The aim of this work was to develop a new method for studies of diesel spray behavior based on the shadowgraph technique using high speed photography.
This novel technique is based on the construction of two types of images; one averaged and one PPI (Presence Probability Image) spray image. Each average image is a result of averaging the images from a series of test replicates. On the other hand, the PPI images are constructed from addition of the binarized images from the same replicate tests. In these binarized images, a certain gray scale value is defined as the limit for the presence of liquid fuel spray, i.e. areas in the image represented by gray scale values below this threshold is defining the liquid fuel core of the spray. Using these images one track the regions where liquid fuel is most probably present in the spray.
A comparison is then performed of these two approaches, using averaged images and PPI images. The results showed that the averaged images were best suited for digital image analysis, such as numerical representations of the gray scale across the spray. The differences between the spray behaviour of each distinct fuel were easier to interpret visually in the PPI images than in the averaged ones.
Different gray scale threshold values in the images are compared in order to determine the presence of a liquid core, and the influence on the results is presented.
The statistical correlation between the dominating fuel parameters density, viscosity and aromatic content and the spray length and spray angle, respectively, are also presented.
The technique in this paper is applied to six different diesel fuels, representing a wide range of both viscosity and density.