Click here for a printable version of this article.

September 2003: Automatic Sprinkler Performance

Automatic sprinklers are the most effective fire protection in buildings overall available. Based on studies and loss experience, when sprinklers are present, the chances of dying in a fire and the average property loss per fire are both cut by one-half to two-thirds, compared to fires where sprinklers are not present. This basic comparison even understates the value of sprinklers because it lumps together all sprinklers, regardless of type, coverage, or operational status, and is only limited to fires reported to fire departments. If unreported fires were included and if complete (not partially protected facilities), well maintained and properly installed/designed systems (according to NFPA and recognized standards),) could be isolated, sprinkler effectiveness would undoubtedly be noted as even more impressive.

When sprinklers do not perform satisfactorily, which is rare, the reason is usually always due to some type of human error - improper maintenance, improper design for current occupancy/hazard, or failure to provide for other aspects of building fire protection.

What causes unsatisfactory sprinkler performance when it occurs?

NFPA discontinued its sprinkler performance tracking database after 1970 since the data collection was biased toward cases of poor sprinkler performance. This was due to the many unreported instances where fires were controlled or extinguished by automatic sprinklers.

The tables below summarize the causes of unsatisfactory sprinkler performance, based on the 45 years of cases reported to NFPA from 1925 to 1969.

The second table reorganizes these causes into major problem groups: a) failure to maintain the system in operational status (primarily because of closed valves), which accounts for more than half the unsatisfactory-performance cases; b) failure to assure that the system in place is complete and adequate for the current use of the property, which accounts for nearly one-fourth of the unsatisfactory-performance cases; c) defects in other parts of the building's system of fire protection; d) inadequate performance by components of the sprinkler system itself, which accounts for less than 6% of the unsatisfactory-performance cases; and e) other factors.

Leading Reasons When Sprinkler Performance Is Unsatisfactory

Problem
Percentage of Cases
Water shut off
35.4
System not adequate for level of hazard in occupancy
13.5
Inadequate water supplies
9.9
Inadequate maintenance
8.4
Obstruction to water distribution
8.2
System designed for partial protection only
8.1
Faulty building construction
6.0
Antiquated system
2.1
Slow operation
1.8
Defective dry-pipe valve
1.7
Exposure fire
1.7
System frozen
1.4
Other or unknown
1.9
Total
100.0

 

Groups of Leading Reasons for Unsatisfactory Sprinkler Performance

 

Problem Group
% of Cases
Problem
% of Cases
I. Failure to maintain operational status of system
53.4
Water shut off
35.4
Inadequate maintenance
8.4
Obstruction to water distribution
8.2
System frozen
1.4
II. Failure to assure adequacy of system
21.6
System not adequate for level of hazard in occupancy
13.5
System designed for partial protection only
8.1
III. Defects affecting but not involving sprinkler system
15.9
Inadequate water supply
9.9
Faulty building construction
6.0
IV. Inadequate performance by sprinkler system itself
5.6
Antiquated system
2.1
Slow operation of sprinklers
1.8
Defective dry-pipe valve
1.7
V. Other
3.6
Exposure Fire
1.7
Other or unknown
1.9
Total
100.0
Total
100.0

 

Tables Source: "Automatic Sprinkler Performance Tables, 1970 Edition," Fire Journal, July 1970, page 37. Based on 3,134 fires reported to NFPA during 1925 to 1969 for which sprinkler performance was deemed unsatisfactory. Of these, 75.0% were in industrial facilities, 12.0% were in storage facilities, 5.6% were in stores, and 7.4% were in all other properties.

The tables demonstrate three keys to obtaining full benefit from sprinklers and other automatic suppression systems. First and foremost, these systems need to be regularly maintained and tested. In particular, keeping the water turned on and all valves open is a top priority. Some of the most devastating fires (from an industrial, property loss standpoint) in history involved shut valves. Use of a tag/permit system (see our sample Fire Protection Impairment Tag) to provide visual reminders of the on/off status of the system and to alert appropriate personnel of the system impairment is a good way to prevent inappropriate and dangerous water shutoff conditions. Valve supervision using a tamper switch connected to a central alarm monitoring station can also be helpful (valve tamper alarms should also be tested regularly). Some of the recommended system maintenance is highlighted in our March 2002 "Back to Basics" web article.

Second, a sprinkler system needs to be designed to fit the current use/needs of a property. If the occupancy changes, it is essential to review the adequacy of the sprinkler system. Even if the property use has not changed, the passage of time alone can dictate a review of the system (i.e., code/standard changes, new test data, etc.). And no one should confuse a partial sprinkler system with complete sprinkler protection.

Third, even a well-maintained, properly designed, complete sprinkler system will not provide 100% protection in most cases. It requires the support of a well-considered integrated design for all the other elements of the building's fire protection. Unsatisfactory sprinkler performance can result from an inadequate water supply, faulty building construction, sprinkler obstructions, etc.

If you would like further information regarding sprinkler system performance, please contact Risk Logic Inc.