Measurement uncertainties of the calorimeter cone according to the standard ISO 5660-1

The purpose of this article is to describe the various calculations of uncertainties associated with the results of a calorimeter cone test in accordance with the standard ISO 5660-1 in compliance with ISO/IEC 98-1, which describes the calculation of measurement uncertainties, and with the Guide to the Expression of Uncertainty in Measurement JCGM 100:2008 – GUM 1995. ISO 5660-1 specifies a method for calculating the heat output and smoke development rate of horizontally oriented test specimens exposed to externally controlled levels of irradiance. The heat output is determined by measuring the oxygen consumption derived from the oxygen concentration, as well as the flow rate in the exhaust duct of combustion products.

The dynamic smoke obscuration rate is calculated by measuring the attenuation of a laser beam by the flow of combustion products. Smoke obscuration is recorded for the duration of the test, whether or not the test piece ignites.

The standard NF EN 45545-2 is also used to calculate the MARHE parameter, which is important for fire protection in rail vehicles. This article also details the measurement uncertainty associated with this value.

For each characteristic associated with its uncertainty calculation, an uncertainty budget will be established according to GUM procedures, in compliance with ISO/IEC Guide 98-1, which describes the calculation of measurement uncertainties. The methods used to calculate uncertainty will be clearly described, and a list of all uncertainty components will be drawn up.

The uncertainty calculations presented in the article assume by default that the quantities are independent, since no correlation has been identified.

In addition, some of the numerical data used in the article are not universal, but come either from calibrations, manufacturer data or estimates. Some numerical values come from personal feedback.

The uncertainty calculation presented in this article is an example of a calculation that is not intended to be universal. Other users of cone calorimeters may deploy a different calculation with different statistical or mathematical assumptions, or different input parameter values, depending on their own user feedback, or the data at their disposal, or the influencing factors that may have been studied (annual sensor drift, representativeness error, calibration uncertainty, sensitivity to ambient temperature, vibration sensitivities, etc.).