The Winter 2006 issue of this magazine
focused on fire protection in warehouses, big box stores, and other
similar occupancies. In the Viewpoint article "So What Do You Expect,"
Warren Stocker noted that occupant notification is not always required
by model codes for many warehouse-type spaces. 1 However,
when an alarm system is required or desired as part of an overall fire
safety plan, it is challenging to find codes and standards that address
the unique characteristics of storage occupancies. He noted that there
is a "natural tendency to apply the general requirements, which in some
cases may border on excess."
This article is the first of a two-part
series on visual signaling in large spaces. In this first part, the need
for research on strobe effectiveness in large-volume spaces is
presented, and an engineering research project aimed at supplying
information for a code committee is described. That project directly
addressed the concerns expressed by Warren Stocker in his Viewpoint
article. In Part 2, which will appear in the Fall 2006 issue, the strobe
project results will be described. Key factors affecting system design
and performance will also be discussed and illustrated.
The NFPA 72 Technical Committee on
Notification Appliances recognized that the effectiveness of the
existing prescriptive and the performance-based requirements for visible
appliances had not been tested in large-volume spaces. The requirements
for the installation and performance of visible signaling in NFPA 72, The National Fire Alarm Code, are based on occupants being alerted by indirect signaling effects. 2 That
is, they are alerted by the illumination of their surroundings, not
necessarily by direct viewing of the signaling appliance. The testing
that led to those requirements was limited to classroom and office-type
spaces. 3 The methodology was never tested in large, well-lit
spaces such as warehouses, large "super stores," etc. Nevertheless,
because strobes are being used for occupant notification in these
spaces, the installation and performance requirements of NFPA 72 are
being enforced despite the lack of any technical foundation. In some
cases, authorities are imposing their own requirements such as not
allowing ceiling-mounted appliances. The Annex of NFPA 72 states that
there may be more efficient methods of visible signaling in large spaces
such as warehouses and distribution centers.
NFPA 72 permits a performance based design
approach that actually exceeds the prescriptive requirements for
visible signaling. 2 Ad hoc testing in a large home supply
store showed that such an approach may be effective, but not necessarily
for the same reasons that it works in smaller spaces. The tests
appeared to show that high ambient light levels resulted in little or no
indirect signaling effect (reflected light) in some parts of the space.
The signal-to-noise ratio produced by the operating strobes was low in
many locations. However, with strobes located over the aisles or
unobstructed by stock, direct signaling and some indirect signaling was
achieved. From that testing, a working hypothesis was formed: In large
spaces, occupant notification is achieved in a large part due to direct
viewing of the strobes, and indirect signaling (reflected light) is
minimized due to high ambient light levels. That is, because of the long
viewing distances, it might be possible for an individual to see
strobes in direct line-ofsight at a distance, while not seeing those
closest. See Figure 1.
The Fire Protection Research Foundation
was asked to consider sponsoring a project to address some of the
committee's concerns. The project sought to identify key factors
affecting occupant notification by strobes in large spaces and to
determine if the existing performance-based requirements of NFPA 72
would be effective. A proposal for limited engineering research and
testing was submitted to the Fire Protection Research Foundation. The
proposal was accepted, and the project was funded.
The project was intentionally designed
for quick but meaningful results. Work began in mid-July 2005 and needed
to be substantially complete by the end of October 2005 to provide
feedback to the code committee. The project did not seek to define all
variables and parameters required for success. Instead, the strobe
alerting systems were studied as a whole to determine if they were
effective. Because of the limited scope and small sample sizes, it was
not possible, nor intended, to determine why a particular system or
component was effective in providing occupant notification.
Nevertheless, engineering analysis of the tests resulted in
identification of several variables that affect the success of the
systems in alerting occupants. As a result, some information on
suggested design and installation practices was added by the technical
committee for inclusion in the Annex of the next (2007) edition of NFPA
The Fire Protection Research Foundation
was the principle sponsor for the project. They provided a grant to
conduct tests, draft code language, and write a report. 4 Several
companies and organizations provided targeted financial support to the
Foundation. One of the goals of this project was to keep costs low and
to actively involve several stakeholders. To achieve these goals, the
help of organizations other than funding participants was solicited.
Project partners were solicited to provide facilities for testing,
technicians to assist in the tests, and other services, including an
American Sign Language interpreter, necessary to move the project
forward. Several project partners provided in-kind donations of time,
facilities, and services.
The results show that it is possible to have
effective occupant notification by strobes installed per the
requirements of the performance-based section of NFPA 72. Occupant
alerting is achieved by a combination of direct and indirect signaling.
The tests highlighted additional factors that designers, installers, and
owners should consider in order to increase the effectiveness of
systems in large spaces. As a direct result of this project, the NFPA 72
Technical Committee on Notification Appliances drafted a Committee
Comment revising the Annex text regarding visible signaling in large
spaces to incorporate ideas and concepts found in the testing.
Tests were conducted in three different warehouse type stores. 4 In
summary, for each test, participants were solicited and asked to walk
around the store. The fire alarm system was then activated. Pre-and
post-test surveys were used to gather data.
For each site, information was gathered
concerning the design and installation of the strobe alerting system.
Ceiling heights were either obtained from plans or measured. The
mounting height of strobes relative to the floor, the ceiling, and
lighting fixtures was measured. Stock heights, aisle widths, and rack
widths were measured in several locations. Ambient light measurements
were taken throughout the stores. Lighting in each of the locations was
provided by fluorescent lamps. Store #3 also had skylights. Lighting
levels were on the order of 430 - 1900 lux (40-180 ft-candles).
Information about each location, including ceiling heights, strobe
heights, stock heights, aisle widths, and aisle spacings, are listed in
the complete report. 4
At Store #1, the fire alarm system was
designed and installed to permit the audible signal to be disabled
separately from the visible signals. This allowed the strobe lights to
be activated without any audible signal.
Strobes were located below the ceiling,
at about the same level of the hanging fluorescent lights. The original
design called for the strobes to be located over the aisles, between
racks. Within each aisle, the strobes were spaced approximately 45 ft
(14 m) to 48 ft (15 m). Rack spacing varied, with most 16 ft (4.9 m) on
center and some as much as 30 ft (9.1 m) on center. Thus, strobe
coverage might be 45 ft (14 m) x 16 ft (4.9 m) in order to provide a
line of strobes in each rack aisle. However, after the system was
installed, the rack layout was altered, resulting in many lines of
strobes not falling directly over an aisle.
The strobes at this location are the
multi-candela-type that is field-adjustable. After the test, it was
found that at least one strobe was never changed from the nominal 15 cd
eff. out-of-the-box setting.
The fire alarm system at Store #2 did not permit
the audible signal to be disabled separately from the visible signals.
The strobes were located on a suspended
acoustical tile ceiling at the same level of the building's fluorescent
lights. The design and installation resulted in most strobes being
located over the aisles, between racks. Within each aisle, the strobes
are spaced approximately 48 ft (15 m). Rack spacing varied, with most 16
ft (4.9 m) on center and some as much as 30 ft (9.1 m) on center. Thus,
strobe coverage might be as low as 48 ft (15 m) x 16 ft (4.9 m) in
order to provide a line of strobes in each rack aisle.
The test at Store #3 coincided with the
Report on Comments meetings of the NFPA 72 Technical Committees. The
participants were all Technical Committee members and included almost
all members of the Notification Appliances Committee.
The fire alarm system at this location
was designed and installed to permit the audible signal to be disabled
separately from the visible signals. However, the manner in which this
was effected resulted in a single audible chirp when the system was
activated. After that first chirp, the audible signals stopped and the
strobes continued to operate.
The strobes at this location were located on
the bottom of the bar joists supporting the ceiling/roof. The
fluorescent light fixtures were approximately 8-12 in. (200-300 mm)
below the bar joists. The design and installation resulted in strobes
being located over most of the main aisle and circulation areas.
However, not every merchandise aisle had a row of strobes overhead.
Typically, the strobes were over the main aisles and over every third to
fifth stock aisle.
There were 13 participants at the Store #1 test, 12 at
Store #2, and 22 at Store #3. Most participants were from the fire
protection and fire service communities. Several participants in each
test had mild to severe hearing impairments. None were considered to be
For each location, participants were asked to rate the effectiveness of
the fire alarm strobe light system. The results for the three test
locations are shown in Figure 2 through Figure 4 and combined for all
locations in Figure 5.
In general, the strobe systems were judged to be
effective. However, clearly there were differences that made the Store
#2 system stand out as the most effective and the Store #1 system as the
least effective system. The Store #2 system used higher intensity
strobes on a reduced spacing and located over almost all aisles between
racks. In Store #1, the strobe locations were designed to be over
aisles. However, before completion, the rack layout changed, resulting
in most aisles not having a row of strobes directly overhead. It is also
possible that many of the strobes in Store #1 were not set to the
correct output intensity. In Store #3, most respondents felt the system
was effective. Based on strobe intensity, ceiling height, and strobe
spacing, the system in Store #3 was over-designed. Nevertheless, when
superimposed on aisle/stock layout, there were aisles where coverage was
minimal or nonexistent. These locations received poor evaluations.
In Part 2 of this article (Fall 2006),
additional test results are presented and the factors affecting system
performance are discussed. Finally, guidelines for system design and
installation are presented.
Stocker,W., "So What Do You Expect," Fire Protection Engineering, Society of Fire Protection Engineers, Bethesda, MD 20814, Winter 2006.
NFPA 72, National Fire Alarm Code, National Fire Protection Association, Quincy, MA, 1993 through 2002 editions.
DeVoss, F., "Report of Research on
Emergency Signaling Devices for Use by the Hearing-Impaired,"
Underwriters Laboratories, Inc., Northbrook, IL, 1991.
Schifiliti, R., "Direct Visual Signaling
as a Means for Occupant Notification in Large Spaces Research Project,"
the Fire Protection Research Foundation, Quincy, MA, 2006.
Editor's Note About This Article
This is a continuing series of
articles that is supported by the National Electrical Manufacturer's
Association (NEMA), Signaling Protection and Communications Section, and
is intended to provide fire alarm industryrelated information to
members of the fire protection engineering profession.