In October 1996, Dave Thomas, P.E., the head of the Plans Review Section of the Fairfax County (VA) Fire and Rescue Department`s Fire Prevention Division, noticed an article in a sprinkler trade journal indicating that Underwriter Laboratories Inc. (UL) was investigating reports that higher than expected water pressure had been required to operate some of Central Sprinkler Corporation`s Omega series of sprinklers. After initial investigations, it was discovered that the sprinklers had failed to operate at a fire in a Michigan hotel. Thomas believed that this was a potentially serious problem that warranted further investigation. Thomas; the head of our Testing Section, Captain Francis Teevan; and I met to discuss how to proceed. We began by calling the manufacturer to arrange a meeting.


We met at Central`s offices in Lansdale, Pennsylvania. The manufacturer`s representatives explained the problem as they saw it. They believed that the high operating pressures were caused by the use of “Stop-Leak”-type products in the sprinkler systems or excessive cutting oil in the sprinkler water. They also indicated that only steel pipe systems were involved. After the meeting, it became apparent to our representatives that we had a serious problem and that we had to identify the locations of the Omega heads in Fairfax County.


Our primary concern was that sprinkler systems installed in buildings in our jurisdiction may not function as designed. Since most buildings equipped with sprinklers use them to reduce other fire protection requirements, it is imperative that these systems work as designed. The BOCA National Building Code, used in Virginia, offers incentives for installing sprinkler systems in buildings. Among the incentives is allowing the exit access travel to be lengthened from 50 to 100 feet farther than it could be in a nonsprinklered building and to lower the corridor fire resistance rating from one hour to zero hours.

These trade-offs can make a fire in a fully sprinklered building where the sprinklers fail to control the fire much more hazardous to occupants and firefighters than they would be if the building had not been sprinklered. The increase in travel distance also increases the distance required to advance hoselines into the building, and the lower rating of the corridor walls may make the corridor untenable more rapidly than would otherwise be expected.

Fairfax County Fire and Rescue Department`s Fire Prevention Division has a long history of aggressive enforcement of the fire prevention and building codes. Our Testing Section performs acceptance testing on all new fire protection systems installed in Fairfax County, and our program for witnessing the retesting of fire protection systems in existing buildings is one of the most comprehensive in the nation. We annually witness the retesting of fire protection systems in all of the high-rise buildings, other large structures, and child care facilities in the county under a program that allows us to charge a fee for our services. So it was natural that we would rapidly move to remedy what we recognized as a major problem with a fire protection system.

The first obstacle we faced was how to identify where the heads had been installed in our county. Thomas ran a list of all the buildings for which we had approved sprinkler plans since the head was listed by UL in 1983. We found that more than 2,000 plans had been approved.

We assigned personnel to review each plan and identify the type of head that had been installed in each structure. It soon became apparent that this would be an enormous task, since many buildings used multiple types of heads and tenant build-outs resulted in even more head variations.

Given the scale of the project, we prioritized buildings by occupancy type and the type of sprinkler pipe used. We soon discovered that the Omega heads had been used extensively throughout the county. They were located in several schools, libraries, hospitals, and board-and-care facilities. When we realized that the county owned many of these structures, we met with Acting County Executive Anthony Griffin to outline the depth of the problem and our plan to fix it. Griffin, in turn, informed our Board of Supervisors about the issue.

Fortunately, we were supported by the Fairfax County Board of Supervisors. It approved our request for six more inspectors and funded the purchase of vehicles and other required supplies for the project. The additional personnel were critical in allowing us to rapidly move forward in resolving the problem.


The first step of our plan involved removing and testing sprinkler heads from the 33 schools we had identified as having Omega heads. We contacted Central Sprinkler, which agreed to send a crew to remove and replace the sample heads. Since we were concerned that the scope and nature of the problem were not known, we decided to treat the removed heads as evidence. As each head was removed, it was capped in accordance with the manufacturer`s instruction, marked with a number, and boxed. Following the rules of evidence, we maintained the chain of custody for all removed heads.

The sample heads were then divided into two lots. One half of the heads went to Central Sprinkler`s testing laboratory in Lansdale. The 7 psi operating pressure specified for quick-response sprinklers in the UL standard was used for these tests. We sent personnel along to witness the tests and record the results. These tests showed that 29 of the 33 schools had heads that operated at pressures higher than 7 psi, some much higher.

Given the serious implications of test failures, it was important that an independent third-party testing lab conduct the tests. To ensure impartiality, samples were tested by Factory Mutual at its testing laboratory in West Greenwich, Rhode Island. These tests were also witnessed by members of the Fairfax County Fire Prevention Division. The failure rate and pressures were similar to the results we had received from Central`s labs. Further testing was also accomplished at the University of Maryland and at Hughes Associates in Baltimore, Maryland. In each case, the test results showed failure rates similar to those in the Central tests.

Analysis showed that at least one sprinkler operated at pressures in excess of 40 psi in 51 percent of the buildings tested and that operating pressures exceeded 30 psi in 60 percent of the buildings (see box). Five of the test buildings used plastic pipe to supply the sprinklers. These results showed that two of the five buildings had operating pressures in excess of 7 psi and that the buildings that had failures all exceeded a 40 psi operating pressure. The only copper pipe building we tested also had head operating pressures in excess of 40 psi.

Several more buildings in Fairfax County have been tested as part of the remediation program. However, because we did not witness the testing, they are not included in our database. An analysis of these tests has shown that their results closely parallel our results.

After these tests were completed, Central Sprinkler began replacing heads in county-owned buildings. This replacement process is still ongoing.


Because we knew we had a problem in Fairfax County, we were obligated under the Fire Prevention Code to take action to fix it. A careful review of the code failed to show anything in the code or in the referenced standards that allowed us to directly order property owners to have their sprinkler heads tested. The code section we used requires the proper maintenance of water-based fire protection systems. We believe that the testing we had performed conclusively proved that failure to test any system equipped with Omega heads would be tantamount to improper maintenance of the system.

The test samples are removed using the criteria found in NFPA 25, Inspection, Testing, and Maintenance of Water-Based Fire Protection Systems. This standard also stipulates that if one head in a system fails, all the heads in that system must be replaced.

To date in Fairfax County, we have found Omega heads in 302 buildings using about 50,000 heads in all. When buildings with Omega sprinklers are discovered, we issue a Notice of Violation (NOV) to the owner-manager ordering that the heads be tested. On completion of the tests, the results must be sent to the Fire Prevention Division so we can order the proper course of action for resolving the problem. In some of the tested properties, the operating pressures were so high that we ordered the property management to institute a fire watch until the heads were replaced. Where no failures occurred, no further action was required. High operating pressures have been associated with heads removed from systems that used steel, copper, and plastic piping.

We found it important that our staff issuing the NOVs fully understand the issue and be prepared to explain the problem to building owners and managers. In many cases, on-site managers failed to fully appreciate the severity of the problem and attempted to deal with it at their level. When building managers are issued the NOVs, we make it a point to tell them that the top management of their company should become involved as soon as possible.

We have found that building managers and owners need the support of the Fire Prevention staff in resolving this issue. This is a technical problem, and someone who does not understand fire protection systems needs all the assistance we can offer.


We used the UL 199 standard of 7 psi for testing the heads for several reasons. The Omega was originally listed under this standard and, most importantly, some systems are designed to this pressure. We do not have the expertise or authority to establish test standards. While the static pressure will be above 7 psi on every system, the most important pressure is the residual pressure for which the system was designed.

Sprinkler piping is the same as a hose layout in that it has friction loss. The overall pressure loss due to friction will increase with the distance the water flows if the pipe size remains constant. A bid-conscious sprinkler designer will use the smallest pipe possible to deliver the water at the appropriate pressure and volume to the most remote sprinkler head in the system. Since many fires are not controlled by one head, the argument that high-static pressure will overcome a high operating pressure is faulty. The sprinkler head must be able to operate at the residual flow pressure if more than one head is needed to control the fire.

Central Sprinkler has been invaluable throughout this process. On request, it has worked with building owners to test heads, and it has replaced problem heads as needed. I strongly urge any authority having jurisdiction finding Omega heads in its locality to contact Central at (800) 523-0700 for additional information.

Each fire department needs to fully evaluate this problem in its own jurisdiction. The safety of our firefighters and the citizens they protect depend on the proper operation of all installed fire protection systems. n

This article and the investigation it details were made possible by the hard work of the Fairfax County Fire and Rescue Department`s Fire Prevention Division and especially Deputy Chief Stephen Smith, Captain II Francis Teevan, Captain I Geoff McNamara, and Dave Thomas, P.E.

Click here to enlarge image

CARL A. MAURICE is a 27-year veteran of the Fairfax County (VA) Fire and Rescue Department. A battalion chief for 10 years, he spent most of his career in operations until he was assigned to the Fire Marshal`s Office in 1996.

No posts to display