A large-scale experiment was performed to measure heat transfer to a massive cylindrical calorimeter engulfed in a 30 minute circular-pool fire. This test simulated the conditions of a truck-sized nuclear waste transport package in a severe fire. The calorimeter inner surface temperature and the flame environment emissive power were measured at several locations as functions of time. An inverse heat conduction technique was used to estimate the net heat flux to the calorimeter. Tall porous fences surrounded the test facility to reduce the effect of wind on the fire. Outside the fences, 2.9 m/s winds blew across the calorimeter axis at the beginning of the test but decreased with time. The wind tilted and moved the fire so that the initial flame environment emissive power was substantially less on the windward side than the leeward side. The calorimeter became more uniformly engulfed as the winds decreased. The maximum heat flux to the calorimeter was 150 MW/m2 on the leeward side at the beginning of the fire, and generally decreased with time. The local variations of calorimeter temperature and heat flux were closely related to the local flame environment emissive power.

A series of experiments were performed to measure heat transfer to a cylindrical steel calorimeter engulfed in a 30-minute pool fire. The calorimeter inner surface temperature history was measured at 46 locations. A one-dimensional inverse heat conduction technique was used to determine the net heat flux to the calorimeter as a function of time and location. The uncertainty in heat flux caused by three-dimensional effects is estimated using finite element computer simulations. A Monte Carlo uncertainty simulation is used to estimate the uncertainty in heat flux from propagated uncertainties in dimensions, temperature measurements, and material properties. The estimated uncertainty in the measured heat flux over the 30-minute fire test and the entire calorimeter was found to be ±18 kW/m2, or 27% of the average heat flux of 66.6 kW/m2. The uncertainties for the early times of the fire test are less than those at later times in the test due to the instability of the inverse conduction calculations caused by the Curie effect of the carbon steel calorimeter material.

The goal of this work was to measure the temporally varying heat flux and surface temperature of a pipe calorimeter in a pool fire, and assess its uncertainty. Three large-scale fire tests were conducted at the Sandia National Laboratories outdoor fire test facility. In each test a cylindrical calorimeter was suspended above a water pool with JP8 fuel floating on top. The calorimeter was a 2.4 m diameter, 4.6 m long, and 2.5 cm wall thickness pipe with end-caps suspended 1 m above the 7.2 m diameter pool. 58 thermocouples were attached to the calorimeter interior surface and backed with 8 cm of insulation. The Sandia One-Dimensional Direct and Inverse Thermal (SODDIT) code was used to determine the calorimeter external surface heat flux and temperature from the measured interior surface temperature versus time. To determine the uncertainty of the SODDIT results, a simulation of the calorimeter in a fire similar to the experiments was performed using the Container Analysis Fire Environment (CAFE) computer code. In this code, a Computational Fluid Dynamics (CFD) fire model applies a temporally and spatially varying heat flux to the exterior surface of a Finite Element (FE) calorimeter model. Flux is similar but not identical to the flux in the experiment. The FE model calculates the internal calorimeter surface temperature, which is used by SODDIT to calculate heat flux which was compared to the applied values. The absorbed heat flux and surface temperature at one calorimeter location was calculated by SODDIT and then compared to the CAFE applied heat flux and surface temperature. From this comparison a base case uncertainty due to inherent inverse calculation errors and frequency smoothing methods is presented. Uncertainties in temperature measurements, calorimeter material properties and wall thickness were applied to the SODDIT calculation and iterated using the Monte Carlo method to determine the overall heat flux and surface temperature uncertainty. The total absorbed heat flux uncertainty at the one studied location is ±4.8 kW/m2 at 95% confidence. The outer surface temperature uncertainty for all data at the one studied location is ±6.6°C at 95% confidence. For all 58 measurement locations, the overall combined total absorbed heat flux uncertainty is ±13.8 kW/m2 at 95% confidence, surface temperature uncertainty is ±7.6°C. These uncertainties are valid only when the calorimeter temperature is not within the Curie temperature range of 999 to 1037K.

A series of large-scale experiments were recently performed to measure heat transfer to a massive cylindrical calorimeter engulfed in a 30-minute circular-pool fire [1]. The calorimeter inner surface temperature was measured at several locations and an inverse conduction technique was used to determine the net heat flux. The flame emissive heat flux was measured at several locations around the calorimeter. Light winds of around 2 m/s blew across the calorimeter axis at the beginning of the test but diminished and stopped as the test continued. The winds tilted the fire so that the windward side of the calorimeter was only intermittently engulfed. As a result, the measured flame emissive power near the windward side was substantially less than the leeward surface. The variation of calorimeter temperature and heat flux was closely correlated with the measured flame emissive power.