SILENT SENTINELS UNDER FIRE
BY BILL MANNING
“A lot of hubbub,” said the manufacturer.
“We are surprised at these sudden inflammatory media stories,” said the industry association.
“All manufactured products have occasional problems,” said the other industry association.
“Failure does not mean failure to operate,” said the advocacy group.
Four documented failures of Omega brand automatic sprinklers to discharge near the area of origin in actual fires, plus the sprinklers` very poor track record in recent field and laboratory sample testing have the fire safety community concerned–and the fire sprinkler industry circling the wagons.
The “hubbub” began March 1995 in a Marriott Courtyard guestroom in Michigan: the fusible link gave way on an Omega sprinkler closest to the fire but it never discharged. It happened again that month in an upstate New York hospital. Marriott began replacing its Omega sprinkler heads. In March 1996, Factory Mutual, which never listed the Omega sprinklers in its approvals guide, issued a potential failure-to-operate warning and recommended against its insured customers installing them. About three months later, Omega`s manufacturer, the Central Sprinkler Corporation of Lansdale, Pennsylvania discontinued its line of Omegas and reissued it with a component replacement to solve the discharge problem.
But there were–and are–about eight million “old” Omegas in buildings across the U.S. A very popular head first marketed in 1983, its quick-response capability made it especially appealing for large residential and public facilities such as hotels, jails, hospitals, apartments, and schools.
In August 1996, Underwriters Laboratories, whose distinctive mark graces the Omega, announced that operating pressures measured for some samples of Omega sprinklers “…could exceed the water pressure available for the installation and affect the sprinkler`s intended operation.” UL tested 800 samples from around the country and found that 80 percent operated at 40 psi or less, while 69 percent operated at five psi, meaning 31 percent of the “old” Omegas tested did not meet UL`s own test standard requirement that sprinklers discharge at five psi head pressure.
In early 1997, Omega sprinklers failed to respond in fires in California and Indiana. Fire officials from several states and then the regional and national media reacted.
Field tests conducted by the Fairfax County (VA) Fire & Rescue Department`s Fire Prevention Division paint a grim picture. It tested a total of 696 Omega sprinklers from 59 buildings and found that 250 heads failed to meet the National Fire Protection Association standard of seven psi for sprinkler activation–a 36 percent failure rate. In 56 buildings for which tests were completed, 49 had at least one sprinkler head that failed to meet the NFPA standard. In one building, 30 of 38 sampled sprinklers failed to operate under the required pressure; in another, 13 of 26; in another, 12 of 24; and so on. Perhaps even more disturbing, many sprinklers in Fairfax County failed to operate at pressures of 80, 100, and 105 psi.
Fire Engineering has learned that tests on a small sample of Omegas from a single building elsewhere in the country showed failure to operate at pressures up to 500 psi.
Central Sprinkler and UL maintain that contaminants in the water–namely cutting oil used to assemble steel pipe systems and “stop-leak”-type pipe sealer used by unscrupulous or ignorant individuals to plug system leaks–cause a rubber O-ring in the head assembly to swell, inhibiting an internal plunger from dropping down under the force of required or even typical water pressures, thereby obstructing water flow.
When the story finally hit the newspapers, Central was quick to blame installers for the swelling problem. This piqued the angst of the American Fire Sprinkler Association (AFSA) and the National Fire Sprinkler Association (NFSA), both of which moved to defend the virtue of reputable sprinkler installers among their memberships. Indeed, cutting oils have been used in steel pipe fabrication since God made U.S. Steel, and while UL and other sources evidence the possible presence of “stop-leak” in some of the sample heads, it is difficult to imagine a party of “stop-leak” bandits conspiring from New York to Los Angeles.
Hydrocarbon-based cutting oils` swelling the rubber O-ring is the more plausible and frequent culprit. However, this cannot explain the fact that Omega failures have been found in plastic and copper pipe systems as well. On August 13, 1997, UL issued a bulletin urging that Omega sprinklers installed “in all types of piping systems” be tested. One explanation offered for failures in copper and plastic systems is that metals in the water react with the brass head in pre-1990 Omegas to produce green-colored copper deposits that plug up the works.
As yet, though, no one has dared utter the words “design failure.” Alas, we must leave the details and the heavy brain work to the engineers, but could it be that the sprinkler`s tapered barrel, O-ring, and plunger setup just isn`t, or wasn`t right for real field conditions–conditions that include commonplace cutting oils and system pressures far less than 80 or 100 or 200 psi?
Relief would come were we able to conclude simply: Test the things, get a new sprinkler with a silicone O-ring or a newly UL-listed retrofit kit from the manufacturer, and let our sprinklers across the fruited plain do their job as Frederick Grinnell intended. Still, there are troubling aspects to this mess.
Comments published by the sprinkler industry were shocking. Operation Life Safety, AFSA, and NFSA recognized problems with Omega (how could they not?) but under fire became life-safety spin doctors–shameless apologists for non-NFPA-compliant sprinklers, echoing the manufacturer`s contention that the critics were just purists bound to an archaic standard. They said, in effect, “It`s really not so bad. `Failure` is not really failure. It`s okay not to achieve seven psi activation pressure because most of the time the system pressures are higher than that. The standard`s not that important.” Can they be serious when they protest, “Hey! Eighty percent of the sprinklers tested activated at normal system pressures!”–leaving, if one were to extrapolate the UL test figures, some 1.5 million heads in serious doubt, anyway? Have we gotten to the point in this crazy, mixed-up world where “almost” counts with automatic fire sprinklers, as it does with horseshoes and hand grenades?
We must not–not if we want automatic fire sprinklers to remain a fabulous fire safety success story and the true salvation of America`s fire problem. The vast majority work as intended and save lives despite a chain of human variables, from design to discharge. In large part, that`s precisely because standards are followed and tolerance for error is low.
Still, the Omega case has painfully revealed what can happen when there`s a breakdown at the front end of the process. It has shown that when you need it most, there is no such thing as the sprinkler police. In theory, the AHJ has the ultimate authority over what sprinklers are placed where and how. In reality, the AHJ in many ways is at the mercy of the third-party agency, which is by no means a policeman but a service contractor that operates in a virtual vacuum.
UL may well scramble to regroup as it has before, but it must excuse fire service personnel if they`re too busy correcting a big mistake to swallow all that business about “the Mark” being “the defining moment in the fire inspector`s day.”
Now is definitely a good time for fire officials across America to ensure that their sprinklers of every brand and model are in working condition, so that the silent sentinels are on the fire, not the other way around.