Uncertainty of calorimetry measurement for medium-scale fire experiments

Designing controlled fire experiments to be conducted under an exhaust hood requires designing the instrumentation to estimate flow rate, heat release rate (HRR) and smoke extinction coefficient. These measurements are associated with uncertainties that are difficult to quantify or predict. The objective of this research is to quantify uncertainty of measurements in a medium-scale fire experiment (300 kW limit). A variance-based sensitivity analysis was conducted to quantify the impact of uncertainties in measured quantities on quantities of interest (QOIs). Flow rate was found to be influenced primarily by variance in pressure drop followed by gas temperature. The presence of a flow mixer reduced the average velocity by 58% but the velocity profile skewed with maximum velocity shifting closer to the duct perimeter. Calorimetry was conducted using two standardized approaches and a Gamma-type approach that considers the molecular structure of the fuel. The HRR uncertainty was between 2 to 7% for the range of fires studied here and the calculated HRR deviated between −7 to 5% from theoretical HRR. HRR was primarily influenced by the estimation of change in oxygen mass fraction, followed by the mass flow rate. The influence of mass flow rate on HRR measurement increased with increasing HRR. • Impact of mass flow and oxygen depletion accuracy proportional to HRR. • Flow mixer reduces velocity by 58% but slightly reduces variation across the duct. • Calibration-independent Gamma-type calorimetry estimate HRR within 5% accuracy. • Influence of flow and oxygen depletion variance increase and decrease as HRR increases. • Influence of expansion factor and a fuel parameter more dominant as HRR increases.