Turbulent structure and emissions of strongly-pulsed jet diffusion flames

This current research project studied the turbulent flame structure, the fuel/air mixing, the combustion characteristics of a nonpremixed pulsed (unsteady) and unpulsed (steady) flame configuration for both normal- and microgravity conditions, as well as the flame emissions in normal gravity.;The unsteady flames were fully-modulated, with the fuel flow completely shut off between injection pulses using an externally controlled valve, resulting in the generation of compact puff-like flame structures. Conducting experiments in normal and microgravity environments enabled separate control over the relevant Richardson and Reynolds numbers to clarify the influence of buoyancy on the flame behavior, mixing, and structure. Experiments were performed in normal gravity in the laboratory at the University of Washington and in microgravity using the NASA GRC 2.2-second Drop Tower facility. High-speed imaging, as well as temperature and emissions probes were used to determine the large-scale structure dynamics, the details of the flame structure and oxidizer entrainment, the combustion temperatures, and the exhaust emissions of the pulsed and steady flames. Of particular interest was the impact of changes in flame structure due to pulsing on the combustion characteristics of this system.;The turbulent flame puff celerity (i.e., the bulk velocity of the puffs) was strongly impacted by the jet-off time, increasing markedly as the time between pulses was decreased, which caused the degree of puff interaction to increase and the strongly-pulsed flame to more closely resemble a steady flame. This increase occurred for all values of injection time as well as for constant fuelling rate and in both the presence and absence of buoyancy. The removal of positive buoyancy in microgravity resulted in a decrease in the flame puff celerity in all cases, amounting to as much as 40%, for both constant jet injection velocity and constant fuelling rate.;The mean flame length of the strongly-pulsed flames was not strongly impacted by buoyancy. This lack of sensitivity to buoyancy was consistent with offsetting changes in flame puff celerity and time to burnout for the microgravity versus normal-gravity cases.;The emissions of CO and NO were examined in the vicinity of the visible flame tip and at the combustor exit for strongly-pulsed flames. The highest exhaust-point emission indices of CO for compact, isolated puffs were as much as a factor of six higher than those of elongated flames with longer injection times. The amount of CO decreased substantially with a decreased amount of flame puff interaction. The higher CO levels for pulsed flames with the shortest injection times were consistent with quenching due to the very rapid mixing and dilution with excess air for the most compact flame puffs.;The injection time for which steady-flame emission levels were attained was comparable to the injection time for which the visible flame length approached the flame length of steady flames. The CO emissions, for a given fuelling rate, were strongly dependent on both the injection time and jet-off time for a jet-on fraction less than approximately 50%. The NO levels were generally proportional to the fuelling rate.;This work indicates that there are specific combinations of injection time and jet-off time that considerably change the fuel/air mixing, resulting in emissions comparable to those of the steady flame while the flame length is significantly shorter. This points the potential utility of the strongly-pulsed injection technique in the development of compact, low emissions combustors involving turbulent diffusion flames. (Abstract shortened by UMI.)...