SUPPLEMENT 4

Residential Sprinkler Systems

John L. Bryan

Editor’s Note: Supplement 4 is a reprint of Chapter 10 in the fourth edition of NFPA’s book, Automatic Sprinkler and Standpipe Systems. This material provides a historical perspective of the development of residential sprinklers and standards. This supplement provides details of specific residential fire tests used in the development of the initial residential sprinkler technology and reviews residential sprinkler incentives, tradeoffs, activations, and code adoptions across the United States. This supplement also provides an overview of NFPA residential sprinkler requirements and applications.

The first edition of NFPA 13D, Standard for the Installation SPRINKLER SYSTEMS FOR DWELLINGS of Sprinkler Systems in One- and Two-Family Dwellings Fire Death and Injury Data and Manufactured , issued in 1975, exempted cer- tain portions of dwellings from sprinkler coverage based on More than four of every five dwelling fire fatalities (ap- the established fire hazard record. The 1980 13D standard proximately 86 percent between 1986 and 1990) occurred radically changed sprinkler system standards and sprinkler in fires that began in the normal-use areas of a residence: technology by focusing the attention on the ineffectiveness living , , , , , or an of the standard sprinkler for protecting the lives of occu- area typically found in the such as laundry or pants in dwellings. For the first time a sprinkler standard heating areas (NFPA 13D, 2002). Karter reported a total required a sprinkler designed to protect the occupant in of 3190 civilian fire deaths in the in 2004, approxi- the room of fire origin with a fast activation and a unique mately 82 percent of the total civilian fire deaths of 3900 water distribution. The development of the residential individuals.1 The National Association sprinkler resulted in subsequent refinement and improve- classifies ‘‘homes’’ as one- and two-family dwellings and ment of the quick-response sprinkler and the development apartment occupancies. Table S4.1 presents civilian fire of the quick-response extended coverage sprinkler and of deaths and injuries for 2004, with percentage variations the early suppression fast response (ESFR) sprinkler. from 2003. The 1980 edition of NFPA 13D was the first effective Hall2 reviewed the fire incident and loss record for design guide for a residential life safety sprinkler system. manufactured homes for the period from 1980 to 2002. This edition of NFPA 13D brought about changes to NFPA Table S4.2 illustrates the fire incidents, civilian fire deaths, 13 relative to protecting dwellings in occupancies other and injuries with the direct fire loss from manufactured than one- and two-family dwellings and manufactured home fires for the years 1980 to 2002. Table S4.3 compares homes and resulted in the development and issuance of civilian fire deaths and injuries in manufactured homes NFPA 13R, Standard for the Installation of Sprinkler Sys- and conventional one- and two-family dwellings for the tems in Residential Occupancies up to and Including Four years 1999 to 2002. indicated that although the fire Stories in Height, in 1989. NFPA 13R provides a sprinkler incident and fire injury rate is lower in manufactured system designed to protect occupants of low-rise multiple homes, the fire death rate is higher, when examined by the dwelling-unit occupancies. number of housing units, as indicated in Table S4.3.

223 224 Supplement 4 • Residential Sprinkler Systems

TABLE S4.1 Estimates of 2004 U.S. Civilian Fire Deaths and Injuries by Property Use

Civilian Deaths Civilian Injuries Percent Change Percent Change from Percent of All from Percent of All Property Use Estimate 2002 to 2003 Civilian Deaths Estimate 2002 to 2003 Civilian Injuries

Residential (total) 3,225 1.9 82.7 14,175 0.7 79.3 One- and two-family dwellings1 2,680 2.0 68.7 10,500 5.0 58.7 Apartments 510 24.4 13.1 3,200 12.3 17.9 Other residential2 35 75.0 0.9 475 11.8 2.7 Nonresidential structures3 804 63.64* 2.1 1,350 11.5 7.5 Highway vehicles 520 14.3 13.3 1,300 7.1 7.3 Other vehicles5 30 50.0 0.7 200 0. 1.1 All others6 45 30.7 1.2 850 8.1 4.8 Total 3,900 0.6 17,785 1.4

Note: Estimates are based on data reported to NFPA by fire departments that responded to the 2004 National Fire Experience Survey. Note that most changes were not statistically significant; considerable year-to-year fluctuation is to be expected for many of these totals because of their small size. *Statistically significant at the .05 level. 1Includes manufactured homes. 2Includes hotels and motels, college , boarding , etc. 3Includes public assembly, educational, institutional, store and office, industry, utility, storage, and special structure properties. 4This decrease reflects 100 fire deaths in the Station Nightclub fire in Rhode Island, and 31 deaths in two nursing home fires in Connecticut and Tennessee that occurred in 2003. 5Includes trains, boats, ships, aircraft, farm vehicles, and construction vehicles. 6Includes outside properties with value, as well as brush, rubbish, and other outside locations. Source: Michael J. Karter, Jr., ‘‘United States Fire Loss for 2004,’’ NFPA Journal, Vol. 99, No. 5, 2005, pp. 49.

Most fatal dwelling fires occur at night, when people occurred in the kitchen or bedroom. The and tend to be asleep. In addition, most victims of dwelling accounted for an additional 11 percent of total fires were not in the room of fire origin; fire spread out sprinkler activations. of that room, typically as a result of reaching flashover. Thus, fatal dwelling fires, which account for the largest Dwelling Sprinkler System Standard share of total U.S. fire deaths, pose certain unique opportu- nities and challenges for sprinkler system design. Specifi- In 1973 the NFPA Committee on Automatic Sprinklers cally, a system that covers normal-use areas and prevents appointed a subcommittee to develop a standard for the flashover prevents most fires from reaching the stage at installation of sprinkler systems in one- and two-family which most fatal injuries occur. Many victims are so vulner- dwellings. Basic design objectives for the dwelling sprin- able—not only asleep but often elderly, preschoolers, hand- kler system were to provide: icapped, or impaired by alcohol or drugs—that anything • A system that allows occupants sufficient time to en- less than automatic suppression may not be enough to save sure their survival them. • An inexpensive system Operation Life Safety 1995 Annual Residential Sprin- kler Activation Report provides data on 551 residential The NFPA Committee on Automatic Sprinklers devel- sprinkler activations with NFPA 13D or 13R systems from oped the initial 13D standard for the installation of sprinkler 1983 to July 31, 1995.3 This activation data indicates that systems in one- and two-family dwellings in 1975. NFPA approximately 50 percent of the total sprinkler activations 13D recommended that a sprinkler system provide a dis-

2007 Automatic Sprinkler Systems for Residential Occupancies Handbook Supplement 4 • Residential Sprinkler Systems 225

TABLE S4.2 Manufactured Home Fires in the United States, 1980 to 2002, Reported to Fire Departments

Direct Property Damage (in millions) Civilian Civilian Current In 2002 Year Fires Deaths Injuries Dollars Dollars

1980 29,700 410 860 $135 $295 1981 27,100 530 850 $142 $280 1982 28,300 460 1,020 $145 $270 1983 26,300 480 870 $167 $301 1984 26,000 380 820 $193* $334* 1985 25,800 550 860 $174 $291 1986 25,400 410 840 $169 $278 1987 22,900 450 800 $140 $221 1988 23,600 510 960 $157 $238 1989 20,200 430 920 $140 $203 1990 19,100 380 750 $170 $235 1991 19,800 370 920 $190 $251 1992 19,300 380 850 $157 $201 1993 20,100 400 920 $201 $250 1994 19,200 350 870 $148 $179 1995 18,200 430 860 $154 $182 1996 17,900 440 800 $173 $199 1997 17,500 320 650 $158 $177 1998 15,500 190 640 $144 $159 1999 13,600 (13,300) 200 (200) 560 (560) $125 ($124) $135 2000 12,900 (12,100) 270 (270) 510 (510) $128 ($128) $134 2001 14,600 (12,200) 270 (270) 630 (600) $119 ($118) $120 2002 17,200 (13,300) 210 (210) 550 (510) $135 ($134) $135

Note: All fires are fires reported to U.S. municipal fire departments and so exclude fires reported only to federal or state agencies or industrial fire brigades. National estimates are projections. Casualty and loss projections can be heavily influenced by the inclusion or exclusion of one unusually serious fire. Fires are estimated to the nearest hundred, deaths and injuries to the nearest ten, and direct property damage to the nearest million dollars. Inflation adjustment to 2002 dollars is done using the consumer price index. Statistics shown in parentheses are without allocation of confined fires. ‘‘Confined fires’’ refer to fires confined to chimneys, cooking equipment (e.g., ovens, pots, pans on stoves), furnaces, or trash containers. Such fires do not have to be coded for most details, including manufactured home versus other dwelling. Confined fires in manufactured homes are estimated based on the manufactured home share of all other fires in one- and two-family dwellings. *This reflects the effect of one National Fire Incident Reporting System (NFIRS) fire reported to have a loss of $10 million, which therefore added $24.2 million to the national estimates total in current dollars. Source: John R. Hall, Jr., Manufactured Home Fires, Fire Analysis & Research Division, National Fire Protection Association, Quincy, MA, 2005.

charge density rate from the sprinklers of 0.1 gpm/sq ft of sprinkler flow of 25 gpm for 10 minutes, resulting in the area for the single sprinkler with the largest area of 250-gal stored water supply. This original standard also coverage. The standard also recommended a stored water established the precedent allowing omission of sprinkler supply of 250 gal if an adequate public water supply is protection in , attached structures, , closets, not available. and entrance . Sprinklers were not required in This original residential sprinkler standard was devel- areas not used for storage or as living areas, nor in open oped for standard 1⁄2-in. orifice sprinklers with temperature attached , , and similar unheated attached ratings of 135 to 225F. The standard was based on a total structures. Bathrooms with floor areas of 40 sq ft or less

Automatic Sprinkler Systems for Residential Occupancies Handbook 2007 226 Supplement 4 • Residential Sprinkler Systems

TABLE S4.3 Fire Experience Rates for Manufactured Homes and Other Dwellings, 1999 to 2002

Manufactured Other One- and Measure Homes Two-Family Dwellings

A. Occupied Year-Round Housing Units Fires (1999–2002 annual average) 14,000 260,500 Civilian deaths (1999–2002 annual average) 240 2,190 Civilian injuries (1999–2002 annual average) 550 9,970 Number of housing units (weighted average of 1999 7.1a 73.7–82.0b and 2001, in millions) Civilian deaths per 100 fires 1.7 0.8 Civilian injuries per 100 fires 3.9 3.8 Fires per 1,000 housing units 2.0 3.2–3.5 Civilian deaths per 100,000 housing units 3.3 2.7–3.0 Civilian injuries per 100,000 housing units 7.7 12.2–13.5 B. Seasonal and Occupied Year-Round Housing Units Fires (1999–2002 annual average) 14,000 260,500 Civilian deaths (1999–2002) 240 2,190 Civilian injuries (1999–2002) 550 9,970 Number of housing units (weighted average of 1999 7.7a 75.7–84.1b and 2001, in millions) Civilian deaths per 100 fires 1.7 0.8 Civilian injuries per 100 fires 3.9 3.8 Fires per 1,000 housing units 1.8 3.1–3.4 Civilian deaths per 100,000 housing units 3.0 2.6–2.9 Civilian injuries per 100,000 housing units 7.1 11.8–13.2

Note: These are fires reported to U.S. municipal fire departments and so exclude fires reported only to federal or state agencies or industrial fire brigades. National estimates are projections. Casualty and loss projections can be heavily influenced by the inclusion or exclusion of one unusually serious fire. Fires are estimated to the nearest hundred, and civilian deaths and injuries are estimated to the nearest ten. Property damage has not been adjusted for inflation. Totals may not equal sums because of rounding. aThis is an upper bound because the definition includes ‘‘trailers,’’ which probably is not limited to the trailer coaches or other manufactured housing referred to by terms including the word ‘‘trailer.’’ bThe lower and higher figures in this range reflect the exclusion and inclusion of housing units in housing 2–4 housing units. Buildings having just two housing units, which correspond to the fire statistics on one- and two-family dwellings, cannot be isolated. Source: John R. Hall, Jr., Manufactured Home Fires, Fire Analysis & Research Division, National Fire Protection Association, Quincy, MA, 2005; based on data from NFIRS and NFPA Survey; Statistical Abstract of the United States, Washington, DC: U.S. Bureau of the Census, 2001 and 2003, Tables 952 and 959, respectively.

were exempted from sprinkler coverage. Sprinklers were family dwellings and mobile homes are shown in the fol- not required in closets whose smallest dimension did not lowing list:5 exceed 3 ft and whose floor area did not exceed 24 sq ft, nor in entrance halls, which were not the sole means of • 0.10 gpm/sq ft (one sprinkler) egress.4 The concept of omitting these auxiliary areas from • 10 minute-water supply sprinkler coverage is present in the current NFPA 13D • 25 gpm flow standard. Basic characteristics of the original NFPA 13D • 250 gal stored water standard for residential sprinkler systems in one- and two- • Standard sprinkler

2007 Automatic Sprinkler Systems for Residential Occupancies Handbook Supplement 4 • Residential Sprinkler Systems 227

• 256 sq ft coverage 50% will 10% fear accidental accept system • Omit attic, , hall, bath discharge • Local alarm [p. 5] 10% do not like system appearance Residential Sprinkler Acceptance and Cost A telephone survey conducted in 1977 concerned public acceptance of residential sprinkler systems and the per- ceived reasonable cost of such a system.5 The survey ques- 30% perceive no threat tioned owners of 400 single-family residences, occupants of fire of 400 multifamily residences, and occupants of 200 rural single-family residences. Acceptance. The survey indicated that approximately 50 50% will not percent of occupants would accept a residential sprinkler accept system system. Of the 50 percent who would not accept a residen- EXHIBIT S4.1 Results of 1977 of 1000 Occupants tial sprinkler system in their dwelling, 30 percent of these Concerning Acceptance of Residential Sprinkler occupants did not believe a fire threatened their dwelling, Systems. (Source: Rolf Jensen & Associates, Inc., Study thus they did not need a sprinkler system. The remaining to Establish the Existing Automatic Fire Suppression 20 percent of the occupants would not accept a residential Technology for Use in Residential Occupancies, A sprinkler system because half of them feared accidental Summary Report, National Fire Prevention and Control discharge and half disliked the appearance of a sprinkler Administration, Washington, DC, 1978, p. 7) system in their dwelling. Exhibit S4.1 presents the survey results. 50% will Cost. For the 50 percent of the study population who accept system indicated they would accept a residential sprinkler system, cost became an important variable. Approximately 30 per- cent of those occupants indicated that they would accept 30% will a sprinkler system at no cost to them. Approximately 9 accept system at percent of the study population were willing to pay up to no net cost $0.20 per sq ft, 7 percent were willing to pay up to $0.40 per sq ft, and 4 percent were willing to pay up to $0.75 per sq ft. Exhibit S4.2 illustrates this distribution. Remember, this study of the acceptance of a residential 9% will pay up sprinkler system was conducted in 1977, it concerned the to $0.20/sq ft 7% will pay up original NFPA 13D residential sprinkler system standard, to $0.40/sq ft it was a telephone survey, and 80 percent of the respondents 4% will pay up were from the cities of Chicago, Atlanta, Philadelphia, to $0.75/sq ft 50% will not Dallas/Fort Worth, or Los Angeles. accept system Wolf reported that approximately 20 years later these EXHIBIT S4.2 Results of 1977 Study of 1000 Occupants attitudes were still prevalent as indicated by a survey of Concerning Acceptance of the Cost of Residential 200 homeowners in Connecticut:6 Sprinkler Systems. (Source: Rolf Jensen & Associates, Inc., Study to Establish the Existing Automatic Fire Evidence of the public’s attitude is more than anecdotal. Suppression Technology for Use in Residential A recent survey of 200 single-family homeowners in Occupancies, A Summary Report, National Fire Connecticut sponsored by the Home Co- Prevention and Control Administration, Washington, DC, alition, a group made up of the American Fire Sprinkler 1978, p. 8) Association (AFSA), the National Fire Sprinkler Asso- ciation (NFSA), and NFPA, found that only four had residential sprinklers. Of the remaining respondents, as an option by builders. Thirty percent of all respon- only 4 in 10 had even heard of them. With the exception dents had heard that sprinklers go off accidentally, and of 1 respondent out of the 41 who built their own 40 percent agreed with this statement to some degree. homes, none had been offered in-home fire sprinklers [p. 59]

Automatic Sprinkler Systems for Residential Occupancies Handbook 2007 228 Supplement 4 • Residential Sprinkler Systems

Foehl reported on design and cost aspects of providing and $0.28 per sq ft for homeowner-installed systems. Thus, complete automatic sprinkler protection using copper tub- the 1986 costs for Foehl’s dwelling would be $1344 for a ing in the distribution system in a single-family dwelling.7 contractor-installed system and $294 for a homeowner- He reported that 3⁄4-in. copper tubing was utilized in the pre- installed system. engineered or packaged ‘‘junior’’ sprinkler system Grinnell Lehn reported on the sale of 36 town homes with manufactured in 1932 to protect in dwellings NFPA 13D residential sprinkler systems in Harrisburg, and other light hazard occupancies. Copper tubing was PA.9 These homes vary in size from 1800 to 2000 sq ft, provided in 12-ft lengths, with a tee fitting at one end of with two or three , full basements, and two sto- the tubing to accommodate the 1⁄2-in. pendent 135F-rated ries. Sales prices of the homes range from $69,900 to sprinkler. Domestic water supplied the system, and an $75,000, with the residential sprinkler system costing $1.00 alarm valve provided shutoff and alarm capabilities. per sq ft. Protecting a one-story, three-bedroom home with full Chapter 6 (in the section on residential sprinklers) basement and attached two-car with an automatic indicated that the National Fire Prevention and Control sprinkler system cost $1000 in 1974.7 The first floor of Administration initiated a research program to develop an Foehl’s design dwelling had approximately 1050 sq ft of efficient, cost-effective, residential sprinkler system. This space used for living room, , kitchen, three federally funded research program resulted in studies in bedrooms, and two baths. The garage area was 480 sq ft, the following areas:10 and the undivided basement area was 1200 sq ft. In 1974 payments on a conventional 20-year mortgage To assess the probable impact of using sprinklers to reduce the incidence of deaths and injuries in residential with an interest rate of 81⁄2 percent would increase by $8.68 fires. Other studies evaluated the design, installation, per month ($104.16 per year) to pay for a $1000 automatic practical usage, and other acceptance factors that would sprinkler system. Foehl imposed design limitations to keep impact on accomplishing reliable and acceptable sys- the cost at $1000 and assumed availability of a municipal tems; studies to evaluate minimum water discharge water supply with residual pressure between 30 and 50 psi rates, automatic sprinkler flow, response sensitivity and at the service entry to the home and a maximum 60 gpm design criteria, and full-scale tests of prototype residen- flow. The sprinkler system was designed with copper tub- tial sprinkler systems. [p. 70] ing to provide a Hazen-Williams friction-loss formula C factor of 150. The sprinkler piping system was independent Thus, by 1979, as a result of this extensive research of the domestic water supply and beyond the water meter effort, a prototype residential sprinkler was developed. This or the service entrance to the residence. Foehl used type-M prototype was the initial fusible-link type of quick-response copper tubing throughout his design. The system’s design (QR) sprinkler with a response time index (RTI) below 50. purpose was increasing the life safety of residents from NFPA conducted full-scale tests of this prototype sprinkler with 60 fires in a two-story single-family dwelling in Los fire. 11 Foehl’s final design (Exhibit S4.3) specified installa- Angeles in 1980. tion of eight horizontal sidewall sprinklers and six pendent sprinklers protecting all areas of the dwelling except the Residential Sprinkler System Tests in Los Angeles garage, closets, and stairwell, thus providing a system that Full-scale fire tests in Los Angeles dwellings, conducted 3 cost (in 1974) $894 (including 210 ft of ⁄4-in. type-M by the NFPA and funded by the U.S. Fire Administration, copper tubing). This system’s total cost was $63.86 per had these objectives:12 sprinkler, or $0.40/sq ft of protected area. Foehl indicated that one sprinkler protecting the bed- To test the performance of alternative sprinkler designs room end of the might be eliminated in favor of in ‘‘controlling’’ the development of fire in single- a particle-of-combustion detector located in the hall. The family dwellings and mobile homes. alarm function of the sprinkler system could be delegated To improve the ability to predict the performance to the detector. The total cost of the sprinkler system with of a sprinkler design under varying fuel loadings and this modification increased to $904. configurations. To field test the performance of alternative low-cost Chapter 3 [of Automatic Sprinkler and Standpipe Sys- sprinkler design configurations. tems] discussed Adams’s and Born’s presentation of the To identify performance parameters of low-cost costs of residential sprinkler systems using both plastic sprinkler system so that a desired level of and steel pipe in one-story, two-story, and split-level dwell- can be established by the NFPA technical committee ings.8 Their 1986 cost calculations indicated that average responsible for NFPA 13D, ‘‘Sprinkler Systems for One- costs were $1.28 per sq ft for contractor-installed systems and Two-Family Dwellings.’’

2007 Automatic Sprinkler Systems for Residential Occupancies Handbook Supplement 4 • Residential Sprinkler Systems 229

EXHIBIT S4.3 Sprinkler System in Dwelling with Eight Sidewall and Six Pendent Sprinklers. (Source: John M. Foehl, ‘‘In Kitchen Dining Quest of an Economical Auto- 11 ft 6 in. x 12 ft 10 ft x 12 ft matic Fire Suppression System Garage for Single-Family Residences,’’ 20 ft x 24 ft Fire Journal, Vol. 68, No. 5, 1974, p. 46) M. bedroom M. 11 ft x 14 6 in.

¹⁄₂-in. sidewall sprinkler Bedroom Bedroom ¹⁄₂-in. pendant sprinkler 10 ft 6 in. x 12 10 ft x Living room Riser or drop ft 6 in. 10 ft 6 in. 16 ft x 16 ft 6 in. P.C. detector

First Floor Plan in. ³⁄₄ ³⁄₄ in. in.

³⁄₄ in. ³⁄₄ in. ³⁄₄ in. ³⁄₄ in.

³⁄₄ ³⁄₄ in. 1 in. 1 in. ³⁄₄ in. ³⁄₄ in. ¹⁄₂-in. pendant sprinkler Riser or drop in. 1

Basement Floor Plan

To develop and make available information relating dwelling fire tests was of pendent design. This design re- to the design criteria, performance and costs of residen- ceived initial listing as a residential sprinkler in June 1981, tial sprinkler systems. fromUnderwritersLaboratories(UL) astheGrinnellF954 To facilitate the adoption of NFPA 13D by federal, residential sprinkler. (See Figure 6.27 in Chapter 6 [of Auto- state and local agencies having fire protection and pre- matic Sprinkler and Standpipe Systems].) In the test series vention responsibilities. [p. 224] this sprinkler’s temperature rating was 140F and its RTI The full-scale fire tests were conducted in a wood was approximately 28. The sprinklers were installed with frame and stucco two-story dwelling with basement. Fires 100-, 144-, or 150-sq ft spacing with the deflector 4 in. from were conducted in the living room and kitchen on the first the . Exhibit S4.5 illustrates the prototype residential floor and in the largest of three bedrooms on the second sprinkler used in the Los Angeles dwelling tests. floor. Exhibit S4.4 shows diagrams of the three The Los Angeles dwelling fire tests consisted of 60 involved in the fire tests. full-scale fire evaluations. The initial 20 tests involved The prototype residential sprinkler used in all these both flaming- and smoldering-initiated fires in all three

Automatic Sprinkler Systems for Residential Occupancies Handbook 2007 230 Supplement 4 • Residential Sprinkler Systems

EXHIBIT S4.4 Living Room, 13 ft 7 in. Kitchen, and Bedroom Involved 2 ft 11 in. in the Los Angeles Dwelling Fire Tests. (Source: Arthur E. Cote, Kitchen 9 ft ‘‘Field Test and Evaluation of First floor 6 ft 8 in. entry way 6 ft 8 in. 4 ft 1 in. Residential Sprinkler Systems,’’ landing Fire Journal, Vol. 19, No. 4, 1983, 11 ft p. 225) 8 ft 3 in. 4 ft

Living room 3 ft 7 in.

Stairway down 30 ft 5 in. 16 ft 10 in. 10 ft 10 in. 5 ft 6 in.

Closet 5 ft 6 in. Bedroom 1 16 ft 4 ft

3 ft 16 ft 2 in. 18 ft 5 in.

Sixteen test fires were conducted to verify the shielded-corner living room fire test FM Global previously conducted.13 This test involved a fire source located in a corner of a ventilated living room with combustible and ceiling. consisted of a three-seat sofa, an upholstered chair, an end table, a plastic wastebasket, a lamp, floor carpeting, and drapes behind both the chair and sofa. The flaming fire was initiated in newspapers in the wastebasket placed under the end table, which shielded the fire source from the sprinkler discharge. Exhibit S4.6 shows the furniture arrangement in the ventilated living room fire test. The circled ‘‘W’’ in the diagram represents the wastebasket under the end table. Compare the furniture arrangement in the ventilated living room fire test (with a shielded fire source) with the original UL residential pendent sprinkler fire test arrange- ment illustrated in Exhibit S4.7. Exhibit S4.7 shows the full-scale test room arrangement for testing pendent resi- EXHIBIT S4.5 Prototype Residential Sprinkler Used in dential sprinklers. Note the arrangement of the simulated the Los Angeles Dwelling Fire Tests. (Source: Courtesy furniture and the fire source in the corner. It appears that of Tyco Fire & Products) the UL developed its test arrangement from the FM Global and Los Angeles dwelling prototype residential sprinkler test rooms, with noncombustible interior finish in the tests.12,13,14 rooms. These test fires involved the following fires: The final 24 fire tests in the Los Angeles dwelling flaming-initiated kitchen fire, smoldering-initiated bed- were developed to obtain data to examine and validate the room fire, flaming-initiated closet fire, smoldering- and design parameters of the residential sprinklers for the 1980 flaming-initiated living room fires, with the fire source edition of NFPA 13D. These tests evaluated fires under shielded and unshielded from sprinkler discharge. the following conditions:

2007 Automatic Sprinkler Systems for Residential Occupancies Handbook Supplement 4 • Residential Sprinkler Systems 231

6 in. EXHIBIT S4.6 Arrangement of Furniture in the Ventilated Living 8 in. Room Shielded-Corner Fire Curtains Test. (Source: Hsiang-Cheng 24 in. L Kung, Robert D. Spaulding, and Edward E. Hill, Jr., Sprinkler C W 6 in. 8 in. Performance in Residential Fire E 54 in. Tests, Factory Mutual Research, FMRC No. 22574, Norwood, MA, 20 in. 1980, p. 21) CP 33¹⁄₂ in.

Sprinkler

Dummy sprinkler 1 in. S

Floor Plan

• They occur in a smaller room. room-fire tests involved sprinklers with 135 or 140F- • There is a lintel-limiting effect on the sprinkler dis- rated sprinklers. The fire was effectively controlled in 4 charge. of the 11 tests, or approximately 36 percent of the room- • Other sprinklers are also in operation. fire tests. • A fire is initiated between sprinklers. Criterion for effectively controlling the fire and pre- venting its extension was established as maintaining a ceil- Chapter 6 previously described Moore’s summary of ing-level temperature below 500F. Cote explained the human tenability criteria in the Los Angeles dwelling criterion:14 tests.11 Cote presented these criteria as follows:14 With the limited amount of water to supply an automatic Temperature—150F at 5 ft from the floor sprinkler system in a residence (15 gpm to 30 gpm), Oxygen—14 percent at 5 ft from the floor only a few sprinklers (one to three) can be supplied CO—10,000 ppm and 25 percent COHb with sufficient water to affect fire control. When more Smoke—11 percent per ft [pp. 49–51] than a few sprinklers open, the reduced water supply Of the 60 full-scale room-fire tests conducted in the to each sprinkler generally results in a lack of fire Los Angeles dwelling, 49 involved using the prototype control with resultant fire damage to the structure. Cel- pendent residential sprinkler. The arrangement of these lulosic and plastic materials usually begin to pyrolyze residential sprinklers effectively controlled fire in 39 of vigorously at a temperature of about 500 F. If the gas temperature beneath the ceiling near the center of a the 49 tests or approximately 80 percent of the room-fire 14 room is maintained below 500 F, the likelihood that tests. the ceiling and the building structures will ignite will Eleven of the full-scale room-fire tests in the Los be small. Therefore, in the Los Angeles test series, a Angeles dwelling were conducted with the spray sidewall limit of 500F for the ceiling gas temperature was se- sprinkler and the prototype pendent residential sprinkler lected in order for sprinkler protection of property to with a standard-response fusible link and a spray concealed be considered adequate. Tests, however, were not termi- sprinkler. All these standard-response spray sprinkler nated if this temperature was reached. [p. 49]

Automatic Sprinkler Systems for Residential Occupancies Handbook 2007 232 Supplement 4 • Residential Sprinkler Systems

L EXHIBIT S4.7 UL Original Resi- Wood crib dential Sprinkler Fire Test Ar- 8 ft (2.4 m) Plywood 19 in. 35 in. 8 ft 3 ft Thermocouple, rangement for Pendent or ¹⁄₄ (483 mm) (0.89 mm) (457 mm) (0.9 m) in. (6.35 mm) Upright Sprinklers. (Source: UL above ceiling 1626, 1994; used with permis- sion from Underwriters Labora- 1 in. (25 mm) 3 in. (0.9 m) tories, Inc.) Simulated furniture Excelsior 8-in. (2.4-m) L/2 Plywood 18 in. (457 mm)

8 in. (203 mm)

Thermocouples, 3 in. (76.2 mm) below ceiling and 5¹⁄₄ ft (1.6 m) above floor

L W/2 2L

Pendent or upright sprinkler (typ.)

4 in. (102 mm)

3 ft (0.9 m)

41 in. L Coverage length (1.04 m) W Coverage width

The most severe of the test fires in the Los Angeles of the development of the modified UL room fire test dwelling tests was the furniture in the corner scenario. The follows:15 original UL fire test was developed from this scenario and deviated from the fire tests of previous sprinkler crib fire Fire tests were conducted using the Underwriters Labo- tests by duplicating a room fire test with simulated furni- ratories (UL) UL 1626 Fire Test for residential sprin- ture. The arrangement of the original UL fire test room is klers. The results indicated that sprinkler performance illustrated in Exhibit S4.7. Over a period of about 20 years in UL 1626 could not be reproduced due to differences concern began to be expressed about the low flow of some in material flammability properties. This conclusion was reached in a cooperative effort with UL. Factory listed residential sprinklers. Bill et al. conducted replication Mutual Research’s Approval fire test that had been tests of the original UL residential room fire tests and, redesigned based on UL 1626 was thus subject to the with the cooperation of UL and the NFPA sprinkler com- same variability. A new fire test was developed using mittees, a problem with the reliability of the room fire tests fuel with controlled material flammability parameters was identified and the room fire test was modified relative based upon the use of the ASTM E2058 Fire Propaga- to the fuel materials and their configuration. The summary tion Apparatus. The new fuel package consists of a

2007 Automatic Sprinkler Systems for Residential Occupancies Handbook Supplement 4 • Residential Sprinkler Systems 233

wood crib (one half the height of the one which was The 2002 edition of NFPA 13D* has the following require- used in UL 1626) supported over a pan with heptane, ment on sprinkler discharge density: two polyether foam cushions (about 60% greater in density than the foam previously used in UL 1626) 8.1.1.2.2* The system shall provide at least the flow measuring 34 in by 30 in by 3 in (864 mm by 762 mm required to produce a minimum discharge density of 2 by 76 mm) and 1⁄4 in (6 mm) Douglas Fir plywood 0.05 gpm/ft (2.04 mm/min) to the design sprinklers. paneling. The new fire test was shown in a series of sprinklered fire tests to provide a reproducible challenge Charlotte, North Carolina, Mobile Home Tests to residential sprinklers comparable to that observed in the Factory Mutual Research and the Los Angeles Following the Los Angeles dwelling room-fire tests of the Residential Test Programs. [p. 101] prototype pendent residential sprinkler, 16 full-scale fire tests were conducted in a mobile home in 1980. The interior These replication tests also identified the need to estab- finish of the mobile home had flame-spread ratings below lish a minimum waterflow arrangement. Therefore, as pre- 200. In the full-scale room-fire tests, both ionization- and viously indicated in Chapter 6, the modified residential photoelectric-type smoke detectors were compared with fire test with a minimum flow requirement of 0.05 gpm/ 2 the prototype residential sprinkler. The tests included com- ft was initiated effective July 12, 2002. Golinveaux indi- parative tests of smoke detectors because Department of cated the replication tests for the modified room fire test Housing and Urban Development’s (HUD) Mobile Home consisted of several hundred tests over a period of two Construction and Safety Standard (1975) required at least years. In addition to the modification of the fuel materials, one single-station smoke detector in mobile homes. These the panels in the room were modified, the time of smoke detectors were grouped in the hallway, mounted on ignition of the simulated foam furniture was changed, and the wall leading to each bedroom. Exhibit S4.9 shows the the configuration of the fuel arrangement in the corner was 16,17 location of both groups of smoke detectors and sprinkler altered. Exhibit S4.8 illustrates the modified UL room locations in the test mobile home. residential sprinkler fire test utilized today. Huggins has Similar to the Los Angeles dwelling tests, the 16 full- summarized the differences between the original UL 1626 scale fire tests were conducted in the living room and room fire test and the modified UL 1626 room fire tests 18 kitchen areas and in the larger bedroom, identified as the as follows: ‘‘test bedroom area’’ in Exhibit S4.9. Exhibit S4.10 shows the arrangement of test areas in relation to the remainder The wood paneling and polyether foam now have more tightly defined burning characteristics. The wood crib of the mobile home. The circled numbers 1, 2, and 4 in is lighter at 5.5 to 7 pounds (2.49 to 3.17 kilograms). Exhibit S4.10 identify the mobile home areas involved in The foam cushions are slightly smaller at 30 by 32 the full-scale room-fire tests. As Exhibit S4.10 indicates, inches (7.62 by 81.28 centimeters). The space between the mobile home consisted of the living room and kitchen the wood crib and simulated furniture has increased by areas, three bedrooms (one of which had a closed 6 inches (15.24 centimeters). The shavings have been during the test fires), and other small rooms and closets. replaced by cotton wicks soaked in heptanes. There’s The mobile home was 70-ft long and 14-ft wide, with a no longer a delay before igniting the simulated furni- ceiling height of 71⁄2 ft. ture. And the length of the test was increased to 30 The mobile home tests’ objectives were to validate minutes, although it may be stopped at 10 minutes if design parameters established in the Los Angeles dwelling the fire is extinguished or only the wood crib is burning. tests by applying them to the unique fuel arrangement and [p. 73] interior configuration of a mobile home. One departure Mitchell has indicated that the resulting water density from the test criteria used in the dwelling fire tests, as requirement of 0.05 gpm/ft2 instituted in both NFPA 13D adapted for the mobile home tests, was reducing ceiling and 13R in actual system design may limit the sprinkler temperature level from 500 to 450 F. The reduced ceiling designer to coverage areas for each sprinkler below 16 ft temperature level helped maintain the mobile home’s integ- by 16 ft or 256 ft2. He also indicated there might be an rity for the total test series. Thus, when ceiling temperature issue for the calculation of the pressure loss through the reached 450 F, a reserve deluge sprinkler system extin- water meters usually involved in residential occupancies. guished the fire. Room-fire tests conducted in the mobile Thus he collected data on the pressure losses induced with 5 five manufacturers’ meters varying in size from ⁄8 in. to *An asterisk accompanying a section number in a code extract indi- 2 in. with maximum flows from 20 gpm to 170 gpm. This cates that the requirement in that section is accompanied by nonman- meter pressure loss information is presented in Table S4.4. datory explanatory material in Annex A of the code or standard.

Automatic Sprinkler Systems for Residential Occupancies Handbook 2007 234 Supplement 4 • Residential Sprinkler Systems

Thermocouple, ¹⁄₄ in. W EXHIBIT S4.8 UL Residential 4-ft (1.4-m) (6.3 mm) above ceiling, Sprinkler Fire Test Arrangement 10 in. diagonally from Plywood corner 42 in. 35 in. 19 in. for Pendent, Flush, Recessed (475 mm) Wood crib (1.05 m) (0.89 m) Pendent, and Concealed Sprin- klers. (Source: UL 1626, Resi- dential Sprinklers for Fire Protection Service, Underwrit- 42 in.

(1.05 m) ers Laboratories, Northbrook,

Plywood Simulated 4-ft (1.4-m) IL, 2003, p. 28; used with permis- furniture L/2 sion from Underwriters Labora- tories Inc.)

8 in. (203 mm)

Thermocouples, 3 in. (76.2 mm) below ceiling and 5¹⁄₄ ft (1.6 m) above floor

Thermocouple, 3 in. (76.2 mm) below ceiling (room center) L 2L W/2

Sprinkler (typical)

3 ft (0.9 m) 4 in. (102 mm)

41 in. (1.04 m)

home involved arrangements adapted from the dwelling source test fires in the living room. However, two of the tests: flaming-initiated kitchen fires, flaming-initiated bed- living room test fires were invalid because the water supply room fires, and flaming-initiated living room fires with a was not adjusted properly. Thus, the living room fires shielded fire source.19 Due to the mobile home’s interior consisted of 10 valid tests, and the mobile home test series finish as well as the typical interior finish, all of these fires consisted of 14 valid tests. Three test fires exceeded the involved combustible walls and . ceiling temperature criterion, and the reserve deluge system Test fires consisted of one test fire in the kitchen area, extinguished fire in the one kitchen test and two living room three test fires in the bedroom, and 12 corner-shielded tests. The residential sprinkler system with the pendent

2007 Automatic Sprinkler Systems for Residential Occupancies Handbook Supplement 4 • Residential Sprinkler Systems 235

TABLE S4.4 Pressure Loss and Flows Through Selected Residential Water Meters

Pressure Loss at

Maximum Pressure Loss at 24 gpm 7⁄16 32 gpm 7⁄16 46 gpm 1⁄2 56 gpm 1⁄2 Size Rated Flow Maximum Rated Flow 12 12 16 16 18 18 sloped 20 20 Manufacturer (in.) (gpm) (psi) (psi) (psi) (psi) (psi)

5 Hersey ⁄8 209 NNN N 5 Sensus SRII ⁄8 207 NNN N Sensus SR 5⁄8 20 10.8 N N N N 5 Schlumberger ⁄8 208 NNN N 5 BadgerMeter ⁄8 25 7 6.8 N N N

Hersey 3⁄4 30 7.5 4.5 N N N 3 Sensus SRII ⁄4 309 NNN N 3 Sensus SR ⁄4 30 11 8.0 N N N 3 Schlumberger ⁄4 30 10 6.3 N N N 3 BadgerMeter ⁄4 35 10.5 5.2 8.5 N N Hersey 1 50 8 2.0 3.4 6.4 N Sensus SRII 1 50 7.3 N N N N Sensus SR 1 50 10.9 2.3 4.5 9.0 N Schlumberger 1 50 8 2.0 3.4 6.5 N BadgerMeter 1 70 12.5 1.5 2.9 5.9 8.0

1 Hersey 1 ⁄2 100 13 0.8 1.3 2.9 4.0 1 Sensus SR 1 ⁄2 100 11.4 1.5 1.5 3.0 4.5 1 BadgerMeter 1 ⁄2 120 10.5 0.5 0.9 1.7 2.5 Hersey 2 160 13 0.6 0.8 1.0 1.6 Sensus SR 2 150 12.1 0.6 0.8 1.0 1.5 BadgerMeter 2 170 9.5 0.2 0.3 0.5 0.9

N Data not available. Source: J. Scott Mitchell, ‘‘Increased NFPA 13D Density, Can Common Water Meters Support It?’’ Sprinkler Age (official publication of the American Fire Sprinkler Association), Vol. 22, No. 1, 2003, p. 20.

prototype residential sprinkler was effective in 11 full-scale tions: All tests were flaming-initiated fires. In one bedroom room-fire tests of the 14 valid tests, approximately 78 smoldering-fire test in the dwelling, fire progressed for 4 percent of the mobile home fire tests. hours, exceeding human tenability criteria, without activat- As expected, smoke detectors activated before the pro- ing the prototype residential sprinkler. This test validates totype residential sprinkler in all of the room fire tests. the need for smoke detectors in dwellings. The additional alert time the smoke detectors provided Cote summarized results of both the Los Angeles before the first sprinkler activated varied according to the dwelling and the mobile home prototype residential sprin- fire’s location.19 In the one kitchen fire, smoke detectors kler tests:14 activated between 2 and 2.5 minutes before the first sprin- kler activated. In the three bedroom fires, smoke detectors In the flaming-started fire tests, the critical limits for activated between 0.5 and 1.8 minutes before the first tenability were not exceeded for those tests in which sprinkler. In the 10 living room fires, the smoke detectors the sprinkler system controlled the fire. In the smoldering-started fire tests, the critical limits activated between 0.08 and 1.65 minutes before the first established for tenability were not exceeded for those sprinkler. Thus, in all 14 valid mobile home tests, smoke tests in which the transition to flaming occurred and detectors activated between 0.08 and 2.5 minutes before the sprinkler system controlled the fire. the first prototype residential sprinkler. In one smoldering-started fire test the transition to Remember that, unlike the Los Angeles dwelling tests, flaming never occurred and the critical limits estab- these mobile home tests included no smoldering fire situa- lished for tenability were exceeded.

Automatic Sprinkler Systems for Residential Occupancies Handbook 2007 236 Supplement 4 • Residential Sprinkler Systems

Z\x 2 10 in. A-A Section 7 ft 6 in. 6 in. Smoke Smoke detector 2 group

C 30 in. ET Z\x

4 22 in. x 18 in. 22 in. Z\x

27 13 ft 2 in. Z\x CW SPK 5 CT SPK 3 SPK 1 74 in. 44 in. x 18 44 in. S 3 in. 12 in. 4 9 ft 1 in. 30 in. Area closed during testing CW Living room area 3 5 13 ft 12 in. 12 in. A A Kitchen area SPK 4 SPK 2 SPK 6 2 , open–23Window, x 31 in. 23 ft 1 in. Smoke Smoke detector 1 group 1 (under ) 8 ft in. Closet in. Test Test bedroom area ¹⁄₂ 6 72 in. 9 ft 6 Curtains maker Coffee SPK 10 SPK 11 47 in. PCPI PCPI BS, BM, FS BS, SH, BL SPK 12 Wastebasket Wastebasket ignition ET 6 in. 22 in. x 18 in. 22 in. 2 in. Living room area Kitchen area area Hallway bedroom area Test Instrument room 2 area Bedroom no. 2 in. 3 5 1 2 4 EXHIBIT S4.9 Living Room/Kitchen and Bedroom Areas 6 of the Mobile Home Where Fire Tests Were Conducted. EXHIBIT S4.10 Interior Arrangement of the Mobile Home (Source: Arthur E. Cote, ‘‘Field Test and Evaluation of for the Sprinkler Fire-Test Series. (Source: Arthur E. Cote, Residential Sprinkler Systems: Part III,’’ , ‘‘Field Test and Evaluation of Residential Sprinkler Vol. 20, No. 2, 1984, p. 42) Systems: Part III,’’ Fire Technology, Vol. 20, No. 2, 1984, p. 42)

2007 Automatic Sprinkler Systems for Residential Occupancies Handbook Supplement 4 • Residential Sprinkler Systems 237

In all other tests in which the sprinkler system was This standard recommends, but does not require, sprin- not able to control the fire the critical limits established klering of all areas in a dwelling; it permits sprinklers to for tenability were exceeded. [p. 58] be omitted in certain areas. These areas have been proved In developing the 1980 edition of NFPA 13D, results by NFPA statistics to be those in which the incidence of and conclusions of the Los Angeles dwelling and the mo- life loss from fires in dwellings is low. Such an approach bile home sprinkler fire tests were the basic data that the provides a reasonable degree of fire safety. Greater protec- NFPA Committee on Sprinklers and its subcommittee on tion in both life and property is achieved by sprinklering residential sprinklers considered. all areas. Guidance for the installation of smoke detectors and fire detection systems is found in NFPA 72, National Fire Alarm Code. NFPA 13D Requirement Changes in NFPA 13D Editions Objectives The 1980 edition of NFPA 13D differed radically from the The 1980 edition of NFPA 13D provided the basis for 1975 edition of NFPA 13D. The concept of a minimum subsequent editions of the standard and also provided the discharge density was replaced by a minimum flow require- for NFPA 13R, examined later in this chapter. ment of 18 gpm to the first sprinkler operating and a The 1975 edition of NFPA 13D design criteria should be minimum flow of 13 gpm each for a maximum of two reviewed. The 1980 edition of NFPA 13D was developed sprinklers operating. The 1980 standard thus established a with two primary objectives:20 total flow requirement of 26 gpm for the system, as opposed 1. It should be a low-cost system economical enough to to the 1975 edition’s total flow requirement of 25 gpm. be installed in dwellings on a fairly widespread basis. The 1980 standard also mandated use of listed residential 2. It should be capable of maintaining life safety in the sprinklers rather than the spray sprinkler. The 1980 edition room of fire origin for at least 10 minutes. of NFPA 13D continued to recommend the 10-minute water supply quoted in the 1975 edition, but with the increased The performance requirement of the 1980 edition of flow requirement a minimum 260-gal water supply instead NFPA 13D as developed from both the Los Angeles dwell- of 250 gal. The 1980 edition reduced the maximum cover- ing tests and the mobile home tests was ‘‘to prevent flash- age area per sprinkler from 256 sq ft per sprinkler to a over in the room of fire origin, when sprinklered, and to maximum of 144 sq ft per sprinkler. The 1975 edition of improve the chance for occupants to escape or be evacu- 10 NFPA 13D specified a local alarm on the sprinkler system. ated.’’ However, the 1980 edition recommended smoke detec- The 2002 edition of NFPA 13D* includes in the pur- tors:10 pose the similar wording from the 1980 edition of NFPA 13D: This standard assumes that one or more smoke detectors will be installed in accordance with NFPA 74. [p. 110] 1.2* Purpose. The purpose of this standard is to provide a sprinkler system that aids in the detection and control The 2002 edition of NFPA 13D* contains requirements of residential fires and thus provide improved protection for smoke detectors in two sections as follows: against injury, life loss, and property damage. A sprin- 6.3* Multipurpose Piping System. kler system designed and installed in accordance with A piping system this standard shall be expected to prevent flashover serving both sprinkler and domestic needs shall be con- (total involvement) in the room of fire origin, where sidered to be acceptable by this standard where the sprinklered, and to improve the chance for occupants following conditions are met....(2)Smoke detectors NFPA 72, National to escape or be evacuated. The layout, calculation, and are provided in accordance with Fire Alarm Code. installation of systems installed in accordance with this standard shall only be performed by people knowledge- 7.6* Alarms. Local waterflow alarms shall be provided able and trained in such systems. on all sprinkler systems in homes not equipped with NFPA 72, National A.1.2 Variouslevels of fire safety are available to dwell- smoke detectors in accordance with Fire Alarm Code. ing occupants to provide life safety and property protec- tion. The omission of sprinklers in attics, bathrooms, clos- ets, entrance halls, and attached structures continued; how- *An asterisk accompanying a section number in a code extract indi- ever, added wording restricted the use of combustible cates that the requirement in that section is accompanied by nonman- fixtures in bathrooms and closet combustible datory explanatory material in Annex A of the code or standard. walls and ceilings. The 1989 edition modified this,

Automatic Sprinkler Systems for Residential Occupancies Handbook 2007 238 Supplement 4 • Residential Sprinkler Systems

allowing limited combustible walls and ceilings in closets without sprinklers. The 1980 edition of NFPA 13D estab- 10 ft 10 ft lished spacing requirements for the residential sprinkler system with a minimum 8-ft distance between sprinklers, 11 ft

a maximum 12-ft distance between sprinklers, and a maxi- Bedroom in. 12 ft Bedroom ³⁄₄ mum 6-ft distance between a sprinkler and a wall. These in. ³⁄₄ spacing requirements reduced those of the 1975 edition of NFPA 13D, which allowed maximum 16-ft spacing be- -in. risers from basement -in. in.

tween sprinklers and a maximum 8-ft spacing from walls ³⁄₄ ³⁄₄ Bath 5 ft Six due to the use of the spray sprinkler. in. Basic design criteria established for residential sprin- ³⁄₄ kler systems in the 1980 edition of NFPA 13D are the 12 ft following:10 Bedroom Closet 10 ft

• 18 gpm (one sprinkler) 10 ft • 13 gpm (each of two sprinklers) Lavatory 8 ft

• 26 gpm flow in. ³⁄₄

• 260 gal stored flow Dining area 24 ft

• Listed residential sprinkler in. • 144 sq ft coverage 1 in. in. ³⁄₄ • Omit attic, closet, hall, bath ³⁄₄ B C Kitchen

• Smoke detector (NFPA 74) [p. 70] First-floor plan Living room Exhibit S4.11 shows a residential sprinkler system on in.

the first floor of a dwelling in conformance with the 1980 14 ft ³⁄₄ edition of NFPA 13D. Note that the and garage areas are not sprinklered. It should be recognized that the 1980 edition of NFPA

13D was designed as a life safety system, and residential A 11 ft sprinklers are still the only sprinklers tested for listing Porch Family room Family based on human tenability criteria. NFPA 13D is still a life safety standard designed to provide adequate time for 12 ft a limited number of occupants to escape from a fire, even when they are in the area of fire origin.20 The definition of the residential sprinkler in the 2002 edition of NFPA 13D and 13R emphasizes this special nature of the residential sprinkler:

3.3.8.2 Residential Sprinkler. A type of fast-response Garage

sprinkler that meets the criteria of NFPA 13, Standard SI units For 1 in. = 25.4 mm 1 ft = 0.305 m for the Installation of Sprinkler Systems, that has been specifically investigated for its ability to enhance sur- vivability in the room of fire origin and is listed for use in the protection of dwelling units. EXHIBIT S4.11 Typical First Floor Residential Sprinkler Underwriters Laboratories standard UL 1626, Resi- System Installed According to 1980 Edition of NFPA 13D. dential Sprinklers for Fire Protection Service, contains the (Source: Arthur E. Cote, ‘‘Update on Residential Sprinkler only sprinkler fire test with criteria for thermal human Protection,’’ Fire Journal, Vol. 77, No. 6, 1983, p. 71) tenability as follows21 (refer to Exhibit S4.8): 28 Fire Test 28.1 General referenced in the installation instructions. Additionally, a maximum of two residential sprinklers shall operate. 28.1.1 When tested as described in 28.1.2–28.4.1, a The sprinklers shall limit temperatures as follows: residential sprinkler shall limit temperatures as speci- fied in 28.1.1 (a)–(d) when tested at each rated spacing a) The maximum temperature 3 inches (76 mm) below

2007 Automatic Sprinkler Systems for Residential Occupancies Handbook Supplement 4 • Residential Sprinkler Systems 239

the ceiling at either location as illustrated in Figures long as there is another means of egress from the dwell- 28.1 to 28.3 shall not exceed 600 F (316 C). ing unit. b) The maximum temperature 51⁄4 feet (1.6 m) above the floor shall not exceed 200 F (93 C). Although sprinklers are not required in garages under c) The temperature at the location described in (b) NFPA 13D, the 2002 edition has an annex section indicat- shall not exceed 130 F (54 C) for more than any ing some local authorities have required the extension of continuous 2-minute period. the residential sprinkler system to garage areas, especially 1 d) The maximum ceiling material temperature ⁄4 inch with attached garages. The 2002 edition of NFPA 13R (6.4 mm) behind the finished ceiling surface shall has provided requirements for the sprinklers in attached not exceed 500 F (260 C). garages, which are covered in detailed later in this chapter. Additional criteria relative to the reduction of the ther- The annex statement relative to sprinklers in garages in mal tenability criteria for the human in the room of fire NFPA 13D is as follows: origin relative to the unusual sprinkler water distribution A.8.6.4 Although NFPA 13D does not require garages were discussed in Chapter 6. Also all of the UL residential to be sprinklered, some authorities having jurisdiction sprinkler test procedures are examined in Appendix D. take it upon themselves to add this requirement locally. In addition to the full-scale fire test data previously In such circumstances, residential or quick-response examined, the sprinkler committee, in its decisions to ex- sprinklers with a two-sprinkler design in the garage empt areas from sprinkler coverage, examined statistics to with the same piping used in the rest of the dwelling determine areas of risk for occupants of one- and two- may be used. It is recognized that residential sprinklers family dwellings and manufactured homes. The annex to have not been tested specifically for fires in garages, the 2002 edition of NFPA 13D presents these statistical but field experience has shown that the sprinklers help data. Table S4.5 presents these statistical data that justify to alert occupants to the fact that there is a fire, can areas omitted from sprinkler coverage in the 2002 edition of reduce the possibility of flashover, and can improve the chances for occupants to escape. NFPA 13D. The areas exempted from automatic sprinkler coverage in NFPA 13D* follow: The statistics relative to the number of fires and the civilian deaths and injuries relative to the area within resi- 8.6 Location of Sprinklers. dential occupancies protected by NFPA 13D systems from 8.6.1 Sprinklers shall be installed in all areas except where omission is permitted by 8.6.2 through 8.6.6. 1986 to 1990 are presented in Table S4.6. 8.6.2 Sprinklers shall not be required in bathrooms of 55 ft2 (5.1 m2) and less. Limited Area Dwelling Sprinkler System 8.6.3 Sprinklers shall not be required in clothes closets, linen closets, and that meet all of the following The 1994 edition of NFPA 13D initially established re- conditions: quirements for sprinkler system installation for small one- and two-family dwellings and manufactured homes. This (1) The area of the space does not exceed 24 ft2 (2.2 m2). sprinkler system applied to dwellings and homes not ex- (2) The least dimension does not exceed 3 ft (0.9 m). ceeding 2000 sq ft in area, one story high with smooth (3) The walls and ceiling are surfaced with noncom- ceilings with a slope not over 10 degrees, 8-ft horizontal bustible or limited-combustible material as identi- ceiling height, and 9-ft sloped ceiling height.22,23 Fleming fied in NFPA 220, Standard on Types of Building indicated that the limited area dwelling sprinkler system Construction. modifies the standard NFPA 13D system as follows:22

8.6.4* Sprinklers shall not be required in garages, open Sprinklers for use in this type of system must be specifi- attached porches, carports, and similar structures. cally listed for limited area dwelling use. 8.6.5 Sprinklers shall not be required in attics, crawl Where only one sprinkler is installed in a compart- spaces, and other concealed spaces that are not used ment, the water demand is to be not less than 10 gpm or intended for living purposes. at a flowing pressure of 25 psi. 8.6.6 Sprinklers shall not be required in covered un- When two or more sprinklers are installed in a com- heated projections of the building at entrances/exits as partment, the water demand for the system is to be not less than 6.5 gpm to each of two sprinklers at a flowing pressure of 11 psi. Verification of the single sprinkler *An asterisk accompanying a section number in a code extract indi- demand point must also be provided. cates that the requirement in that section is accompanied by nonman- The definition of compartment will be a space en- datory explanatory material in Annex A of the code or standard. closed by walls and a ceiling. It can have openings to

Automatic Sprinkler Systems for Residential Occupancies Handbook 2007 240 Supplement 4 • Residential Sprinkler Systems

TABLE S4.5 Statistical Data Used to Justify Omitting Areas from Sprinkler Coverage in NFPA 13D

Causal Factors in One- and Two-Family Dwelling Fires That Caused One or More Deaths Area of Origin Form of Heat of Ignition

Living room 41% Based on 6,066 Smoking materials 36% Based on 5,016 Bedroom 27% incidents where area Heat from fuel — fire or 25% incidents where form Kitchen 15% of origin was powered object of heat of ignition Storage area 4% reported Heat from miscellaneous 15% was reported Heating 3% open flame (including Structural area 2% match) Other areas 8% Heat from electrical 14% equipment arcing or overload

Form of Materials Ignited Furniture 27% Based on 5,080 Hot objects, including 7% Bedding 18% incidents where form properly operating Combustible liquid or gas 13% of material ignited electrical equipment Interior finish 9% was reported Other 3% Structural member 9% Waste, rubbish 4% Clothing (on a person) 3% Cooking materials 3% Electrical insulation 2% Curtains, draperies 2% Other 10% Total number of incidents reported: 10,194

Source: NFPA 13D, 2002, Table A.1.2(a); data from 1986–1990 NFIRS and NFPA Survey.

an adjoining space if the openings have a minimum two-sprinkler demand. This change reduced the minimum lintel depth of 2 inches from the ceiling and provided of 260 gal to 182 gal for the NFPA 13D standard system the openings do not exceed 20 square feet from each when protecting similar dwellings or manufactured homes. compartment. These reduced water duration requirements are continued The water supply must be capable of supporting the in the current NFPA 13D as follows: system demand for ten minutes for one sprinkler op- erating and seven minutes for two-sprinkler operation. 6.1.3 Where stored water is used as the sole source of The area of coverage for sprinklers will not be permit- supply, the minimum quantity shall be permitted to ted to exceed 64 square feet per sprinkler, although a equal the two-sprinkler water demand rate times 7 mi- single sprinkler will be permitted to cover a compart- nutes where dwelling units meet the following criteria: ment with an area not exceeding 100 square feet with (1) One story in height no dimension exceeding 10 feet. (2) Less than 2000 ft2 (186 m2) in area The perpendicular distance to a wall or partition from the sprinkler must not exceed 5 feet, with the minimum A modified NFPA 13D system, the limited area dwell- distance between sprinklers not less than 6 feet. ing sprinkler system had to meet all NFPA 13D require- Sprinklers are required in bathrooms and entrance ments other than those previously identified and initially foyers. stated in the 1994 standard. A listed strainer will be required in risers or feed Bill and Kung conducted a series of eight full-scale mains supplying sprinklers with orifices less than 3 fire tests to evaluate the performance of a prototype sprin- ⁄8-inch in diameter. [p. 1] kler for the limited area dwelling sprinkler system. Tenabil- Reducing water supply requirements for the limited ity for occupants was maintained by keeping 8-ft ceiling area dwelling system to a minimum of 100 gal from the temperatures over the ignition area below 500F and at the required minimum of 260 gal in the NFPA 13D system 5-ft level in the room’s center below 200F. Tenability was resulted in an exception to 7 minutes for the NFPA 13D maintained when one sprinkler activated for 10 minutes at

2007 Automatic Sprinkler Systems for Residential Occupancies Handbook Supplement 4 • Residential Sprinkler Systems 241

TABLE S4.6 Fires, Civilian Deaths, and Injuries by Area of Origin in NFPA 13D Protected Occupancies, 1986–1990

Civilian Area of Origin Deaths (%) Fires (%) Injuries (%)

Living room, family room, or den 1,330 37.1 42,600 10.5 2,546 18.6 Bedroom 919 25.6 50,200 12.4 3,250 23.7 Kitchen 541 15.1 92,670 22.9 3,987 29.1 Dining room 83 2.3 3,780 0.9 189 1.4 Heating equipment room or area 62 1.7 15,130 3.7 374 2.7 Hallway or corridor 48 1.3 3,690 0.9 155 1.1 or area 47 1.3 15,370 3.8 363 2.7 Garage or * 45 1.2 14,580 3.6 524 3.8 44 1.2 8,040 2.0 271 2.0 Unclassified structural area 43 1.2 4,530 1.1 104 0.8 or substructure space 41 1.2 11,200 2.8 317 2.3 Multiple areas 41 1.1 3,350 0.8 96 0.7 Ceiling/floor assembly or concealed 32 0.9 3,470 0.9 64 0.5 space Wall assembly or concealed space 27 0.8 7,090 1.8 93 0.7 Closet 23 0.6 5,020 1.2 186 1.4 Exterior or open porch 22 0.6 5,570 1.4 121 0.9 Exterior wall surface 22 0.6 14,620 3.6 118 0.9 Unclassified area 21 0.6 2,590 0.6 87 0.6 Attic or ceiling/ assembly or 21 0.6 10,740 2.7 98 0.7 concealed space Tool room or other supply storage room 20 0.5 4,160 1.0 133 1.0 or area or entrance way 17 0.5 1,410 0.3 44 0.3 Interior stairway 17 0.5 1,100 0.3 41 0.3 17 0.5 60,530 14.9 75 0.5 Unclassified function area 17 0.5 1,090 0.3 43 0.3 Unclassified storage area 14 0.4 2,460 0.6 80 0.6 Area not applicable 11 0.3 1,180 0.3 22 0.2 Exterior stairway 8 0.2 1,090 0.3 25 0.2 Lawn or field 7 0.2 1,670 0.4 24 0.2 Trash room or area 5 0.1 1,140 0.3 14 0.1 Product storage area 5 0.1 780 0.2 23 0.2 Unclassified means of egress 5 0.1 610 0.2 15 0.1 Unclassified service or equipment area 4 0.1 380 0.1 12 0.1 Library 3 0.1 180 0.0 11 0.0 Other known area 26 0.7 12,880 3.2 195 1.4 Total 3,589 100.0 404,900 100.0 13,691 100.0

Note: Fires are estimated to the nearest 10; civilian deaths and injuries to the nearest 1. *Does not include garages coded as a separate property, which averaged 19 deaths, 259 injuries and 21,170 fires per year. Source: NFPA 13D, 2002, Table A.1.2(b).

10 gpm in five tests and when multiple sprinklers activated The design of the LWS prototype sprinkler is based for 10 minutes at 13 gpm. This research established charac- upon three concepts. First, reduced sprinkler spacing teristics desired in the limited water supply (LWS) sprin- will result in sprinkler actuation when the fire size is klers, as ultimately required by NFPA 13D:24 smaller than that encountered at residential spacing.

Automatic Sprinkler Systems for Residential Occupancies Handbook 2007 242 Supplement 4 • Residential Sprinkler Systems

Second, reduced coverage areas will allow water flux ing designation. Remember that NFPA 13 in 1978 densities to be greater, even at lower flow rates, than permitted the combined use sprinkler system for heating those found in typical NFPA 13D residential systems. and cooling, as discussed in Chapter 3. (See Figure 3.6.) And third, discharge characteristics, such as ADD and However, combined use of sprinkler system piping had not wall wetting capability, should be comparable to or previously been allowed in residential sprinkler systems. greater than those of residential sprinklers, while drop Exhibit S4.12 illustrates 1⁄2-in. fittings for the sprinkler with size should be comparable to that of residential sprin- klers. [pp. 222–223] a minimum of three water flow paths and no dead ends.

Bill and Hill evaluated the prototype limited water Top view of fittings supply sprinkler in three full-scale fire tests in a 14 ft wide for ¹⁄₂-in. network sprinklers by 60 ft long manufactured home with 10 degree, sloped, 9-ft ceilings.25 Tenability for occupants was maintained with criteria identical to those used in the Los Angeles residential sprinkler system tests previously discussed in this chapter.14 All three fires were in the living room/ kitchen area. The sprinkler flow was 10 gpm for one sprin- kler and totaled 13 gpm for two sprinklers. These research results established the NFPA 13D limited-area dwelling sprinkler system requirements allowing sprinklers 6 ft apart Each fitting needs minimum of three separate paths. on a 10 degree, sloped ceiling. EXHIBIT S4.12 One-Half Inch Fittings for Network The 2002 edition of NFPA 13D allows the use of System Sprinklers with Multiple Waterflow Paths. 1 ⁄2-in. plastic or copper tubing in a network/multipurpose (Source: Residential Fire Safety Institute, ‘‘One-Half Inch piping sprinkler system design, which was first allowed in Piping Nearing Approval for Revision to NFPA 13D,’’ the 1996 edition. Puchovsky summarized conditions for Operation Life Safety Newsletter, Vol. 11, No.3&4,1996, 26 using 1⁄2-in. pipe in this design: p. 2. Reprinted by permission. www.firesafe home.org.)

In order to utilize 1⁄2 inch pipe a number of conditions must be met. The system must be hydraulically calcu- The 2002 edition of NFPA 13D deleted the whole lated in accordance with the provisions of NFPA 13, chapter on limited area dwelling sprinkler systems. How- and friction loss through fittings must be accounted ever, the multipurpose system design and the 1⁄2-inch plastic for. Hydraulic calculations must be prepared for each pipe with the gridded three-way water supply was retained sprinkler within the system flowing individually, and with the cross-linked polyethylene (PEX) plastic pipe as for each pair of sprinklers flowing within the same previously described in Chapter 5. The NFPA 13D* provi- compartment. The location of the most demanding sin- sions for this type of piping arrangement are as follows: gle sprinkler and pair of sprinklers along with their pressure and flow requirements need to be indicated 8.4.3.3* 1⁄2-in. (12.7-mm) nonmetallic pipe and 1⁄2-in. on the plan review documents. (12.7-mm) copper pipe along with listed special fittings In addition to verifying the system’s hydraulic shall be permitted to be used only in network systems 1 performance, supply piping in a system utilizing ⁄2 inch under the following conditions: pipe must be arranged so that each sprinkler is supplied (1)* Each sprinkler shall be supplied through a mini- by at least three separate flow paths with no supply mum of three separate paths from the supply lines terminating in a dead end. Methods for joining shutoff valve assembly within the dwelling unit. pipe must be addressed by product listings, and insert (2) Calculations shall clearly indicate the pipes that fittings are not to be installed between sprinkler fitting create the paths to each sprinkler. inlet ports. Piping to any plumbing fixtures is required (3) Sprinkler supply shall not terminate in a dead to be copper or to be listed. Finally, water sources are end. required to be potable or be equipped with a strainer (4) Hydraulic calculations shall be prepared for each at the connection to the supply line. [p. 18] sprinkler flowing individually within the system As significant as the use of 1⁄2-in. pipe is, probably and for each pair of sprinklers within the same more significant is the allowance of looped and gridded compartment. hydraulically calculated systems, thus the network designa- 27,28 tion. The sprinkler piping in this system may serve as *An asterisk accompanying a section number in a code extract indi- the cold water distribution network for residential plumb- cates that the requirement in that section is accompanied by nonman- ing to reduce installation costs, thus the multipurpose pip- datory explanatory material in Annex A of the code or standard.

2007 Automatic Sprinkler Systems for Residential Occupancies Handbook Supplement 4 • Residential Sprinkler Systems 243

(5) The location of the most demanding single sprin- 8.4.5.3 Where residential sprinklers are installed in a kler and pair of sprinklers, including their pres- compartment as defined in 3.3.6, all sprinklers within sure and flow requirements, shall be indicated the compartment shall be of the fast-response type that on the plan review documents. meets the criteria of 3.6.2 (a)(1). (6) The system shall be hydraulically calculated in accordance with the provisions of NFPA 13, The NFPA 13 Sprinkler Committee developed a very Standard for the Installation of Sprinkler Sys- important formal interpretation of Section 1-1 of NFPA tems, except that the friction loss straight through 13D, 1980 edition, and Section 3-16.2.9 of NFPA 13, 1983 a fitting shall be included. edition, in response to a request from the National Fire (7) The system shall be supplied from a potable Sprinkler Association.10 water source, or it shall be equipped with a strainer at the connection to the supply line. Question: Is NFPA 13D appropriate for use in multiple (8) The method of joining the pipe to fittings or to (three or more) attached dwellings under any condition? other pipe shall be covered by the listing. Answer: No. NFPA 13D is appropriate for use only (9) When inert tees are installed, each sprinkler shall in one- and two-family dwellings and mobile homes. have four separate paths from the water supply. Buildings which contain more than two dwelling units A maximum of one inert fitting shall be permitted shall be protected in accordance with NFPA 13. Section in each pipe section between sprinklers to serve 3-16.2.9 of NFPA 13 permits residential sprinklers to only domestic fixtures. be used in residential portions of other buildings pro- (10) The piping to the plumbing fixtures shall be of vided all other requirements of NFPA 13, including copper or listed pipe. water supplies are satisfied. Note: Building codes may contain requirements such Minnick reported on the design of sprinkler systems as 2-hour fire separations which would permit adjacent combined with the domestic plumbing system as installed dwellings to be considered unattached. [p. 109] in a planned community in DuPont, Washington.29 This design reduced sprinkler system costs by 40 to 60 percent, In the 1985 edition of NFPA 13, the sprinkler commit- to approximately $0.60 to $0.70 per sq ft. tee provided the option of installing a four-sprinkler design with residential sprinklers when hydraulically designing a sprinkler system for dwelling units and adjacent corridors. NFPA 13 ADAPTATIONS The 2002 edition of NFPA 13* includes this provision:

Residential Sprinkler Use in Other Occupancies 11.2.3.3 Room Design Method. Section 3-16.2.9 of the 1983 edition of NFPA 13 was first 11.2.3.3.1* The water supply requirements for sprin- to clarify the use of residential sprinklers in occupancies klers only shall be based upon the room that creates the greatest demand. other than one- and two-family dwellings and mobile 11.2.3.3.2 The density selected shall be that from Figure homes: 11.2.3.1.5 corresponding to the room size. Residential sprinklers may be used in residential por- 11.2.3.3.3 To utilize this method, all rooms shall be tions of any occupancy provided they are installed in enclosed with walls having a fire-resistance rating equal accordance with their listing and the positioning re- to the water supply duration indicated in Table quirements of NFPA 13D, Sprinkler Systems in One- 11.2.3.1.1. and Two-Family Dwellings. 11.2.3.3.4 If the room is smaller than the smallest area shown in the applicable curve in Figure 11.2.3.1.5, The 2002 edition of NFPA 13 continued this provision the provisions of 11.2.3.1.8(1) and 11.2.3.1.8 (2) shall with minor wording modifications: apply. 11.2.3.3.5 Minimum protection of openings shall be as 8.4.5.1* Residential sprinklers shall be permitted in follows: dwelling units and their adjoining corridors provided they are installed in conformance with their listing. (1) Light hazard—Non-rated automatic or self-closing 8.4.5.2 Residential sprinklers shall be used only in wet . systems unless specifically listed for use in dry systems (2) Light hazard with no opening protection—Where or preaction systems. openings are not protected, calculations shall in- clude the sprinklers in the room plus two sprinklers in the communicating space nearest each such un- *An asterisk accompanying a section number in a code extract indi- protected opening unless the communicating space cates that the requirement in that section is accompanied by nonman- has only one sprinkler, in which case calculations datory explanatory material in Annex A of the code or standard. shall be extended to the operation of that sprinkler.

Automatic Sprinkler Systems for Residential Occupancies Handbook 2007 244 Supplement 4 • Residential Sprinkler Systems

The selection of the room and communicating space sprinklers to be calculated shall be that which produces the greatest hydraulic demand. (3) Ordinary and extra hazard—Automatic or self- closing doors with appropriate fire-resistance rat- ings for the enclosure. Ӈ (a) 11.2.3.5.1* The design area shall be the area that in- cludes the four hydraulically most demanding sprin- klers. 11.2.3.5.2* Unless the requirements of 11.2.3.5.3 are met, the minimum required discharge from each of the four hydraulically most demanding sprinklers shall be the greater of the following: (1) In accordance with minimum flow rates indicated (b) in individual listings (2) Calculated based on delivering a minimum of 0.1 gpm/ft2 (4.1 mm/min) over the design area in ac- cordance with the provisions of 8.5.2.1 11.2.3.5.3 For modifications or additions to existing systems equipped with residential sprinklers, the listed (c) discharge criteria less than 0.1 gpm/ft2 (4.1 mm/min) shall be permitted to be used. [p. 102] When the residential sprinklers are utilized in the NFPA 13 adaptations, the minimum design discharge den- sity is 0.1 gpm/ft2 rather than 0.5 gpm/ft2 as in NFPA 13D or 13R systems. Exhibit S4.13 presents four examples of (d) the four-sprinkler, hydraulically designed area for dwelling EXHIBIT S4.13 Examples of the Four-Sprinkler Design units based on the provisions of NFPA 13 (2002). Area for Dwelling Units in Conformance with NFPA 13. Table S4.7 presents data from the Holmes’ comparison (Source: NFPA 13, 2002, Figure A.11.2.3.5.2.) of basic design features and philosophies of NFPA 13D and NFPA 13.30 Remember that the sprinkler committee developed the NFPA 13D residential sprinkler system as in the NFPA 13D standard for listed residential sprinklers a life safety system based on the Los Angeles dwelling to occupancies containing fuel configurations and fuel tests and the Charlotte mobile home tests. The tests in- loads not intended by the sprinkler committee. The appen- volved typical limited-occupancy fuel loads; rooms with dix of the 2002 edition of NFPA 13D contains wording fairly low, smooth, flat ceilings, and a corner-shielded fire that reflects Fleming’s concerns: source. Fleming commented on risks assumed when applying NFPA 13D system concepts to sprinkler systems Sprinkler systems designed and installed according to in occupancies other than one- and two-family dwellings, this standard are expected to prevent flashover within mobile homes, and dwelling units of other occupancies:20 the compartment of origin where sprinklers are installed within the compartment. A sprinkler system designed When 13D is applied to occupancies with fuel loads and installed according to this standard cannot, how- of geometric arrangements that differ from those of a ever, be expected to completely control a fire involving typical dwelling, it becomes more likely that the level fuel loads that are significantly higher than average for of protection provided by the system will not be appro- dwelling units [10 lb/ft2 (49 kg/m2)] and where the priate. [p. 55] interior finish has an unusually high flame spread rating Some communities and jurisdictions have extended (greater than 225). [p. 16] the concept of the residential sprinkler system developed Sprinkler systems were installed in residential occu- pancies in accordance with the NFPA 13 standard before *An asterisk accompanying a section number in a code extract indi- the NFPA 13D standard was developed in 1975. Innovative cates that the requirement in that section is accompanied by nonman- and specific designs were developed for individual build- datory explanatory material in Annex A of the code or standard. ings because of the perceived need to reduce the cost of

2007 Automatic Sprinkler Systems for Residential Occupancies Handbook Supplement 4 • Residential Sprinkler Systems 245

TABLE S4.7 Comparison of NFPA 13D Provisions and NFPA 13 Four-Sprinkler Design Provisions

Minimum Duration of Minimum Required Minimum Design Design Flow and Field Standard Scope Design Area Flow Testing

NFPA 13-85, Standard Any occupancy Four sprinklers for 52 gpm for sprinklers 30 minutes for light for the Installation of dwelling units.* (four sprinklers at 13 hazard occu- Sprinkler Systems Additionally, large gpm per sprinkler), pancy; 60, 90, and combustible and 100 gpm for 120 minutes for concealed spaces hose streams. higher hazard must be either Design calculations occupancy. sprinklered or filled subject to review by Certified entirely with non- authority having acceptance combustible jurisdiction. testing required insulation. (hydrostatic — 200 psi for 2 hr; etc.).

NFPA 13D-84, One- and two-family One or two most 1 sprinkler — 18 gpm; 10 minutes. Field Standard for the dwellings and remote sprinkler 2 sprinklers — 26 testing not Installation of mobile homes. heads within a gpm. No allowance required except for Sprinkler Systems in Numerous areas dwelling unit. for fire department visual observation One- and Two- within dwelling Sprinklering of hose streams. No of leaks, usually Family Dwellings units* are not concealed spaces stipulations for under domestic and Mobile Homes required to be not required. review of design water pressure. sprinklered. calculations by authority having jurisdiction.

*Dwelling unit: One or more rooms arranged for the use of one or more individuals living together as in a single housekeeping unit normally having cooking, living, sanitary, and sleeping facilities. This definition was added to NFPA 13-85 to clarify protection requirements for residential occupancies other than one- and two-family dwellings and mobile homes. Source: Craig A. Holmes, ‘‘NFPA 13D Residential Sprinklers vs. NFPA 13 Standard Sprinkler Systems — A Difference in Levels of Protection,’’ Building Standards, Vol. 14, No. 5, 1985, p. 12.

sprinkler systems for residential occupancies, which are NFPA 13D were modified to give the systems a discharge light hazard. density of 0.10 gpm/sq ft with a small orifice 7⁄16-in. sprin- A design for sprinkler protection of an eight-story kler. The average number of sprinklers installed in each high-rise office and residential building was proposed.31 building to protect the single means of egress (stairway) It combined a domestic cold-water system and sprinkler varied from 8 to 14 for the three-dwelling building to 15 system using a concept similar to the combined HVAC or 16 for the six-dwelling building. Exhibit S4.14 shows and sprinkler system discussed in Chapter 3 (see Figure the sprinklers’ arrangement in the first-floor entry and stair- 3.6). By designing a combined system according to NFPA way of a three-dwelling building. 13 provisions, Foehl reduced the costs from 2.7 percent to Another study involved retrofitting a residential sprin- 2.2 percent of total building costs. kler system in a two-story frame dwelling used as a board Foehl also studied the concept of providing retrofitted and care facility for eight residents.33 The system used in sprinklers in multifamily dwellings (apartment buildings) the design was a modification of the 1975 edition of NFPA in New York City to protect means of egress and common- 13D and the proposed 1980 edition of NFPA 13D. Because use areas.32 Buildings involved in these studies were frame, the facility was in an area without a water system, the on- row, multifamily dwellings, typically two and a half or site water supply was designed to deliver 800 gal (30 three stories high. Dwelling units per building varied from minutes as opposed to 10 minutes). The design was based three to six. The sprinkler systems were designed and on three sprinklers operating at 0.60 gpm/sq ft or one installed as partial systems, retrofitted into the existing sprinkler at 0.80 gpm/sq ft. Small orifice 3⁄8-in. sprinklers structures. Basic design criteria of the 1975 edition of were used instead of standard 1⁄2-in. orifice sprinklers. The

Automatic Sprinkler Systems for Residential Occupancies Handbook 2007 246 Supplement 4 • Residential Sprinkler Systems

maximum coverage per sprinkler was 144 sq ft instead of 256 sq ft, and 135F rated sprinklers were installed. One of the most successful programs for the installa- tion of residential sprinkler systems involving fire depart- ments, sprinkler manufacturers, and volunteers was initiated in the late 1980s and early 1990s for the Habitat for Humanity homes. Schirmer reports on a successful program in Pinehurst, North Carolina:34 Entry Where 21 of 57 existing or planned Habitat homes are already protected by residential sprinklers. This has been accomplished through the cooperative efforts of the Pinehurst Fire Department, Central Sprinkler Com- pany and the Viking Corporation; volunteers from the local Habitat for Humanity affiliates; the Carolinas Chapter of the American Fire Sprinkler Association; and local members of the Society of Fire Protection Engineers. [p. 37] Wolf has indicated the importance of the education aspects of installing residential sprinklers in the Habitat Homes in this statement from David Walencewicz:35 As Walencewicz points out, sprinkler houses require more communication. A lot of sprinkler contractors in. ³⁄₄ 45 ft aren’t willing to do residential, he says: It’s lower volume, lower profit—at least they view it that way. But we’ve sought it out, and we’ve been successful at it. Its different kind of work however, Bedroom #2 Bedroom #1 Living room Bath In a residential job, you have to educate builders and other tradespeople about what you are doing— you have to go through the effort. You have to educate the homeowner because it’s so new to them. With commercial jobs, you don’t have to go through the educational steps. [p. 87] Successful and continuing programs for the sprin- klering of Habitat for Humanity homes have been devel- oped and are continuing throughout the country.34,35,36 Dining room First-floor plan Kitchen Water Supply Source NFPA 13D* recognizes four types of water supply sources in the 2002 edition for a residential sprinkler system: 6.2* Water Supply Sources. The following water sup- ply sources shall be considered to be acceptable by this

19 ft 6 in. standard: EXHIBIT S4.14 Partial Sprinkler System Designed for (1) A connection to a reliable waterworks system with Protecting Means of Egress (Stairway and Entry) on First or without an automatically operated pump. Floor of Existing Multifamily Buildings. (Source: John M. (2) An elevated tank. Foehl, Fire Hardening of Old Residential Buildings in High (3) A pressure tank designed to American Society of Risk Urban Communities: Sprinkler System Aspects, Mechanical Engineers (ASME) standards for a TR80-6, Society of Fire Protection Engineers, Boston, MA, pressure vessel with a reliable pressure source. 1980, p. 8) *An asterisk accompanying a section number in a code extract indi- cates that the requirement in that section is accompanied by nonman- datory explanatory material in Annex A of the code or standard.

2007 Automatic Sprinkler Systems for Residential Occupancies Handbook Supplement 4 • Residential Sprinkler Systems 247

(4) A stored water source with an automatically oper- ated pump. [p. 9]

Gagnon reports on the residential sprinkler system installed with 57 sprinklers in a new two story 7354 ft2 residence during construction, which was hydrostatically tested, flow tested, and accepted by the authority having jurisdiction in April 2003.37 The two-story residence in western Howard County, Maryland, is in an area without a public water supply and is supplied water by a well. A waterflow test from the 185-ft deep drilled well showed that the well pump supplied 15 gpm for three hours. An American Society of Mechanical Engineers (ASME) pres- sure water tank and pressure source was therefore selected, as permitted by NFPA 13D. A 281-gal fiberglass water storage and pressure tank was installed on the ground floor level of the residence. Capacity of the tank was determined by hydraulic calcula- tions for two flowing Central 16 ft 16 ft (256 ft2) coverage quick-response residential horizontal sidewall sprinklers, K 3.5, at 0.05 gpm/ft2. The hydraulically most demanding area (based on three sets of calculations) indicated that 26.69 gpm was required with 63.98 psi at the pressure tank. NFPA 13D requires a 10-minute water supply, thus 266.9 gal was the minimum required. A 281- gal fiberglass water storage and pressure tank was selected and installed on the ground floor level of the residence. Exhibit S4.15 illustrates the 281-gal water/pressure tank as EXHIBIT S4.15 Fiberglass Water/Pressure Tank, installed, with the tank discharge pipe, waterflow indicator, Discharge Steel Pipe, Waterflow Indicator, and and the nitrogen cylinder cabinet. Cylinder Cabinet. (Source: Robert M. Gagnon) The nitrogen cylinder cabinet contains a 220 ft3 nitro- freezing temperatures. Exhibit S4.19 shows the waterflow gen cylinder. The needed cylinder capacity was determined horn and strobe light unit, which is central station moni- by multiplying the 281 gal in the water/pressure tank by tored. the system pressure required at the tank, 63.98 psi by a factor of 0.009, resulting in the nearest cylinder size of 220 ft3. The cylinder has a regulator that permits a system NFPA 13R pressure from 5 to 120 psi at the cylinder. The cylinder Standard Development cabinet has gauges for cylinder and system pressure, light- emitting diode (LED) status lights, and alarms for power A new residential sprinkler standard, NFPA 13R, Standard status, low water level, low nitrogen pressure, and low for the Installation of Sprinkler Systems in Residential battery. Exhibit S4.16 shows the nitrogen hose from the Occupancies up to and Including Four Stories in Height, cylinder cabinet to the top of the tank and the tank discharge was developed by the NFPA Sprinkler Committee, ap- pipe connection to the 1⁄2-in. chlorinated polyvinyl chloride proved at the 1988 Fall NFPA meeting, and issued in 1 (CPVC) sprinkler piping. The ⁄2-in. sprinkler piping was January 1989. NFPA 13R can be considered an extension verified by hydraulic calculation. Exhibit S4.17 shows the of the concepts of NFPA 13D and the residential sprinkler 1 ⁄2-in. CPVC sprinkler piping with the quick-response pen- concepts of NFPA 13 to include a broader range of occu- dent residential sprinklers on the lower level. Exhibit S4.18 pancies and applications. illustrates the quick-response horizontal residential sprin- The sprinkler committee developed NFPA 13R to pro- klers on the wall in the upper level. The horizontal residen- vide a consensus standard approach to protecting larger resi- tial sprinklers were selected and installed throughout the dential occupancies and to prevent the continued upper level to avoid having to install the sprinkler piping misapplication of NFPA 13D to these buildings by various in the attic area where it could possibly be subject to jurisdictions.

Automatic Sprinkler Systems for Residential Occupancies Handbook 2007 248 Supplement 4 • Residential Sprinkler Systems

Standard Requirements The new standard adopted provisions from both NFPA 13 and NFPA 13D. Piping and hanger requirements are identical to those in NFPA 13, and a 30-minute water supply is required in conformance with the light hazard occupancies in NFPA 13. Thus, the minimum water supply considered adequate to supply four residential sprinklers, at 13 gpm each (or 52 gpm) for 30 minutes, is 1560 gal. Residential sprinklers are mandated within the dwelling units. Areas exempted from sprinkler coverage within NFPA 13D are also exempted in NFPA 13R with specific modifications, including the NFPA 13D exemption of gar- ages. Other areas exempted from sprinkler protection not considered in NFPA 13 but exempted in 13R due to the difference in the occupancies are attics, penthouse equip- ment rooms, machine rooms, concealed spaces only for dwelling unit ventilation equipment, crawl spaces, elevator shafts, and concealed spaces not used or intended to be used for living areas or storage and have no fuel- fired equipment. NFPA 13R is similar to NFPA 13 and differs from NFPA 13D since it requires a Contractor’s Material and Test Certificate for Aboveground Piping, as illustrated in Exhibit S4.20. NFPA 13R is basically designed as a life safety system in the tradition of NFPA 13D and thus, provides less prop- erty protection than NFPA 13. The annex of NFPA 13R EXHIBIT S4.16 Nitrogen Hose from Cylinder Cabinet to recognizes this design concept in stating its basic intent: 1 Top of Tank, Tank Discharge Pipe Connection to ⁄2-in. CPVC Sprinkler Piping. (Source: Robert M. Gagnon) A.1.2 Various levels of sprinkler protection are avail- able to provide life safety and property protection. This standard is designed to provide a high, but not absolute,

EXHIBIT S4.17 CPVC Sprinkler Piping with Quick-Response Pendent Residential Sprinklers in Residence Lower Level and Supply Pipe to Upper Level. (Source: Robert M. Gagnon)

2007 Automatic Sprinkler Systems for Residential Occupancies Handbook Supplement 4 • Residential Sprinkler Systems 249

EXHIBIT S4.18 Quick- Response Horizontal Sidewall Residential Sprinklers in Resi- dence Upper Level. (Source: Robert M. Gagnon)

6.7.2* Design Criteria—Outside Dwelling Unit. 6.7.2.1 For areas outside the dwelling unit, the follow- ing criteria shall comply with the specifications in NFPA 13, Standard for the Installation of Sprinkler Systems, unless permitted by 6.7.2.2 and 6.7.2.3: (1) Design discharge (2) Number of design sprinklers (3) Sprinkler coverage (4) Position of sprinklers 6.7.2.2 The system demand of areas outside the dwell- ing unit shall be permitted to be limited to the number of sprinklers in the compartmented area but shall not be greater than the demand for a total of four sprinklers where all of the following conditions are met: (1) The area is compartmented into areas of 500 ft2 (46 m2) or less by 30-minute fire-rated construction. (2) The area is protected by quick-response or residen- tial sprinklers not exceeding 130 ft2 (12 m2) per sprinkler for ordinary hazard, 225 ft2 (20.9 m2) for light hazard, or the allowable coverage of the EXHIBIT S4.19 Horn/Strobe Local Waterflow Alarm for sprinkler listing. Residential Sprinkler Activation. (Source: Robert M. (3) Openings have a lintel at least 8 in. (203 mm) in Gagnon) depth. (4) The total area of openings does not exceed 50 ft2 (4.6 m2) for each compartment. level of life safety and a lesser level of property protec- (5) Discharge densities are in accordance with NFPA tion. Greater protection to both life and property could 13, Standard for the Installation of Sprinkler Sys- be achieved by sprinklering all areas in accordance with tems. NFPA 13, Standard for the Installation of Sprinkler 6.7.2.3 The following types of spaces are permitted to Systems, which permits the use of residential sprinklers be protected by residential sprinklers where they have in residential areas.

NFPA 13R* requires that spaces outside dwelling units *An asterisk accompanying a section number in a code extract indi- be protected with sprinklers meeting the NFPA 13 require- cates that the requirement in that section is accompanied by nonman- ments, with the following modifications: datory explanatory material in Annex A of the code or standard.

Automatic Sprinkler Systems for Residential Occupancies Handbook 2007 250 Supplement 4 • Residential Sprinkler Systems

Contractor’s Material and Test Certificate for Aboveground Piping

PROCEDURE Upon completion of work, inspection and tests shall be made by the contractor’s representative and witnessed by an owner’s representative. All defects shall be corrected and system left in service before contractor’s personnel finally leave the job. A certificate shall be filled out and signed by both representatives. Copies shall be prepared for approving authorities, owners, and contractor. It is understood the owner’s representative’s signature in no way prejudices any claim against contractor for faulty material, poor workmanship, or failure to comply with approving authority’s requirements or local ordinances. Property name Date

Property address

Accepted by approving authorities (names)

Address Plans Installation conforms to accepted plans Ye s No Equipment used is approved Ye s No If no, explain deviations

Has person in charge of fire equipment been instructed as Ye s No to location of control valves and care and maintenance of this new equipment? If no, explain

Instructions Have copies of the following been left on the premises? Ye s No 1. System components instructions Ye s No 2. Care and maintenance instructions Ye s No 3. NFPA 25 Ye s No Location of Supplies buildings system Year of Orifice Temperature Make Model manufacture size Quantity rating

Sprinklers

Pipe and Type of pipe fittings Type of fittings

Maximum time to operate Alarm Alarm device through test connection valve or Type Make Minutes Seconds flow Model indicator

Dry valve Q. O. D. Make Model Serial no. Make Model Serial no.

Dry pipe Time to trip Time water Alarm operating through test Water Air Trip point reached operated † † test connection pressure pressure air pressure test outlet properly Minutes Seconds psi psi psi Minutes Seconds Ye s N o Without Q.O.D. With Q.O.D. If no, explain

† Measured from the time inspector’s test connection is opened EXHIBIT S4.20 Sample Contractor’s Material and Test Certificate for Aboveground Piping. (Source: NFPA 13R, 2002, Figure 6.2.2)

2007 Automatic Sprinkler Systems for Residential Occupancies Handbook Supplement 4 • Residential Sprinkler Systems 251

Operation Pneumatic Electric Hydraulic Piping supervised Yes No Detecting media supervised Yes No

Does valve operate from the manual trip, remote, or both Yes No control stations? Deluge and If no, explain preaction Is there an accessible facility in each circuit for testing? valves Yes No Does each circuit operate Does each circuit operate Maximum time to Make Model supervision loss alarm? valve release? operate release Yes No Yes No Minutes Seconds

Location Make and Setting Static pressure Residual pressure Flow rate Pressure- and floor model (flowing) reducing Inlet (psi) Outlet (psi) Inlet (psi) Outlet (psi) Flow (gpm) valve test

Hydrostatic: Hydrostatic tests shall be made at not less than 200 psi (13.6 bar) for 2 hours or 50 psi (3.4 bar) above static pressure in excess of 150 psi (10.2 bar) for 2 hours. Differential dry-pipe valve clappers shall be left open during the test to prevent damage. All aboveground piping leakage shall be stopped. Test description Pneumatic: Establish 40 psi (2.7 bar) air pressure and measure drop, which shall not exceed 1¹⁄₂ psi (0.1 bar) in 24 hours. Test pressure tanks at normal water level and air pressure and measure air pressure drop, which shall not exceed 1¹⁄₂ psi (0.1 bar) in 24 hours.

All piping hydrostatically tested at psi ( bar) for hours If no, state reason Dry piping pneumatically tested Yes No Equipment operates properly Yes No Do you certify as the sprinkler contractor that additives and corrosive chemicals, sodium silicate or derivatives of sodium silicate, brine, or other corrosive chemicals were not used for testing systems or stopping leaks? Yes No

Drain Reading of gauge located near water Residual pressure with valve in test Tests test supply test connection: psi ( bar) connection open wide: psi ( bar) Underground mains and lead-in connections to system risers flushed before connection made to sprinkler piping verified by a copy of the “Contractor’s Material and Test Certificate for Underground Piping.” Yes No Other Explain

Flushed by installer of underground sprinkler piping Yes No

If powder-driven fasteners are used in concrete, Yes No If no, explain has representative sample testing been satisfactorily completed?

Blank testing Number used Locations Number removed gaskets

Welding piping Yes No If yes . . .

Do you certify as the sprinkler contractor that welding procedures comply Yes No with the requirements of at least AWS B2.1?

Do you certify that the welding was performed by welders qualified in compliance with the requirements of at least AWS B2.1? Yes No Welding Do you certify that the welding was carried out in compliance with a documented quality control procedure to ensure that all discs are retrieved, that openings in piping are smooth, that slag and other welding residue are removed, and that Yes No the internal diameters of piping are not penetrated?

Cutouts Do you certify that you have a control feature to ensure that Yes No (discs) all cutouts (discs) are retrieved?

EXHIBIT S4.20 (Continued)

Automatic Sprinkler Systems for Residential Occupancies Handbook 2007 252 Supplement 4 • Residential Sprinkler Systems

Hydraulic Nameplate provided If no, explain data nameplate Yes No

Date left in service with all control valves open Remarks

Name of sprinkler contractor

Tests witnessed by Signatures For property owner (signed) Title Date

For sprinkler contractor (signed) Title Date

Additional explanations and notes

EXHIBIT S4.20 (Continued)

flat, smooth ceilings not exceeding 10 ft (3.0 m) in facilities; large, ‘‘slow’’ board and care facilities; and large, height and are protected in accordance with the require- ‘‘impractical’’ board and care facilities. Obviously, the one- ments for residential sprinklers: and two-family dwellings are covered by NFPA 13D and (1) Lobbies not in hotels and motels the excluded board and care facilities are covered by NFPA (2) Foyers 13. (3) Corridors The 2006 edition of NFPA 101, Life Safety Code, (4) Halls requires the provision of an approved automatic sprinkler (5) Lounges system in all new one- and two-family dwellings for the (6) Other areas with fire loads similar to residential first time. The provisions in Chapter 24 are presented as NFPA 13R provides requirements for sprinkler protec- follows: tion for residential occupancies beyond the scope of NFPA 13D for one- and two-family dwellings and manufactured 24.3.5.1 All new one- and two-family dwellings shall homes. The sprinkler committee intended that NFPA 13R be protected throughout by an approved automatic sprinkler systems be installed in these residential occupanc- sprinkler system in accordance with 24.3.5.2. ies: 24.3.5.2 Where an automatic sprinkler system is in- stalled, either for total or partial building coverage, the 3.3.6 Residential Occupancies. Occupancies, as speci- system shall be in accordance with Section 9.7; in fied in the scope of this standard, that include the fol- buildings up to and including four stories in height, lowing, as defined in NFPA 101, Life Safety Code: systems in accordance with NFPA 13R, Standard for (1) Apartment buildings, the Installation of Sprinkler Systems in Residential Oc- (2) Lodging and rooming houses, cupancies up to and Including Four Stories in Height, (3) Board and care facilities (prompt and slow evacua- and with NFPA 13D, Standard for the Installation of tion type), Sprinkler Systems in One- and Two-Family Dwellings (4) Hotels, motels, and dormitories. [p. 6] and Manufactured Homes, shall also be permitted. It should be recognized that occupancies considered The Services Office, Inc. (ISO) proposed residential in NFPA 101 have been excluded from NFPA an abbreviated evaluation method for sprinkler systems 13R definitions. These specific occupancies are one- and designed according to NFPA 13R for a maximum credit two-family dwellings; small, ‘‘impractical’’ board and care of 25 percent from the commercial fire rating schedule.38

2007 Automatic Sprinkler Systems for Residential Occupancies Handbook Supplement 4 • Residential Sprinkler Systems 253

The water supply must meet the minimum total supply of RESIDENTIAL SPRINKLER SYSTEM 1560 gal (52 gpm for 30 minutes). Residential sprinklers INCENTIVES are recognized in residential areas, and nonresidential areas Fatality and Injury Reductions should meet NFPA 13 provisions. The ISO recognizes NFPA 13R as a substandard (not an NFPA 13) system but A principal incentive for installing residential sprinkler still grants protection credit. systems is the potential reduction in loss of human life and The 2002 edition of NFPA 13R* established sprinkler injuries. Installation of NFPA 13D sprinkler systems with protection requirements for the first time for garages, vary- smoke detectors can reduce fatalities in major fires by 73 ing with the configuration arrangement to the dwelling percent.42 Ruegg and Fuller estimated the probability of the units as follows: deaths and injuries in one- and two-family homes relative to the use of smoke detectors and residential sprinkler systems 6.7.3 Design Criteria—Garages. based on the 1980 edition of NFPA 13D.43 These probabili- 6.7.3.1 Garages that are completely separated from ties, presented in Table S4.8, indicate that adding residen- the residential portion of the building by fire resistive tial sprinklers and smoke detectors to a dwelling that had construction sufficient to have them considered as sepa- neither cuts the fire death rate by 82 percent, while adding rate buildings under the local code shall be protected in accordance with NFPA 13, Standard for the Installa- residential sprinklers to a dwelling that already had smoke tion of Sprinkler Systems. detectors cuts the fire death rate by an additional 63 percent. 6.7.3.2 Garages that are accessible by people from more AsTableS4.9demonstrates,thesewere33reportedsuc- than one dwelling unit, and are not covered by 6.7.3.1, cessful operations of residential sprinkler systems in four shall be considered part of the building and shall be and one-half years that saved a total of 34 lives.44 Exhibit protected in accordance with 6.7.2. Garage doors shall S4.21 presents the Operation Life Safety record of 551 suc- not be considered obstructions and shall be permitted to cessful sprinkler activations of NFPA 13D or 13R systems be ignored for placement and calculation of sprinklers. from 1983 to July 31, 1995, with the occupancies involved.3 6.7.3.3* Garages that are only accessible from a single dwelling unit shall be considered as part of that dwell- ing unit. Such garages shall be sprinklered with residen- Loss and Insurance Savings tial sprinklers in accordance with 6.7.1 or quick- Ten valid residential sprinkler system tests were conducted response sprinklers designed to produce a density of 45 2 in Scottsdale, Arizona. Systems conforming to the 1980 0.05 gpm/ft (2.04 mm/min) over the area of the garage, edition of NFPA 13D installed in a one-family dwelling but not to exceed four sprinklers. Garage doors shall reduced fire loss 85 percent during these tests. Ford has not be considered obstructions and shall be permitted to be ignored for placement and calculation of sprinklers. presented the fire loss record in Scottsdale, Arizona, from 1986, when their residential sprinkler system ordinance There is a significant difference in the requirements was implemented, to 1995 as follows:46 for sprinklers in garages between NFPA 13D, in which the garages are exempted, and NFPA 13R, as indicated in the As of January 1, 1996, 19,649, or 15 percent of Scotts- dale single-family homes were sprinklered, as were 2002 versions of both documents. Brown indicated that 13,938 or 49 percent, of the city’s multifamily homes. many communities have added local provisions for ex- Between 1986 and 1995, residential sprinklers activated tending the NFPA 13D system to protect the attached gar- in 44 of the 589 home fires that occurred in Scottsdale. ages. These installations are more economical for structures Forty-one of these fires were controlled or contained located in milder climates where protection of the wet pipe by one or two sprinklers. Two of the three fires in residential sprinkler system from freezing is not a necessary which more sprinklers were needed to extinguish the consideration.39 Lawrence reviewed the changes initiated blaze were flammable liquid arson fires. in the 2002 edition of NFPA 13R for garages.40 Kelly No one died in these 44 fires. If the death rate per compared the requirements for sprinklers in parking gar- fire in the sprinklered homes had matched the rate in ages and repair garages in NFPA 13 and the residential the unsprinklered homes, however, the fatalities that garages in NFPA 13R as presented in the foregoing ex- did occur in home fires during that period might have nearly doubled. As it was, 10 people died in 8 fires, all tract.41 in unsprinklered single-family homes. Smoke detectors had been installed in seven of the homes, and at least four of the detectors were working. The high-risk *An asterisk accompanying a section number in a code extract indi- groups for fire safety were well-represented: three of cates that the requirement in that section is accompanied by nonman- the victims were elderly, two were impaired, and two datory explanatory material in Annex A of the code or standard. were children.

Automatic Sprinkler Systems for Residential Occupancies Handbook 2007 254 Supplement 4 • Residential Sprinkler Systems

TABLE S4.8 Estimated Impact of Sprinklers on Fire Deaths and Injuries in One- and Two-Family Houses

Average Predicted Predicted Predicted Predicted Reported Fires Deaths and Deaths and Deaths and Deaths and Deaths and and Losses in Injuries per Fire Injuries per Fire Injuries per Fire Injuries per Fire Injuries per Fire One- and Two- Given Existing Assuming Assuming Assuming Assuming Family Houses Use of Fire Neither Smoke Sprinklers but Smoke Smoke in the U.S. in Protection Detectors nor Not Detectors but Detectors and Type of Event 1981a Devicesc Sprinklersd Detectorsd,e Not Sprinklersd Sprinklerse,f

Fires in one- 522,175 and two- family houses Civilian deaths 3,895b 0.00746 0.00821 0.00253 0.00390 0.00146 Civilian injuries 13,851 0.02653 0.02676 0.01446 0.02546 0.01436

aFire in the United States, 4th ed., Federal Emergency Management Agency, Washington, DC, September 1982. There were an estimated 6.1 million one- and two-family houses in the U.S. in 1981, such that the frequency of reported fires for this type of dwelling was 522,175 0.0856 6,100,000 bThe national estimate was reduced by 6 percent to exclude estimated unreported deaths in order to base the analysis on reported data. cThe rate of deaths and injuries per fire in the reported fire data reflects an existing use of smoke detectors by approximately two in three households on the average; the rate is found by dividing the number of deaths or injuries by the number of fires in 1. dBased on simulated test data from the NBS/SRI Fire Loss Model. See A. Gomberg et al., A Decision Model for Evaluating Residential Fire-Risk Reduction Alternatives, NBSIR, 1984. eThe data reflect the assumption of a 0.08 probability that the sprinkler system will not operate effectively due to improper installation or maintenance. fThe data reflect the assumption of a 0.153 probability that the smoke detector will be nonfunctional at a given time, based on the results of the field survey. Source: Rosalie T. Ruegg and Sieglinde K. Fuller, A Benefit-Cost Model of Residential Fire Sprinkler Systems, NBS Technical Note 1203, National Bureau of Standards, Gaithersburg, MD, 1984, p. 71.

Reported activations EXHIBIT S4.21 NFPA 13D and 345 13R Successful Sprinkler Acti- 350 vations from 1983 to July 31, 315 1995, with the Occupancies In- 280 volved. (Source: Residential Fire 245 Safety Institute, ‘‘Residential Sprinkler Activations—Annual 210 Report,’’ Operation Life Safety 175 Newsletter, Vol. 10, No. 9, 1995, 140 p. 2. Reprinted by permission. 105 www.firesafehome.org) Number of activations 73 70 37 34 32 20 35 10 0 Multi- Single- Dorm Motel/ Rooming Nursing Others family family hotel home Types of Occupancies

2007 Automatic Sprinkler Systems for Residential Occupancies Handbook Supplement 4 • Residential Sprinkler Systems 255

TABLE S4.9 Reported Successes of Residential Sprinkler Systems from October 1983 to March 1988

Fire Loss Number of Lives Location Date Structure Room of Origin ($) Sprinklers Saved

Cobb Co., GA 0/10/83 Apartment Kitchen 300.00 one 1 Cobb Co., GA 12/8/83 Motel Kitchen 400.00 one 1 Cobb Co., GA 6/29/84 Apartment Porch 2,000.00 one 0 Washington, DC 1/29/85 Rooming Living room 1,000.00 one 10 Cobb Co., GA 4/2/85 Apartment Bedroom 400.00 one 4 Greenburgh, NY 11/1/85 Single-family Kitchen Minimal two 0 Dover, NH 12/28/85 Condominium Basement Minimal one 0 Cobb Co., GA 5/3/86 Apartment N/R 2,000.00 one 0 Charlotte, NC 6/16/86 Apartment Bathroom 8,000.00 one 0 Aiken, SC 7/14/86 Apartment Kitchen Minimal one 0 Ft. Myers, FL 9/4/86 Apartment Living room 1,500.00 one 0 Barker, CO 11/7/86 Single-family Chimney 0.00 one 0 Cobb Co., GA 11/21/86 Motel Kitchen 200.00 one 1 Anne Arundel Co., MD 11/25/86 Home Care N/R Minimal one 0 Cobb Co., GA 12/13/86 Apartment Kitchen 2,000.00 one 1 San Clemente, CA — Single-family Kitchen — one 0 Cobb Co., GA 4/8/87 Apartment Kitchen 10,000.00 one 0 Cobb Co., GA 4/9/87 Apartment Kitchen 9,500.00 one 0 Altamonte Spr., FL 4/12/87 Apartment Closet 5,000.00 one 0 Scottsdale, AZ 4/13/87 Condominium Kitchen 200.00 one 0 Cobb Co., GA 4/18/87 Apartment Dining room Minimal one 0 Cobb Co., GA 4/18/87 Apartment Kitchen Minimal one 1 Cobb Co., GA 4/26/87 Apartment Kitchen Minimal one 0 Aberdeen, MD 5/8/87 Apartment Living room 5,000.00 one 1 Elmsford, NY 5/27/87 Single-family Kitchen 350.00 one 0 Lane Co., OR 6/8/87 Apartment Living room 1,500.00 one 0 Cobb Co., GA 8/7/87 Apartment Living room Minimal two 3 Cobb Co., GA 8/7/87 Apartment Living room Minimal one 2 Cobb Co., GA 8/20/87 Apartment Kitchen Minimal one 3 Orange Co., GA 11/30/87 Single-family Kitchen Minimal one 3 Largo, MD 12/9/87 Apartment Living room Minimal two 0 Marietta, GA 1/14/88 Apartment Kitchen Minimal two 2 Coon Rapids, MN 1/20/88 Apartment Kitchen Minimal one 1

Source: Paul Teague and Jim Pothier, ‘‘Adoption by the Model Groups of Residential Sprinkler Requirements Will Make the Inclusion of Such Provisions in Local Building Codes More Likely,’’ Fire Journal, Vol. 82, No. 5, 1988, p. 59; courtesy of Operation Life Safety.

The estimated amount of water discharged when a piping. Second, sprinkler systems may be uneconomi- residential sprinkler system activated was 209 gallons, cal for homeowners who purchase them independently as opposed to an estimated 3,690 gallons released by and whose probability of fire and risk of death, injury, extinguishing a house fire with hoses. The and property loss are low to average. However, they average loss per incident was lower, too: $1,544 in must be cost-effective for homeowners who are subject sprinklered properties as opposed to $11,624 in unsprin- to a higher risk of fire and are in greater peril of death klered properties. [p. 42] and injury if fire occurs. Ruegg and Fuller examined the cost-effectiveness of It may also be inferred that sprinkler systems are more likely to be cost-effective for homeowners who a residential sprinkler system and concluded:47 are part of a community of sprinkler users and who First, residential sprinklers are more cost-effective receive the benefits of cost avoidance by local govern- when code changes allowed the use of approved plastic ments or the home builder. [p. 117]

Automatic Sprinkler Systems for Residential Occupancies Handbook 2007 256 Supplement 4 • Residential Sprinkler Systems

Insurance premium reductions traditionally have been areas.49 Dalton compared insurance premium savings for a factor in the cost-effectiveness of NFPA 13 sprinkler sprinklered and nonsprinklered properties in Montgomery systems. In many states the ISO has insurance rate filings County, Maryland.50 Table S4.10 presents data from the that allow a 5 percent discount on a homeowners policy report involving multifamily garden apartment buildings for a dwelling entirely covered by a sprinkler system.48 (three or four stories), back-to-back (three sto- ISO also allows a 2 percent discount for a system that ries), separated townhouses, and single-family dwellings. meets the requirements of the 1975 edition of NFPA 13D Ford indicated the insurance discounts in Scottsdale de- without closet and bathroom coverage. pended on the design of the system and varied from 5 Following the report of the Ad Hoc Committee on to 45 percent. The higher discounts required additional Residential Sprinklers, ISO permitted discounts of 15 per- features, such as system monitoring, fire extinguishers, cent for insuring a dwelling completely protected by a smoke detectors, and deadbolt locks. Most insurers in residential sprinkler system and an 8 percent discount for Scottsdale appeared to offer some insurance discount, on an NFPA 13D system with the permitted nonsprinklered average about 10 percent, for an approved residential sprin-

TABLE S4.10 Insurance Premium Reductions in Residential Buildings in Montgomery County, Maryland

Multifamily Garden Apartments — 548 Units (30% savings) Insurance Premiums Insurance Premiums Unsprinklered $150,000 Sprinklered $105,000 per Year per Year

1 yr 5 yr 10 yr 15 yr 20 yr Per unit ($) 82 410 821 1,232 1,642 Per project ($) 45,000 225,000 450,000 675,000 900,000

Back-to-Back Townhouses — 14 Units, Three-Story Frame (28% savings) Insurance Premiums Insurance Premiums Unsprinklered $6,000 per Sprinklered $4,300 per Year Year ($428.57 unit/yr) ($307.14 unit/yr)

1 yr 5 yr 10 yr 15 yr 20 yr Per unit ($) 121 607 1,214 1,821 2,429 Per project ($) 1,170 8,500 17,000 25,000 34,000

Townhouses — 72 Units, Three-Story Frame (36% savings) Insurance Premiums Insurance Premiums Unsprinklered $18,000 per Sprinklered $11,500 per Year Year ($428.57 unit/yr) ($159.72 unit/yr)

1 yr 5 yr 10 yr 15 yr 20 yr Per unit ($) 90 451 903 1,354 1,806 Per project ($) 6,500 32,500 65,000 97,500 130,000

Single Family (25% savings) Insurance Premiums Insurance Premiums Unsprinklered $410 per Year Sprinklered $306 per Year

1 yr 5 yr 10 yr 15 yr 20 yr Savings per unit ($) 104 520 1,040 1,560 2,080

Source: Jim Dalton, ‘‘Insurance Reductions Support Residential Fire Sprinklers,’’ Operation Life Safety Newsletter, Vol. 2, No. 8, 1987, p. 9.

2007 Automatic Sprinkler Systems for Residential Occupancies Handbook Supplement 4 • Residential Sprinkler Systems 257

kler system.46 Skinner indicated in his survey of six insur- in Table S4.11.52 Siarnicki reported that as of January 1, ance companies—Allstate, Hartford, Liberty Mutual, 1992, all new residential structures (including one- and Prudential, State Farm, and Kemper—that the discounts two-family homes) were required to have residential sprin- for a residential sprinkler system varied from 5 to 15 per- kler systems. An analysis of the performance of the residen- cent based on the homeowner’s total premium.51 tial sprinkler systems in the period from January 1, 1992, through December 31, 1999, is presented as follows:53 Incentives and Tradeoffs Residential sprinkler system reported 121 Coughlin identified a total of 35 financial incentives that incidents benefited the developer, the builder, the installer, and the Residential sprinkler system reported 117 homeowner when a residential sprinkler system was in- fire incidents stalled. The distribution of these incentives are presented Residential sprinkler activations 143

TABLE S4.11 Benefits to Developers, Builders, Installers, and Owners of Automatic Sprinkler Systems

Incentive Developer Builder Installer Owner

Reduced impact fees X X — — Low-cost loans X X — X Increased density (narrower setbacks) X — — — Reduced fire flow requirements (smaller main size, X — — — less storage, less maintenance) More space between hydrants X — — — Longer access road distance X — — — Longer distance from fire stations X — — X Reduced access to building sides X — — — Narrower streets X — — — Fewer parking restrictions X — — — Longer cul-de-sacs X — — — Reduced radius of turnarounds at dead ends X — — — Reduced permit fees — X X — Reduced or exempted plan review fees — X X — Reduced or exempted fees for field inspections — X X — Reduced fire resistance ratings, no parapet walls — X X — Increased distance to exits — X — — Single water line for domestic and sprinkler system — — X — No separate meter for sprinkler system — — X — No fee increase for larger meter — — X — No special connection charge — — X — Check valves instead of low-pressure-principle — — X — backflow preventer Lower insurance premiums (fire portion for home- — X — — owners, fire and liability for building owners) Lower ISO rating (primarily applies to buildings, — — X — does not affect 1-family homes when rating drops below 7) Reduction in annual fire service assessment — — — X Property tax reductions for sprinklers — — — X No meter rental or monthly maintenance fees from — — — X water purveyors

Source: Pat Coughlin, ‘‘Incentives a Powerful Tool for Winning Fire Sprinkler Ordinances,’’ Sprinkler Age (official publication of the American Fire Sprinkler Association), Vol. 19, No. 1, 2000, p. 23.

Automatic Sprinkler Systems for Residential Occupancies Handbook 2007 258 Supplement 4 • Residential Sprinkler Systems

Total fire loss $401,220 pression resources are utilized in the manual fire suppres- Potential fire loss $38,230,000 sion comparisons. Table S4.12 presents the comparison of Reported lives saved 154 the resources for the NFPA 13D residential sprinkler sys- Injuries reported (all minor in nature) 7 [p. 2] tem and the manual fire suppression activities at a single- Jelenewicz summarized a variety of studies relative family manufactured home fire. Table S4.13 presents this to the benefits and costs of residential sprinkler systems comparison at a single-family dwelling fire. not only in the United States but also in some other coun- In addition to insurance rate reductions and premium tries. He indicated that over a seven-year period after a credits, government offices can provide financial incen- residential sprinkler ordinance was implemented in the city tives to promote installation of residential sprinkler sys- of Vancouver, British Columbia, the fire death rate declined tems. The state of Alaska has provided the longest and by 69 percent, whereas the fire death rate declined by 42 most effective support with a tax incentive. Since January percent for the same period in the rest of Canada. In New 1, 1981, Alaska has exempted 2 percent of the assessed Zealand one study estimated a 72 percent reduction in the value of any structure with a sprinkler system from taxa- fire death rate with sprinklers in the homes. Another New tion. This law applies the term ‘‘structure’’ to dwellings Zealand study estimated that 100 lives would be saved in and manufactured homes. 30 years with residential sprinkler systems. The current Model building codes have provided incentives for annual fire death rate in New Zealand is approximately 18 installing residential sprinkler systems through the concept persons per year.54 of tradeoffs. Tradeoffs allow reductions in certain speci- Davis compared the value of a residential sprinkler fied requirements regarding access to, or construction of, system in the reduction of the time for fire development a building when a sprinkler system is installed. Viniello and the resources needed with the time for fire department in Table S4.14 summarizes some typical tradeoffs and response and manual suppression activities.55 Davis pre- building code modifications for residential sprinkler sys- sented an estimate of the manual fire suppression reflex tems. time in Exhibit S4.22 and an estimate of the NFPA 13D The Uniform Building Code was the first model build- ing code to require sprinklers in low-rise residential occu- residential sprinkler system reflex time in Exhibit S4.23. 56 He presented the comparison between the resources identi- pancies. The 1988 edition of the code required fied with the rate of water application, the time of applica- 1. Sprinkler systems in hotels that are three or more tion, the total water supply needed, the number of 125 gpm stories high or have 20 or more guest rooms. hose lines needed, the number of fire fighters needed, 2. Sprinkler systems in apartment buildings that are and the number of engines required for suppression. Four three or more stories high or have more than 15 nationally recognized standards for the manual fire sup- dwelling units.

Free Permitted Transmission Alarm Turn-Out Travel Set-Up Control Overhaul EXHIBIT S4.22 Manual Fire Burning Burn Time Handling Time Time Time Time Time Suppression Reflex Time. Time Time Time 12345678910(Source: Larry Davis, ‘‘First Re- Ignition Recognition Detection Notification Alert Get-out Arrival Attack Black-out Extinguishment sponders Residential Sprinklers Point Point Point Point Point Point Point Point Point Can Douse Fires Long Before Rescuers Arrive on Scene,’’ Fire Reflex Time Rescue Magazine, Vol. 21, No. 5, 2003, p. 102)

Free Permitted Transmission Alarm Tur n-Out Travel Set-Up and Overhaul EXHIBIT S4.23 Automatic Burning Burn Time Handling Time Time Time Sprinkler Reflex Time. (Source: Time Time Time 1234567 10Larry Davis, ‘‘First Responders Ignition Recognition Detection Notification Alert Get-out Arrival Extinguishment Residential Sprinklers Can Point Point Point Point Point Point Point Point Douse Fires Long Before Rescu- ers Arrive on Scene,’’ Fire Res- Control Overhaul Time Time cue Magazine, Vol. 21, No. 5, 8910 2003, p. 103) Attack Black-out Extinguishment Reflex Time Point Point Point

2007 Automatic Sprinkler Systems for Residential Occupancies Handbook Supplement 4 • Residential Sprinkler Systems 259

TABLE S4.12 Comparison of Resources Required for Manual Fire Suppression with an NFPA 13D Sprinkler System at a Single-Family Manufactured Home Fire

Total Fire Rate of Time Water 125-gpm Fighters Pumpers Application Frame Supply Handlines Required Required Method (gpm) (min) (gal) Required for Attack* for Attack*

NFPA 13D 10 10 100 0 0 0 Iowa ROF 60 .5 30 1 2 1 NFA NFF 250 — — 2 4 1 NFPA 1231 250 8 2,000 2 4 1 ISO 500 120 60,000 4 8 2

*Assumption for comparison purposes 2 fire fighters per 125-gpm handline, 250-gpm flow from each pumper. Source: Larry Davis, ‘‘First Responders Residential Sprinklers Can Douse Fires Long Before Rescuers Arrive on Scene,’’ Fire Rescue Magazine, Vol. 21, No. 5, 2003, p. 104.

TABLE S4.13 Comparison of Resources Required for Manual Fire Suppression with a NFPA 13D Sprinkler System at a Single-Family Dwelling Fire

Total Fire Rate of Time Water 125-gpm Fighters Pumpers Application Frame Supply Handlines Required Required Method (gpm) (min) (gal) Required* for Attack* for Attack*

NFPA 13D 26 10 260 0 0 0 Iowa ROF 1900 .5 950 8 24 4 NFA NFF 5835 — — 24 72 12 NFPA 1231 1000 23 23,000 4 12 2 ISO 500 120 60,000 2 6 1

*Assumption for comparison purposes 3 fire fighters per 250-gpm handline, 500-gpm flow from each pumper. Source: Larry Davis, ‘‘First Responders Residential Sprinklers Can Douse Fires Long Before Rescuers Arrive on Scene,’’ Fire Rescue Magazine, Vol. 21, No. 5, 2003, p. 105.

3. Quick-response or residential sprinklers in guest Incentives provided in the model codes for the installa- rooms of hotels and in dwelling units of apartment tion of sprinkler systems in buildings are listed below:56 buildings. [p. 7] Construction and fire protection incentives: Endthoff57 reported that the 1991 edition of NFPA Height and area increases 101 required installation of sprinklers in most new hotels, Construction type of corridors and tenant separa- motels, and dormitories. Remember that if these occupan- tions cies have four or less stories, they may be sprinklered with Interior and floor finish an NFPA 13R design. Occupancies higher than four stories Travel distance to exits Exit width may be sprinklered with a NFPA 13 design using the room Standpipe requirements, hose stations, and water design concept previously explained in this chapter. The flow 1994 edition of NFPA 101 mandated application of quick- Fire detection systems response or residential sprinklers in the guest rooms of Draft stopping in attic spaces these occupancies. This requirement is maintained in the Typical site development incentives: 2006 edition. Fewer fire hydrants with greater spacing in between

Automatic Sprinkler Systems for Residential Occupancies Handbook 2007 260 Supplement 4 • Residential Sprinkler Systems

TABLE S4.14 Typical Building Code Access Tradeoffs and Modifications for Buildings with Residential Sprinkler Systems

Access Tradeoffs with Without Residential Fire Protection Residential Fire Protection Savings/Advantage

Dwelling units must be within 5-min response of nearest fire Waived Increased development station potential Dwelling units must not be located more than 150 ft from Waived Maximum use of site closest vehicular access way Standard street widths required Reductions permitted Reduced cost Street grade of 15 percent or greater not permitted Waived Reduced cost Must adhere to hydrant spacing requirements Greater hydrant spacing Reduced cost permitted Fire flow requirements must be met 50 percent reduction in Smaller water mains required fire flow Hazardous areas — 75 dwelling units for single means of Number may increase Increased density ingress or egress Nonhazardous areas — 150 dwelling units for single means Number may increase Increased density of ingress or egress Access for ladder trucks — 50 ft from rear of parking stalls Extended to 75 or 100 ft Allows maximum use of to face of unit site and landscaping When any portion of a building is in excess of 150 ft from a Waived Reduced cost water supply, it may be required that onsite fire hydrants and mains capable of supplying the required water flow be installed Maximum tenant travel to a fire exit must not exceed 20 ft Exits may exceed the Design freedom maximum required travel for a tenant with one exit exceeding 15 ft for a maximum travel of 35 ft Interior finish must be Class A Interior finish may be Class C Design freedom or better No can be within 15 ft of other tenants’ travel Windows may be placed Design freedom anywhere desired

Building Code Modifications 5 1 Tenant separation walls must be made of ⁄8-in. Type X fire- ⁄2-in. standard gypsum $23.00/1,000 sq ft rated gypsum wallboard wallboard may be substituted 5 3 1 ⁄8-in. Type X rated gypsum wallboard and ⁄4-in. plywood ⁄2-in. standard gypsum $55.00/1,000 sq ft 5 must be used in 1-hr floor/ceiling assemblies wallboard and ⁄8-in. plywood may be substituted in floor/ceiling assemblies Fire stops are required in the attic at every unit Fire stops in attic required $123.00 each fire stop every 3,000 sq ft Tenants may not travel by other tenants’ doors unless those Standard doors and frames $111.70 each door 3 doors are ⁄4-hr fire rated in 20-min frames with three UL- with no closers may be used listed spring hinge closers 1-hr rating on walls and doors of hazardous areas No 1-hr rating of doors or walls $23.00 per 100 sq ft on of hazardous areas walls; $111.70 each door Hazardous areas must be 1-hr rated walls with 3⁄4-hr rated Standard 1⁄2-in. sheetrock and $23.00 per 1,000 sq ft on doors and three UL spring hinge closers (hazardous areas standard doors with no walls; $55.00 per 1,000 are common storage, common mech. rooms, etc.) closers may be used sq ft on ceilings; $111.70 each door

(continues)

2007 Automatic Sprinkler Systems for Residential Occupancies Handbook Supplement 4 • Residential Sprinkler Systems 261

TABLE S4.14 Continued

Access Trade-offs with Without Residential Fire Protection Residential Fire Protection Savings/Advantage

Every 6,000 sq ft a 4-hr masonry fire separation is required The fire-rated separation $8,345 each 4-hr wall; would not be required until note: cost of 2-hr sheet- 10,000 sq ft, and after 8,000 rock wall $750 sq ft a 2-hr separation would be permitted

Source: John A. Viniello, ‘‘Residential & Quick Response Fire Sprinklers — You Need to Know the Difference,’’ New Technology Update, National Fire Sprinkler Association, Patterson, NY, 1985, p. 8.

Reduced fire flow, smaller supply pipe Straight percentage of property tax rebate for build- Increased acceptable response time to nearest fire ings with sprinkler systems station Elimination of water department tap fees and Increased allowable distance from public access- standby charges way Insurance reductions for sprinklered properties Street width reduction [p. 8] Cul-de-sacs allowed Other incentives (tax or insurance) include: Approximately 200 communities in the United States Elimination of the value of the sprinkler system have legislative or incentive programs for installing resi- when calculating property tax dential sprinkler systems.58 Table S4.15 lists provisions

TABLE S4.15 Active Residential Sprinkler Programs as of January 1987

Hotel and Health One- and State Community All Multifamily Motel Care Lodging Two-Family

Alaska I I I I I I Arizona Peoria V V V V V V Scottsdale O O O O O O Tucson V V V V California Anaheim C* C C Arroyo Grande O O O O O O Belmont O* O* O* O* O* Beverly Hills O* O* Campbell V V V V V V Carpinteria O* O* O* O* O* O* Clovis V V O* O* O* V Corte Madera O* O O O* El Monte V V V V V V Escondido O* O* O* O* O* O* Fontana O O O O O O Glendale O* O* O* O* O* O* Hawthorne C C C C C C Huntington Beach O O O O Livermore O O O O O O Los Angeles O* O* O* O* O* O* Millbrae O O O O O O Modesto O* O* O* O*

(continues)

Automatic Sprinkler Systems for Residential Occupancies Handbook 2007 262 Supplement 4 • Residential Sprinkler Systems

TABLE S4.15 Continued

Hotel and Health One- and State Community All Multifamily Motel Care Lodging Two-Family

California Montclair O* (continued) Monterey O* O* O* O* O* O* Orange County O* O* O* O* O* O* Piercy V V V V V V Porterville O Redlands V V V V V V Redwood City O* O* O* O* O* O* Riverside O* O* O* O* O* Salinas O O O O O O Salinas Rural O O O O O O San Bernadino V San Carlos O* O* O* O* O* San Clemente OC OC OC OC OC OC San Diego C San Francisco OC Sanger OC OC OC OC OC OC Santa Cruz OI OI OI OI Sausalito VI Tiburon O* O* O* O* Union City O O O Ventura O* O* O* O* O* O* Ventura County O* O* O* O* O* O* Colorado Boulder OI OI OI OI OI OI Fort Collins I I I I I I CONNECTICUT LC LC Fairfield O FLORIDA L L Altamonte Springs VI VI Broward County L L L L Longboat Key O O O O O O Naples O O O O O O* Palm Beach Gardens O O St. Petersburg VI VI VI VI VI VI Tampa C C C C Georgia Cobb County VI VI VI VI VI VI De Kalb County O O O O Thunderbolt O O O O O* Illinois Champaign VI VI VI VI VI VI Oak Lawn O* O* O* O* Indiana Hobart O O O O O O Maine Bangor VI VI VI VI VI Maryland Montgomery County L L* Prince Georges County C C C C Massachusetts Cohasset V I I I I V Haverhill V Wenham VI Michigan Portage V O O O O V Montana Billings V

2007 Automatic Sprinkler Systems for Residential Occupancies Handbook Supplement 4 • Residential Sprinkler Systems 263

TABLE S4.15 Continued

Hotel and Health One- and State Community All Multifamily Motel Care Lodging Two-Family

Nevada Carson City OC OC Henderson VI L L L L VI Reno O O O New Hampshire Dover O O O O O O* Hudson OC New York Dobbs Ferry L L L L L L Fairview L L L L L L Harrison V V V V V V Onondaga V V North Carolina Cary VI VI Charlotte I Wrightsville Beach I* I* I* I* I* I* OKLAHOMA L L L L Pennsylvania Lower Merion Township O O O O O* South Carolina Hilton Head O* O* O* O* Texas Beaumont C C Conroe O* O* O* O* Friendswood L Plano I Utah Salt Lake City O Virginia Chesterfield County VI VI VI VI VI Richmond O O Washington Auburn O* O* O* O* O* O* Port Angeles I I O I I Wisconsin Madison V C*

States listed in all capital letters have a program in effect or under consideration on the state level. *Indicates that special conditions affect that particular area of the communities program (i.e., square footage requirements, waterflow requirements, all residences except one- and two-family, etc.). C Code — A program that is developed as a code that will govern the community (i.e., building codes, fire codes, etc.). I Incentives — A program established to provide enticements or code variance for the homeowner or builder to promote the use of residential sprinkler systems. L Law — Any program that will be established as a law by the governing body of the community. O Ordinance — Any program created as an ordinance for the community. V Voluntary — A program established to promote the use of residential sprinklers but not requiring their use, nor providing incentives for their use. Table reprinted with permission by Teague from Sprinkler Quarterly, Spring 1987, published by the National Fire Sprinkler Association, Inc., Robin Hill Corporate Park, Route 22, PO. Box 1000, Pat- terson, NY 12563.

for the installation of residential sprinkler systems in 91 predecessor to the limited area dwelling sprinkler system communities and occupancies. Exhibit S4.24 presents a requirements, which were first included in NFPA 13D’s survey listing 236 communities in the United States in 1994 edition and previously reviewed in this chapter, be- 1999 with automatic fire sprinkler requirements for one- cause these systems were installed for nine years even and two-family homes. though these requirements were withdrawn as a separate The Fire Safety Forum (1989) developed ‘‘Design chapter in the 2002 edition. Guide for Rural Fire Sprinkler Systems,’’ a procedural The Fire Sprinkler Advisory Board of Southern Cali- guide for installing sprinkler systems in rural areas. This fornia’s five-year fire sprinkler ordinance survey obtained document proposes four levels of protection.59 It was a responses from 274 jurisdictions protecting 87.5 percent

Automatic Sprinkler Systems for Residential Occupancies Handbook 2007 264 Supplement 4 • Residential Sprinkler Systems

ALASKA Irvine (c;d) San Gabriel Ketchikan (a;d) Kentfield San Juan Capistrano (c;d) Laguna Niguel (c;d) San Luis Obispo (b) ARIZONA Laguna woods (c;d) San Marcos (c;d) Chandler (a) La Habra Heights San Mateo Mesa (c) Lake Forest (c;d) San Miguel Peoria (c;d) Lakeside San Rafael Scottsdale (a;d) La Palma (c;d) Sanger Sedona (c;d) Larkspur Santa Cruz (a) Livermore (c) Santa Monica (a) CALIFORNIA Loma Linda Santee (a) Agoura Hills (c;d) Los Alamitos (c;d) Saratoga Albany (c;d) Los Angeles (c;d) Sausalito Alhambra Los Angeles County (c;d) Sierra Madre Alto-Richardson Bay (a) Malibu Seal Beach (c;d) Aptos/La Selva (a) Manteca-Lathrop (c;d) Sonoma (a) Arcadia (b) Marin County Stanton (c) Arroyo Grande Millbrae (c;d) Sunnyvale (b) Auburn (a;d) Mill Valley Tamalpais Valley Aurora (a;d) Milpitas Thunderbolt Bakersfield (b;d) Mission Viejo (c;d) Tiburon (a) Belmont Montclair Tustin (c;d) Beverly Hills (a) Montebello Union City (a) Buena Park (c) Monterey Vacaville Burbank (c) Napa (b) Ventura Carmel Highlands Newark (a;d) Ventura County Carpinteria Norco (a;d) Villa Park (c;d) Central North County (b) West Hollywood Cloverdale Novato Fire (b) Westminster (c;d) Claremont Orange County (c;d) Woodland (b) Corte Madera (a;d) Oxnard (a) Yolo County Covina Pacific Grove Yorba Linda (e;d) Culver City (a) Palm Springs (a;d) Cypress (c;d) Petaluma COLORADO Dana Point (c) Pismo Beach (b;d) Aspen (a;d) Daly City (a) Placentia (c) Basalt (a;d) Desert Hot Springs Rancho Santa Fe (a) Boulder El Cerrito (a;d) Redlands Fort Collins El Monte (a;d) Redondo Beach (a;d) CONNECTICUT Encinitas Redwood Wethersfield (a) Escondido Richmond (a;d) Fontana (c) Ross Valley FLORIDA Foothill (b;d) Sacramento County (c) Altamonte Springs (c;d) Foster City (a) Salinas Boca Raton (a;d) Fremont (c;d) Salinas Rural (b) Casselberry (a;d) Gilroy (a;d) San Bernadino Flagler Beach (a;d) Glendale (b) San Bernadino County (c;d) Fort Myers Hawthorne San Carlos Hillsborough Healdsburg San Clemente (c) Jackson/Teton County (a;d) Huntington Beach San Diego Jacksonville Beach (a;d)

EXHIBIT S4.24 U.S. Cities with Sprinkler Ordinances for One- and Two-Family Homes. (Source: Operation Life Safety Newsletter, Vol. 14, Nos. 10 & 11, 1999, pp. 6–7)

2007 Automatic Sprinkler Systems for Residential Occupancies Handbook Supplement 4 • Residential Sprinkler Systems 265

Longboat Key MASSACHUSETTS SOUTH CAROLINA Naples Marblehead (a;c;d) Mount Pleasant (b;d) Orlando (a;d) North Andover (b) Oviedo (a;d) TENNESSEE Palm Beach MINNESOTA Germantown (c;d) Palm Beach County (c;d) Plymouth (a;d) St. Lucie County (b) TEXAS NEW JERSEY Addison (a) GEORGIA Franklin County Bellaire (c;d) Marrietta Lewisville (c;d) Morrow (c) NEW MEXICO Piney Point Village (c) Santa Fe Plano IDAHO Santa Fe County Sugar Land (a;d) Ketchum (a;d) NEW YORK UTAH ILLINOIS Dobbs Ferry Salt Lake City Buffalo Grove Fairview Deerfield-Bannockburn (b;d) Greenburgh (a) WASHINGTON Hazel Crest (a;d) Harrison Auburn Lombard (a;d) Issaquah (c;d) Bothell Long Grove (a) Dupont (c) Mount Prospect OHIO King County Fire Dist. 4 (a;d) Oak Park (a;d) Colerain Twp. (a;d) King County Fire Marshal (c;d) Tinley Park Olympia (a;d) Wheeling (b) PENNSYLVANIA Port Angeles Wood Dale (b;d) Broomall (b;d) Redmond (a;d) Lower Merion Twp. (c;d) KANSAS Mt. Lebanon WISCONSIN Leawood (a;d) Philadelphia Madison Lenexa (c;d) Phoenixville Shorewood Hills (a) Plymouth MARYLAND Upper Dublin Twp. (c) Visit the RFSI website at Howard County (c) Upper Merion Twp. (b) www.firesafehome.org for Montgomery County Warrington Twp. (c) specific restrictions as Prince Georges County (b) York indicated by the letter “d”

Note 1.This data represents information gathered by the Residential Fire Safety Institute (RFSI) from fire departments and other sources. Based on the response to our inquiries, we estimate that the actual number of ordinances is substantially higher. Note 2.Jurisdictions printed in italics may have ordinances covering 1-2 family homes, but this has not been verified. The RFSI staff is verifying the information and will continue to revise the list. Note 3.This list does not include: • Jurisdictions that offer incentives for sprinkler installation but do not require them by ordinance. • Jurisdictions that require sprinklers on a case-by-case basis, such as a decision by a zoning board. • Jurisdictions that only require sprinklers in model homes and leave sprinklers as a homebuyer’s option. Special Requirements (a) means that sprinklers are required in all structures including 1- and 2-family homes. (b) means that sprinklers are required in all residential occupancies including 1- and 2-family homes. (c) means that sprinklers are required in all 1- and 2-family homes. (d) indicates a restrictive. For example, in Corte Madera, CA, sprinklers are required in all occupancies, but the requirement only applies when the water flow is less than 500 gpm.

EXHIBIT S4.24 Continued

Automatic Sprinkler Systems for Residential Occupancies Handbook 2007 266 Supplement 4 • Residential Sprinkler Systems

of California’s population.60 Of the 262 jurisdictions re- response extended coverage sprinkler, and the early sup- sponding to survey items concerning residential sprinkler pression fast response (ESFR) sprinkler. In the past 30 requirements for one-and two-family dwellings, apart- years, sprinkler technology has radically changed due to ments, and townhouses, 189 or 72.1 percent had residential the NFPA Sprinkler Committee’s established goal to pro- sprinkler requirements. However, although 258 jurisdic- vide an economically feasible and technically effective tions reported requirements for sprinklers in all new single- residential sprinkler system for life safety. family dwellings, only 76 or 29.4 percent had this require- The 1980 edition of NFPA 13D was the first effective ment in effect. Table S4.16 presents factors cited as justi- design guide for a residential life safety sprinkler system. fying sprinklers in single-family dwellings. This standard is also unique because it is one of the NFPA standards based on a specific research effort involving TABLE S4.16 Factors Justifying Requiring Sprinklers in approximately 100 full-scale laboratory and field fire tests. All New Single-Family Dwellings in California The 1980 edition of NFPA 13D resulted in changes made Access problems 118 to NFPA 13 relative to protecting dwelling units in occu- Inadequate water supply 103 pancies other than one- and two-family dwellings and man- Floor area 83 ufactured homes. This edition of NFPA 13D also resulted Building height 52 in the development and issuance of NFPA 13R in 1989. Response time in excess of 44 NFPA 13R provides a sprinkler system designed to 3 min 1 protect occupants of low-rise multiple dwelling-unit occu- 5 min 15 pancies. NFPA 13R attempted to provide a design that 6 min 4 prevents misapplication of NFPA 13D to residential occu- 7 min 1 pancies of up to and including four stories, consisting of 10 min 7 apartment buildings, lodging and rooming houses, hotels, other 16 motels, dormitories, and some board and care facilities. Other reasons 61 NFPA 13D sprinkler design concepts and residential Source: Fire Sprinkler Advisory Board of Southern California, The sprinkler technology resulted in the adoption of economic Residential Summary Report on the 1993 Fire Sprinkler Ordinance and legislative incentives to promote residential sprinkler Survey, Cerritos, CA, 1993, p. 2. systems. SUMMARY ACTIVITIES The concept of residential sprinkler systems is not new. These systems have been available since 1932 when Grin- 1. Explain the justification of the NFPA Sprinkler Techni- nell’s ‘‘junior’’ system was developed. However, the initial cal Committee to allow sprinklers to be omitted in NFPA 13D, Standard for the Installation of Sprinkler Sys- certain areas based on the NFPA 13D and NFPA 13R tems in One- and Two-Family Dwellings, issued in 1975, standards. radically changed sprinkler system standards and, more 2. How do the basic philosophies established in the NFPA importantly, sprinkler technology. This edition of NFPA 13 and NFPA 13D standards differ? 13D introduced the concept of exempting certain dwelling 3. Compare the sprinkler system design criteria estab- areas from sprinkler coverage. These exemptions were lished in the 1975 edition of NFPA 13D and in the based on the statistically perceived hazard and, as a way of 1980 edition of NFPA 13D. Explain the discrepancies. reducing costs, aimed to provide an economically feasible 4. Review Underwriters Laboratories’ test criteria for system. More importantly, the 1975 edition of NFPA 13D residential sprinklers as presented in Appendix D, and focused attention on the standard sprinkler’s unsuitability compare the fire test criteria with the shielded-corner as an effective device for protecting occupants’ lives. Thus, room fire in the full-scale fire tests discussed in this with the federal government’s funded research program, chapter. the sprinkler committee’s guidance, NFPA’s organization, 5. Review the list of active residential sprinkler programs and FM Global’s research, this effort culminated in the presented in Table S4.15 with the occupancies and development of the residential sprinkler based on the full- programs involved. Are any of these types of programs scale Los Angeles dwelling tests and the Charlotte mobile being initiated in your community? If not, explain why home tests. not. If so, explain why and how. The development of the residential sprinkler resulted 6. List at least 10 building code modifications or trade- in subsequent refinement and improvement of the quick- offs typically allowed in model building codes to pro- response sprinkler, and the development of the quick- mote installation of sprinkler systems.

2007 Automatic Sprinkler Systems for Residential Occupancies Handbook Supplement 4 • Residential Sprinkler Systems 267

7. The 1980 edition of NFPA 13D mandated the use of dential Sprinkler Systems,’’ Fire Journal, Vol. 74, No. listed residential sprinklers, yet a review of Chapter 6, 1980, pp. 44–47. 6 indicates that the first residential sprinkler was ap- 12. Cote, Arthur E., ‘‘Field Test and Evaluation of Resi- proved and listed by Underwriters Laboratories in June dential Sprinkler Systems: Part I,’’ Fire Technology, 1981. What was the philosophy of the NFPA Sprinkler Vol. 19, No. 4, 1983, pp. 221–232. Technical Committee in requiring a sprinkler that was 13. Kung, Hsiang-Cheng, Robert D. Spaulding, and Ed- not yet available? ward E. Hill, Jr., Sprinkler Performance in Residential 8. Explain the difference in water supply requirements Fire Tests, Factory Mutual Research, FMRC No. for NFPA 13D and NFPA 13R. What justifies this 22574, 1980, Norwood, MA. difference? 14. Cote, Arthur E., ‘‘Field Test and Evaluation of Resi- dential Sprinkler Systems: Part II,’’ Fire Technology, Vol. 20, No. 1, 1984, pp. 48–58. BIBLIOGRAPHY 15. Bill, Robert G., Jr., Hsiang-Cheng Kung, and Scott Anderson, ‘‘A New Test to Evaluate the Fire Perfor- References Cited mance of Residential Sprinklers,’’ Fire Technology, 1. Karter, Michael J., Jr., ‘‘U.S. Fire Loss for 2004,’’ Vol. 38, No. 3, 2002, pp. 101–124. NFPA Journal, Vol. 99, No. 5, 2005, pp. 44–49. 16. Golinveaux, James, ‘‘Listings and Applications of 2. Hall, John, R., Jr., ‘‘Manufactured Home Fires,’’ Fire Residential Sprinklers, A Technical Analysis, Part 1,’’ Analysis & Research Division, National Fire Protec- Sprinkler Age, Vol. 22, No. 3, 2003, pp. 15–16. tion Association, Quincy, MA, 2005. 17. Golinveaux, James, ‘‘Listings and Applications of 3. lnternational Association of Fire Chiefs, ‘‘Residential Residential Sprinklers, A Technical Analysis, Part 2,’’ Sprinkler Activations—Annual Report,’’ Operation Sprinkler Age, Vol. 22, No. 4, 2003, pp. 25–26. Life Safety Newsletter, Vol. 10, No. 9, 1995, p. 2. 18. Huggins, Roland, ‘‘Home Safe,’’ NFPA Journal, Vol. 4. Jensen, Rolf H., ‘‘Development of the NFPA 13D 98, No. 3, 2004, pp. 68, 70–73. Residential Sprinkler Standard,’’ in U.S. Fire Adminis- 19. Cote, Arthur E., ‘‘Field Test and Evaluation of Resi- tration, Proceedings of the Fourth Conference on Low- dential Sprinkler Systems: Part III,’’ Fire Technology, Cost Residential Sprinkler Systems, Washington, DC, Vol. 20, No. 2, 1984, pp. 41–46. 1980. 20. Fleming, Russell P., ‘‘A Closer Look at the NFPA 5. Rolf Jensen & Associates, Inc., Study to Establish the Residential Sprinkler Standards,’’ Fire Journal, Vol. Existing Automatic Fire Suppression Technology for 82, No. 2, 1988, pp. 51–55, 91–92. Use in Residential Occupancies, A Summary Report, 21. UL 1626, Residential Sprinklers for Fire Protection National Fire Prevention and Control Administration, Service, Underwriters Laboratories, Northbrook, IL, Washington, DC, 1978. 2003. 6. Wolf, Alisa, ‘‘A Tough Sell,’’ NFPA Journal, Vol. 92, 22. Fleming, Russell P., ‘‘NFPA Sprinkler Committee Ex- No. 2, 1998, pp. 59–63. pands Limited Area Concept for 13D,’’ Sprinkler Tech- 7. Foehl, John M., ‘‘In Quest of an Economical Auto- Notes, Vol. 15, No. 4, 1993, pp. 1–2. matic Fire Suppression System for Single-Family Res- 23. Isman, Kenneth E., ‘‘NFPA 13D—A New Look,’’ idences,’’ Fire Journal, Vol. 68, No. 5, 1974, pp. Sprinkler Quarterly, No. 86, 1994, pp. 12–14. 42–48. 24. Bill, Robert G., Jr., and Hsiang-Cheng Kung, ‘‘Lim- 8. Adams, Scott W., and Christopher H. Born, Cost Esti- ited-Water-Supply Sprinklers for Manufactured (Mo- mations of Residential Sprinkler Systems, Including bile) Homes,’’ Fire Technology, Vol. 29, No. 3, 1993, Cases Where Municipal Water Supply Is Inadequate pp. 203–225. or Non-Existent, Department of Fire Protection Engi- 25. Bill, Robert G., Jr., and Edward E. Hill, Jr., ‘‘Sprinkler neering, University of Maryland, College Park, MD, Protection of Manufactured Homes with Sloped Ceil- 1987. ings Using Prototype Limited Water Supply Sprin- 9. Lehn, Kenneth W., Regional Report, Mid-Atlantic klers,’’ Fire Technology, Vol.31, No. 1, 1995, pp. 5–16. States, National Fire Sprinkler Association, Inc., Feb- 26. Puchovsky, Milosh T., ‘‘Changes to NFPA 13D 1996 ruary, 1996, p. 3. EditionSeesAllowanceof 1⁄2 InchPipe,’’SprinklerAge, 10. Cote, Arthur E., ‘‘Update on Residential Sprinkler Vol. 15, No. 3, 1996, p. 18. Protection,’’ Fire Journal, Vol. 77, No. 6, 1983, pp. 27. Fleming, Russell P., ‘‘One-Half-Inch Sprinkler Pipe 69–71, 109–113. Debate Heats Up for NFPA13D,’’ Sprinkler TechNotes, 11. Moore, David A., ‘‘Field Test and Evaluation of Resi- Vol. 18, No. 5, 1995, pp. 1–2.

Automatic Sprinkler Systems for Residential Occupancies Handbook 2007 268 Supplement 4 • Residential Sprinkler Systems

28. International Association of Fire Chiefs, ‘‘One-Half 44. Teague, Paul, and Jim Pothier, ‘‘Adoption by the Model Inch Piping Nearing Approval for Revision to NFPA Building Code Groups of Residential Sprinkler Re- 13D,’’ 1996, Operation Life Safety Newsletter, Vol.11, quirements Will Make the Inclusion of Such Provisions No.3&4,pp.1–2. in Local Building Codes More Likely,’’ Fire Journal, 29. Minnick, Benjamin, ‘‘DuPont Development Using Vol. 82, No. 5, 1988, pp. 57–59. Unique Home Fire Sprinklers,’’ Operation Life Safety 45. Evans, DavidP., ‘‘SprinklersCome Homeat Last,’’Fire Newsletter,Vol.10,No.2&3,1995, p. 2. Service Today, Vol. 49, No. 10, 1982, pp. 14–16. 30. Holmes, Craig A., ‘‘NFPA 13D Residential Sprinklers 46. Ford, Jim, ‘‘One City’s Case for Residential Sprinkler vs. NFPA 13 Standard Sprinkler Systems—A Differ- Systems,’’ NFPA Journal, Vol. 91, No. 4, 1997, pp. ence in Levels of Protection,’’ Building Standards, Vol. 41–44. 14, No. 5, 1985, pp. 11–12. 47. Ruegg, Rosalie T., and Sieglinde K. Fuller, ‘‘The Eco- 31. Foehl, John M., ‘‘In Quest of an Economical Automatic nomics of Fast-Response Residential Sprinkler Sys- Fire Suppression System for Multi-Family Residential tems,’’ Fire Journal, Vol. 79, No. 3, 1985, pp. 18–22, Complexes,’’ Fire Journal, Vol. 70, No. 1, 1976, pp. 115–118. 48–53, 57. 48. Viniello, John A., ‘‘Residential & Quick Response Fire 32. Foehl, John M., Fire Hardening of Old Residential Sprinklers—You Need to Know the Difference,’’ New Buildings in High Risk Urban Communities: Sprinkler Technology Update, National Fire Sprinkler Associa- System Aspects, TR80-6, 1980, Society of Fire Protec- tion, Patterson, NY, 1985, pp. 4–10. tion Engineers, Boston, MA. 49. Ad Hoc Insurance Committee on Residential Sprin- 33. Hammerman, David M., ‘‘Project Home—A Pilot Pro- klers, Report on 1980 Property Loss Comparison Fires: gram for a Low-Cost Residential Sprinkler System,’’ An InsuranceIndustry Assessmentof the Impactof Resi- Fire Journal, Vol. 75, No. 2, 1981, pp. 66–69. dential Sprinklers on Fire Loss, U.S. Fire Administra- 34. Schirmer, Chester W., ‘‘Residential Sprinklers in Habi- tion, Washington, DC, 1981. tat for Humanity Homes,’’ Fire Protection Engi- 50. Dalton, Jim, ‘‘Insurance Reductions Support Residen- neering, No. 3, 1999, pp. 37–38. tial Sprinkler Systems,’’ Operation Life Safety Newslet- 35. Wolf, Alisa, ‘‘Sprinkler Giveaway,’’ NFPA Journal, ter, Vol. 2, No. 8, 1987, p. 9. Vol. 93, No. 5, 1999, pp. 83–87. 51. Skinner, Richard, ‘‘Insurance Discounts—Residential 36. Montaivo, D’Arcy G. (ed.), ‘‘Building Piece of Mind, Fire Sprinklers, ‘‘Regional Report, Mid-Atlantic States, Volunteers Sprinkler Habitat for Humanity Homes,’’ National Fire Sprinkler Association, Inc., October, Sprinkler Age, Vol. 21, No. 1, 2002, pp. 12–13. 2000. 37. Gagnon, Robert M., Personal communication, 10-18- 52. Coughlin,Pat,‘‘IncentivesaPowerfulToolforWinning 04. Fire Sprinkler Ordinances,’’ Sprinkler Age, Vol.19, No. 38. Fleming, Russell P., ‘‘ISO Proposes Insurance Credit 1, 2000, pp. 22–23. for NFPA 13R Systems,’’ Sprinkler TechNotes, Vol.12, 53. Siarnicki, Ronald Jon, ‘‘Residential Sprinklers: One No. 2, 1989, pp. 3–4. Community’s Experience Twelve Years After Manda- 39. Brown, Phillip, ‘‘Garages and NFPA 13D, Some Com- tory Implementation,’’ Operation Life Safety Newslet- munities Extend 13D Requirements with Local Ord- ter,Vol.17,Nos.7&8,2002, p. 2. nances,’’ OLS Newsletter,Vol.14,Nos.7&8,1999, p. 54. Jelenewicz, Chris, ‘‘What Have We Learned About the 7. Benefits and Costs of Residential Fire Sprinkler Legis- 40. Lawrence, Tom, ‘‘NFPA 13R Update, Technical Com- lation?’’ Fire Protection Engineering, No. 25, 2005, pp. mittee Clarifies Garage Protection,’’ Sprinkler Age, 34–39. Vol. 21, No. 4, 2002, p. 18. 55. Davis, Larry, ‘‘First Responders Residential Sprinklers 41. Kelly,Kevin, ‘‘SprinklersinGarages,’’ SprinklerQuar- Can Douse Fires Long Before Rescuers Arrive on terly, No. 121, 2003, pp. 31, 33. Scene,’’ Fire Rescue Magazine, Vol. 21, No. 5, 2003, 42. Budnick, Edward K., ‘‘Estimating Effectiveness of pp. 101–106. State-of-the-ArtDetectorsandAutomaticSprinklerson 56. Isman, Kenneth, ‘‘Sprinklers and the Model Codes: Life Safety in Residential Occupancies,’’ Fire Technol- What’s Required? What’s Allowed?’’ SFPE Bulletin, ogy, Vol. 20, No. 2, 1984, pp. 5–22. 1989 (May/June), pp. 6–9. 43. Ruegg, Rosalie T., and Sieglinde K. Fuller, A Benefit- 57. Endthoff, Gene B., ‘‘Life Safety Code to Require QR Cost Model of Residential Fire Sprinkler Systems, NBS or Residential Sprinklers in New Hotels, Motels and Technical Note 1203, 1984, National Bureau of Stan- Dorms,’’ CodeWatch, No. 27, 1994, p. 1. dards, Gaithersburg, MD. 58. Teague, Paul E., ‘‘Residential Sprinklers, An Idea

2007 Automatic Sprinkler Systems for Residential Occupancies Handbook Supplement 4 • Residential Sprinkler Systems 269

Whose Time Has Almost Come,’’ Fire Journal, Vol.82, NFPA 13D, Standard for the Installation of Sprinkler Sys- No. 5, 1988, pp. 45–54. tems in One-and Two-Family Dwellings, and Manu- 59. Fire Safety Forum, ‘‘Design Guide for Rural Fire Sprin- factured Homes, 2002. kler Systems,’’ Building Standards, Vol. 58, No. 3, NFPA 13R, Standard for the Installation of Sprinkler Sys- 1989, pp. 28–30. tems in Residential Occupancies up to Four Stories 60. Fire Sprinkler Advisory Board of Southern California, in Height, 1989. The Residential Summary Report on the 1993 Fire NFPA 13R, Standard for the Installation of Sprinkler Sys- Sprinkler Ordinance Survey, Cerritos, CA, 1993. tems in Residential Occupancies up to and Including Four Stories in Height, 2002. NFPA Codes, Standards, and Recommended NFPA 101, Life Safety Code, 2006. Practices National Fire Protection Association, 1 Batterymarch Park, Suggested Readings Quincy, MA 02169-7471. Cote, Arthur E., ‘‘Update on Residential Sprinkler Protec- NFPA 13, Standard for the Installation of Sprinkler Sys- tion,’’ Fire Journal, Vol. 77, No. 6, 1983, pp. 69–71, tems, 1983. 109–113. NFPA 13, Standard for the Installation of Sprinkler Sys- Fleming, Russell P., ‘‘A Closer Look at the NFPA Residen- tems, 1985. tial Sprinkler Standard,’’ Fire Journal, Vol. 82, No. NFPA 13, Standard for the Installation of Sprinkler Sys- 2, 1988, pp. 51–55, 91–92. tems, 2002. NFPA 13D, Standard for the Installation of Sprinkler Sys- NFPA 13D, Standard for the Installation of Sprinkler Sys- tems in One- and Two-Family Dwellings and Manufac- tems in One- and Two-Family Dwellings, 1975. tured Homes, National Fire Protection Association, NFPA 13D, Standard for the Installation of Sprinkler Sys- Quincy, MA, 2002. tems in One- and Two-Family Dwellings and Mobile NFPA 13R, Standard for the Installation of Sprinkler Sys- Homes, 1980. tems in Residential Occupancies Up to and Including NFPA 13D, Sprinkler Systems in One- and Two-Family Four Stories in Height, National Fire Protection Asso- Dwellings and Manufactured Homes, 1994. ciation, Quincy, MA, 2002.

Automatic Sprinkler Systems for Residential Occupancies Handbook 2007