The Interaction of Roof Vents and Draft Curtains with Standard Sprinklers, Part 1

By Richard Schulte

Schulte & Associates, Evanston, Ill.

Section 910.1 in the International Building Code (IBC) requires that automatic smoke and heat (roof) vents be provided in single-story portions of buildings with an undivided room or space that exceeds 50,000 square feet in floor area, which is classified as either a moderate hazard industrial (F-1) or storage (S-1) occupancy. This section further requires that smoke and heat vents also be provided in buildings that contain high-piled storage per the International Fire Code. Given that the IBC requires that sprinkler protection be provided in fire areas that contain moderate hazard industrial or storage occupancies that exceed 12,000 SF in floor area, the IBC, in effect, mandates that smoke and heat vents be provided in large single-story industrial and storage buildings that are protected by standard sprinklers.

The issue of whether or not smoke and heat vents should be provided in buildings that are protected by standard sprinklers has been the subject of debate in the fire protection field for well over 30 years. The debate is over whether open roof vents will delay the operation of sprinklers in the vicinity of a fire and whether this delay in sprinkler operation will be detrimental to the ability of the sprinkler system to achieve control of the fire. While the interaction of sprinklers and smoke and heat vents is still debated, two other issues regarding the interaction of sprinklers and smoke and heat vents have been resolved. These issues are whether the operation of sprinklers affects the opening of automatic roof vents and whether the use of draft curtains affects the operation of the sprinklers.

Despite containing numerous obvious flaws in logic, a paper titled "Interaction of Sprinklers with Smoke and Heat Vents"(1) is an excellent overview of the issue of installing roof vents and draft curtains in buildings that are protected by standard sprinklers. The paper reviews research projects that have been conducted on the interaction of sprinklers, roof vents and draft curtains since 1955 and also discusses the various pros and cons of using roof vents and draft curtains in buildings protected by sprinklers. The following are excerpts from the Beyler/Cooper paper:

"The experimental studies have shown that early vent activation has no detrimental effects on sprinkler performance and have also shown that current design practices are likely to limit the number of vents operated to one, and vents may in fact not operate at all in very successful sprinkler operations. Design practices should move to methods that assure early operation of vents, and vent operation should be ganged so that the benefit of roof vents is fully realized."

"Positive Claim: Smoke and heat vents improve visibility: The benefit of improved visibility is a result of the fundamental action of the venting. Smoke that is vented from the building does not contribute to the reduction of visibility within the building."

“Because the buoyancy and smoke concentration is greatest in the curtained area of the fire, smoke and heat vents provided within the draft curtain area of fire origin will most effectively vent the smoke and heat of the fire, hence improving visibility with the building. The enhanced visibility benefits escaping occupants of the building and firefighters who need to locate the fire to complete fire extinguishment."

"Positive Claim: Smoke and heat vents reduce temperatures and hazardous gas concentrations: The above explanation for improved visibility, i.e., removal through vents of the smoke and replacement with cool, uncontaminated air, also explains how vents generally lead to reduced temperatures and reduced toxic and combustible gas concentrations within the space. The reduction in temperatures and hazardous gas concentrations benefit escaping occupants of the building and firefighters who need to locate the fire to complete fire extinguishment."

"Positive Claim: Smoke and heat vents contain damage to the curtained space: The combined action of draft curtains and smoke vents not only allows for the removal of smoke and heat from the building but also acts to limit the spread of heat and smoke outside the curtained area. The smoke and heat are trapped within the curtained area and are directly vented to the outside. In the absence of the curtains and vents, the smoke would spread throughout the facility, causing additional damage to the building contents."

"Positive Claim: Smoke and heat vents assist the fire department in identifying the location of the fire within the facility and reduce the need for hazardous manual roof venting: The opening of the vents will lead to a flow of smoke through the roof of the facility, but only within the bounds of this curtained compartment of fire origin. Thus, the location of the fire inside the facility is revealed to the fire department from outside the facility. In the absence of the curtain/vent system, the smoke would spread through the volume of the entire facility and flow to the outside through all randomly spaced leaks in the upper building envelope. ... "

"Positive Claim: Smoke and heat vents provide protection even if the sprinklers do not work: It is generally recognized that sprinkler systems are operational and effective in 90 to 95 percent of the fires, depending on the statistical source used and the definitions and qualifications applied. If the sprinkler system is not operational or effective, then manual firefighting needs to be relied upon for fire control. The smoke and heat vents will be effective in limiting damage to the building, providing firefighter access to the fire and aiding in the escape of building occupants. In short, the benefits of heat and smoke vents can be realized in the absence of an effective sprinkler system."

"Positive Claim: Smoke and heat vents prevent an excessive number of sprinklers from operating: By limiting the spread of heat and smoke to the curtained area of fire origin, the operation of sprinklers remote from the fire is prevented. While sprinkler systems are designed to perform adequately without the benefit of smoke vents and draft curtains, in marginal fire control situations, the prevention of the activation of remote sprinklers can allow successful fire control by the sprinklers where control might otherwise not be achieved."

"Negative Claim: Smoke and heat vent flow rates are insufficient to realize any benefit: The claim here is that the action of discharging sprinklers is so effective in cooling the smoke that the remaining forces of buoyancy will not be strong enough to successfully drive a significant amount of smoke out of the roof vents. As such, the benefits posed for smoke and heat venting will not be realized."

"The FMRC fire test facility at West Gloucester was used for a full-scale test program to determine whether existing or new technology fire sprinkler systems are capable of providing acceptable protection for storage found in warehouses and warehouse-type retail stores. ... The authors indicated that neither of the two tests with draft curtains [without roof vents] met the above criteria."

" ... While there have been many attempts to model all or part of the interactions of sprinklers and vents, the issues are more complex than can be dealt with using even the most sophisticated modeling methods available today. The clearest indication of this is the recent NFPRF research project. While modeling of the fluid mechanical aspects of the problem were quite successful in predicting aspects of sprinkler activation in the first heptane spray fire series, the model was unable to predict the corresponding results in the rack storage tests beyond first sprinkler activation. Similarly, there have been many studies of portions of the problem through experimentation and analysis. None of that work is sufficiently comprehensive to rise to the level of insight provided by the experimental studies in the prior section. ... It is notable that in the time since the 1974 FMRC model study, FMRC has conducted hundreds of full-scale sprinkler tests and has published no additional modeling studies of the type used in the 1974 report."

"None of the testing programs reviewed used a test building of sufficient size to fully evaluate the interactions of sprinklers and roof vents. As large as some of the test facilities were, they are dwarfs beside the buildings in which sprinklers and vents are used. The FMRC facility (4650 m2 (50,000 ft2)) has no capabilities to include roof vents, and as such, FMRC has never performed a full-scale sprinklered test with roof vents. The UL facility has a test area of only 1393 m2 (15,000 ft2), only about three times a typical curtained area, and that facility cannot be operated without ventilation due to environmental concerns. As such, we must realize that the data available to us at this time are not complete and require great care in assessing our understanding of the issues."

"Limiting the extent of smoke spread is the key physical process that allows emergency egress and firefighter access and that limits spatial extent of smoke and heat damage."

"The claim that venting assists the fire department in locating the fire and reduces the need for manual venting relates to operational characteristics of vents. That automatically operated vents or even manually operable vents reduce the demands on firefighters venting the building are not matters for fire research. Similarly, that fire plumes are visible from roof vents is not a matter for research."

"The claim that smoke and heat vents operate effectively when sprinklers do not operate is clearly a valid one. The smoke venting studies reviewed in this paper and others clearly provide the basis for the claim. The real question here is how relevant is the claim, i.e., how reliable are sprinkler systems. While it is outside the scope of this paper to review sprinkler system reliability studies, sprinkler systems are widely reported to be 90 - 95 percent reliable, with strong indications that the actual reliability is even lower. In the remaining cases, manual firefighting must be relied upon, and the support of an automatic venting system has clear value in these cases."

"The claim that the use of smoke and heat vents will enhance burning rates has been actively made by Factory Mutual (e.g., Battrick 1986; Ward 1985]. This view has also been the basis for advising firefighters to not enter or vent a building protected by sprinklers, but rather the building should be "buttoned up," and the sprinkler system should be left to do its work. Entry should only be attempted after the fire is clearly controlled, though guidance on how this is to be determined is not clearly given. This guidance clearly contradicts normal fire service practices, and the FM guidance does not seem to be followed in general."

"The claim that smoke and heat vents will delay sprinkler activation is not supported by the available data, except when the fire is directly below the vent. Tests in which vents were manually operated at the start of the test by FMRC [Heskestad 1974], IITRI [Waterman et al. 1982; Waterman 1984], Ghent [Hinkley et al. 1992a], 1998 UL [McGrattan, Hamins, and Stroup 1998] all showed no effect on the activation of early sprinklers. Similarly, the 1998 UL rack tests, where vents were opened at the first sprinkler activation, showed no effect on the timing of subsequent sprinkler operations. Where the fire is not directly beneath the vent, there are no data that indicate this claim is valid. When the fire is directly beneath the vent, the FMRC tests [Heskestad 1974] found no notable effect of the vent on sprinkler activations. In the 1998 UL heptane tests, some delays in early sprinkler activations were noted. No serious effects were noted. The 1998 UL rack tests intended to explore this phenomenon, but the vent fusible link failed to operate the vent due to cold soldering. The overwhelming evidence is that vents do not affect sprinkler operations even if opened at the start of the test. ... Early activation of vents and ganging vents are viable strategies that should be employed to improve venting reliability."

"The negative claim is that smoke and heat vent flow rates are insufficient to realize any benefit. ... It is well known that vent flow rate is reduced at temperatures below 200/C (392/F) [Hinkley 1995] and that sprinklers can cause cooling of upper layer smoke to well below this level. For example, in sprinklered fires, it would not be unreasonable for smoke layer temperatures to be 70/C (158/F). At such a temperature, the theoretical flow rate relative to the maximum possible high temperature flow rate would be halved."

"The final negative claim that smoke and heat vents are not cost effective has never been seriously studied. Any such study would need to consider the cost of installation, the energy/lighting savings that may be realized through natural lighting and the reduction in heat, smoke, and fire damage that results from the use of vents. While the first two are reasonably well known, the latter has not been studied in any investigation reported in the fire literature. As such, this claim has no clear basis and must be regarded as mere speculation."

"First, it is clear that the current focus on assuring that vent operation is delayed has an adverse effect on [vent] system performance. It is important that design attention be paid to causing vents to operate more rapidly and in greater numbers. ... Second, it has been noted that draft curtains represent obstructions and should be dealt with in sprinkler design as obstructions. Draft curtains should be provided in the center of aisles and not directly over the storage. Dealing with these issues will improve fire protection design."

In order to understand the interaction between sprinklers, roof vents and draft curtains, it is first necessary to understand the mechanisms of how sprinklers control and extinguish a fire. The following description of how sprinklers work is contained in an article titled "Meeting the Challenges of an Ever-Changing Storage Industry," written by James Golinveaux and Joe Hankins of Tyco Fire Protection. The article appeared in the Winter 2006 issue of Fire Protection Engineering magazine.

"Cooling takes place at the roof/ceiling, where relatively small drops are lifted by the fire plume and cool the gas layer at the ceiling. This has the positive effect of preventing collapse of the building structure and sprinkler piping but also can delay operation of adjacent sprinklers (commonly known as "skipping")."

"Prewetting takes place away from the actual fire area, where discharge from sprinklers falls onto unburned combustibles, preventing ignition."

"Penetration of the fire plume by water is the only one of the three mechanisms that actually reduces the heat release rate (HRR) of a fire and, if sufficient, can completely extinguish a fire. Penetration is a function of the momentum of water discharge from sprinklers and drop size, as well as the intensity of the fire plume."

"Control mode (CM) [standard] sprinklers are designed to rely on cooling and prewetting, allowing the fire to continue to burn in the area of ignition while controlling roof/ceiling temperatures and preventing fire spread until firefighters arrive or until the fire burns itself out. Control mode sprinkler protection is characterized by a relatively large area of sprinkler operation (15 - 50 sprinklers) [in storage occupancies]."

In addition to understanding the mechanisms of how sprinklers control and extinguish a fire, it is also helpful to understand the capabilities of sprinklers with respect to controlling fires in storage occupancies. The following excerpts are from explanatory material regarding the requirements for sprinkler systems protecting storage racks contained in NFPA 13:

"Sprinkler protection installed as required in this standard is expected to protect the building occupancy without supplemental fire department activity. Fires that occur in rack storage occupancies are likely to be controlled within the limits outlined in B-1.1, since no significant building damage is expected. The first fire department pumper arriving at a rack storage-type fire should connect immediately to the sprinkler siamese fire department connection and start pumping operations."

"During the testing program, the installed automatic extinguishing system was capable of controlling the fire and reducing all temperatures to ambient within 30 minutes of ignition. Ventilation operations and mop-up were not started until this point."

An understanding of firefighting tactics for large buildings is also useful in determining the need for roof vents and draft curtains. A recent publication issued by the National Institute for Occupational Safety and Health (NIOSH), titled "Preventing Injuries and Deaths of Fire Fighters due to Truss System Failures" (NIOSH 2005-132), dated April 2005, contains the following excerpts regarding firefighting tactics:

"Steel trusses are also prone to failure under fire conditions and may fail in less time than a wooden truss under the same conditions."

"The number of fire-fighter fatalities related to structural collapse could be significantly reduced through proper education and information concerning truss construction. Fire fighters should be discouraged from risking their lives solely for property protection activities."

"Lives will continue to be lost unless fire departments make appropriate fundamental changes in fire-fighting tactics involving trusses."

"NIOSH recommends that fire departments, fire fighters, building owners and managers take steps to minimize the risk of injury and death to fire fighters during fire fighting operations involving structures with truss roof and floor systems."

"Use defensive strategies whenever trusses have been exposed to fire or whenever structural integrity cannot be verified. Unless life-saving operations are under way, evacuate fire fighters and use an exterior attack [Brannigan 1999; Dunn 2001]."

Since most large single-story industrial and storage buildings utilize the unlimited area building provisions contained in the IBC, these buildings are normally constructed utilizing unprotected noncombustible (Type 2B) construction. (The term "unprotected" refers to the lack of protection of the structural members by fire resistive materials, not to whether or not the building is protected by a sprinkler system.) Given this, large industrial and storage buildings are typically constructed utilizing steel trusses to support the roof construction. Hence, the NIOSH recommendations would be applicable to most buildings where roof vents are required by the IBC/IFC.

A discussion of these excerpts will appear in the March issue of Plumbing Engineer.

Richard Schulte is a 1976 graduate of the fire protection engineering program at the Illinois Institute of Technology. After working in various positions within the fire protection field, he formed Schulte & Associates in 1988. His consulting experience includes work on the Sears Tower and numerous other notable structures. He has also acted as an expert witness in the litigation involving the fire at the New Orleans Distribution Center. He can be contacted by sending e-mail to rschulte@plumbingengineer.com.