Measurement uncertainty guide
The current global economy relies on measurements and tests that can be trusted and accepted at an international level without creating technical trade barriers. This calls for a sound metrological foundation. The LVD project supported measurement and testing (M&T) laboratories and related infrastructures in Newly Associated States (NAS) in order to fully implement the Low-voltage Directive (LVD). The LVD applies to all electrical products which operate in the voltage range of 50 to 1000 Volts AC and 75 to 1500 Volts DC. The general underlying principle of the directive is that such electrical products must not jeopardise the safety of people, domestic animals or property. Given this, the importance of testing is apparent. Testing determines the characteristics of a product, a process or a service according to set procedures, methodologies or requirements. It can therefore assess whether or not a product conforms to such specifications. Measurement uncertainty in testing involves the quantitative measure of the quality of measurement results allowing those results to be compared with other results, references, specifications or standards. Measurements are subject to error since the result of a measurement may differ from the true value of what is being measured. Unfortunately there is usually limited time and resources to determine and correct measurement errors. In light of this, the LVD project has provided a written practical guideline for the evaluation of measurement uncertainty in testing. The guideline is based on the GUM method, a widely-used method which is accepted and recommended by accreditation bodies such as the ISO (International Organisation for Standardisation). The guide summarises the GUM method, suggests model functions and explains and illustrates the use of generic model functions relevant for electrical measurements. Additionally the guide includes examples of uncertainty evaluation.
