Issue 56: Challenges in Estimating Smoke Detector Response
By James A. Milke, Ph.D., P.E., FSFPE
An engineering approach for estimating the response of
smoke detectors to flaming fires using temperature rise, obscuration and
velocity is included in Annex B of NFPA 72.1 Given that
neither ionization nor photoelectric smoke detectors respond to
conditions represented by any of these three parameters, inherent errors
might be anticipated when using these guidelines for estimating the
response of smoke detectors.2
While it's easily recognized that smoke detectors do not
respond to thermal or velocity conditions, it may not be so apparent
that light obscuration is also irrelevant. Contemporary photoelectric
smoke detectors respond based on the scattering of light caused by smoke
particles.3 The response of smoke detectors incorporating
ionization and photoelectric technologies depends on the distribution of
particle sizes and concentration of particles in the smoke, as
presented in Table 1.4
Table 1. Relationship of Smoke Particle Characteristics and Smoke Detector Sensitivity
Light scattering (LS)
Ionization chamber (MIC)
* ni, and di are the number count (density) and particle diameter for a
small range of particle size "i", referred to as a "bin" in the UL
The relationship between optical density (i.e. light obscuration) and the size and concentration of smoke particles is:
Comparing the proportionalities in Table 1 with that in
equation (1), the disconnect between light obscuration, light scattering
and ionization chamber response is readily apparent.
Heskestad and Delichatsios suggested values of the
optical density that coincided with smoke detector response in their
experimental program. These optical densities are noted in NFPA 721 and
can be converted into the obscuration levels included in Table 2.
Table 2. Obscuration levels for response of smoke detectors to flaming fires [%/f (%/m)]
EXPERIMENTAL DATA OF SMOKE DETECTOR RESPONSE
Data of smoke parameters and smoke detector response from five
large-scale experimental programs were reviewed (see Table 4). A wide
variety of fuels were included in the programs.
Table 4. Summary of Experimental Programs
Naval Research Laboratory (NRL)7,8,9
None, 12 ACH
Home Smoke Alarm Project (NIST)10
Smoke Characterization Project (UL)4
University of Maryland and Underwriters Laboratories (UMD-UL) 11,12
None, 12 ACH
In all of the experimental programs, smoke detectors activated at a
wide range of obscuration levels. As an example, obscuration
measurements in the vicinity of the smoke detectors at the time of
response of ionization and photoelectric smoke detectors from the NIST
project are indicated in Figure 1.
Figure 1. Obscuration Levels at the Time of Smoke Detector Response10
While the wide variation in obscuration levels is due to the lack of a
relationship between light obscuration and the detection mechanisms,
another factor is the transient nature of the experiments. The transient
variation in light obscuration during the shredded paper test in the
UMD/UL experimental program is depicted in Figure 2. Ionization and
photoelectric smoke detectors were in close proximity to each other on
the ceiling of the test room. The "I" and the "P" indicate the
obscuration levels in the vicinity of the exterior of an ionization
smoke detector ("I") and a photoelectric smoke detector ("P") when they
If only the obscuration levels at the time of smoke detector response
were reported, a conclusion could be developed to suggest that an
appropriate threshold obscuration level to depict the response of a
photoelectric smoke detector is 2.5%/ft. However, the first time the
light obscuration reached 2.5%/ft (8%/m) in this experiment was
approximately 80 seconds after ignition. The photoelectric smoke
detector responded at approximately 140 sec after ignition, with the
maximum obscuration level prior to activation being well in excess of
The smoke detector responses in the five programs strongly depend on
the characteristics of the smoke and in some cases the detector
technology. As such, proposing a single set of guidelines to estimate
detector response is very difficult. Guidelines are proposed using 80th
percentile values of the smoke parameters measured at the time of
response for combinations of mode of combustion (flaming vs.
non-flaming) ventilation (none, 6 ACH and 12 ACH) and type of detector
technology (ionization vs. photoelectric).
Figure 2. Light Obscuration in Shredded Paper Test11,12
For flaming fires, the 80th percentile values of the obscuration
level at the time of detector response in the experimental programs are
presented in Figure 3.
Figure 3. 80th Percentile Values of Light Obscuration at the Time of Smoke Detector Response11,12
Suggestions on guidelines for estimating smoke detector
response using temperature rise and velocity are included in the
James A. Milke is with the University of Maryland
NFPA 72, National Fire Alarm Code, National Fire Protection Association, Quincy, MA, 2010.
Schifiliti, R. and Pucci, W., "Fire Detection Modeling, State of the Art," Fire Detection Institute, 1996.
Custer, R, , Meacham, B., and Schifiliti, R., "Design of Detection
Systems," SFPE Handbook of Fire Protection Engineering, 4th Ed.,
National Fire Protection Association, Quincy, MA, 2008.
Fabian,T., Gandhi,P., Patty, P and Chapin, J., "Smoke Characterization
Project: Technical Report," Fire Protection Research Foundation,
Quincy, MA, April 2007.
Heskestad, G. and Delichatsios, M., "Environments of Fire Detectors –
Phase 1: Effect of Fire Size, Ceiling Height and Material," Measurements
Vol I (NBS-GCR-77-86), Analysis Vol II (NBS-GCR-77-95), National
Institute of Standards and Technology, Gaithersburg, MD, 1977.
Su, J. Crampton, G., Carpenter, D., McCartney, C. and .Leroux, C.
"Kemano Fire Studies - Part 1: Response of Residential Smoke Detectors,
Research Report," National Research Council Canada, Ottawa, 2003.
Gottuk, D., et al., "Identification of Fire Signatures for Shipboard
Multi-criteria Fire Detection Systems," Naval Research Laboratory,
Rose-Pehrsson, S. et al., "Multi-Criteria Detection Systems Using a
Probabilistic Neural Network," Sensors and Actuators, B 69, 325-335,
Wong,, J. et al., "Results of Multi-Criteria Fire Detection System Tests," Naval Research Laboratory, Washington, 2000.
Bukowski, R., et al., "Performance of Home Smoke Alarms Analysis of
the Response of Several Available Technologies in Residential Fire
Settings," NIST TN 1455-1, National Institute of Standards and
Technology, Gaithersburg, MD, 2008.
Mowrer, F., Milke, J., and Gandhi, P., "Validation of a Smoke
Detection Performance Prediction Methodology, Volume 2. Large-scale room
fire tests," Fire Protection Research Foundation, Quincy, MA, 2008.
Mowrer, F., Milke, J., and Gandhi, P., "Validation of a Smoke
Detection Performance Prediction Methodology, Volume 3. Evaluation of
Smoke Detector Performance," Fire Protection Research Foundation,
Quincy, MA, 2008.
1st Quarter 2011 - Detection of Forest and Wildlife Fires -- NEMA
An introduction to automated forest/wildland fire detection, along with
a summary of some of the work being done by scientists and engineers. READ MORE
1st Quarter 2011 - Fixed Fire Fighting Systems in Road Tunnels -- Andreas HÃ¼ggkvist
The use of Fixed Fire Fighting Systems (FFFS) in road tunnels is a
controversial issue in many parts of the world. This article presents an
overview of the current research, standards, debates, and attitudes
regarding FFFS in road tunnels. Negative attitudes, the author contends,
are often not based in reality. READ MORE
Winter 2009 - Fire Detection and Alarm Systems in Building Under Construction -- NEMA
This article discusses several scenarios where detection and alarm
systems can play a valuable role in protecting buildings that are under
construction and describes how these systems can be configured. It poses
numerous questions to consider, including how to ensure that fire
detection indeed will contribute to fire protection. READ MORE
For questions concerning delivery of this e-Newsletter, please contact our Customer Service Department at (216) 931-9934 or magazine.sfpe.org.