This report summarizes the research effort undertaken by the FAA to determine any differences in occupant survivability during a simulated post-crash fire when using thermoplastic paneling located in the lower seating area that meets current heat release rate requirements versus paneling that does not meet the current heat release rate requirements. The heat release requirement is based on the Ohio State University (OSU) Rate of Heat Release Test Method.

Two full-scale tests were conducted in the Full-Scale Fire Test Facility at the FAA’s William J. Hughes Technical Center in Atlantic City, New Jersey. The full-scale tests were conducted with a large external fuel fire adjacent to a B-707 narrow-body aircraft fuselage, which simulated a severe but survivable accident in which the fire entered the cabin through a simulated fuselage rupture. The fuselage was instrumented with thermocouples, gas-sampling lines, heat flux transducers, and smoke meters to monitor conditions during the test. The fuselage was also outfitted with flat honeycomb panels installed as sidewalls, stowage bins, and ceiling, and four simulated triple seats constructed of steel angle.

The upper section of the seats contained fire-hardened seat-cushion bottoms and backs that met current FAA flammability requirements, and the lower area contained thermoplastic sheet material on the aft and side areas. During the initial test, thermoplastic paneling that met the current FAA heat release requirements was used, whereas, during a second test, the thermoplastic paneling did not meet current requirements. The tests determined that the use of the noncompliant paneling resulted in more hazardous conditions late in the test.

These conditions were determined using a fractional effective dose model, using temperature and gas data collected during the tests.

The ASTM standard method for measuring heats of combustion of plastics in microscale combustion calorimetry by the oxygen consumption principle uses only the volumetric flow rate and O2 volume fraction exiting a premixed combustor in the calculation. The carbon dioxide (CO2) generated by complete combustion replaces some or all of the O2 consumed from the dry gas stream, depending on the atomic composition of the fuel, so it can change the volumetric flow rate and affect the flow meter response, which is typically calibrated for pure nitrogen. Consequently, the presence of CO2 in the combustion stream causes a systematic error (bias) in the heat of combustion measurement that increases monotonically with the initial O2 concentration. Accounting for volume changes using the combustion stoichiometry is sensitive to the atomic composition of the fuel and is still subject to the CO2 bias in flow meter response, which can be up to 3%. Accounting for volume changes using both the initial and instantaneous flow rates measured at the terminal flow meter in the calculation and correcting the flow meter response for CO2 using an average combustion stoichiometry is more accurate and less sensitive to material composition. Heats of combustion computed by the new method are in quantitative agreement with theoretical values.

The main deck cargo compartment of a freighter (all cargo) aircraft is not required to have a fire extinguishing system. Upon detection of an in-flight fire in a freighter, an approved procedure is to depressurize the aircraft in order to utilize the lower concentration of oxygen at high altitudes to suppress the fire. However, the reduced density of air at high altitude results in less cooling which increases the ignitability. The relative importance of these opposing physical phenomena at high altitude - slower burning rate but faster ignition - has not been studied. To this end, the Federal Aviation Administration initiated a study to measure the burning behavior of materials at high altitude using a pressure vessel modified to control and independently vary the pressure and oxygen concentration. Measurements were made of the burning rate, soot yield and ignition delay of acrylic plastic, a material with an intense but relatively uniform burning rate, at pressures ranging from sea level (1 atmosphere) to the approximate maximum cruise altitude of cargo airplanes (40,000 feet, 0.2 atmospheres) and at oxygen concentrations ranging from 21% (ambient) to 12%. It was found that the ignition delays and burning rates could be described by a new theory that accounts for the separate effects of pressure and oxygen concentration.

This study has been carried out at the request of the Federal Aviation Administration (FAA) and the United Kingdom Civil Aviation Authority (UK CAA) under the provisions of a UK CAA contract. The broad objectives of the study are to collect and analyze data relating to in-service occurrences involving fire, smoke or fumes on US registered aircraft. This involved the compilation of data into a Fire, Smoke or Fumes Occurrence (FSF) Database compiled in Microsoft Excel. The analysis compares genuine and false occurrences by source of fire, smoke, fumes or odors and consequences (diversions, overweight landings, etc.). The data has also been analyzed to derive any likely trends in rates of occurrence. These objectives have now been achieved for data collected over the period 2002 to 2011 and are addressed in this report. A further objective of the study is to analyze the data to determine the monetary impact of the occurrences and any trends in these impacts, which is also addressed in this report.

Download the FSF Database here.

Fire tests were conducted on lithium-ion, lithium-pouch, and lithium-metal battery cells of various cathode chemistries and sizes to evaluate their failure effects. First, tests were performed with a single cell in thermal runaway. Next, a thermal runaway propagation test with five cells was conducted. Finally, a vent gas ignition test to determine the flammability of the vent gases was performed. The tests showed a large variation in the fire hazard characteristics of the thermal runaway event. The characteristics depended on cell size, chemistry, construction, and orientation. As a result of the tests, it is recommended that each battery cell be evaluated on an individual basis dependent on its specific application and operating environment.