Is Historic Data Reliable for Assessing Sprinkler and Fire Alarm Performance?

Private protection should be carefully evaluated to see if it still has the loss reduction impact that it had in the past. Central station fire detection systems were routinely added to buildings for a credit in fire insurance premium. If you look at the current trends in fire behavior, by the time the alarm is activated, the central station notifies the public safety answering point and the fire department is dispatched the fire will flashover before the apparatus arrives. Perhaps the call routing should include the claims department?

The effectiveness of sprinklers in large, high piled storage environments should also be carefully evaluated. The history of automatic sprinkler systems trumpets the effectiveness of these systems. From the introduction of the modern sprinkler head in 1874 and its adoption by Factory Mutual, the insurance industry began the recordkeeping around these systems.

In the early days there was a single design standard for installation. As time evolved this developed into there being three designs for automatic sprinklers, light, ordinary and extra hazard. There were a limited number of sprinkler heads available and a limited number of temperature ratings. This set of design standards remained consistent until the 1980’s.

These sprinkler systems were intended to control a fire until the fire department arrived and applied hose streams to complete extinguishment. The highest water demand for a single sprinkler was around 40 gallons per minute (GPM). Hose stream allowances for extra hazard and storage systems was 500 GPM.

In the mid 1980’s the first storage protection standard was released, and sprinkler systems began to be specifically designed to protect high piled storage. These early standards included multiple requirements for in-rack sprinklers. This made sprinklers less attractive to warehouse operators due to water damage concerns.

The design criteria for storage were based on full scale fire tests run at Factory Mutual and Underwriters Laboratories.  These tested developed curves and adjustments that resulted in very specific design criteria. These systems had to be calculated to prove the design and water supply were adequate.

The use of computers to do the complex calculations allowed designers to use more piping layouts than before. Sprinkler systems became specific to the hazard. In the late 1980’s a new type of sprinkler was introduced, the large drop sprinkler. This was the first sprinkler to be used specifically for storage and allowed the omission of some in rack sprinklers. Large drop sprinklers were allowed until they were not. Later editions of NFPA 13 removed the designs criteria for large drop sprinklers for some storage arrays. Subsequent testing no longer supported the ability of these sprinklers to be effective. A quiet withdrawal of previously approved protection. What happened to buildings that had these orphan systems?

The next generation of sprinklers were Early Suppression Fast Response (ESFR) sprinklers. These sprinklers had larger orifices and were designed to suppress fires versus controlling them. For the next thirty years ESFR became the primary sprinkler of choice for storage exposures. The size of the sprinkler orifice increased and the amount of water per sprinkler head has increased to over 195 GPM per sprinkler for some storage designs. As these systems intended to suppress the fire, hose stream allowances were cut to 250 GPM.

ESFR was the answer until the 2022 edition of NFPA 13 was released and the design criteria for certain storage arrays no longer included the original K-14 and early adopted K16.8 sprinklers. New testing indicated that the designs for these sprinklers were inadequate for the hazards. The approvals were quietly withdrawn. What happened to the risks who installed these systems in the prior periods? Codes are generally not retroactive.

NFPA 13 now had to address the very specific installation rules for these new sprinkler heads. My 1987 edition of NFPA 13 was 112 pages including index and appendices. The 2022 edition had grown to 619 pages. Sprinkler systems have become exponentially more complex. The number of conditions that affect performance of these engineered system is long and complex.

There have been some famous fires where a fully protected warehouse with ESFR sprinklers burned down. The Plainfield, IN Wal Mart fire in 2022 that destroyed a 1.2 million square foot warehouse. Litigation is underway around this loss.

At the risk of being kicked out of the HPR club, I contend that sprinkler protection has gotten fragile and can be adversely affected by very common conditions. The fire tests that are used to “prove” these systems work is done under ideal conditions. Real life is not that pristine. The focus should be on installation issues, storage practices that impact sprinkler water delivery and the training of fire departments on how to respond to these mega buildings. Sprinkler systems designs that exceed the height of the test facilities should be viewed with a healthy dose of skepticism. In my career I described these systems as SWAG, a scientific wild ass guess. 

You cannot rely on the rearview mirror with about 150 years of sprinkler experience. For many installations since the arrival of the designed system, you should be asking how these systems have performed. I am not sure there is a reliable answer to that question.