Using a one-dimensional spray model to improve liquid length and ignition delay estimations for diesel flames

Abstract

In the current paper, a methodology based on the combination of a one-dimensional spray model and experimental correlations has been proposed to predict the physical time associated with ignition delay in diesel diffusion flames. This physical time depends significantly on the nozzle geometry, and its influence is not captured in traditional Arrhenius-like correlation. To assess this influence, three multi-hole nozzles with different degrees of conicity (expressed in terms of k-factor) have been tested on an optically accessible 2-stroke single-cylinder engine. First, the hydraulic behavior of the nozzles is assessed from the point of view of injection rate and spray momentum. Later, the effect of the geometry on vapor spray angle has been analyzed through a Schlieren visualization technique. Mie-scattering has allowed to determine the stabilized liquid length. Then, chemiluminescence imaging was used to obtain the temporal and spatial appearance of OH radicals, which are used as indicators to the ignition delay. Finally, all the results are combined with a one-dimensional spray model to determine the physical induction time and include it into a new ignition delay correlation, which shows up to 4% accuracy improvement compared to a traditional Arrhenius equation.