Search

Understanding drop breakup will optimize aircraft engine performance, reduce agro-chemical overspray, and improve pharmaceutical tablet efficacy. Large fuel fragments in engines lead to lowered fuel economy and higher pollutant emissions, while small drops yield more agro-spray drift into surrounding residential and environmental zones. Better pharmaceutical tablets will improve drug uptake and patient comfort. Engineers and scientists are currently unable to predict the number, size, and velocity of fragments formed during important drop breakup processes. Therefore, we are required to measure these quantities. We use digital inline holography (DIH) to record three-dimensional diameter and position data for fragments formed during multi-mode breakup. DIH provides 3D images at framing rates 300 times faster than in an IMAX theater. A laser is used as the light source and a high speed camera records the breakup events to video files. A MATLAB script is used to extract the diameters and positions of all fragments in the spray. The data is sorted into bins and histograms are produced which describe the probability of observing a fragment of any particular size and speed. Results show size histograms with more than one peak, a finding in direct contradiction to the last 40 years of spray research. Multiple peaks are indicative of fragmentation processes that occur due to multiple breakup mechanisms, with the number of histogram peaks corresponding to the number of mechanisms (some combination of bag, rim, and/or stamen breakup modes). The histograms will be useful to those modeling sprays in gas turbine engines and industrial sprayers.

Digital micro holography offers an in-situ, non-contact and three-dimensional way to explore the microscopic world. However, as it is difficult to focalize the whole object in one single reconstructed image, the application of digital micro holography to cases with a large longitudinal object volume is limited by the microscopes depth of field. By extending the depth of field in reconstructed micro holograms in the wavelet domain, this paper fully takes advantage of numerical reconstruction algorithms to solve this problem. First, a recorded hologram is rebuilt using the wavelet transform approach by setting up an appropriate longitudinal interval to obtain a series of reconstructed hologram planes. Then each plane is decomposed with wavelet into its sub-images of both high and low frequencies. Furthermore, the local variance of the maximum intensity gradients of the high- and low-frequency coefficients is calculated and utilized as the focus criterion. Finally, the image planes are fused into a single one with the depth of field extended to a large extent. The feasibility and robustness of this reconstruction procedure for both continuum and particle fields are investigated. One of the demonstrations is made in an experiment of a tilted continuum:carbon fiber. It is different from most of the previous applications where the interrogated is the particles and where the area involved is parallel to the CCD. The carbon fiber gets successfully reconstructed in three dimensions, and the measurement errors of its diameter are presented together with the reconstruction distances. Another is an experiment of a dispersed particle field:micro transparent particles are generated by an ultrasonic atomizer, for which the reconstruction procedure achieves an extended depth of field. In addition, a numerical model based on generalized Lorenz-Mie theory is used to simulate the holograms of both opaque and transparent particles of 1-15 m in diameter. Variations of the longitudinal location errors with the Fraunhofer number are analyzed, and comparisons are made between the results of opaque and transparent particles. Both the experimental and simulation outcomes show that this reconstruction procedure is a reliable one to acquire an extended-depth-of-field hologram for both the continuum and the dispersed particle fields, and then to accurately measure the objects.