Fourier Transform Infrared Diagnostics for Improved Fire Detection Systems.
Fourier Transform Infrared Diagnostics for Improved Fire
Serio, M. A.; Bonanno, A. S.; Knight, K. S.; Newman, J.
NISTIR 5904; October 1996.
National Institute of Standards and Technology. Annual
Conference on Fire Research: Book of Abstracts.
October 28-31, 1996, Gaithersburg, MD, 115-116 pp, 1996.
Sponsor:National Institute of Standards and Technology,
Available from: National Technical Information Service
Order number: PB97-153514
fire research; fire science; fire detection systems;
FT-IR; false alarms; photoelectric detectors;
A major advance in the past two decades is the
availability of low cost smoke detectors based on either
ionization or photoelectric detectors. However, these
detectors have some drawbacks because of the high
frequency of false alarms. Other types of detector
technologies have been developed for specific gases,
such as CO2, CO, or O2, based on metal oxide
semiconductors, electrochemical sensors, or optical
sensors. However, all single parameter methods are
hindered by the lack of generality for several types of
fires and a lack of "intelligence," i.e., not always
being able to discriminate against false signals. The
objective of this study was to demonstrate the
feasibility of an Fourier Transform Infrared (FT-IR)
spectroscopy based fire detection system. This work
involved four tasks:  modification of an FT-IR
spectrometer system to investigate three different
detection modes (open-path, cross-duct, extraction into
a multi-pass cell);  FT-IR measurements in the three
modes for several types of flames;  investigation of
advanced signal processing techniques for data analysis;
 preliminary design of a prototype fire detection
system. The types of fires examined included polymethyl
methacrylate (PMMA), polystyrene (PS), polyvinyl
chloride (PVC), polyurethane, douglas fir, methanol,
hexane and heptane.