8811 
Sami Koivisto
MSc(Tech) EUR ING
Technical subcommittee
CEN/TC 332
European Committee for Standardization
As a concept, calibration can be understood in several ways and the interpretation can vary between different devices. When associated with pH meters, calibration means adjusting the device to give a specific value with a defined standard solution. With balances the same word means only the determination of the difference between the reference weight mass and the reading obtained from the device display. The understanding of these concepts can vary even within the same organisation. It is important that the concepts used in discussion are understood. In this article the following definitions apply:
- Calibration is the determination of the difference between the mean volume of the measurement series and the selected volume on the pipette display.
- Adjustment is the alteration of pipette settings for the actual volume to correspond to the selected volume.
This means that during calibration no physical changes are made to the device itself and the process gives only the measurement result. During the adjustment phase the device is physically altered to deliver a different volume. To confirm the effect of adjustment the device needs to be recalibrated.
Calibration in practice
In practice, calibration means the functional verification of the pipette. This begins at the factory, where the pipettes are adjusted and calibrated for quality assurance. It is recommended that the clients perform an inspection of the devices before they are released for operator use. This is due to the different operating conditions, which can have an effect on the calibration results. However, very often the pipettes are not calibrated until the first maintenance session.
The calibration procedures are usually performed according to a standard method and only the functionality of the pipette is verified. The standard check is always performed using water. Since pure water is seldom used in laboratory applications, the actual amount of liquid can differ from the standard calibration value. The result obtained in calibration is bound to time, place, ambient conditions, the liquid used and the operator. Therefore, in addition to the reference calibration it is suggested that a calibration is performed using the liquid to be transferred. This should be done in laboratory conditions by each operator using the pipette to verify that a desired volume amount is obtained. The pipetting method and the equipment setting (pipette + tip) demanded by the application should also be included in the test.
However, adjustment is not recommended for liquids other than water due to the insufficient or approximate correlation data (mass to volume conversion). The operator should also bear in mind that pipettes will be used to transfer different liquid types with deviating properties. Adjustments would have to be done each time a liquid is changed. Water is officially accepted as a standard calibration liquid due to thorough research and statistical results. This gives reliable reference values for the devices. The follow-up of the mechanical functioning of the pipette is therefore easier for the user and better accepted by official parties.
The calibration values obtained using liquids other than water can be referred to the obtained reference values and necessary corrective actions can be taken accordingly. For example the statistical deviation of these results can be added to the uncertain calculations used by accredited laboratories.
International standard ISO 8655 for POVA devices
In January 2003 the new EN ISO 8655 standard for piston-operated volumetric apparatus (POVA) was published as a result of the work of joint technical subcommittees ISO TC 48/SC1 and CEN/TC 332/WG1. This standard replaced the well-known DIN 12650 standard, used since 1997. The previous DIN standard was mainly directed at manufacturer use, but the present ISO standard is also directed at service houses and end-users. ISO 8655 contains seven parts covering different kinds of piston-operated liquid transfer devices. The sections that relate to piston pipettes are:
- Part 1: Terminology, general requirements and user recommendations.
- Part 2: Piston pipettes.
- Part 6: Gravimetric methods for the determination of measurement error.
- Part 7: Nongravimetric methods for the assessment of equipment performance.
Part 1 contains valuable general information for all parties including vocabulary. Part 2 describes the structure and operation principle of a piston pipette and also contains the acceptance tolerances for calibration – according to nominal volumes. The actual gravimetric testing procedure is described in Part 6. The demanded calibration conditions are also stated. Part 7 describes alternative calibration methods and the terms when these methods can be used to fulfil the standard requirements. Present methods include the photometric method (horizontal direction) and the titrimetric method.
Coverage of ISO 8655
The standard states the acceptance specifications, test methods and testing conditions for pipettes. It also gives some information about the basic pipetting errors affecting the procedure. The standard is directed to manufacturers as a basis for product type testing and quality control, including, when appropriate, the issuance of manufacturer’s declarations of conformity. The standard also addresses the needs of test houses and other bodies as a basis for independent certification of conformity, and the needs of equipment users to enable routine performance checking.
The test method described in ISO 8655-6 can be applied in every calibration procedure regardless of the ambient conditions. However, when the functionality of the equipment or the conformance to official acceptance limits are concerned, only the conditions mentioned in the standard are acceptable. The acceptance limits in the standard do not apply to liquids other than water or pipetting methods different from the forward pipetting technique.
Manufacturer’s acceptance specifications versus ISO 8655 acceptance specifications
One of the puzzling questions about ISO 8655 is the difference between the manufacturer’s acceptance limits and acceptance specifications. The specifications in the standard are designed to serve all parties from manufacturers to end-users. Since the acceptance limits are the same for everyone, the manufacturers continue to use their internal quality control limits to demonstrate the performance capabilities of the pipettes. The manufacturer’s acceptance limits are usually much stricter than the specifications mentioned in the standard, especially where adjustable pipettes and small volumes are concerned.
The standard specifications are wider than the corresponding manufacturer’s specifications since there is no actual need for more accurate or precise dosing in applications where POVA devices, such as pipettes, are used. Although the sensitivity of laboratory applications has increased, in many cases the required level of measurement accuracy can still be more than 10-fold compared with device acceptance limits. Naturally, it is good to minimise known error sources, but most are not caused by the device itself. Knowing and understanding the most critical error factors demands both theoretical and practical knowledge from the operator.
The main difference between the manufacturer’s acceptance limits and standard specifications concerns the variable volume pipettes. Manufacturers give volume-bound specifications for certain test volumes. These volumes are usually the nominal volume of the pipette (maximum volume) and the minimum volume or 10% of the nominal volume. Sometimes additional specifications, such as the middle volume (50% of the nominal value), are also published. The standard, however, gives specific values that are bound to the nominal volume values of the pipettes. All other selectable volumes have the same absolute acceptance limit values as the model specific nominal volume. For example, a pipette model of 100–1000µl has an acceptance value of 8µl at the maximum volume for systematic error (inaccuracy). This is 0.8% of the nominal volume. At the setting of 100µl, the acceptance limit is the same (8µl). However, the relative value is 8%. The limits are always defined according to the absolute error values.
Special requirements of ISO 8655
ISO 8655 standard describes many requirements that end-users and independent test-house operators should follow to fulfil the standard. These requirements can be classified under different categories.
Use of pipettes
To ensure safe and efficient use of the device the standard requires the user to follow instructions provided by the manufacturer. Special care should be taken to prevent the entry of liquid into the pipette interior. The user is also responsible for ensuring that used pipettes and pipette tips are resistant to the transferred liquids.
Calibration performed by the user
To achieve the needed reliability of volume measurement, users should establish their own acceptance maximum error limits for both systematic (inaccuracy) and random (imprecision) error. These requirements are determined by the applications where pipettes are used. Conformity to the defined acceptance limits should be tested by the user at regular intervals as part of their own quality protocol. The required time interval should be based on the frequency of use, number of operators using the device, nature of the liquids and finally the demands of the applications. The time interval can be, for example, three months but no more than one year, and the defined protocols should be recorded into the user’s quality system. In case pipette tips different from those of the manufacturer are used, the operator should confirm the suitability and functionality of the tips with the pipette.
Maintenance of the pipette
Many laboratories send their pipettes to other test houses or the manufacturer for maintenance and calibration procedures. In these cases it is the responsibility of the user to perform pipette decontamination. Nondecontaminated devices should not be sent for servicing.
Other uses of the standard method
When the standard method is used by the manufacturer for purposes other than type testing, such as quality control or functional verification as part of the after-sales service, the standard allows the manufacturer to define:
- The number of needed test volumes.
- The number of measurements used per volume.
- For multichannel pipettes, the number of channels tested.
According to these definitions, many manufacturers use their own testing protocols in quality assurance. Both one- and two-test volumes are used. The results are determined using four or more measurement values per volume. Results from multichannel pipette calibration can be reported for two or more individual channels. These choices are mainly based on the reference measurements and extensive statistical experience of the manufacturer. The present quality systems require increasing attention to the equipment performance and to traceable reference calibration records. This can be achieved by using a standardised calibration method such as the procedure described in the ISO 8655 standard. It is recommended that pipettes are also adjusted according to a standard procedure.
[[HHE06_fig1_L36]]
Although the best way to assure the functional performance of the pipette is to follow a standard procedure, the real situation within the laboratory can be different. To ensure reliable results, it is recommended that pipettes are calibrated in a “real” laboratory environment in addition to undertaking traceable standard quality control. This, however, demands both good theoretical and practical knowledge from the operators.
Further reading
International standard EN ISO 8655. Piston-operated volumetric apparatus. Geneva: ISO; 2002.
Technical report. ISO/TR 20461. Determination of uncertainty for volume measurements made using the gravimetric method, Geneva: ISO; 2000.
