Automatic sprinklers have come a long way since Henry H. Parmalee introduced the first practical automatic sprinkler in 1874.1 Parmalee's sprinkler led to a major step in the advancement of industrial fire protection. Prior to the introduction of his sprinkler, sprinkler systems typically consisted of steel pipes equipped with perforated holes through which water would flow, similar to today's deluge-type sprinkler system. Roughly seven years after the introduction of the Parmalee sprinkler, Frederick Grinnell began modifications to the sprinkler that allowed for it to be more effective and produced at a lower cost.

With the advent of the automatic sprinkler system came guidelines for sprinkler system installations as well as guidelines for sprinkler system designs. On November 16, 1891, the Associated Factory Mutual Insurance Company (now known as FM Global) released the first automatic sprinkler system installation guideline entitled, Location and Spacing for Automatic Sprinklers. The design for automatic sprinkler systems became rooted on a pipe schedule basis where the size of the sprinkler system piping was based on the number of sprinklers located downstream of the pipe. The pipe schedule method was divided into three categories: light hazard, ordinary hazard and extra hazard pipe schedule. Based on the anticipated hazard of the occupancy to be protected, the size of the sprinkler system pipe was then determined by the number of sprinklers that were installed downstream of it.

The pipe schedule design method offered a simple means for determining the proper size of a sprinkler system. It, however, did not take into account the water supply available for the automatic sprinkler system, nor did it allow for flexibility when the occupancy hazard protected by an existing sprinkler system increased. Even with these drawbacks to the pipe schedule design method, automatic sprinkler systems prior to the 1950s did an excellent job of keeping fires under control until the local fire service was able to arrive and manually extinguish the blaze. This was due in large part to the combustible loading of the stored materials being relatively low, coupled with the relatively low storage and ceiling heights maintained in warehouse areas.

However, at the start of the 1950s, changes in industrial practices demonstrated the limitations of the pipe schedule design method. At this time came (1) an increased use of steel supported building structures, (2) the invention of the fork-lift truck and (3) an increased use of plastic materials.

Although the use of steel allowed buildings to be built higher than before, steel weakens at elevated temperatures. Since industrial fires can exceed these elevated temperatures, they create a condition where a building structure could possibly collapse due to the failure of a steel column even when automatic sprinkler protection was provided at ceiling level.

The invention of the fork-lift truck allowed storage height to be dramatically increased, which prior to the 1950s was only about 6 to 8 ft (2.0 to 2.4 m) high, or as high as a person could lift the stored item. In addition, most commodities maintained in storage areas prior to this timeframe consisted of ordinary combustibles, such as materials made from metal, glass or wood. The introduction of plastic materials increased the fire hazard within industrial facilities as the heat of combustion is two to three times higher than ordinary combustibles.2

To account for these changes, research conducted at FM Global in the 1950s3 led to two major changes in fire protection. The first major change was the introduction of the standard spray automatic sprinkler, which modified the sprinkler deflector to discharge nearly all of the water towards floor level in a parabolic shape. The second major change was the introduction of the density/area design concept. This concept identified a specific flow rate per sprinkler for all sprinklers operating within an indicated area. Unlike the pipe schedule design method, the density/area design concept required the water supply to be evaluated to verify that it could provide the necessary flow and pressure for the required design.

Although the design/area design concept worked well, testing at FM Global in the 1960s and 1970s demonstrated that the sprinkler technology at that time was not very effective for storage-type occupancies. As a result, research was initiated at FM Global in the 1970s to develop a sprinkler specifically intended for the protection of storage. This research led to the development of the "large-drop" sprinkler. This advancement in sprinkler performance also led to a new design format, one based on a specified minimum operating pressure at the most remote sprinkler while simultaneously opening an indicated number of sprinklers. A decade later, FM Global used the knowledge gained from the large-drop sprinkler program, coupled with another project from the 1970s that helped develop the quick-response thermal element, to develop the sprinkler concept that would eventually lead to the development of the "early suppression fast response" sprinkler, or ESFR for short. The design format for the ESFR sprinkler was also based on the same design format used with the large-drop sprinkler.

By the start of the 21st century, sprinklers were commercially available in various K-factor sizes, orientations, nominal temperature ratings, RTI ratings, finishes and spacing coverage. They had been grouped into three categories, known today by the terms "control mode density area" (CMDA), "control mode specific application" (CMSA) and "suppression mode" (formerly called ESFR) sprinklers. The first two categories group sprinklers by an assumed performance during a fire event (i.e., control of a fire) where as suppression mode sprinklers are assumed to suppress any fire that they protect. The assumed suppression performance allows for a reduced number of sprinklers in the design area (typically 12 sprinklers) as well as a reduced hose stream allowance (250 gpm [950 Lpm]) and sprinkler system duration (1 hour). The CMDA sprinklers differ in design format as they utilize the density/area design format whereas both the CMSA and suppression mode sprinklers use the number of sprinklers at a given minimum pressure design format.

Automatic sprinkler protection is the best line of defense against fire within an industrial facility; however, since the release of the first installation and design guidelines back in 1891, the requirements for automatic sprinklers have become much more complex. Prior to 2010, FM Global's installation guidelines for automatic sprinklers were provided in the following three data sheets: Data Sheet 2-2, Installation Rules for Suppression Mode Automatic Sprinklers,4 Data Sheet 2-7, Installation Rules for Sprinkler Systems Using Control Mode Specific Application (CMSA) Ceiling Sprinklers for Storage Applications,5 and Data Sheet 2-8N, NFPA 13 Standard for the Installation of Sprinkler Systems, 1996 Edition,6 encompassing a total of 344 pages. In addition, the design guidelines for typical warehouse occupancies, covered in FM Global Data Sheet 8-9, Storage of Class 1, 2, 3, 4 and Plastic Commodities,7 consisted of 123 pages.

To help reduce the complexity of automatic sprinkler installation and design, FM Global established a new method of classifying automatic sprinklers in 2010. Instead of categorizing sprinklers based on an assumed performance during a fire event, such as control mode or suppression mode sprinklers, FM Global now categorizes sprinklers based on intended application using the terms "storage sprinklers," "non-storage sprinklers" and "special protection sprinklers." The intended application of storage sprinklers is for protection of storage-type occupancies as well as other high heat release occupancies. The intended application of non-storage sprinklers is for the protection of non-storage occupancies, such as offices as well as manufacturing or other moderate heat release rate occupancies. The intended application of special protection sprinklers is for the protection of occupancies not generally covered by the other two categories.

This new method allows for a clearer understanding of the compatibility of the sprinklers with the occupancy they are to protect and allows for a single design format for all sprinklers. FM Global has chosen the number of sprinklers at a given minimum pressure design format to allow the design of a sprinkler system to be based on the actual performance of the chosen sprinkler as opposed to an assumed performance or, in the case of the density/area design method, the performance of the least efficient sprinkler.

Based on this new approach, FM Global has taken its three data sheets for sprinkler system installation and combined them into a single document entitled Data Sheet 2-0, Installation Guidelines for Automatic Sprinklers.8 The installation guidance provided within this new document addresses the specific requirements for storage sprinklers, non-storage sprinklers or special protection sprinklers, coupled with the installation guidelines that are common to all three types of sprinklers.

Also, as a result of this new approach, FM Global Data Sheet 8-9 now references the use of FM Approved storage sprinklers at ceiling level and when needed, as in-rack sprinklers. In addition, the ceiling-level designs offered in Data Sheet 8-9 are now based on five attributes associated with a sprinkler: (1) K-factor, (2) orientation, (3) response time index (RTI) Rating, (4) sprinkler spacing and (5) temperature rating.

 

Figure 1. Example protection table from FM Global Data Sheet 8-9, Storage of Class 1, 2, 3 4 and Plastic Commodities (c)2010 Factory Mutual Insurance Company. All rights reserved.

*Based on maximum water delivery time of 25 seconds **Based on maximum water delivery time of 20 seconds ***The protection options indicated in the protection table can be based on any ceiling height equal to or higher than the actual maximum ceiling height of the protected area. ****The protection options indicated in the protection table for upright sprinklers can also be used as an alternative option for pendent sprinklers having the same K-factor, RTI rating, nominal temperature rating and spacing requirements as the upright sprinkler. *****The design of 8 @ 40 (2.8) has a hose stream allowance of 250 gpm (950L/min) and a duration of 60 minutes when the maximum linear spacing is up to 12 ft (3.6 m); for linear spacing over 12 ft (3.6 m) the hose stream allowance is 500 gpm (1,900 L/min) and the duration is 120 minutes.

By going to the storage sprinkler concept, the design approach for sprinklers within Data Sheet 8-9 is now based on the number of sprinklers operating at a given minimum pressure. This means that the design approach of density/area has been eliminated from Data Sheet 8-9. To many in the fire protection community, this may appear illogical as sprinkler systems that have been installed using the density/area design format have performed very well since the design concept was introduced in the 1960s. However, FM Global feels there are limitations with this design approach. These limitations include: (1) current density/area protection tables must be based on the performance of the least effective sprinkler listed for the table, and (2) the ability of the most remote sprinkler's design pressure must be dependent on the installed sprinkler's spacing.

For the first point, consider two full-scale fire tests conducted at FM Global's Research Campus.9 The tests involved open-frame rack storage of cartoned expanded plastics to 15 ft (4.5 m) high under a 30 ft (9.1 m) high ceiling with an 8 ft (2.4 m) wide aisle provided between storage racks. The tests used CMDA standard-response K11.2 (K160) 160°F (70°C) nominally rated sprinklers on 10 x 10 ft (3.0 x 3.0 m) spacing. For the first test, this arrangement was protected using a 1.00 gpm/ft2 (40 mm/min) density with an upright sprinkler, whereas the second test was conducted using a 0.60 gpm/ft2 (24 mm/min) density with a pendent sprinkler. With the 1.00 gpm/ft2 (40 mm/min) density and an upright sprinkler, the test resulted in 32 sprinklers opening,3 whereas only 10 sprinklers opened using the 0.60 gpm/ft2 (24 mm/min) density with a pendent sprinkler.4 These two tests help to demonstrate that density is not a driving factor for sprinkler system design. In addition, using today's density/area design concept, the design for both sprinklers would be the same and would have to be based on the results of the K11.2 (K160) upright sprinkler, which had the poorer performance in this particular test.

The test comparison outlined above is representative of many of the tests that FM Global has conducted over the decades when comparing various sprinkler attributes. In general, test results will differ when a sprinkler's K-factor, orientation, RTI rating and nominal temperature rating is changed. What tests over the past 40 years have demonstrated is that the amount of water that is discharged from ceiling-level sprinklers in terms of an applied density is not as important as the amount of water that actually reaches the fire area, which can be thought of as an actual delivered density (ADD). What helps increase the ADD during a fire event can be found in the aforementioned attributes of a sprinkler, namely orientation, K-factor, RTI rating, temperature rating and, in some degree, sprinkler spacing. Because of this, FM Global now uses these five attributes to define the protection required for storage arrangements handled by Data Sheet 8-9 using the number of sprinklers at a given minimum pressure design format. A protection table from Data Sheet 8-9 is shown in Figure 1.

In addition to being easy to read, this protection table actually replaces a total of 9 protection tables from the June 2009 version of Data Sheet 8-9, making it less complicated than prior versions.

By moving to the new categorization of Storage sprinklers, FM Global has created a new method by which the hose stream allowance and the duration of a sprinkler system is determined. FM Global now bases the hose stream allowance and the required duration of a sprinkler system, in general, on the number of sprinklers in the ceiling design chosen. Some of the protection options shown in Figure 1 are highlighted with a green background. What these highlighted options represent are options that require only a 250 gpm (950 Lpm) hose stream allowance and a duration of only 1 hour.

With these changes, coupled with anticipated future changes in both FM Global Data Sheets 2-0, 8-9 and other design-based data sheets, FM Global aims to provide the most effective installation and design protection options, which are intended to be simpler to understand, less costly to install, but a more sustainable choice.

Weston C. Baker Jr. is with FM Global.

References:

 

  1. Richardson, K. (Ed.) History of Fire Protection Engineering, National Fire Protection Association, Quincy, MA, 2003.
  2. Tewarson, A. "Generation of Heat and Gaseous, Liquids and Solid Products in Fires," SFPE Handbook of Fire Protection Engineering, National Fire Protection Association, Quincy, MA, 2008.
  3. Kung, H. C., "A Historical Perspective on the Evolution of Storage Sprinkler Design," Fire Protection Engineering, First Quarter, 2011.
  4. FM Global Property Loss Prevention Data Sheet 2-2, Installation Rules for Suppression Mode Automatic Sprinklers, FM Global, Norwood, MA, 2002.
  5. FM Global Property Loss Prevention Data Sheet 2-7, Installation Rules for Sprinkler Systems Using Control Mode Specific Application (CMSA) Ceiling Sprinklers for Storage Applications, FM Global, Norwood, MA, 2005.
  6. FM Global Property Loss Prevention Data Sheet 2-8N, NFPA 13 Standard for the Installation of Sprinkler Systems, 1996 Edition, FM Global, Norwood, MA, 2004.
  7. FM Global Property Loss Prevention Data Sheet 8-9, Storage of Class 1, 2, 3, 4 and Plastic Commodities, FM Global, Norwood, MA, 2009.
  8. FM Global Property Loss Prevention Data Sheet 2-0, Installation Guidelines for Automatic Sprinklers, FM Global, Norwood, MA, 2011.
  9. Sienkiewicz, S. "Comparison of Cartoned Standard Commodities in Large-Scale Fire Tests," FM Global, Norwood, MA, 2009.