Masters Theses

Date of Award

5-1995

Degree Type

Thesis

Degree Name

Master of Science

Major

Electrical Engineering

Major Professor

Bruce W. Bomar

Committee Members

Roy Joseph, Montgomery Smith

Abstract

The Center for Laser Applications at The University of Space Institute required a computer program to calculate the period and phase shift between two digitized photodetector signals. The phase shift and period can be used to determine the size and the velocity of jet fuel particles falling through a laser probe volume. Particles are dropped in the probe volume where two laser beams intersect. A particle scatters light as it passes through the probe volume and the scattered light is detected by two photodetectors. The photodetector signals are digitized by 8-bit analog-to-digital converters and read into the RAM of two digital signal processing cards which can hold the data from approximately 1000 particles. The data placed in RAM is used to determine the phase shift and period using the crosscorrelation function. The limitations on using the crosscorrelation for this purpose are discussed as well as methods to reduce the error introduced during the data reduction process. The performance of the data acquisition system was verified by checking that the DSP cards were triggering simultaneously so that no erroneous phase shift is introduced during data acquisition. This was accomplished by introducing simulated signals into the DSP cards via different length cables. The time lag was calculated and it was determined that the cards were functioning correctly. Program checkout consisted of verifying the code on simulated data and determining under what conditions the program would give valid results. Simulated signals similar to the signals produced by the experiment were input into the data reduction portion of the program. The signal visibility and background light power were varied to produce signals with different levels of noise. These signals were input to the program to test the conditions under which the program would fail. The program yielded accurate results if the signal visibility is greater than 75% and the background light power is less than 25%. It was determined that an accurate phase and period measurement results if the number of periods used in the crosscorrelation calculation is between five and twenty. If the signals do not contain five periods, there are not enough periods to calculate an accurate crosscorrelation. If there are more than twenty periods, the determination of the peaks is difficult because they may fall between two points on the crosscorrelation curve. The peaks of the crosscorrelation curve are crucial in the determination of phase and period.

Download
Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS