Measurement uncertainty analysis for calorimetry apparatuses

The absence of a standardized uncertainty analysis methodology for fire calorimetry apparatuses makes significant comparison of test results between fire laboratories difficult. From a testing (pass/fail) perspective, the lack of a common means to estimate the uncertainty is problematic. From a material development viewpoint, the ability to differentiate versions of material systems is uncertain which potentially results in economic penalties for the developer. Probability and statistics methodologies recommended by International Organization for Standardization and National Institute of Standards and Technology were applied to begin developing a standardized measurement uncertainty method for fire calorimetry apparatuses. The method was applied to each component instrument (intrinsic and calibration uncertainty only) and each related physics quantity measured indirectly. Results were reported at the 95% confidence level. The results were compared with various reference cases as well as those from a Monte Carlo Simulation technique modelling the experiments. The consistency verified our measurement uncertainty method. The results for heat release rate and average heat of combustion provide the user of Cone Calorimeter data with the ability to estimate potential experimental error inherent in the operation of the Cone. This information can be used to make informed decisions related to the matters cited above. Copyright © 2007 John Wiley & Sons, Ltd.