Infrared thermometer calibration: blackbody, emissivity and spot size

What does an infrared thermometer actually measure?

An infrared thermometer, also called a radiation thermometer or pyrometer, works like a camera. A lens and a band-pass filter focus thermal radiation onto a detector, and the instrument converts the signal into a temperature (NPL Good Practice Guide No. 125, Chapter III). Because it reads radiation and not the temperature of a sensor in contact with the object, it must be calibrated against a known source of thermal radiation.

The waveband depends on the temperature it is made for. Instruments for room temperature commonly work between 8 µm and 14 µm, others between 3 µm and 5 µm, and instruments for high temperatures, above about 700 °C, between 0.65 µm and 1.0 µm (NPL GPG 125).

Which standards cover infrared thermometer calibration?

ASTM E2847 does not cover instruments above 1000 °C or narrowband instruments, and it does not treat the design of infrared thermometers. It may also be applied to other radiation thermometers and to thermal imagers.

Blackbody cavity or flat-plate source?

NPL Good Practice Guide No. 125 calls the blackbody the reference condition: a calibration uses a source of known temperature with an emissivity of, or close to, 1, and calibrations at other emissivities can then be calculated. A well-designed cavity comes close. NPL's variable-temperature sources are generally at least 0.999, and NPL notes that a well-built cavity can exceed 0.9999. The temperature comes from a calibrated contact sensor, such as a platinum resistance thermometer or a thermocouple, which gives traceability to the ITS-90. Often the cavity and sensor are calibrated together against a reference blackbody.

Cavities can be large and slow, so blackened plates, usually grooved, are common. A plate is heated from behind with a sensor near its surface. Because the surface is exposed, its emissivity is likely to be substantially below 1, and there are temperature differences across the plate and between the sensor and the surface. NPL's guide says the plate itself must be calibrated against a reference blackbody.

• Metrological traceability explained
• Uncertainty of the source's contact sensor
• How to build a measurement uncertainty budget

How should the emissivity be set during calibration?

Set the thermometer to match the source where the setting can be changed. IEC TS 62492-1:2008 clause 4.1.1 gives data sheet values at an emissivity setting of 1 unless stated otherwise, and for an instrument with a fixed setting other than 1 it requires that value to be stated with the specification.

A real surface also reflects radiation from its surroundings. NPL Good Practice Guide No. 125 writes the radiance reaching the thermometer as L = εL(Ts) + (1 − ε)L(Tb), where ε is the emissivity, Ts the surface temperature and Tb the background temperature. With a cavity and a setting of 1, the second term is close to zero. With a flat plate, or a thermometer fixed at a setting that differs from the source, use that relation to work out what the thermometer should read from the source emissivity and the background radiance. ASTM E2847 gives the test setup and a structure for the uncertainty calculation.

A certificate says nothing about the emissivity of your own part. NPL warns that errors of tens of degrees are common when the emissivity of a surface is wrongly estimated, so service use needs its own control: a known value for that material, or a high-emissivity coating where one is allowed (NPL GPG 125).

Distance, spot size and the size-of-source effect

The distance ratio, often written D:S, is the measuring distance divided by the diameter of the field of view (IEC TS 62492-1:2008 clause 3.1.6). A thermometer rated 12:1 at 300 mm sees a spot 25 mm across. The specification should state from which part of the instrument the distance is measured (clause 4.1.1.4.2), and the calibration should use that point.

The field of view has no sharp edge. IEC TS 62492-1:2008 clause 4.1.1.5 therefore states it as the diameter that contains a stated fraction of the signal, at least 90% and typically 90%, 95% or 99%. Some radiation from outside that circle still reaches the detector, so the reading changes with the size of the radiating area. That is the size-of-source effect (clause 3.1.7). The BIPM radiation thermometry guide attributes it to scatter and reflections in the optics and notes that dust on the objective can change it.

Two practical rules follow. Use a source significantly larger than the stated spot, as NPL's guide advises for any surface measured. And test for the effect: change the source aperture or the distance and see whether the reading moves. IEC TS 62492-1:2008 clause 4.1.1.7 compares the nominal field of view with twice its area.

What ambient conditions matter?

• Warm-up. Let the thermometer stabilize in the room first. IEC TS 62492-1:2008 treats its specifications as valid after the warm-up time and under stable ambient conditions, and defines an added uncertainty when the instrument's own temperature departs from them (clause 3.1.10).
• Reflections. Lights, heaters and even people add to the signal at low temperatures (NPL GPG 125). A flat plate reflects part of what surrounds it, so keep its surroundings steady.
• Line of sight. Dust, water vapor and carbon dioxide can absorb or scatter radiation (NPL GPG 125). Keep the path short and clear.
• Lens. Dust on the objective can change the size-of-source effect (BIPM radiation thermometry guide). Clean it the way the maker says to.
• Record the room temperature and humidity, since a humidity effect may be stated too (IEC TS 62492-1:2008 clause 3.1.11).

What is the difference between an accuracy check and a calibration?

ASTM E2847 is titled for both calibration and accuracy verification. A calibration compares the thermometer with a traceable source under stated conditions and reports the error and its uncertainty at each point. A check confirms that the thermometer still behaves, at one or a few points, and it does not replace a calibration or reset the due date.

A check only tells you something when it is repeated under the same conditions: the same stable source, the same distance and the same emissivity setting, compared with earlier results. ISO/IEC 17025:2017 clause 6.4.10 requires a procedure for intermediate checks where they are needed to keep confidence in the equipment.

• Calibration vs verification
• Intermediate checks between calibrations

Pyrometers, narrowband instruments and very high temperatures

ASTM E2847 stops at 1000 °C and excludes narrowband instruments. ASTM E1256 gives the test methods for single waveband radiation thermometers. Instruments for high temperatures use short wavelengths, where the temperature error may be 10 to 20 times smaller than the emissivity error. An emissivity error of 1% then gives less than 0.1% of the temperature in kelvin (NPL GPG 125).

Above the freezing point of silver, 961.78 °C, the ITS-90 is defined through radiation thermometry, with the silver, gold or copper fixed point as the reference (BIPM Guide to the Realization of the ITS-90, Radiation Thermometry). Pyrometers in that range trace to fixed-point or high-temperature blackbodies through a calibration laboratory.

Two-color or ratio pyrometers help with small targets that do not fill the field of view and in dirty atmospheres, but they do not remove the emissivity problem: the emissivity cancels only if it is the same at both wavelengths (NPL GPG 125).

How often should an infrared thermometer be calibrated?

No standard sets the interval, and NIST does not recommend a set recalibration interval. ILAC-G24:2022 / OIML D 10:2022 lists the factors to weigh, including the tendency to drift and the severity of use. Start from the maker's recommendation, then adjust it from your as-found history and the results of your checks. Shorten it after a drop, damage to the lens or use near the limits of the range, and record the reason for each change.

• How to set calibration intervals
• Calibration interval calculator

What should the certificate and record show?

ASTM E2847 specifies the elements of a calibration report (section 5.3), and ISO/IEC 17025:2017 clause 7.8 sets the general content of a report for accredited work. Keep at least the following.

• The thermometer and its emissivity setting during the calibration.
• The source type, cavity or flat plate, with its size, its emissivity and the reference used for its temperature.
• The measuring distance and the point on the thermometer it was measured from.
• The temperatures calibrated, with the source value, the reading, the error and the uncertainty at each point.
• The room temperature and humidity.
• The readings before and after any adjustment, the date, who did the work and the next due date.

• Calibration certificate requirements
• Calibration certificate template (Excel)

What if an infrared thermometer is out of tolerance?

Record the as-found readings before anyone adjusts it. If a reading is outside its limit, the thermometer may have been out of tolerance for an unknown part of the time since its last good calibration or check. ISO 9001 clause 7.1.5.2 expects you to look back at what it measured and deal with anything that may be wrong. For an infrared thermometer, look at what its readings decided, such as an oven or furnace temperature, a process limit or a safety check.

• What to do when a gage fails calibration
• As-found and as-left readings

Where Axiospec fits

Axiospec is calibration management software. It keeps each infrared thermometer and pyrometer with its calibration history, certificates and next due date. Each calibration names the reference standard used, as free text or as a link to a master instrument, such as your blackbody source. One calibration record holds up to 100 test points, each with its nominal temperature, its own tolerance and the as-found and as-left readings, and the record also keeps the room temperature and humidity.

On an in-house calibration, a final reading outside its tolerance sets the result to Fail, and recording a pass anyway takes a written reason. An accepted Fail or Damaged puts the thermometer in Quarantined. Any approved Fail, Damaged or Pass with as-found out of tolerance needs an impact written, and unrecorded ones show in Needs attention. Axiospec does not run the comparison or pick the points, the limit or the interval. Those stay with you or your lab.

The free plan covers up to 50 active instruments. A new workspace starts with a 45-day trial of Professional, with room for 1,500 active instruments and no credit card. We load your existing records for you at no charge on every plan.

• Free measurement uncertainty budget calculator
• Free as-found and as-left worksheet
• See the live demo, no signup
• Start a 45-day trial
• Compare plans