The high-precision thermometer from Anton Paar is designed for the most accurate temperature measurements, comparison calibrations and fixed point calibrations in industry, laboratories and research. In combination with calibrated Platinium Resistance Thermometers, Anton Paar's high-precision thermometer achieves a measuring uncertainty of 1 to 10 mK.
The world-class high-precision thermometer is extremely flexible:
As all electrical conductors change their resistance with temperature, the resistance value can be used as an indicator for the temperature.
As the resistance of a material depends on its mechanical dimensions and purity, platinum is commonly used because it is chemically quite inert (which means there is no chemical reaction which may change the sensor's behavior).
A Pt100 sensor changes its resistance by about 0.4 Ohm per °C (per K).
The most common platinum sensor has 100 Ohms resistance at 0°C and is therefore called Pt100.
Other common sensors are Pt1000 and Pt25.
The correlation between temperature and resistance is not linear. Therefore, for high-accuracy applications, calibration coefficients have to be determined for each sensor. To do this, various calibration standards have been developed. The two most common ones are mentioned below.
There are two common standards that are used to calculate a temperature from a resistance value:
The most important unique parameter of a sensor is its resistance at 0°, the R0 value. This is needed for both IEC 751 and ITS-90.
IEC 751 (Callendar-Van Dusen equation)
The IEC 751 also needs three further coefficients: a, b, c
The constant c is only needed if temperatures below 0 °C are measured. Above 0 °C it has no influence.
For the range between -200 °C to 0 °C the equation is
R(t) = R(0)[1 + A * t + B * t2 + (t - 100)C * t3]
For the range between 0 °C to 661 °C the equation is
R(t) = R(0)(1 + A * t + B * t2)
The IEC 751 is the most common calibration standard and all sensors provided by Anton Paar come with the three or four coefficients needed: R0, a, b, c
ITS-90
The ITS-90 employs up to 7 coefficients depending on the temperature range. This standard is mostly used by high-end calibration services.
For more details on these calibration standards, see the Appendix "Temperature Calculation Methods" in the MKT 50 instruction manual.
Temperature is the most often measured quantity in the world!
Almost all chemical and physical processes are influenced by temperature. This makes a reliable measurement of temperature so important. To guarantee that a temperature sensor gives correct readings, it has to be calibrated regularly. The calibration interval depends on the type, use and expected accuracy of the sensor.
Example: 0.1°C comparison calibration
Temperature sensors in many fields of applications have to be calibrated regularly. MKT 50 has two channels to measure the reference sensor and device under test simultaneously. The sensor under test and a reference sensor are placed in a temperature-stable system. After reaching thermodynamic equilibrium the two temperatures are compared.
A common uncertainty that is expected here is 0.1 °C.
For a legitimate calibration, the reference instrument should be 3 to 10 times more accurate than the instrument to be calibrated. The basic uncertainty of MKT 50 is 0.001 °C. Some standard sensors are calibrated to 0.01 °C accuracy. As the system accuracy (MKT 50 and sensor) depends mostly on the sensor, it is 0.01 °C (10 mK) in this case. Therefore, the MKT 50 is perfectly suitable for calibration applications of 0.1 °C accuracy.
For special high-accuracy applications, SPRTs (Standard Platinum Resistance Thermometers) are available that achieve a system accuracy (MKT 50 and sensor) of up to 1.4 mK.
For more details about comparison calibration, see the Application Flash "Comparison Calibration with the MKT 50" on the "Applications" tab.
Many famous names have contributed to the field of temperature measurement throughout the centuries. In 1593, Galileo Galilei (1564-1642) invented a water thermometer which made it possible to measure temperature variations for the first time. Another Italian inventor, Santorio Santorio (1561-1636), is credited with the innovative idea of using a numerical scale on the thermometer. However, the accuracy of this thermometer was poor as it did not take the differences in air pressure into consideration.
In 1654 the Grand Duke of Tuscany, Ferdinand II (1610-1670), produced the first sealed liquid-in-glass thermometer. Over half a century later, in 1714, the Dutch physicist and engineer Gabriel Fahrenheit invented the first mercury thermometer. He called it a "thermoscope". Fahrenheit's thermoscope was much more accurate than any previous devices for measuring temperature had been.
A few decades later, in 1741, Swedish scientist Anders Celsius (1701-1744) proposed the idea of a thermometer scale which divided the range from the freezing to the boiling point of water into 100 degrees. With the help of Jean Pierre Cristin (1683-1755) this scale would later become the Centigrade scale used today in many parts of the world.
In 1871, Sir William Siemens first presented the idea of making use of electrical conductors in thermometers. Electrical conductors increase their electrical resistance with increasing temperature. The construction of these thermometers was made possible by methods developed by Callendar, Griffiths, Holborn and Wein in the last years of the nineteenth century.
Throughout the twentieth century, the accuracy of thermometers increased. In the 1960s, Anton Paar, in cooperation with the University of Technology in Graz, began developing high-precision temperature measurements as an indispensable part of its production of high-quality digital density meters. The density of materials changes considerably with temperature, making temperature measurement particularly important when measuring density. These years of research have resulted in the know-how needed to develop Anton Paar's high-end thermometers which achieve absolute accuracies of up to 0.001°C (1 mK).