|Contact thermometry FAQs|
1. What is a temperature scale?
2. How do you convert Fahrenheit or Kelvin temperatures into Celsius temperatures?
3. What is the ITS-90?
4. What are fixed points?
5. What is a temperature calibration ?
6. What are the standards used in thermometry?
7. What is a thermocouple?
8. What colours are the lead wires of each thermocouple type?
9. What are the benefits of grounded, ungrounded, and exposed probe junctions?
10. What does heterogeneity of a thermocouple mean?
11. What is a thermopile?
12. What is the difference between an RTD and a thermistor?
13. What is a bimetallic thermometer?
14. How to carry out a surface temperature measurement?
A temperature scale is a system of measuring temperature; it is formed by placing two reference points and evenly subdividing the points into temperature intervals. There are three temperature scales in use today, Celsius, Kelvin and Fahrenheit.
The Celsius (ºC) temperature scale is the scale based on 0 degrees for the freezing point of water and 100 degrees for the boiling point of water. It is sometimes called the centigrade scale because of the 100-degree interval. The Celsius scale is used in scientific work everywhere.
The Kelvin (K) temperature scale is the international temperature scale; its zero is the absolute zero point, the theoretical temperature at which the molecules of a substance have no thermal energy. The Kelvin scale is related to the Celsius scale. The difference between the freezing and boiling points of water is 100 degrees in each, so that the Kelvin has the same magnitude as the degree Celsius. The Kelvin is the unit of thermodynamic temperature measurement in the International System (SI) of measurement. It is defined as 1/ 273.16 of the triple point (equilibrium among the solid, liquid, and gaseous phases) of pure water.
The Fahrenheit (ºF) temperature scale is, like the Celsius scale, based on the freezing and boiling points of water. The freezing point of water is 32 degrees and the boiling point is 212 degrees.
To convert Fahrenheit or Kelvin to Celsius; or Fahrenheit to Kelvin; use the following conversion formulas:
- Celsius to Fahrenheit: t / °F = 9/5 . (t / °C) +32
- Kelvin to Fahrenheit: t / °F = 5/9 . T / K - 459.67
- Celsius to Kelvin: T / K = t / °C + 273.15
ITS-90 is the International Temperature Scale and was created in 1990. It is the scale which accredited laboratories refer to in their calibration certificates.
ITS-90 is based on 17 fixed points (phase transitions of pure materials) and interpolating equations between the fixed points or pressure/temperature relations of gas. The ITS-90 extends upwards from 0.65 K to the highest temperature practically measurable in terms of the Planck radiation law using monochromatic radiation.
The units used for the ITS-90 are: the Kelvin symbol (K) (temperature T90) and the degree Celsius symbol (ºC) (temperature t90).
For further information see the ITS website.
Fixed points are phase transitions of pure substances whose temperature is fixed and highly reproducible. These are:
melting point, the transition from a solid to a liquid at normal atmospheric pressure
freezing point, the transition from a liquid to a solid at normal atmospheric pressure
triple point, the temperature at which the solid, liquid and vapour phases are in thermodynamic equilibrium with each other
boiling point, the transition from liquid to vapour at normal atmospheric pressure.
These fixed points are realisable with numerous pure substances. However, many of them do not have the necessary stability and reproducibility and some others require a complex procedure and special laboratory facilities. The greatest difficulty in realising a fixed-point temperature is due to the influence of impurities in the fixed-point material.
The triple point of water is the most important defining thermometric fixed point used in the calibration of thermometers to the International Temperature Scale of 1990 (ITS-90), it is one of the most accurately realisable of the defining fixed points. It provides a useful check of the stability of a thermometer when investigating whether a shift in calibration has occurred.
In the specific area of temperature, a calibration is a comparison between a thermometer to be calibrated and a “standard” related to National standards. The result of a calibration is a correction to apply to the reading of the calibrated thermometer and its associated uncertainty.
There are two calibration methods:
1. The fixed points method
This absolute method is used for the realisation of the International temperature scale: ITS-90. The thermometer is calibrated by measurements at a series of temperature fixed points, e.g. freezing/melting points, triple points, vapour pressure points. This method gives calibration results with high accuracy. It is relevant only for high quality thermometers.
2. The comparison method
The thermometer is calibrated by comparison with a reference / standard thermometer in a thermally stabilised bath or furnace suitable for the calibration. This method allows coverage of a wide range of temperature during a calibration operation, point-by-point or continuously, in a short time, and if required to calibrate simultaneously a large number of thermometers. However, this method is less accurate than the fixed point method because of the lesser accuracy and stability of the reference standard and enclosure.
Depending on the type of thermometer, the calibration can be carried out by considering the thermometer either as a whole system (including the detector, connectors and readout instrument) or as just the detector.
For example, in the case of a liquid-in-glass thermometer, the three components cannot be separated and so the calibration is of the system as a whole. Contrarily, in the case of a thermoelectric or resistance thermometer, the sensing element can be separated from its connecting wires and the readout instrument, and calibrated separately. However, it is generally better to calibrate a thermometer system as a whole.
The standards used in thermometry are fixed points and standard thermometers, described as follows:
Thermometers are calibrated at fixed temperatures (e.g. freezing / melting points, triple points, vapour pressure points) specified in the International Temperature Scale of 1990 (ITS-90). Interpolation equations depending on the temperature range are determined for the thermometer.
Standard thermometers are principally Platinum Resistance Thermometers (PRTs) or thermocouples (type S, R or D), ideally calibrated at fixed points.
A thermocouple consists of two wires (thermoelement) made of dissimilar metals and welded at one end. This junction is called the hot junction or the measuring junction. At the other end of the wires a cold junction is or reference junction is made and this is connected to the output device (voltmeter, temperature indicator).
The resulting voltage or electromotive force (emf) of typically a few millivolts is generated by thermal gradients between the hot and cold junctions. It is a function of the difference in temperature between the measurement junction and the reference junction. The relationships between the emf and temperature are available for the most commonly used thermocouples in reference tables (type R, S, B, J, T, E, K, N).
Thermocouples are the most widely used sensors in industry due to their low cost, simplicity, robustness, size and temperature range of use.
The standards define an identification system based on colours to identify the types of thermocouple and the polarity of the conductor cables.
Grounded junctions - are welded to the tip of the sheath with wires completely sealed from contaminants. They offer good response times and are ideal for measuring the temperature of static or flowing corrosive gases and liquids.
Exposed junctions - have the fastest response times and are ideal for measuring rapid temperature changes. Clear coating on most models provides a humidity barrier for the thermocouple. Do not use exposed junctions with corrosive fluids or atmospheres.
Changes in the material composition of the thermocouple conductors induce structural inhomogeneities which modify the magnitude of the electromotive force (emf) generated by the thermocouple. The emf is affected by the thermal gradient which exists along the length of the exposed conductors. In homogeneous conductors, the emf depends only on the temperature of the two junctions. The inhomogeneities in the conductors produce parasitic emf in regions where temperature gradients exist. It is a major source of measurement errors with thermocouples and is very difficult to detect.
A thermopile is made of several thermocouples connected in series. The resulting voltage is the sum of the individual thermocouple voltages and therefore increases the measured signal. Thermopiles with a large number of junctions are used in heat flux sensors. Thermopiles are made in a number of ways, either as a wound wire, printed circuit or etched metal foils.
The sensing element for both an RTD (Resistance Temperature Detector) and a thermistor has an electrical resistance that changes with temperature. The sensing element of an RTD is metallic, while thermistors have semiconductors made from mixtures of oxides of nickel, manganese, etc. The resistance of an RTD increases approximately linearly with temperature. Typically, the resistance of a thermistor drops with increasing temperature and is highly non-linear and can usually be expressed as an exponential function of temperature.
The most common temperature range for a thermistor is 0°C - 100°C. At higher temperatures, it is subject to large drifts in output.
The most common type of RTD is the PRT (Platinum Resistance Thermometer) and it is the most reproducible type because platinum is highly corrosion and oxidation resistant and stable over a wide temperature range (-250°C to 850°C). It is used as interpolating standard between the fixed points of the International temperature Scale ITS-90.
Thermistors are not as reproducible as PRTs but are much more sensitive. The high sensitivity of thermistors can be an order of magnitude greater than that of the RTDs. Thermistors are useable over a smaller temperature range than RTDs but are very accurate (usually better than 0.05°C or 0.1°C).
Compared with thermistors, RTDs do not allow point measurement of temperature because of the large sensing element. Contrarily, the small size of thermistors makes them very adaptable and they are often included in electronic circuits.
For both RTDs and thermistors, the self-heating due to the current flow through the resistance is a source of measurement error. To minimise this effect, the current used must be fitted with the value of the measured resistance.
A bimetallic strip thermometer is a mechanical thermometer. It consists of two strips made of dissimilar metals and bonded together with one end fixed and the other free.
The principle of operation is that as the temperature changes one strip expands more than the other, causing motion of the free end of the strip, i.e. a bending caused by the different expansions. Two constructions are available: a spiral strip and a cantilever strip. They are the most widely used in industry for temperature control, due to their robustness, temperature range and simplicity. They are used in thermostats for measuring and controlling temperature, or in ovens.
A surface temperature measurement can be difficult to carry out and requires a lot of precautions to be accurate. It can be achieved either with a contact thermometer or by a non-contact technique using an optical pyrometer.
With contact thermometry it is possible to employ an indirect or a direct method. In the indirect method the surface temperature Ts of a body is deduced from two temperature measurements, T1 and T2, performed inside the material by means of two sensors placed at different depths x1 and x2 into the material with a thickness e.
The surface temperature Ts may be determined by extrapolating the T1 and T2 temperature measurements as follows:
In the direct measurement method the surface temperature is determined via direct contact of a sensor with the surface. Two approaches are possible:
The thermoelement can be a thermocouple, a resistance or a thermistor. The main sources of uncertainty with the direct measurement method come from the sensor itself (sensor calibration, heat conduction in the leads, self-heating etc.) and the possibly poor thermal contact between sensor and surface. These uncertainties can be quantified by means of a relevant calibration of the surface thermometer.
Non-contact thermometry or radiation thermometry is a non-invasive technique. This approach is useful in determining the surface temperature of moving bodies, or for high temperatures (too high for a contact thermometer) or chemically reactive applications. The main sources of uncertainty are due to the thermometer itself, from the emissivity of the body surface if not precisely known (it usually isn't!) and from parasitic radiation (i.e. radiation reflected between the surface and the thermometer or background radiation reflected from the surface into the thermometer).