Ground-based inerting or GBI involves fuel tank inerting on the ground only. This provides protection during ground operations and during the initial phases of flight where the exposure of the fuel tank to flammable vapors is usually greatest. It consists of displacing most of the oxygen dissolved in the fuel with nitrogen by a process called fuel scrubbing. It also requires displacing the air in the fuel tank empty space, also known as ullage, with nitrogen enriched air (NEA) in a process called ullage washing. GBI could be accomplished by ground-based equipment, that resides at the airport, which would allow for increased cost effectiveness of the process with minimal impact on the airplane. The purpose of this research effort was to more accurately quantify the cost of implementing GBI for all US flights carrying more than 19 passengers. The cost of implementing GBI only for airplanes with heated center wing tanks (HCWTs) was also determined. Airplanes that have the air conditioning equipment, or packs, located below the center wing fuel tanks are considered to have heated center wing tanks. This analysis considered the cost of GBI over 10 years, with a 3-year start up period. The cost of modifying the airplane to allow for GBI was not considered in this analysis.

To determine the cost of GBI, an industry team was assembled, lead by an FAA representative. The team possessed expertise in airports and aircraft ground operations, and state-of-the-art hollow fiber membrane (HFM) nitrogen generation technology. The cost of implementing GBI at Atlanta Hartsfield International Airport (ATL) and Atlantic City International Airport (ACY) were first determined. Using DOT and FAA data, the cost was then extrapolating to all US airports that operate airplane that carry more then 19 passengers.

The results of the cost analysis indicated that the cost of GBI for all flights over 19 passengers, at the nation's 400 largest airports would be approximately 1.6 billion dollars. The cost of GBI for all airplanes with heated center wing tanks, which did not include fuel scrubbing with nitrogen, was determined to be approximately 800 million dollars. For the purpose of this study, aircraft with HCWTs were considered to be all Boeing airplane models manufactured in Boeing's facilities in the Seattle, Washington area (that is, not including the models manufactured in Long Beach, California) and all Airbus airplanes. Nonrecurring costs accounted for 37% of the cost of GBI for all US departures over 19 seats and 50% of the cost of GBI for HCWTs only. Labor accounted for 46% of the total cost of GBI for total applicable departures and 37% of the cost for inerting HCWT departures only. The cost of nitrogen for all applicable departures and HCWT departures only was 14% and 10% of the total cost, respectively.

The purpose of this research effort is to more accurately quantify the amount and purity of NEA required to inert a simple vented rectangular tank with a single NEA nozzle. For the testing, a rectangular fuel tank with approximately 88 cubic feet of volume was employed which was instrumented to allow for temperature, oxygen, and hydrocarbon concentration to be measured in the ullage space. The ullage space of the tank was heated or cooled as required and small portions of heated fuel provided for elevated hydrocarbon concentrations in the tank in accordance with the test plan. The tank was inerted with NEA from an industrial Hollow Fiber Membrane gas generator which provided for both variable NEA oxygen concentration (or purity) and flow rate. All tests were performed at the William J. Hughes Technical Center in the Aircraft Components Fire Test Facility. Each series of tests was designed to determine a specific relationship or to highlight the ability to nondimensionalize the data in terms of a certain parameter. Each test consisted of a single inerting of the test article. Shop air was used to purge the tank to ensure a consistent initial oxygen concentration (approximately 20.9 percent oxygen by volume). At the start of each test, the data acquisition system was started and the NEA was directed into the tank. The tank was considered inert after the oxygen concentration was less then 8 percent by volume; however, all inerting runs did not necessarily end with the test article ullage space at 8 percent. Results were compared to a model developed by the FAA National Resource Specialist for fuel system design.

Varying NEA oxygen concentration and flow rate had the expected effect. Reduction of tank oxygen concentration was faster with decreased NEA purity (decreasing oxygen concentration) and flow rate. The data was nondimensionalized in terms of NEA oxygen concentration and flow rate to get one universal inerting curve. A polynomial curve fit was developed for the curve. Good agreement with this emperical curve and a model developed by ANM-100 and an exact solution calculated was observed with the model and solution tending to be more conservative in terms of predicting volumes of NEA required. Ullage temperature and hydrocarbon concentration tended to have little or no effect on the ability of the NEA to inert the tank.