Malls, sprinklers and smoke : is smoke in a shopping
mall atrium more of a problem that we bargained for?
Recent tests have been looking at the concerns
Lougheed, G.D.; McCartney, C.
A version of this document is published in / Une version de ce document se trouve dans :
Canadian Consulting Engineer, v. 43, no. 3, May 2002, pp. 23-24
Malls, Sprinklers and Smoke : Is smoke in a shopping maill atrium more of a
problem that we bargained for? Recent tests have been looking at the concerns
G.D. Lougheed and C. McCartney
Fire Risk Management Program
National Research Council of Canada
In North America, it has generally been assumed that communicating spaces
connected to an atrium or mall are sprinklered and, as a result, the sprinklers will
effectively limit the size of a fire in the adjoining space. As a result, engineering design
guides for smoke management systems such as NFPA 92B  have assumed that the
smoke will have minimal effect in the atrium or mall space. However, the design guides
do allow for smoke management designs in which the smoke is allowed to spill into the
atrium space. With the introduction of performance-based designs for fire protection
systems, there has been increasing need to address the potential effects of the smoke in
the atrium or mall.
In a recent joint study with the American Society of Heating, Refrigerating and
Air-Conditioning Engineers (ASHRAE), NRC investigated smoke movement from fires in
sprinklered retail spaces linked to a mall [2, 3]. To conduct the study, a large-scale test
facility was set-up to simulate areas of particular concern: a retail outlet on the second
floor of a mall and a section of a pedestrian mall in a shopping centre. Because North
American fire statistics indicate that approximately 90% of retail fires activate four or
fewer sprinklers , four sprinklers were used in the retail portion of the test facility. The
mall portion of the test facility included a mechanical smoke exhaust system.
An initial series of tests were conducted using a propane burner shielded from
the sprinkler water spray to determine the impact of both fire and sprinkler scenarios
typical of retail spaces on smoke movement. In addition, a series of tests were
conducted to simulate retail fire scenarios with the fuel shielded from direct water spray
from the sprinklers. These scenarios were typical of those that occur in retail stores in
malls and included clothing and toys in boxes located in display units (Figure 1), and
stored or displayed bulk goods, such as paper towels (Figure 2).
The resulting fires had three distinct phases: fire growth and sprinkler activation,
steady fire, and decay. During the fire growth phase, all four sprinklers were typically
activated within 5 minutes and hot smoke flowed into the mall portion of the test facility.
During the steady phase, hot smoke continued to flow into the mall section of the
test facility (Figure 3). A smoke layer formed in the mall area even though the smoke
exhaust system was in use. For the simulated fire scenarios used, hot smoke flowed
into the mall area for up to 20 minutes depending on the test conditions. The optical
density of the smoke in the upper portion of the simulated mall and its carbon monoxide
concentration both exceeded tenability limits. Any accumulation of this smoke in exit
routes could limit evacuation.
During the decay phase, the smoke was cooled to near or below ambient
temperature. The cool smoke was mixed throughout the fire compartment (retail portion
of the test facility) and spilled through the opening and descended into lower areas of the
mall portion of the facility (Figure 4).
A primary objective of the project was to address concerns that smoke cooled by
the sprinklers in retail spaces connected to malls could travel downward, where it could
endanger people evacuating the building. Results from the study, however, indicate
that, during the initial stages (fire growth and steady phases) of a sprinklered retail fire
scenario, the smoke entering the mall area is hot and rises towards the ceiling. A smoke
management system using mechanical exhaust could be used to remove this smoke.
During the decay phase of the fire scenario, the smoke optical density for the
smoke in the secondary space approached or exceeded tenability limits. The rapid
mixing of smoke throughout the fire compartment in or near the opening into the mall
area during this phase could trap any occupants still in the area. However, the extent of
the smoke zone was limited and occurred after occupants should have evacuated the
1. NFPA 92B, Guide for Smoke Management Systems in Malls, Atria, and Large Areas,
National Fire Protection Association, Quincy, MA, 2000.
2. Lougheed, G.D.; McCartney, C.; Taber, B.C., Smoke Movement for Sprinklered
Fires, ASHRAE Transactions, Volume 106, pp. 605-619, 2000.
3. Lougheed, G.D., McCartney, C., Taber, B.C., Sprinklered Mercantile Fires, ASHRAE
Transactions, Volume 107, pp. 730-743, 2001.
4. Johnson, P., Shattering the Myths of Fire Protection Engineering, Fire Protection
Engineering, Issue 1, 1999, pp. 18-27.
Figure 1. Toy display with shielded area.
Figure 2. Bulk display of paper towels.
Figure 3. Hot smoke filling upper portion of simulated mall.
Figure 4. Smoke cooled by sprinklers spilling into simulated mall.