Tomographic reconstruction of spray evolution considering multiple light scattering effects: Characterization of an ECN Spray G Injector

Meeting stricter legal emission limits and the simultaneous introduction of new synthetic fuels are key challenges for current and future research in the field of engine combustion. A deep knowledge of spray behavior is mandatory to address these problems, as air‐fuel mixture and spray propagation in gasoline direct injection (GDI) are essential processes to achieve a highly efficient and clean combustion. Thus, a lot of effort is put into the identification of air‐fuel spray distributions. Most of them use lasers or X‐ray sources, which are accompanied by a high experimental complexity and further drawbacks. In this work, the measurement technique application of high speed diffuse back illumination extinction imaging (DBIEI) is used to obtain quantitative information in the form of projected liquid volume fraction (PLV). The DBIEI setup is simplified to enable an easier and quicker application for different experimental environments, using a LED-Panel as light source, which fulfills diffuse back illumination (DBI) criteria. Measurements are done in a constant volume chamber, allowing easy optical access with up to 200 mm in diameter and enabling measurements at real world ambient engine conditions. An engine combustion network (ECN) Spray G injector is used. ECN ambient conditions G1 (3.5 kg/m³ ambient density at 300°C), G2 (0.5 kg/m³ ambient density at 60°C) and G3 (1.01 kg/m³ ambient density at 60°C) are chosen. Isooctane is used as fuel. The injector is mounted in a motorized rotational system, enabling measurements of the spray at defined and exact angles. The DBIEI measurement technique suppresses the effect of beam steering at elevated ambient conditions, allowing the measurement of scattering based light attenuation by spray droplets. This requires a light source radiating uniformly over a certain angle‐range. Nevertheless, an inherent error in the quantification of liquid phase results from the detection of multiple scattered photons. The error is even more enhanced when using a non‐collimated light source. This leads to an underestimation of the optical depth (OD), which further results in a false calculation of the projected liquid volume. Therefore, beside very low‐density regions e.g. at spray boundaries, PLV results have to be assumed wrong. To enable the use of DBIEI in more dense spray regions, we present a simulation-based method correcting the ODs. Derived from this the corrected values indicate an underestimation of the OD of a factor greater than 2. The corresponding PLV data at different viewing angles is then used to reconstruct three-dimensional data of the liquid volume fraction (LVF) with filtered back projection (FBP). Thereby we can obtain time and spatial resolved quantitative spray information using an easy experimental setup, with an approach to correct beam steering and multiple scattering, while the experimental effort is kept low by using LED light sources. This data can be used for comparison, calibration and evaluation of simulation data of transient sprays, leading to detailed knowledge of spray behavior and mixture formation in different conditions.