An investigation into the fire safety of a wing fuel tank has been performed to aid in the effort to eliminate or reduce the possibility of a wing fuel tank explosion in a commercial aircraft. A computational model is built to predict the generation of flammable mixtures in the ullage of wing fuel tanks. The model predicts the flammability evolution within the tank based on in-flight conditions of a wing fuel tank. The model is validated through supporting experiments performed in an altitude chamber, the wind tunnel facility as well as data obtained from flight tests. The results from the experiments are compared to the computational results. Computational results from the altitude chamber follow the general trend of the experimental results, but produce them at a different flash point. This is due to the replenishment of species with lower flash point at the surface of the fuel which emulates the flash point of the entire fuel to be lower. Experimental results for the aluminum wing tests from the wind tunnel experiments are in good agreement with the computational results as well.

A simpler model is developed from a program that calculates fuel air ratio within the ullage of fuel tanks in order to reduce the required number of inputs to the model. This model is applied to the data sets for the experiments performed in the altitude chamber and wind tunnel. For the tests conducted in the altitude chamber, the correlation estimates the hydrocarbon concentrations extremely well during ascent and descent. During the on-ground condition the estimation is good, but not as accurate as the ascent or descent conditions. For the tests conducted in the wind tunnel, the computational values follow the general trend of the experimental values, but the computational values estimates the total hydrocarbon concentration approximately 10% lower than the experimental value consistently.

Flammability studies are also performed in order to track the effects of temperature, pressure, and oxygen concentration on the upper and lower flammability limits. For the temperature and pressure profiles considered in this work, it is found that the temperature and pressure effects on the flammability limits are minimal. In contrast, the oxygen concentration has a significant effect on the flammability limits of the vapor; the flammable region narrows with a decrease in oxygen concentration.

This study has been performed to aid in the effort to minimize the possibility of a fuel tank explosion in a commercial aircraft. An understanding of the mechanisms behind fuel vaporization processes in an aircraft fuel tank is essential to developing accident prevention techniques. An experiment was designed to measure the conditions existing within a heated aluminum fuel tank, partially filled with JP-8 jet fuel, under varying ambient conditions similar to those encountered by an in-flight aircraft. Comprehensive fuel tank data, including all temperatures, pressure, and ullage hydrocarbon concentration, was obtained during testing, and is available for use to validate heat and mass transfer calculations. An existing model was employed in this work to calculate ullage temperature and ullage fuel vapor concentration in the tank and compare with measured values, to explain the transport processes occurring in the tank during testing, and to estimate the flammability of the ullage vapors existing within the tank. The calculations made by the model were in good agreement with the measured data. The model also gave a good indication of the temporal mass transport processes occurring in the tank and gave a reasonable assessment of the ullage vapor flammability in the tank.

This report describes the methodology and results of a study undertaken for the Federal Aviation Administration into the characteristics of Fuselage Breaks and their effects on occupant survival in ground pool fire Accidents. For those accidents where the fuselage remains largely intact, a determination has been made of the nature of any Fuselage Ruptures resulting from the accident sequence. An assessment has also been made as to whether the probability of occurrence of ground pool fires is different for Aircraft with Wing Mounted Engines than for Aircraft without Wing Mounted Engines. The Cabin Safety Research Technical Group Aircraft Accident Database was used to select Survivable Accidents that occurred during the period 1967 to 2000, to Passenger carrying western world turbojet aircraft. Where appropriate to the study these accidents were analyzed in depth using Accident Reports and other data published by National Airworthiness and Investigating Authorities.

The results of the study suggest that further data would be needed to determine any significant difference that might exist between the probabilities of occurrence of a ground pool fire accident for Aircraft with Wing Mounted Engines and for Aircraft without Wing Mounted Engines. All that can be determined, within the constraints of the size of the existing data set, is that any difference that might exist is unlikely to be large. Whilst it is likely that the majority of ground pool fire accidents in which the aircraft remains largely intact sustain Fuselage Ruptures, there are insufficient data available to establish the size of any such ruptures.

For the Fuselage Break accidents studied the majority involved at least two breaks. Whilst the number of Fuselage Breaks does not appear to be influenced by the intensity of the impact the probability of a Fuselage Break tends to increase as the impact becomes more severe. Although no firm conclusions can be made, it is considered likely that approximately half of the Fuselage Breaks occur at a point of structural discontinuity. The occurrence of a Fuselage Break in ground pool fire accidents seems to result in a more severe fire threat to the occupants. However, it is evident that for the majority of ground pool fire accidents studied, involving a Fuselage Break, the occupants used the breaks as an escape route. In order to ascertain the net effects of Fuselage Breaks on occupant survival a Monte Carlo simulation model was developed. The primary value of the model was an assessment of the effects on occupant survival of changes in the probability of occurrence of Fuselage Breaks. Based on the results derived from the model it is considered that Fuselage Breaks have a net adverse effect on occupant survival. The change in the number of Fatal Injuries, F, with changes in the probability of a Fuselage Break ΔB for an aircraft with N occupants may be reasonably well represented by the following equation: