PRIMARY, SECONDARY OR DERIVED STANDARDS... WHAT'S THE DIFFERENCE? Uncertainty and traceability issues in low-flow gas calibrations.

摘要

ISO 9000 and the aggressive move toward more efficient plant operation is placing greater accuracy demands on gas flow measurement devices. The need for clear, objective and unambiguous reporting of measurement results and associated uncertainty will also raise the traceability requirements for gas flow calibration standards. Because gas is a compressible medium and because gas flow can be a complicated phenomena, accurate gas flow meter calibration requires the use of very accurate facilities and flow standards. These standards involve a range of component measurement systems where the potential for error propagation is much greater than with most other measurements. This fact is often misunderstood and frequently leads to invalid or inaccurate gas flow calibrations. For all these reasons, the traceability of the data from meters and calibration facilities to designated standards needs to be realistically established and maintained. ANSI/NCSL Z540-1-1994 specifies that the collective uncertainty of the measurement standards not exceed 25% of the acceptable tolerance (manufacturer's specification) of the device under test. This paper presents a description of primary, secondary and inferred gas flow calibration standards, their uncertainty levels and their adequacy in meeting acceptable tolerances and the traceabilities required to substantiate claims. Traceability is examined in relationship to an uncertainty analysis of the measurement component contributions to total error of the calibration system.