NIST Time|NIST Home|About NIST|Contact NIST

HomeAll Years:AuthorKeywordTitle2005-2010:AuthorKeywordTitle

Particulate Entry Lag in Spot-Type Smoke Detectors.

pdf icon Particulate Entry Lag in Spot-Type Smoke Detectors. (1141 K)
Cleary, T. G.; Chernovsky, A.; Grosshandler, W. L.; Anderson, M.

Fire Safety Science. Proceedings. Sixth (6th) International Symposium. International Association for Fire Safety Science (IAFSS). July 5-9, 1999, Poitiers, France, Intl. Assoc. for Fire Safety Science, Boston, MA, Curtat, M., Editor(s), 779-790 pp, 2000.


fire research; fire safety; fire science; computational fluid dynamics; smoke detectors; test methods; signals; detection time; detection response


It is well documented that alarm signals from spot-type smoke detectors (ionization and photoelectric) are delayed when the threshold value has been achieved outside the detector housing as a result of convective transport of smoke through the detector to the sensing volume. This delay must be understood in order to properly design detection systems, improve fire modeling where detection is a focus, and interpret analog detector signals coming to a central panel that processes the information. Previously researchers have modeled the time-lag as a first order response with a characteristic time proportional to the inverse of velocity and a constant of proportionality defined as a characteristic length. Here, a two-parameter model is presented. The model parameters are a dwell time and a characteristic mixing time. Both parameters are correlated with velocity using a power-law equation. Detector response experiments were performed in the Fire Emulator/Detector Evaluator (FE/DE). Ceiling mounted analog output detectors were exposed to near step changes in smoke concentration over a wide range of flow velocities. Detector signals were compared to laser extinction measurements slightly forward of the detector housing. Over a flow velocity range from 0.02 m/s to 0.6 m/s, the time for detectors to achieve the maximum response ranged from seconds to 100's of seconds. The data were used to fit model parameters that exhibit strong velocity dependence at low velocities.