Development of optical diagnostics for quantitative imaging of evaporating fuel films and soot in combustion
Spray impingement and the resulting non-evaporated liquid fuel films are main causes of soot formation in gasoline direct-injection (GDI) engines. This work presents an experimental investigation of liquid fuel-film formation and evaporation, and fuel-film premixed-flame interaction at conditions representative for engines. Two experimental campaigns were conducted at IFPEN (France) and the University of Duisburg-Essen (UDE, Germany).
The experiments at IFPEN were performed in a constant-volume vessel. One of the 8-hole “Spray G” injectors from the Engine Combustion Network (ECN) was used, with a transparent plate mounted perpendicular to the injector axis at 30 mm from the nozzle. A fuel surrogate consisting of 30% toluene - 70% iso-octane was injected at 200 bar, under non-reacting and reacting conditions. UV-absorption imaging and diffused back-illumination (DBI) techniques were used to measure liquid-fuel film thickness and soot extinction, respectively. Since vapor and liquid cannot be distinguished spectrally, morphological post-processing was developed to separate the diffuse, moving vapor and soot clouds from the sharp, stationary features of the fuel film. Spatio-temporally resolved measurements of the film thickness were obtained by high-speed imaging of the UV absorption of toluene. DBI images at 810 nm provided information on soot formation due to film/flame interaction and its spatial distribution at different oxygen excess percentages.
At the UDE, a UV-vis absorption technique was developed to image the fuel-film thickness after direct injection in a heated constant-flow facility in the presence of fuel-vapor and soot. A six-hole GDI injector sprayed fuel at 100 bar onto a transparent plate 30 mm from the nozzle. The gas and wall temperatures were respectively 103 and 79°C, and the gas pressure 1 bar. Another fuel surrogate consisted of a mixture of 30% toluene - 60% iso-octane - 10% n-octane (boiling points 110, 99 and 125°C, respectively) was also used. Absorption by toluene vapor was estimated similarly as at IFPEN. The contributions of scattering and soot extinction at 265 nm were estimated from absorbance images at 310, 365, and 520 nm. Soot incandescence was accounted from dark frame images. The multi-spectral approach permitted obtaining spatio-temporally resolved fuel-film thickness measurements and additional information on the soot.
The experiments at IFPEN were performed in a constant-volume vessel. One of the 8-hole “Spray G” injectors from the Engine Combustion Network (ECN) was used, with a transparent plate mounted perpendicular to the injector axis at 30 mm from the nozzle. A fuel surrogate consisting of 30% toluene - 70% iso-octane was injected at 200 bar, under non-reacting and reacting conditions. UV-absorption imaging and diffused back-illumination (DBI) techniques were used to measure liquid-fuel film thickness and soot extinction, respectively. Since vapor and liquid cannot be distinguished spectrally, morphological post-processing was developed to separate the diffuse, moving vapor and soot clouds from the sharp, stationary features of the fuel film. Spatio-temporally resolved measurements of the film thickness were obtained by high-speed imaging of the UV absorption of toluene. DBI images at 810 nm provided information on soot formation due to film/flame interaction and its spatial distribution at different oxygen excess percentages.
At the UDE, a UV-vis absorption technique was developed to image the fuel-film thickness after direct injection in a heated constant-flow facility in the presence of fuel-vapor and soot. A six-hole GDI injector sprayed fuel at 100 bar onto a transparent plate 30 mm from the nozzle. The gas and wall temperatures were respectively 103 and 79°C, and the gas pressure 1 bar. Another fuel surrogate consisted of a mixture of 30% toluene - 60% iso-octane - 10% n-octane (boiling points 110, 99 and 125°C, respectively) was also used. Absorption by toluene vapor was estimated similarly as at IFPEN. The contributions of scattering and soot extinction at 265 nm were estimated from absorbance images at 310, 365, and 520 nm. Soot incandescence was accounted from dark frame images. The multi-spectral approach permitted obtaining spatio-temporally resolved fuel-film thickness measurements and additional information on the soot.