Knowing the intensity spectrum of a hydrogen fire helps in evaluating fire detection methods.
The National Aeronautics and Space Administration (NASA) uses large quantities of liquid hydrogen and has expended significant effort in the development, testing, installation, and maintenance of hydrogen fire detectors based on ultraviolet, near-infrared, mid-infrared, and/or far-infrared flame emission bands. Yet, prior to this work, there was no intensity-calibrated broadband hydrogen-air flame spectrum in the literature, making it difficult to compare the merits of different radiation-based hydrogen fire detectors.
A KSC standard hydrogen flame, produced by flowing 20 liters/minute of hydrogen through a 1/16th inch (≈1.6 mm) diameter orifice and burning in air, was placed 4 meters from the spectrometer and used as the emission source. Four different spectrometer configurations were developed — each using different gratings, long pass filters, and detection method — to cover the spectral band from 200 nm in the ultraviolet to 13.5 microns in the far-infrared. Three different calibrated irradiance sources — a deuterium lamp, a tungsten lamp, and a 1,050 ºC blackbody — were used as references, allowing the determination of an intensity- calibrated hydrogen flame irradiance spectrum.
The measurements show that the primary ultraviolet peak of a hydrogen flame (309 nm) has about 1,000 times less irradiance than the primary midinfrared emission band (2.4 to 2.9 microns), contrary to previously published results. But even though the ultraviolet peak was small, it had the best signal-to-noise ratio of any location in the spectrum due to the low noise performance of the detectors used in the ultraviolet region of the spectrum.
As expected, there was minimal visible emission from the flame, though the ubiquitous sodium line at 590 nm was distinctly present. Starting at about 700 nm a series of increasingly strong irradiance peaks was seen, plateauing at 2.4 to 2.9 micron band. Two moderate peaks appeared in the 5-to-8-micron region followed by a rapid drop in irradiance in the far-infrared. The minimal emission of a hydrogen flame in the far-infrared raises concerns about the use of far-infrared cameras to monitor for hydrogen fires, with the data indicating that remote imaging of hydrogen flames might be better done in the mid-infrared.
The calibrated spectral data that was obtained in this effort provides a reference for the intensity of a hydrogen fire as well as a benchmark for the capability of current detectors to detect and distinguish the hydrogen fire peaks from background radiation. This knowledge will aid in the evaluation of the sensitivity of proposed fire detection methods as well as aid in the development of better calibration techniques for existing detectors and the development of field testing devices to verify detector performance at Kennedy Space Center.
This work was done by Robert Youngquist, Ellen Arens, and Stanley Starr of Kennedy Space Center. KSC-13842