In approximately half of the United States, the National Fire Incident Reporting System gathers data about significant fire incidents, particularly when fatalities occur. This information is reported to the National Fire Protection Association and/or the U.S. Fire Administration. After a serious fire, the first questions likely to be asked are, "Were sprinklers present?" "Did the sprinkler system operate satisfactorily?" Or, "Did they extinguish the fire?" The NFIRS forms provided by the U.S. Fire Administration contain spaces for answers to these questions.
However, there is no location on any form that asks investigators to answer other important questions, such as "Did penetration seals, dampers, fire doors, fire-rated walls, floors, ceiling materials or other passive features perform satisfactorily?"
Occasionally, an investigator may provide a narrative that addresses some of these issues, if they are known by that particular investigator to have contributed to (or to have prevented) fire or smoke spread. The NFPA and the U.S. Fire Administration have no means to incorporate such important details in the fire incident database.
Likewise, the National Fire Protection Association has collected its own information on fire incidents for decades, and this NFPA data is often cited to justify sprinkler trade-offs. However, NFPA discontinued the tracking of unsatisfactory sprinkler performance in its database around 1970. At that time, the NFPA determined that it was more valuable to collect information measuring fire loss. That is, they decided to record property damage and details related to loss of life in fires, but not sprinkler performance. Until about 1970, the NFPA measured sprinkler effectiveness by looking at the percentage of fires in sprinklered properties that showed satisfactory sprinkler performance.
Investigations suspendedHowever, the association decided to discontinue monitoring this statistic because information on fewer and fewer small or medium size fires could be captured, and NFPA felt that this created a bias toward cases of poor sprinkler performance. Such cases produced larger fires that were more likely to require fire department intervention or insurance company attention, and were more likely to be reported. Investigators felt that such data would cause some to believe that sprinkler effectiveness was declining. As a result, sprinkler performance is no longer monitored and reported by the NFPA.
While the negative bias argument has merit, the decision to suspend investigation has left a 30-year gap in NFPA's statistical fire record on the history of sprinkler performance. The effects of maintenance (or lack of), human errors, design flaws, the relationship between active and passive protection, aging systems, corrosion and other important factors would have been useful in determining equivalents in justifying sprinkler trade-offs. It may have also inadvertently created a false sense of security due to the conspicuous lack of information on sprinkler system malfunction.
According to the NFPA, data collected from 1925 through 1969 shows that the major causes for unsatisfactory sprinkler performance have been failure to maintain the system in operational status (closed valves). Human error accounts for more than half the cases of unsatisfactory sprinkler performance. Secondly, systems fail to meet expectations when building owners neglect to assure that the system in place is complete and adequate for the current use of the property. This accounts for nearly one-fourth of the instances of unsatisfactory performance.
Historically (through the reporting period ending in 1969), inadequate sprinkler system performance accounted for less than 6% of failures. However, recent instances of sprinkler head malfunction and component recalls may cause some to question that low figure.
The NFPA report quoted numerous conditions, and concluded that during the years of 1988 to 1997, sprinkler systems are highly effective, reducing the chance of death and property loss by 1/2 to 1/3 compared to structures with no sprinklers. However, it is important to note that the average age of a building in the U.S. is approximately 30 years.
It may be assumed that many of the buildings that were included in NFPA data were built during times when passive systems were also present. The presence of sprinklers, combined with compartmentalization, contributed to a very impressive record. It is of concern then, that many of the critical passive features that contributed to that record are now being traded off in the ICC.
Clearly, a well-maintained, properly installed suppression system is a key element in fire safety and property protection. However, when sprinklers do not deliver the desired result, the NFPA report explains, it is due to partial coverage, antiquated installations, and to systems that are poorly maintained or have been inadvertently disabled. In cases of explosion or flash fires, a sprinkler system might be overpowered, and fires occurring close to people or sensitive property may do significant damage despite the activation of a sprinkler system.
Underwriter's Laboratories Inc. (UL) has investigated and listed automatic sprinklers since the early 1900s. The UL 199 Standard (Automatic Sprinklers for Fire-Protection Service) and UL 1626 (Standard for Residential Sprinklers for Fire-Protection Service) pertain to testing of commercial and residential automatic sprinklers for fire protection service. Automatic sprinkler manufacturers must meet applicable test requirements in the Standards in order to apply the UL mark to their products.
On Oct. 14, 1998, the U.S. Consumer Product Safety Commission (CPSC) and a major sprinkler system supplier announced a nationwide recall of approximately 8.4 million units manufactured since 1982 because the CPSC alleged that such sprinklers are defective and could likely fail in a fire.
This major recall of sprinkler components, combined with the voluntary recall of others, serves to dramatize the vulnerability of sprinkling systems to malfunction or human error, and the fallacy of depending on this particular approach to fire protection to the exclusion of a more balanced program of active and passive components.