top of page

Phase Doppler Anemometer - The Shift From Simulation To The Real World

J. F. Meyers (1), G. Wigley (2), J. Jedelsky (3), M. Maly (3), O. Cejpek (3)

(1) (Retired), NASA – Langley Research Center, Hampton, VA, USA

(2) Aero and Auto Engineering, Loughborough University, UK

(3) Mechanical Engineering, Brno University of Technology, Brno, Czech Republic

DOI:

A Monte Carlo type simulation program has been developed to model phase Doppler systems. The work was divided into four main sections, the development of a launch strategy of a known range of droplet sizes and trajectories, the determination of the scattered light fields using Mie theory, the generation of the photomultiplier signals, and then their processing to generate drop size estimates followed by the comparison of these data with the launched data. The signal processing portion of the simulation measured the Doppler frequency in each signal to estimate the droplet velocity. Rather than using measurements of the phase differences between the multiple Doppler-shifted signals, the time shift between the arrival times of each signal was used to estimate the dropsize. This provided a linear relationship between the dropsize and the signal time shift. The missing link in the simulation is the ability to guide the experimentalist as to the best way to acquire the signal bursts and determine how accurate the measurements would be when using a PDA system to determine the droplet size distributions in real spray flows. Thus, the only way to determine measurement accuracy is to include that in the signal processing of an actual PDA system and spray. In an assessment of the capabilities of the signal processing technique, a sequence of Doppler bursts from an actual spray characterization were recorded and used as input to the software signal processor. The Time Shift Technique was based on the use of Weighted Average calculations to determine the exact middle of each of the three photomultiplier signal bursts. However, it was found that good quality signal bursts are required. Since cycle measurement techniques are less effected by noise, a cycle technique based on the samples captured by the AD converter was added to the software. Whereas this technique did show an improvement in measurement accuracy it led to the development of a cycle technique based on the signal burst frequency. This involved a reconstruction of the original captured bursts. The measurement errors were found to be in the micron range.

20th Edition
bottom of page