A closer look at determining burning rates with imaging diagnostics

K. Ryan Bratton, Connor Woodruff, Loudon L. Campbell, Ronald J. Heaps, Michelle L. Pantoya

Research output: Contribution to journalArticle

Abstract

Energetic composites such as pyrotechnics and thermites often produce high-brightness reactions such that measuring burning rate is challenging because of camera sensor saturation. The objective of this study was to compare burning rate measurements of reacting powders utilizing various filtration and illumination techniques. Experiments were designed using aluminum (Al) and molybdenum trioxide (MoO3) powder mixtures. The mixture was loaded into transparent tubes at a constant bulk density and ignited. Burning rate was determined using a high-speed camera and applying a series of neutral density filters to the camera lens. Another approach used a copper vapor laser (CVL) and 511 nm notch filter on the camera lens. Two methods of image analysis were compared. The first was based on propagation of the reaction front using still-frame images processed by the camera software (frame-by-frame analysis). The second used a MATLAB software program to analyze pixel-by-pixel intensity as a function of time (pixel-intensity analysis). Results showed an average 8.31 cm/s burning rate that remained constant for all image filtration techniques using the frame-by-frame analysis. The pixel-intensity analysis showed an average burning rate of 8.05 cm/s for all filtration levels. The two methods of image analysis resulted in approximately the same burning rate when the standard deviation of the samples was considered (∼0.4 cm/s). A sensitivity test to explore the influence of the number of frames analyzed to compute the burning rate was also performed. Results showed large inconsistencies in determined burning rate when less than 100 frames were analyzed but converged with relatively small (i.e., 1%) error once a minimum of ∼150 frames were analyzed. The analysis revealed the stochastic nature of flame propagation that occurs in cyclic stages of stationary heating followed by energy propagation. If the number of frames used for data analysis are limited, it is possible to include only propagation or only heating steps leading to a faster or slower burning rate, respectively.

Original languageEnglish
Article number105841
JournalOptics and Lasers in Engineering
Volume124
DOIs
StatePublished - Jan 2020

Keywords

  • Burning rate
  • Energetic reactions
  • Flame speed
  • High brightness imaging
  • Laser illumination imaging
  • Thermites

Fingerprint Dive into the research topics of 'A closer look at determining burning rates with imaging diagnostics'. Together they form a unique fingerprint.

  • Cite this