X3 expansion tube driver gas spectroscopy and temperature measurements

Abstract

The University of Queensland’s X3 facility is a large, free-piston driven expansion tube used for super-orbital and high Mach number scramjet aerothermodynamic studies. During recent development of new scramjet test flow conditions, experimentally measured shock speeds were found to be significantly lower than that predicted by initial driver performance calculations. These calculations were based on ideal, isentropic compression of the driver gas and indicated that loss mechanisms, not accounted for in the preliminary analysis, were significant. The critical determinant of shock speed is peak driver gas sound speed, which for a given gas composition depends on the peak driver gas temperature. This temperature may be inaccurately estimated if an incorrect fill temperature is assumed, or if heat losses during driver gas compression are significant but not accounted for. For this study, the ideal predicted peak temperature was 3750 K, without accounting for losses. However, a much lower driver temperature of 2400 K is suggested based on measured experimental shock speeds. This study aimed to measure initial and peak driver gas temperatures for a representative X3 operating condition. Examination of the transient temperatures of the driver gas and compression tube steel wall during the initial fill process showed that once the filling process was complete, the steady-state driver gas temperature closely matched the tube wall temperature. Therefore, while assuming the gas is initially at the ambient laboratory temperature is not a significant source of error, it can be entirely mitigated by simply monitoring tube wall temperature. Optical emission spectroscopy was used to determine the driver gas spectra after diaphragm rupture; the driver gas emission spectrum exhibited a significant continuum radiation component, with prominent spectral lines attributed to contamination of the gas. A graybody approximation of the continuum suggested a peak driver gas temperature of 3200 K; uncertainty associated with the blackbody curve fit is ±100 K. However, work is required to quantify additional sources of uncertainty due to the graybody assumption and the presence of contaminant particles in the driver gas; these are potentially significant. The estimate of the driver gas temperature suggests that driver heat losses are not the dominant contributor to the lower-than-expected shock speeds for X3. Since both the driver temperature and pressure have been measured, investigation of total pressure losses during driver gas expansion across the diaphragm and driver-to-driven tube area change (currently not accounted for) is recommended for future studies as the likely mechanism for the observed performance gap.