Thermometer 1/16
10/5/23, 11:12 AM MSL Tech Guide 22 - Calibration Infrared Thermometer | PDF | Infrared | Thermometer

of 8 
MSL Technical Guide 22
Calibration of
Low-Temperature
Infrared Thermometers

The guide begins by briefly outlining the principles of

Introduction infrared thermometry and giving details of the measure-
ment processes employed inside the thermometers
The advent of low-cost handheld infrared (IR) ther- themselves. By necessity, there is some mathematical
mometers has led to a proliferation of non-contact tem- detail that can’t be avoided for any laboratory wanting to
perature measurement in the food, building, and low- set up IR thermometer calibration procedures. However,
temperature processing industries. However, these in- once understood, the mathematics can be easily imple-
struments are not as simple to use as they first appear mented in a spreadsheet application. A sample Excel
due to systematic effects that are present in almost all spreadsheet is available for download with this guide.
measurements.
This technical guide provides information relating to
the calibration of “low-temperature” IR thermometers, How IR Thermometers Measure
which typically measure temperatures in the range Temperature
–50 °C to 500 °C (see Figure 1). These thermometers
commonly use uncooled thermopile detectors that detect Electromagnetic Spectrum
radiation in the 8 µm to 14 µm spectral range (or similar). All objects emit radiation in the form of electromag-
Because these detectors are uncooled, radiation emitted netic waves. This radiation is distributed across the elec-
by the detector itself must be considered in the calibra- tromagnetic spectrum, from radio waves, through micro-
tion process. The emissivity setting on the thermometer, waves, infrared radiation, visible light, ultraviolet light,
which is often fixed at a value of 0.95, and any radiation and x-rays, to gamma rays. The actual distribution and
reflected from the surroundings, must also be taken into intensity of the radiation emitted by a specific object is
account. largely determined by the temperature of the object. For
As a consequence of these systematic effects, cali- objects near room temperature, almost all of the emitted
bration methods are more complicated than for contact radiation is contained within the infrared part of the spec-
thermometers or high-temperature IR thermometers. trum, at wavelengths in the vicinity of 10 microns
The expected reading, even on a perfect IR thermome- (10 µm). For objects near 1000 °C, the radiation is cen-
ter, does not necessarily match the reading of the refer- tred near the shorter wavelength of 2 µm.
ence thermometer. Procedures for determining the re- By measuring radiation at a fixed wavelength, or
quired corrections are given in this guide. more correctly over a fixed wavelength range, an IR
thermometer can determine the temperature of an object
from the intensity of the signal measured at the output of
its radiation detector – the higher the signal the higher
the temperature. Many IR thermometers are designed to
measure radiation over the wavelength range 8–14 µm,
and these instruments are suitable for measuring tem-
peratures in the range –50 °C to 500 °C, and sometimes
higher. For temperatures outside this range, thermome-
ters operating at different wavelengths are better suited.

Thermometer Response Function

The measured signal is usually a current or voltage
at the output of the IR thermometer’s detector. This is
not usually of much use to the user, who wants to know
the temperature that this signal corresponds to. It turns
out that the signal varies in a highly non-linear fashion as
a function of the temperature of the target object. This
non-linearity means, for example, that twice the signal
does not correspond to twice the temperature. The rela-
tionship between detector signal and temperature is
given by a thermometer response function, which is well-
approximated by the equation:

C
Figure 1. A typical low-temperature handheld infrared ther- S(T ) = , (1)
mometer. Photo courtesy of Fluke Corporation, reproduced with  c2 
exp   −1
permission.  AT + B 

Measurement Standards Laboratory of New Zealand • fax: +64 (0)4 931 3003 • e-mail: [email protected] • website: http://msl.irl.cri.nz 1

https://www.scribd.com/document/358553314/MSL-Tech-Guide-22-calibration-Infrared-Thermometer 2/16
10/5/23, 11:12 AM MSL Tech Guide 22 - Calibration Infrared Thermometer | PDF | Infrared | Thermometer

of 8 

00:03 03:41

https://www.scribd.com/document/358553314/MSL-Tech-Guide-22-calibration-Infrared-Thermometer 3/16
10/5/23, 11:12 AM MSL Tech Guide 22 - Calibration Infrared Thermometer | PDF | Infrared | Thermometer

of 8 
where A, B and C are constants related to the properties Influences on the Readings of
of the IR thermometer, and c2 is a universal constant
with the value 14388 µm.K. Note that the value of T in
IR Thermometers
equation (1) has the units of kelvin, which are related to
the more commonly used degrees Celsius units by
Emissivity
The intensity of the radiation emitted by an object
T K = t °C + 273.15 . (2) depends not only on its temperature, but also on a prop-
erty called emissivity. Emissivity is a number ranging
from 0 to 1 that characterises how well an object emits
Thus, a room temperature of 20 °C corresponds to a
radiation. An object with an emissivity of 1 is referred to
temperature on the kelvin scale of 293.15 K.
as a blackbody (an idealised perfect emitter). An object
The relationship represented by equation (1) is de-
with an emissivity of 0.8 emits 80% of the radiation that a
termined by the manufacturer of an IR thermometer, and
blackbody does, an object with an emissivity of 0.5 emits
is processed electronically inside the thermometer to
50% of a blackbody’s radiation, and so on.
produce a reading in degrees Celsius on the thermome-
ter’s display. This conversion of signal to temperature is,
thus, hidden from the user and, in any case, is not usu- Instrumental Emissivity
ally of much interest to the user. Because of the influence of emissivity, different ob-
However, in order to calibrate an IR thermometer, jects at the same temperature will produce different IR
the calibration laboratory does require knowledge of the thermometer signals, and give correspondingly different
details of this conversion process. (This is also true in readings. To account for this, IR thermometers have
many in-use measurement situations, but in-field use of what is called an “instrumental emissivity” adjustment,
IR thermometers is not covered in this guide.) The con- which should be set by the user to the value of the emis-
version from signal to temperature can be represented sivity of the target object’s surface. On some thermome-
by the inverse of equation (1): ter models, the instrumental emissivity cannot be ad-
justed, but is fixed at a value of usually 0.95, but some-
c2 B times 0.97. These instruments have limited application.
T= − . (3)
A ln(C S + 1) A
Reflected Radiation
Evaluation of equations (1) and (3) requires knowl- A further complication in IR thermometry measure-
edge only of the thermometer parameters A and B. Be- ments is that objects that are not blackbodies are partial
cause calibration of an IR thermometer involves the de- reflectors of radiation. For any opaque object, the emis-
termination of temperature corrections, it turns out that sivity and reflectivity always sum to 1. Thus, an object
the value of C is unimportant, as long as the same value with an emissivity of 0.8 has a reflectivity of 0.2. This
is always used for all calculations. Therefore, we can means that 20% of all the radiation that is emitted from
simply assign C = 1 in the evaluation of equations (1) surrounding objects, and falls onto the target object, is
and (3). A and B are both related to the wavelength reflected. This reflected radiation is detected by the IR
range over which the IR thermometer operates: thermometer and is added to the radiation emitted by the
target object. Thus, the reading on the thermometer de-
 ∆λ 2  pends not only on the temperature of the target, but also
A = λ0  1 −  (4) on the temperature of its surroundings. Objects that are
 2λ02 
good emitters (those having an emissivity close to 1)
tend to be black, and the higher the emissivity the lower
c 2 ∆λ 2 the reflectivity and, therefore, the smaller the effect of
B= , (5)
24λ02 the surroundings on the thermometer reading.

where λ0 is the centre wavelength of the range and ∆λ is Detector Radiation

the width of the wavelength range. Thus, for an IR ther- Finally, the detector itself also emits radiation. The
mometer operating from 8–14 µm, we get λ0 = 11 µm signal at the detector’s output corresponds to the differ-
and ∆λ = 6 µm, and equations (4) and (5) give ence between the in-coming radiation from the target
A = 9.36 µm and B = 178 µm.K, respectively. (including reflected radiation) and the out-going radiation
We can illustrate the conversion of temperature to emitted by the detector. All low-cost IR thermometers
signal by using these values (along with C = 1) in equa- use uncooled detectors, so the detector is at, or slightly
tion (1). For example, for a temperature of 50 °C above, room temperature. If the target is below about
(323.15 K), equation (1) gives S = 0.01132. (Signal-to- 200 °C then the radiation emitted by the detector is a
temperature conversion can then be checked by using significant fraction of the in-coming radiation and must
this value of S in equation (3) to re-calculate the value of be accounted for. Some IR thermometers use thermoe-
T = 323.15 K.) lectric devices to cool the detector well below ambient. In
these more expensive models, the detector radiation can
While many low-temperature IR thermometers oper-
normally be neglected for most target temperatures, but
ate over this wavelength range of 8–14 µm, there are
these instruments are not covered in this guide.
many other ranges also in use, such as 8–13 µm and
7–18 µm. It is important to check the specifications to
determine the actual wavelength range used. Measurement Equation
Including all of these effects – the emissivity of the
target, the reflected radiation, and the radiation emitted
by the detector – results in a measured signal, Smeas, at

Measurement Standards Laboratory of New Zealand • fax: +64 (0)4 931 3003 • e-mail: [email protected] • website: http://msl.irl.cri.nz 2

https://www.scribd.com/document/358553314/MSL-Tech-Guide-22-calibration-Infrared-Thermometer 4/16
10/5/23, 11:12 AM MSL Tech Guide 22 - Calibration Infrared Thermometer | PDF | Infrared | Thermometer

of 8 
the output of the detector, given by a measurement Armed with this information, the IR thermometer
equation that consists of the sum of three terms: processes the measured signal as follows: first the
measured signal is divided by the instrumental emissivity
Smeas = ε sS(Ts ) + (1 − ε s )S(Tw ) − S (Td ) , (6) setting; then a quantity corresponding to the signal at the
detector temperature is added (i.e., an amount given by
where Ts is the temperature of the target (the quantity of equation (1) with T = Td inserted); finally, the resulting
interest), Tw is the temperature of the surroundings (of- signal value is converted to a measured temperature
ten the walls of the room), Td is the temperature of the value, Tmeas. This is represented mathematically as:
detector, εs is the emissivity of the target’s surface, and
1 – εs is its reflectivity. Each of the three terms on the Smeas
S(Tmeas ) = + S(Td ) . (7)
right-hand side of equation (6) corresponds to a different ε instr
component of radiation: the first, εsS(Ts), corresponds to
the radiation emitted by the target; the second,
Measurement Errors
(1 – εs)S(Tw), to the radiation reflected from the sur-
roundings; and the third, S(Td), to the radiation emitted What are the consequences of this signal process-
by the detector. Each of the S(T) functions can be calcu- ing? To answer this, we first substitute Smeas from equa-
lated by inserting the appropriate temperature value into tion (6) into equation (7):
equation (1), remembering to first convert any degrees
Celsius temperatures into kelvin temperatures. ε sS(Ts ) + (1 − ε s )S(Tw ) − (1 − εinstr )S(Td )
S(Tmeas ) = . (8)
ε instr

Processing the Measured Signal

We can rewrite this as the sum of three terms:
If the measured signal, Smeas, given by equation (6) is
substituted into the signal-to-temperature conversion
equation (3), the result will not be the target temperature S(Tmeas ) = S(Ts )
Ts, because of the influences of the target emissivity, the (1− ε instr )
temperature of the surroundings, and the temperature of
+ [S(Tw ) − S(Td )] (9)
ε instr
the detector (εs, Tw, and Td). In order to produce a read-
(ε s − εinstr )
ing that better represents the target temperature, the IR + [S(Ts ) − S(Tw )].
thermometer pre-processes the measured signal before ε instr
conversion to temperature, in effect applying corrections
for the above influences. To do this reliably, the ther- If the second and third terms (lines) of this equation are
mometer must somehow obtain values for the three in- both zero, then the equation is simply S(Tmeas) = S(Ts).
fluence variables, εs, Tw, and Td. This would imply that Tmeas = Ts, that is, that the meas-
The detector temperature, Td, can be accurately de- ured temperature is equal to the target temperature,
termined by using an internal temperature probe which is what we are striving for. However, when either
mounted directly on the detector. Its measurement oc- of the second or third lines of the equation is not zero,
curs automatically as part of the IR thermometer’s meas- they represent error terms. In this case, the measured
urement process and is completely hidden from the user. temperature is no longer equal to the target temperature.
For IR thermometers with an adjustable instrumental The second line is zero when either εinstr = 1 or
emissivity setting, εinstr, the user can inform the ther- Tw = Td. The condition Tw = Td is the manufacturers’ as-
mometer of the value of the target emissivity by setting sumption mentioned above. Equation (9) allows us to
εinstr correctly. For fixed-emissivity instruments, the quantify the error when this condition doesn’t hold. Set-
manufacturer designs the thermometer to carry out ting εinstr = 1 (if possible) is often a good strategy be-
measurements only on specific objects (i.e., objects hav- cause then the measured temperature, Tmeas, is inde-
ing a specific emissivity). Many food products, plastics, pendent of Td, as can be seen from both equations (8)
materials made from organic compounds (such as pa- and (9). However, this may introduce error through the
per, wood, and skin), and most paints have emissivities third line in equation (9).
close to 0.95 in the 8–14 µm range (i.e., they are almost This third line is zero when either εinstr = εs or Ts = Tw,
black), so 0.95 is often chosen as the fixed instrumental that is, when the instrumental emissivity matches the
emissivity setting. emissivity of the target, or the target temperature is the
Finally, the temperature of the surroundings, Tw, will same as the temperature of the surroundings. Incorrectly
depend on the measurement situation, and will vary from setting the instrumental emissivity leads to an error that
measurement to measurement. For this influence vari- increases as the difference between these two tempera-
able, most IR thermometer manufacturers make the im- tures increases. The user has no control over this error
plicit assumption that Tw will be approximately the same for fixed-emissivity instruments, unless the condition
as the detector temperature, Td. In other words, they as- Ts = Tw holds.
sume that all measurements are to be performed in am- To summarise, it is only when Tw = Td and εinstr = εs
bient surroundings. This assumption is usually fine dur- that the reading on an IR thermometer will be equal to
ing calibration in a well-controlled laboratory, but it can the true target temperature. If either of these conditions
be quite misleading in other measurement situations. does not hold, then the thermometer reading will be in
These include cool stores, where the temperature of the error. The only exception is when the target is a black-
surroundings is well below the temperature of the ther- body (εs = 1), or blackbody conditions prevail (Ts = Tw),
mometer, and situations where products to be measured and the instrumental emissivity has been also set to 1, in
are surrounded by hot objects, such as heaters, which which case the reading depends on neither Tw nor Td.
are well above the temperature of the thermometer.

Measurement Standards Laboratory of New Zealand • fax: +64 (0)4 931 3003 • e-mail: [email protected] • website: http://msl.irl.cri.nz 3

https://www.scribd.com/document/358553314/MSL-Tech-Guide-22-calibration-Infrared-Thermometer 5/16
10/5/23, 11:12 AM MSL Tech Guide 22 - Calibration Infrared Thermometer | PDF | Infrared | Thermometer

of 8 

Calibration
The errors discussed above occur in almost all
measurements with IR thermometers, and care must be
taken to ensure that these errors are not excessive.
They also occur during calibration because the condi-
tions for which the errors are zero (Tw = Td and εinstr = εs)
very rarely both hold. So how do we calibrate an IR
thermometer when we expect errors in the readings,
even for a perfect thermometer? The answer is that we
must first calculate the expected readings for an ideal
device under the calibration conditions and see how
close the actual readings are to the expected ones. Or
put another way, since we invariably use blackbodies to
calibrate IR thermometers, we need to calculate “black-
body corrections”, which we apply to our reference ther-
mometer readings before comparing with the readings of
the device under calibration.
Conventional blackbodies are made from cavities so
that their effective emissivity is very close to 1 (see Fig-
ure 2). These blackbody cavities include purpose-built
furnaces and inserts into dry-block calibrators. We can
estimate the effective emissivity of a cavity, εbb, from its
length L, the radius of its aperture, r, and the emissivity
of the material from which it is made, εs:
Figure 3. A flat-plate calibrator being viewed by an infrared
2 thermometer. Photo courtesy of Fluke Corporation, Hart Scien-
r  tific Division, reproduced with permission.
ε bb = 1 − (1 − ε s )   . (10)
L

For example, a cavity made out of a material that has an The reference thermometer, which measures the
emissivity of 0.9 (oxidised stainless steel), whose length true temperature of the blackbody, can be either a con-
is 150 mm and whose aperture radius is 25 mm, has an tact thermometer, such as a platinum resistance ther-
effective emissivity of: mometer, or a reference infrared thermometer. In the
special case of the ice-point blackbody, no separate ref-
2 erence thermometer is required. These three methods of
 25  calibration are discussed below.
ε bb = 1 − (1 − 0.9)  
 150 
= 0.997. Contact Thermometer as Reference
When a contact thermometer is used as the refer-
Plate-flat calibrators are also used as blackbody ence, it is important that it is positioned in such a way
sources (see Figure 3). However, their emissivity is usu- that it measures the true temperature of the blackbody.
ally close to 0.95, so they are not true blackbodies. For This is especially important for flat-plate calibrators,
the purpose of this guide, though, both types of calibra- where temperature gradients can lead to differences be-
tion source will be referred to as blackbodies, and they tween the temperature of the plate where the IR ther-
will be distinguished by their effective emissivities. mometer under calibration is aimed and the temperature
at the location of the reference thermometer (usually be-
hind the surface of the plate).
In a calibration laboratory, the temperature of the
surroundings is usually equal to ambient temperature,
Tamb. Thus, we can rewrite equation (8) for the expected
thermometer reading, Texp, as:

ε bbS(Tref ) + (1 − ε bb )S(Tamb ) − (1 − εinstr )S(Td )

S(Texp ) = , (11)
r ε instr

where εbb is the effective emissivity of the blackbody and

Tref is the true temperature of the blackbody, as deter-
mined by the reference thermometer. The second term
in the numerator of equation (11) corresponds to ambi-
ent radiation that enters the blackbody cavity from the
L surroundings and finds its way back out, or the radiation
that is reflected off the flat plate. For the case of the cav-
Figure 2. A blackbody cavity of length L and aperture radius r, ity, whose effective emissivity is generally very close to
whose walls have an emissivity εs. The effective emissivity of 1, this term will be very small.
the cavity is given by equation (10). In use, the cavity is heated For a given set of conditions, the expected tempera-
uniformly in a furnace or dry-block calibrator. ture can be calculated by evaluating the right-hand side

Measurement Standards Laboratory of New Zealand • fax: +64 (0)4 931 3003 • e-mail: [email protected] • website: http://msl.irl.cri.nz 4

https://www.scribd.com/document/358553314/MSL-Tech-Guide-22-calibration-Infrared-Thermometer 6/16
10/5/23, 11:12 AM MSL Tech Guide 22 - Calibration Infrared Thermometer | PDF | Infrared | Thermometer

of 8 
of equation (11) (with the aid of equation (1) to deter- shown in Figure 6. When the detector temperature also
mine S(Tref), S(Tamb), and S(Td)), thus giving S(Texp), matches the ambient temperature, the corrections are all
then equation (3) can be used to extract Texp from zero regardless of the blackbody temperature.
S(Texp). The “blackbody correction”, ∆Tbb, is the differ-
ence between the expected reading and the reference
thermometer’s reading: 20

15
∆Tbb = Texp − Tref . (12)
10
This blackbody correction can be added to the refer-
ence thermometer reading to give the expected IR ther- 5
mometer reading. These calculations are easily per-
formed in an Excel spreadsheet (see 0 T = 15 °C
d
http://msl.irl.cri.nz/sites/all/files/training-manuals/tg22-
Td = 20 °C
blackbodycorrections.xls). -5
An example calculation is given in Table 1 for the Td = 25 °C

calibration of an 8–14 µm IR thermometer whose in- -10

strumental emissivity is fixed at 0.95, using a blackbody -50 0 50 100 150 200 250 300 350 400 450 500

cavity with an effective emissivity of 0.997. In the exam- Blackbody temper ature / °C
ple, the ambient temperature is 20 °C and the detector Figure 4. Corrections to the temperature of a blackbody calibra-
temperature is 21 °C. Note that the detector temperature tion source, whose effective emissivity is εbb = 0.997, to give the
is not generally known, as it is not displayed on the de- expected reading for an 8–14 µm IR thermometer with a f ixed
vice, so its value must be approximated, or guessed at, emissivity of εinstr = 0.95. The ambient temperature is assumed
in order to calculate the blackbody corrections. The val- to be Tamb = 20 °C and the detector temperature is as given on
ues in Table 1 provide a means for you to check the ac- the graph.
curacy of your spreadsheets.
Figure 4 shows the blackbody corrections for the en- 10
tire range of blackbody temperatures from –50 °C to ε = 0.997
bb
500 °C. Also shown is the effect of various detector tem- 5
ε = 0.95
bb
peratures. The detector temperature may differ from
ambient if the thermometer has been stored in a room at 0
a different temperature to the calibration laboratory and
not given time to equilibrate before measurements are -5
made. The detector may also become warmer than am-
bient if placed in front of a hot blackbody for a length of -10
time. As Figure 4 shows, there is only a weak depend-
ence of the blackbody correction on detector tempera- -15
ture, mainly at lower temperatures. The bulk of the cor-
rection is due to the fact that εinstr ≠ εbb. -20
-50 0 50 100 150 200 250 300 350 400 450 500
If the instrumental emissivity is not fixed, then setting
Blackbody temper ature / °C
it to 1 removes any dependence of the reading on the
detector temperature (see equation (11)). When the cali- Figure 5. The solid line is for the same conditions as Figure 4,
bration is carried out using a blackbody cavity with an except that the emissivity setting on the IR thermometer is
effective emissivity close to 1, then we should expect the εinstr = 1. The dashed line is for a flat-plate calibrator with emis-

blackbody corrections to be much smaller than those sivity of εbb = 0.95, again for εinstr = 1. For both curves, the
blackbody corrections are independent of the detector tempera-
shown in Figure 4 because now εinstr ≈ εbb. This is indeed
ture.
the case, as shown by the solid line in Figure 5. How-
ever, if the blackbody is a flat-plate calibrator with an
emissivity of 0.95, then the larger blackbody corrections 0.8
given by the dashed line in Figure 5 need to be applied.
0.6
For a flat-plate blackbody, the blackbody corrections
would be smaller if the instrumental emissivity were set 0.4
to match εbb, i.e., εinstr = 0.95. These corrections are
0.2

Table 1. Calculation of the blackbody corrections for three val- 0

ues of Tref for an 8–14 µm IR thermometer (A = 9.36 µm,
-0.2
B = 178 µm.K) with εinstr = 0.95, εbb = 0.997, Tamb = 20 °C, and
T = 15 °C
Td = 21 °C. d
-0.4
Td = 20 °C
-0.6 Td = 25 °C
Texp ∆Tbb
Tref S(Tref) S(Tamb) S(Td) S(Texp)
(°C) (°C) -0.8
(°C) [eq (1)] [eq (1)] [eq (1)] [eq (11)] -50 0 50 100 150 200 250 300 350 400 450 500
[eq (3)] [eq (12)]
Blackbody temperature / °C
–50 0.00175 0.00732 0.00744 0.00147 –56.5 –6.5
Figure 6. Same conditions as Figure 4, but for an IR thermome-
100 0.02025 0.00732 0.00744 0.02088 103.0 3.0 ter with an emissivity setting of εinstr = 0.95 viewing a flate-plate
500 0.16773 0.00732 0.00744 0.17566 516.4 16.4 calibrator with emissivity of εbb = 0.95. When Td = Tamb = 20 °C,
the blackbody corrections are zero for all temperatures.

Measurement Standards Laboratory of New Zealand • fax: +64 (0)4 931 3003 • e-mail: [email protected] • website: http://msl.irl.cri.nz 5

https://www.scribd.com/document/358553314/MSL-Tech-Guide-22-calibration-Infrared-Thermometer 7/16
10/5/23, 11:12 AM MSL Tech Guide 22 - Calibration Infrared Thermometer | PDF | Infrared | Thermometer

of 8 

The calibration procedure is summarised by the fol- In this section, we assume that the readings of the
lowing steps: reference IR thermometer are made at the same time as
the measurements of the device under calibration, so
1. Determine the minimum and maximum wavelengths
that the ambient conditions during a sequence of meas-
for the operating wavelength range of the device un- urements are identical. We also assume that the instru-
der calibration. These should be stated in the ther-
mental emissivity setting on the reference thermometer
mometer’s specifications under “spectral response”. is 1, so that its readings are independent of the tempera-
2. Using these values in equations (4) and (5), calculate ture of its detector. Finally, it is also assumed that the
the A and B coefficients of the thermometer re-
operating wavelength range of the reference thermome-
sponse function. ter is the same as that of the device under calibration
3. Determine the blackbody emissivity, εbb, either as an (the A and B values are the same for both instruments).
effective value from equation (10) for a cavity, or di- If this latter condition doesn’t hold, then the calculation of
rectly from the specifications for a flat-plate calibra- the blackbody corrections is more complicated and re-
tor. quires additional information.
4. Set the instrumental emissivity, εinstr, as close as Applying equation (8) to both the reference ther-
possible to εbb (unless requested otherwise by the mometer and the thermometer under calibration, and
client), or if the instrumental emissivity is fixed, de- determining the difference in measured signals, gives:
termine its value from the thermometer’s specifica-
tions. (1− ε instr )
5. Measure the ambient temperature, Tamb, with an air- S(Texp ) = S(Tref ) + [S(Tref ) − S(Td )] . (13)
ε instr
temperature probe.
6. Estimate the detector temperature, Td. This is likely
to be the same as ambient temperature provided the As before, Texp is the expected reading for the device
IR thermometer has had sufficient time to equilibrate under calibration, εinstr is its instrumental emissivity set-
with the calibration laboratory and if the IR ther- ting, and Td is the temperature of its detector. Tref is the
mometer is not heated excessively by radiation from reading on the reference thermometer (which is gener-
the blackbody. If the instrumental emissivity is set to ally not equal to the temperature of the blackbody, par-
1, the value of Td is not required. ticularly when a flat plate is used). The blackbody correc-
7. For each calibration point, read the reference ther- tions are, again, given by ∆Tbb = Texp – Tref. Notice that
mometer, Tref, and calculate the expected IR ther- equation (13) does not depend on the true temperature
mometer reading, Texp, using equation (11). Compare of the blackbody, or on the value of its effective emissiv-
the actual reading on the IR thermometer with the ity, or on the ambient temperature. The fact that these
value of Texp. The difference between the expected values do not need to be known is a major advantage of
temperature and the actual reading is the correction this method. Additionally, when εinstr = 1, the blackbody
that should be reported on the calibration certificate. corrections are zero for all temperatures. This is con-
trasted with equation (11) where, under this condition,
Infrared Thermometer as Reference the blackbody corrections still depend on ambient tem-
perature and the effective emissivity of the blackbody.
In some cases a separate IR thermometer is used as
Figure 7 gives the blackbody corrections for this
the reference device to measure the temperature of the
method for a range of instrumental emissivity settings on
blackbody calibration source. This reference IR ther-
the device under calibration, as a function of the refer-
mometer must itself have already been calibrated. The
ence thermometer reading. The temperature of the de-
blackbody corrections for this method of calibration differ
tector of the device under calibration is assumed to be
to those given in the previous section.
Td = 20 °C. Although not shown, the variation of the cor-
Some flat-plate calibrators have their digital tempera-
rections with changes in detector temperature is ap-
ture readout adjusted by the manufacturer using an IR
proximately the same as that shown in Figure 4.
thermometer as the reference. This is often referred to
as a “radiometric calibration” of the display. This takes 20
into account the emissivity of the plate, ambient reflec-
ε = 0.95
tions, and the instrumental emissivity setting of the de- 15
instr
ε = 0.97
vice under calibration (through a setting on the flat-plate instr

controller). In effect, this type of calibrator automatically ε

instr
= 0.99
10
applies the corrections discussed in the previous sec-
tion, and the reading on the display gives the expected 5
reading of the device under calibration. However, care
must be taken when using this display as the reference 0
temperature, as the conditions under which it was cali-
brated may differ from those present when the plate is -5
used as a calibration source. In particular, the ambient
temperature may not be the same (causing the reflected -10
radiation to be different), the temperature of the detector -50 0 50 100 150 200 250 300 350 400 450 500
of the device under calibration may differ from ambient, Reference IR thermometer reading / °C
and the reference thermometer’s wavelength range may Figure 7. Corrections required to the readings of a reference IR
be different from that of the device under calibration. thermometer aimed at a blackbody calibration source when
Normally, the specified uncertainty for the calibrator will calibrating an 8–14 µm IR thermometer with a detector tem-
include a component to account for modest variations in perature of Td = 20 °C and with instrumental emissivity as given
ambient and detector temperatures. The user should on the graph. The reference thermometer also operates from
check that the operating wavelength ranges match. 8–14 µm and its instrumental emissivity is set to 1.

Measurement Standards Laboratory of New Zealand • fax: +64 (0)4 931 3003 • e-mail: [email protected] • website: http://msl.irl.cri.nz 6

https://www.scribd.com/document/358553314/MSL-Tech-Guide-22-calibration-Infrared-Thermometer 8/16
10/5/23, 11:12 AM MSL Tech Guide 22 - Calibration Infrared Thermometer | PDF | Infrared | Thermometer

of 8 
The calibration procedure when an IR thermometer 0 °C, no reference thermometer is required for a calibra-
is used as the reference is summarised as follows: tion at this temperature.
To determine the expected reading on an IR ther-
1. Determine the minimum and maximum wavelengths
mometer when aimed at an ice-point blackbody, we can
for the operating wavelength range of the device un-
insert the value Tref = 0 °C into equation (11). In this
der calibration. These should be stated in the ther-
case, the blackbody correction is the same as the ex-
mometer’s specifications under “spectral response”.
pected reading when expressed in degrees Celsius. Fig-
2. Check that these minimum and maximum wave-
ure 8 shows the blackbody correction as a function of
lengths are the same for the reference IR thermome-
the instrumental emissivity setting on the device under
ter.
calibration for three different detector temperatures, as-
3. Using these values in equations (4) and (5), calculate
suming that the ambient temperature is Tamb = 20 °C and
the A and B coefficients of the thermometer re-
the effective emissivity of the ice-point blackbody cavity
sponse function (which apply to both thermometers).
is εbb = 0.999.
4. Set the instrumental emissivity, εinstr, on the device
The calibration procedure for the ice-point calibration
under calibration as close to 1 as possible (unless
is identical to that for the contact thermometer as refer-
requested otherwise by the client), or if the instru-
ence, except that in step 7 there is no reference ther-
mental emissivity is fixed, determine its value from
mometer to read; instead a single reference temperature
the thermometer’s specifications.
5. Set the instrumental emissivity on the reference IR of Tref = 0 °C is used in equation (11).
thermometer to 1.
6. Estimate the detector temperature, Td, for the device Detector Temperature
under calibration. This is likely to be the same as A point of difficulty throughout this guide is that the
ambient temperature provided the device under cali- value of Td is not known to the user. It is measured in-
bration has had sufficient time to equilibrate with the ternally by the instrument, but is not displayed on the
calibration laboratory and if the device under calibra- readout. The best assumption is that Td will be the same
tion is not heated excessively by radiation from the as, or close to, ambient temperature, and this may re-
blackbody. If its instrumental emissivity is set to 1, quire some conditioning of the instrument before use.
the value of Td is not required. However, there are some situations where ensuring that
7. For each calibration point, read the reference IR
Td = Tamb is not practicable. One is where the IR ther-
thermometer, Tref, and calculate the expected read-
mometer is used for an extended period in front of a hot
ing, Texp, of the device under calibration using equa-
source (e.g., during calibration at high temperatures),
tion (13). Compare the actual reading on the device
where the radiation from the source may heat the detec-
under calibration with the value of Texp. The differ-
tor above ambient. The second is where the IR ther-
ence between the expected temperature and the ac-
mometer is used inside a walk-in freezer, where the am-
tual reading is the correction that should be reported
bient temperature is well below the specified operating
on the calibration certificate.
temperature for the thermometer. In this case, the ther-
mometer can still be used successfully, but the detector
Ice Point as Reference must be maintained above, usually, 0 °C.
The ice point is an accurate and reliable reference Dependence on the detector temperature is removed
temperature suitable for checking the accuracy and drift if the instrumental emissivity is set to 1 (this is precluded
over time of IR thermometers. Since the emissivity of ice in fixed-emissivity instruments). On the other hand, rela-
is εs = 0.96 in the infrared part of the spectrum, a good tively large errors in the estimate (guess) of the detector
blackbody cavity, with an effective emissivity close to 1 temperature can usually be tolerated without significant
(according to equation (10)), can easily be constructed error in the reading.
using shaved ice (see [1]). Because the temperature of a For adjustable-emissivity instruments, the detector
properly constructed ice-point is defined to be precisely temperature can, in fact, be inferred from two measure-
ments of the same target using two different instrumental
emissivity settings. The temperature and emissivity of
0.2 the target, and the ambient temperature, do not need to
0 be known. All that is required is that these values are
constant during the two measurements. The detector
-0.2
temperature, Td, can be calculated from the equation:
-0.4

-0.6 ε instr1S(Tmeas1 ) − εinstr2S(Tmeas2 )

S(Td ) = , (14)
-0.8 ε instr1 − εinstr2

-1.0 Td = 15 °C
where Tmeas1 and Tmeas2 are the two IR thermometer
-1.2 Td = 20 °C
readings when the instrumental emissivity is set to εinstr1
-1.4 Td = 25 °C and εinstr2, respectively. The accuracy of calculations
-1.6 made with equation (14) is best when εinstr1 and εinstr2 are
0.95 0.96 0.97 0.98 0.99 1 widely spaced; values of 0.5 and 1 are usually adequate.
Instrumental emissivity setting For example, two readings of a blackbody using an
Figure 8. Expected reading for an 8–14 µm IR thermometer as 8–14 µm thermometer might be Tmeas1 = 141.8 °C when
a function of instrumental emissivity setting when viewing an the instrumental emissivity is set to εinstr1 = 1, and
ice-point blackbody cavity with effective emissivity of Tmeas2 = 219.4 °C when εinstr2 = 0.5. Substituting these
εbb = 0.999. The ambient temperature is Tamb = 20 °C. values into equation (14) gives Td = 21.5 °C.

Measurement Standards Laboratory of New Zealand • fax: +64 (0)4 931 3003 • e-mail: [email protected] • website: http://msl.irl.cri.nz 7

https://www.scribd.com/document/358553314/MSL-Tech-Guide-22-calibration-Infrared-Thermometer 9/16
10/5/23, 11:12 AM MSL Tech Guide 22 - Calibration Infrared Thermometer | PDF | Infrared | Thermometer

of 8 

Calculations of this sort are useful for determining It should be stressed that the calibration is designed
the expected variation of detector temperature as the only to determine how well the thermometer conforms to
conditions inside the laboratory change. While it is not its expected behaviour (i.e., how well Tmeas conforms to
possible to do this for fixed-emissivity instruments, it may equation (8)). In use, it is the responsibility of the user to
be possible to infer their behaviour from measurements ensure that the measurement conditions are such that
made with similar adjustable-emissivity instruments. the value of Tmeas (after applying any certificate correc-
tions) is a good approximation to Ts, the true tempera-
Conclusion ture of the target. The necessary conditions are that both
εinstr = εs and Td = Tw hold. That is, the instrumental
Because low-temperature IR thermometers are de- emissivity must be set to the emissivity of the target and
signed to automatically overcome the problem of re- the detector temperature must equal the temperature of
flected radiation when used to carry out temperature the surroundings. When either, or both, of these condi-
measurements, calibration of such devices is not a tions does not hold, then the resulting error must be cal-
straightforward task. Procedures for calibrating IR ther- culated using equation (9). Examples of these calcula-
mometers need to be carefully designed to take into ac- tions are given in [2] and [3].
count the instrumental emissivity setting on the device,
its detector temperature, and ambient temperature, as
well as the properties of the blackbody calibration source References
and the reference thermometer. [1] MSL Technical Guide 2: “Infrared Thermometry Ice
The methods given in this technical guide allow the Point”, http://msl.irl.cri.nz.
calibration laboratory to calculate the expected reading
for an ideal device, or the correction to the reference [2] P Saunders, “Reflection errors for low-temperature
thermometer reading (the “blackbody correction”), radiation thermometers”, in Proceedings of TEMP-
against which the reading on the device under calibra- MEKO 2001, 8th International Symposium on
tion should be compared. A calibration certificate should Temperature and Thermal Measurements in Indus-
state the device corrections with respect to these ex- try and Science, edited by B Fellmuth, J Seidel, G
pected readings, not with respect to the reference ther- Scholz, VDE Verlag GmbH, Berlin, 149–154, 2002.
mometer readings. For example, referring to Figure 8, an [3] P Saunders, “Calibration and use of low-
8–14 µm thermometer with an instrumental emissivity of temperature direct-reading radiation thermome-
0.95 is expected to read –1.2 °C (for detector and ambi- ters”, Measurement Science and Technology, 20,
ent temperatures both equal to 20 °C). If the actual read- 025104, 2009.
ing was –0.9 °C, then the reported correction for this
reading would be –0.3 °C. The certificate should also
state the instrumental emissivity setting. Prepared by Peter Saunders, June 2009.

MSL is New Zealand’s national metrology institute, operating within Industrial Research Limited under the authority of the
New Zealand Measurement Standards Act 1992.

Measurement Standards Laboratory of New Zealand • fax: +64 (0)4 931 3003 • e-mail: [email protected] • website: http://msl.irl.cri.nz 8

Reward Your Curiosity

Everything you want to read.
Anytime. Anywhere. Any device.

https://www.scribd.com/document/358553314/MSL-Tech-Guide-22-calibration-Infrared-Thermometer 10/16
10/5/23, 11:12 AM MSL Tech Guide 22 - Calibration Infrared Thermometer | PDF | Infrared | Thermometer

Read free for 30 days

No Commitment. Cancel anytime. of 8 

Share this document

    

You might also like

Document 9 pages

TG22 Version 3 MSL Technical Guide

BalaMayurPriyan
No ratings yet

Document 8 pages

tg22-july-2009v2.pdf
ShrikantKarwa
No ratings yet

Document 32 pages

IR-Basics
chitturi jagadish
No ratings yet

https://www.scribd.com/document/358553314/MSL-Tech-Guide-22-calibration-Infrared-Thermometer 11/16
10/5/23, 11:12 AM MSL Tech Guide 22 - Calibration Infrared Thermometer | PDF | Infrared | Thermometer

of 8 

Document 9 pages

Temperature Measurement
barry nancoo
No ratings yet

Document 9 pages

Emissivity and accuracy

alexander
100% (1)

Document 9 pages

Emissivity measurement and temperature correction accuracy considerations

Giust
No ratings yet

https://www.scribd.com/document/358553314/MSL-Tech-Guide-22-calibration-Infrared-Thermometer 12/16
10/5/23, 11:12 AM MSL Tech Guide 22 - Calibration Infrared Thermometer | PDF | Infrared | Thermometer

of 8 

Document 27 pages

Thermal Imaging
Satyam Kumar
100% (1)

Document 5 pages

Infrared Thermometer
Lorenz Banada
No ratings yet

Document 34 pages

2
arwa amaira
No ratings yet

https://www.scribd.com/document/358553314/MSL-Tech-Guide-22-calibration-Infrared-Thermometer 13/16
10/5/23, 11:12 AM MSL Tech Guide 22 - Calibration Infrared Thermometer | PDF | Infrared | Thermometer

of 8 

Document 57 pages

embeded based human security

sreeragpm
No ratings yet

Document 4 pages

Industrial Temperature Monitoring System Design Ba

boy sad
No ratings yet

Document 12 pages

Temperature Sensor Guide

river147
No ratings yet

https://www.scribd.com/document/358553314/MSL-Tech-Guide-22-calibration-Infrared-Thermometer 14/16
10/5/23, 11:12 AM MSL Tech Guide 22 - Calibration Infrared Thermometer | PDF | Infrared | Thermometer

of 8 

Related titles

Document Document Document Document Document

TG22 Version 3 tg22-july- IR-Basics Temperature Emissivity and

MSL Technical… 2009v2.pdf Added by chitturi… Measurement accuracy
Added by… Added by… Added by barry… Added by alexander

0 ratings 0 ratings 0 ratings 0 ratings 100%

About Support Legal Social

About Scribd Help / FAQ Terms Instagram

Press Accessibility Privacy Twitter

Our blog Purchase help Copyright Facebook

Join our team! AdChoices Cookie Preferences Pinterest

Contact us Publishers Do not sell or share my

personal information
Invite friends

Gifts

Scribd for enterprise

https://www.scribd.com/document/358553314/MSL-Tech-Guide-22-calibration-Infrared-Thermometer 15/16
10/5/23, 11:12 AM MSL Tech Guide 22 - Calibration Infrared Thermometer | PDF | Infrared | Thermometer

Get our free apps

of 8 
Audiobooks • Books • Documents • Magazines • Podcasts • Sheet music

Language: English

Copyright © 2023 Scribd Inc.

https://www.scribd.com/document/358553314/MSL-Tech-Guide-22-calibration-Infrared-Thermometer 16/16