Methods of thermometer calibration

A thermometer consists of three subsystems:

 

  • The measuring element (resistor, thermocouple, etc)
  • The conversion method (resistance to temperature, emf to temperature, etc)
  • The readout instrument (or temperature indicator)

 

A thermometer can be calibrated either as a whole system (containing all the three above subsystems) or by calibrating each subsystem separately. In the case of a liquid-in-glass thermometer, the three components cannot be separated, so it is calibrated as a whole system. On the other hand, in the case of a resistance thermometer, the sensing element (resistor) can be calibrated separately (by measuring directly resistance at various temperature points) and the temperature indicator can be calibrated on its own by applying known resistor values and checking the indication.

When possible, it is better to calibrate the thermometer as a whole system, since this is how it is used in practice.

Temperature is one of the most widely measured physical quantities. But unlike other quantities, such as mass and time whose SI units are based on physical realisations: the temperature is defined on a theoretical set of conditions. The current working temperature scale is the International Temperature Scale of 1990 (ITS-90) and it is measured in degrees Celsius (oC).

There are two methods widely used for calibrating thermometers:

 

  1. Fixed points calibration
  2. Comparison to standard thermometers

 

  1. Fixed points method

A thermometer is calibrated by measurements at a series of temperature fixed points (freezing/melting points, triple points or vapour pressure points of pure materials). By using this method we insert the thermometer in a fixed point cell which provides the desired temperature point.

In the following table the most common fixed points according to ITS-90 are shown:

 

Fixed Point Physical Property Temperature

(oC)

Argon Triple Point * – 189.3442
Mercury Triple Point * -38.8344
Water Triple Point * 0.010
Gallium Melting Point 29.7646
Indium Freeze Point 156.5985
Tin Freeze Point 231.928
Zinc Freeze Point 419.527
Aluminium Freeze Point 660.323
Silver Freeze Point 961.78
Gold Freeze Point 1064.18

 

* Triple point is defined as the point where liquid, solid and gas are in equilibrium.

The water triple point is the most important and accurately realizable of the fixed points. The apparatus used for the realization is a glass flask nearly filled with very pure water and placed in an ice and water bath that contains the cell at or near the freezing point of water.

The fixed points method is the most accurate calibration method and it is used only in the highest quality calibrations. It is not commonly used by calibration laboratories since it is very complex and high cost.

 

  1. Comparison method – Brannan’s usual method of thermometer calibration

This method of calibration is the most widely used. It is based on comparing the thermometer under test to a higher accuracy standard thermometer. The comparison usually takes place in a liquid bath or a dry block calibrator. Thermometers used as standards by comparison method are usually high accuracy resistance thermometers.

Several parameters must be taken into account when using the comparison method. The most important are immersion depth and homogeneity of the liquid or air where the thermometer is immersed. The immersion depth depends on the construction of the thermometer, the temperature difference between the bath and the surrounding atmosphere, the heat transfer capability of the bath and the temperature stability of the bath. Homogeneity depends on the equipment used. Better homogeneity is achieved by liquid baths, but we are able to improve homogeneity of dry block calibrators by inserting a metal equalising block with thermowells to receive both the standard and the thermometer under test.

 

Temperature probe & thermometer calibration methods

Temperature probes and thermometers are used in all types of industry from food, to aerospace and industrial applications.

This category includes mostly direct reading temperature measuring instruments (glass thermometers, temperature sensors with incorporated indication, etcetera). These instruments are calibrated using the thermal (temperature) calibration method. Depending on their accuracy, either the comparison method can be selected, or the fixed point method (for instruments requiring a high accuracy calibration.

 

Climatic chambers

Climatic chambers are chambers with controlled temperature. They are programmed to a set point and they also contain a temperature indicator. The comparison method is used for their calibration. But in this case, a number of standard thermometers, placed in specific positions inside the chamber, are required. There are special rules and procedures for climatic chambers calibration.

 

Furnaces, ovens, liquid baths

There are several types of instruments that belong in this category. They have a wide range of applications in several sectors such as food industry, medical industry, calibration laboratories, etc. Especially for the liquid baths, the liquid used depends on the desired temperature, so alcohols are used for temperatures below 0°C, water from 0°C to 80°C and oils up to 300°C. Their calibration is similar to the one used for the dry block calibrators and is based on the comparison method.

 

Chart recorders

They provide a hard copy record of the measured temperature. They may contain additional functions such as real-time display or alarm. Their calibration is performed by using the comparison method.

 

Data loggers

They are similar to the chart recorders, but they are electronic measuring instruments and they do not provide paper-based records. They store the measurements into their USB memory. Data loggers may also contain real-time display, alarm outputs and other functions. Their calibration is similar the one used for the chart recorders.

 

Thermocouples

They consist of two dissimilar conductors connected together at the measuring junction. The temperature change at the reference junction causes a voltage to be generated. Depending on the type of the thermocouple (K, J, T, etc.) there are reference tables which correspond the generated voltage into temperature values. Thermocouples are calibrated by measurement either with fixed-point temperatures or, by comparison with a reference thermometer, in thermally stabilised baths or furnaces. Also, a combination of both methods can be used.

There are of course several other instruments used to either produce or measure temperature. All these instruments are used in many different applications and processes. No matter the purpose of the application, the only way to be sure about our temperature measurements is to use instruments which are properly calibrated from accurate reference sources.

 

Calibration of a liquid-in-glass thermometer – an example

Before commencing with the thermometer’s calibration, we must visually inspect the thermometer to ensure that there are no malfunctions such as gaps in the measuring liquid, errors in the measuring scale, etc.

We must perform the calibration in a laboratory with ambient temperature within 23oC ± 3o C and temperature stability, during the testing period, of ± 1oC. We must make sure that the thermometer under test is left in the laboratory for sufficient time period in order to be conditioned to the lab’s environmental conditions.

The thermometer we want to calibrate has a measuring range of -10oC to +50oC and it is a partial immersion thermometer, which means that its bulb and a specified part of its stem must be inserted into the bath, in order to indicate correct temperature readings. The thermometer is a C accuracy class thermometer (± 0.5oC MPE).

The calibration is performed by using a liquid bath, containing ethanol, for temperatures from -10oC to 0oC, a dry block calibrator (with its thermowells) for temperatures from 0oC to +50oC, and a Pt-100 Platinum resistance thermometer alongside a high accurate digital multimeter for measuring the resistance of our standard.

Firstly, we insert the thermometer under test and our standard thermometer in the liquid bath. We set the bath’s temperature at -10oC and leave it to stabilise. When we achieve stabilisation we take the readings of both the under test instrument and the standard. We repeat this procedure for the 0oC temperature point.

We remove the thermometers from the liquid bath and insert them into the dry block calibrator. We set the dry block calibrator at +10oC, leave it to stabilise and record the readings of both instruments. We repeat this procedure for the rest temperature points (+20oC, +30oC, +40oC and +50oC). (We can use different measurements points, less or more, depending on our procedures or on the customer’s specific requirements).

The calibration report will contain the measurement results in a table similar to the following:

 

Standard Reading

(oC)

Test Instrument Reading

(oC)

Deviation

(oC)

Tolerance

(oC)

-9.99 -10.0 -0.01 ± 0.5
0.03 0.0 -0.03 ± 0.5
10.05 10.0 -0.05 ± 0.5
20.05 20.0 -0.05 ± 0.5
30.07 30.0 -0.07 ± 0.5
40.08 40.0 -0.08 ± 0.5
50.10 50.0 -0.1 ± 0.5

 

In order for our calibration report to be complete, we must also include a column containing the uncertainties of measurements, which can be evaluated by using the document EA-04/2 “Expression of the Uncertainty of Measurement in Calibration”.

 

Contact us to discuss temperature calibration for your instrument.

Find out about how we conduct calibration for thermometers and temperature instruments.

Find out how we conduct calibration for pressure instruments.

Find out about calibrating environmental instruments.

Find out about calibrating hydrometers.

Find out about timer calibrations.

View Brannan’s extensive range of quality certifications including UKAS accreditation here.

Subscribe to our newsletter