Experimental analysis of tracer gas injection methods simulating aerosols and their effects on dispersion in aircraft cabins

Abstract

This thesis focuses on experimental environmental engineering in an wide-body aircraft cabin, utilizing injected tracer gas to simulate aerosol dispersion. In this present study, a Boeing 767 mockup cabin section consisting of 11 rows (7 seats per row in a 2-3-2 configuration) is used as the test environment, with a neutrally-buoyant mixture of helium and carbon dioxide used as the injected tracer gas. A total of 123 injection tests were performed at two separate injection source locations in the cabin, using 3 distinct injection methods. The locations tested were the “center” of the cabin and a wall (window) seat in the same row, while the injection methods were a 1-minute puff injection, a 3-minutes puff injection and a (45 minute) continuous injection. Previous work at the Aircraft Cabin Environmental Research Laboratory (ACERL, or simply ACER) has primarily used the continuous injection method, with some studies using a “coughing” type or puff type injection. For comparison purposes, this present study uses a slightly modified puff injection type along with the continuous types. After evaluation of an initial data set, the focus of the study shifted to evaluating the effect the cabin environment (with its inherent chaotic local air currents) had on the three different injection types. The data for each injection type for each testing location were compared. It was found that each injection type had different responses. The responses were evaluated both as the carbon dioxide concentrations versus time graphs, as well in reduced form (the fraction measured, or inhaled, versus the total injected, per location). The reduced responses were not just different, but for the three injection types, they demonstrated a consistent difference between the three injection types that was unexpected. An interesting discovery within the data also showed that there was significant variation in the data from test to test, an unrepeatability and unpredictability, which had not been explored in previous work at ACERL. This variation is best seen in continuous data sets, where shifts in the carbon dioxide concentrations measured at a given location during the injection (a continuous and constant injection) phase change from one quasi-stable steady state condition to a radically different pseudo steady state response. Most previous work relied on testing numerous source locations in the cabin with 3-5 tests per location, as opposed to the approach of this present study, which was 18-22 tests per location, per injection type. This large number allowed better insight into the differences in average behaviors for each injection type, and for insights into the variations in behavior at each injection and measurement location for a given injection type.