Uncertainty budgeting for instrumental methods in emissions. A practical approach for measuring institutes.

Abstract: A procedure based on zero and span readings before and after each measurement campaign has been developed and described. The procedure is rather flexible, and it can be used for the instrument in general, for the instrument when used in certain applications, or when used for one specific measuring task. Furthermore the procedure also gives a procedure for determining the detection limit of the instrument – under field conditions. For normal use the alternative procedure does not require additional activities in the field – just more calculations. The alternative procedure does not cover all uncertainty sources, and consequently additional calculations are needed.

Keywords: Uncertainty calculations, Field tests, Emissions.

1. Introduction

The international standard EN ISO 14956 [1] was accepted as an international standard in August 2002 and shall be implemented as national standards in the CEN countries at the latest by February 2003. Knowing that the accreditation bodies in Denmark and Sweden will request – or at least strongly encourage – the measuring institutes to use this standard for their uncertainty budgeting, the two national environmental protection agencies1 decided to establish a project, where the procedures in the standard are described, and also an alternative and “easier-to-access” procedure is given.

The argumentation for an alternative procedure is, that it is difficult for the individual institute to establish the necessary information required to make a valid calculation. Further more the differences in the uncertainties calculated by different institutes are assumed not to originate from differences in the instruments and procedures used but from differences in thecalculations.

The main task of the project was to establish a procedure, which should be relatively simple – or at least simple to harmonise, that gives a measure for the uncertainty of an instrument under field conditions. The intention of the alternative procedure is to substitute as many uncertainty sources as possible by information of the performance of the instrument during measurement campaigns.