Storage Protection in the Presence of - nfpa.org · test configuration was 6 tiers of rack storage under a 40-foot ceiling. A 25-foot-long ignition array with A 25-foot-long ignition

© 2017 Fire Protection Research Foundation

1 Batterymarch Park, Quincy, MA 02169-7417, USA Email: [email protected] | Web: nfpa.org/foundation

Storage Protection in the Presence of Horizontal Barriers or Solid Shelving - Literature Review FINAL REPORT BY:

Garner Palenske, P.E. Garth Ornelas, M.S. Jensen Hughes San Diego, CA, USA October 2017

—— Page ii ——

—— Page iii ——

FOREWORD There is limited guidance related to storage protection and horizontal barriers in NFPA 13, Standard for the Installation of Sprinkler Systems. In addition, the terms horizontal barrier and shelves are both used in NFPA 13 with differing protection strategies. For example, one form of alternate protection in NFPA 13 is to provide a solid horizontal barrier and provide sprinklers around the perimeter of the barrier, including face sprinklers. On the other hand, the shelf requirements do not require face sprinklers. Another potential issue is that shelves larger than 64 square feet require sprinklers under each shelf, but there is no limit to tiers of storage under each shelf. Therefore, the Foundation initiated a project to holistically review the requirements in NFPA 13 relating to horizontal barriers and shelves and the basis behind them as well as existing test data, with the goal to identify gaps and remaining research needs on this topic in order to develop guidance on the types and configurations of horizontal barriers that stop vertical fire spread and allow for independent flow calculation between the ceiling and the barrier level. With the increasing height of warehouses, there is a benefit to considering protection schemes that use horizontal barriers to stop vertical spread. The Fire Protection Research Foundation expresses gratitude to the report authors Garner Palenske, P.E., and Garth Ornelas, M.S., who are with Jensen Hughes located in San Diego, CA, USA. The Research Foundation appreciates the guidance provided by the Project Technical Panelists, the funding provided by the project sponsors, and all others that contributed to this research effort. The content, opinions and conclusions contained in this report are solely those of the authors and do not necessarily represent the views of the Fire Protection Research Foundation, NFPA, Technical Panel or Sponsors. The Foundation makes no guaranty or warranty as to the accuracy or completeness of any information published herein. About the Fire Protection Research Foundation

The Fire Protection Research Foundation plans, manages, and communicates research on a broad range of fire safety issues in collaboration with scientists and laboratories around the world. The Foundation is an affiliate of NFPA.

About the National Fire Protection Association (NFPA)

Founded in 1896, NFPA is a global, nonprofit organization devoted to eliminating death, injury, property and economic loss due to fire, electrical and related hazards. The association delivers information and knowledge through more than 300 consensus codes and standards, research, training, education, outreach and advocacy; and by partnering with others who share an interest in furthering the NFPA mission.

http://www.nfpa.org/foundation

—— Page iv ——

All NFPA codes and standards can be viewed online for free. NFPA's membership totals more than 65,000 individuals around the world. Keywords: NFPA 13, sprinkler, sprinkler protection, horizontal barriers, solid shelves, storage, warehouse Report number: FPRF-2017-19

http://www.nfpa.org/codes-and-standards/free-access

http://www.nfpa.org/member-access

—— Page v ——

—— Page vi ——

PROJECT TECHNICAL PANEL

Tracey Bellamy, Telgian

Christina Francis, P&G

Ray Grill, Arup

Louis Guerrazzi, NFSA

Roland Huggins, AFSA

Bill Koffel, Koffel Associates, Inc.

Tom Pedersen, IKEA

Rich Pehrson, Pehrson Fire PC

Dan Steppan, UL

Chad Duffy, NFPA Staff Liaison

Matt Klaus, NFPA Staff Liaison

PROJECT SPONSORS

Property Insurance Research Group:

AIG CNA Insurance

FM Global Liberty Mutual Insurance

Tokio Marine America Travelers Insurance

XL GAPS Zurich Insurance Group

—— Page vii ——

11770 Bernardo Plaza Court l Suite 116

San Diego, CA 92128 USA

jensenhughes.com

+1 619-488-9810

Fax: +1 619-488-9811

STORAGE PROTECTION IN THE PRESENCE OF HORIZONTAL BARRIERS OR SOLID SHELVING- LITERATURE REVIEW

Prepared For

Ms. Amanda Kimball, P.E. Research Project Manager The Fire Protection Research Foundation One Batterymarch Park Quincy, MA 02169-7471 617.984.7295 [email protected]

November 6, 2017

Project #: 1GPA16002

STORAGE PROTECTION IN THE PRESENCE OF HORIZONTAL BARRIERS OR SOLID SHELVING PAGE ii November 6, 2017

JENSEN HUGHES

TABLE OF CONTENTS

1. INTRODUCTION .............................................................................................................................. 4

2. LITERATURE REVIEW ................................................................................................................... 4

2.1. Legacy Horizontal Barrier Full-scale Fire Testing ......................................................... 5

2.2. Alternate Protection Scheme Full-scale Fire testing ..................................................... 7

2.3. Contemporary Horizontal Barrier Full-scale Fire Testing ........................................... 10

2.4. Solid Shelf Full-scale Fire Testing ................................................................................ 24

3. DISCUSSION ................................................................................................................................. 31

4. GAP ANALYSIS ............................................................................................................................ 35

5. BIBLIOGRAPHY ............................................................................................................................ 37

APPENDIX A: ALTERNATE PROTECTION SCHEME TESTING DATA SUMMARY .......................... A-1

TABLES

Table 1: 1971-1972 Tests Involving Horizontal Barriers and In-Rack Sprinklers ................................ 6 Table 2: Alternate Protection Scheme Details ......................................................................................... 7 Table 3: Comparison of Alternate Protection Scheme Requirements .................................................. 9 Table 4: Test 1 Parameters and Results ................................................................................................. 12 Table 5: Test 2 Parameters and Results ................................................................................................. 15 Table 6: Test 3 Parameters and Results ................................................................................................. 18 Table 7: Test No. 4 Parameters and Results .......................................................................................... 21 Table 8: Comparison of NFPA 13/ FM Global Solid Shelf Sprinkler Location Requirements ........... 25 Table 9: Comparison of NFPA 13/FM Global Solid Shelf In-rack Sprinkler Design Requirements .. 26 Table 10: Test 98 Summary Part 1 ........................................................................................................... 28 Table 11: Test 98 Summary Part 2 ........................................................................................................... 28 Table 12: Test 147 Summary Part 1 ......................................................................................................... 29 Table 13: Test 147 Summary Part 2 ......................................................................................................... 30 Table 14: NFPA 13 K Factor Requirements ............................................................................................ 33

PHOTOGRAPHS

Photograph 1: Test No. 1 Horizontal Barrier Detail ............................................................................... 10 Photograph 2: Test No. 2 – East View of Fire with Ignition Within Transverse Flue Space, Early Stages of Fire Development ..................................................................................................................... 22 Photograph 3: Test No. 3 – West View of Rack with Ignition at Rack Face, Early Stages of Fire Development .............................................................................................................................................. 22 Photograph 4: Test No. 2 – Fire Movement Under Horizontal Barrier ................................................. 23 Photograph 5: Test No. 2 – In-rack Sprinkler Operation Under Horizontal Barrier ............................ 23

STORAGE PROTECTION IN THE PRESENCE OF HORIZONTAL BARRIERS OR SOLID SHELVING PAGE iii November 6, 2017

JENSEN HUGHES

FIGURES

Figure 1: Test No. 1 Layout Plan View .................................................................................................... 11 Figure 2: Test No. 1 Sprinkler Operating Sequence .............................................................................. 11 Figure 3: Test No. 2 Layout Plan View .................................................................................................... 13 Figure 4: Test No. 2 Sprinkler Operating Sequence .............................................................................. 14 Figure 5: Test No. 3 Layout Plan View .................................................................................................... 16 Figure 6: Test No. 3 Sprinkler Operating Sequence .............................................................................. 17 Figure 7: Test No. 4 Layout Plan View .................................................................................................... 19 Figure 8: Test No. 4 Sprinkler Operating Sequence .............................................................................. 20 Figure 9: Solid Shelf Storage Array Details............................................................................................ 27 Figure 10: Test 98 Fire Damage .............................................................................................................. 28 Figure 11: Test 66 Fire Damage ............................................................................................................... 29 Figure 12: Test 147 Fire Damage ............................................................................................................. 30 Figure 13: Test 89 Fire Damage ............................................................................................................... 30 Figure 14: Percentage of Survey Recipients That Use Horizontal Barrier Scheme ........................... 32 Figure 15: Comparison of Orifice Size (K-Factor) to Sprinkler Performance ..................................... 34 Figure 16: Sprinkler Type Most Commonly Utilized by Survey Recipients ........................................ 35

STORAGE PROTECTION IN THE PRESENCE OF HORIZONTAL BARRIERS OR SOLID SHELVING PAGE 4 November 6, 2017

JENSEN HUGHES

1. INTRODUCTION

Since the 1970’s, horizontal surfaces have been placed in storage racks. In some cases, these surfaces are a component of the sprinkler protection scheme, and in other cases they are installed for storage convenience or efficiency. The former can result in an enhancement to sprinkler protection, the latter a detriment.

One type of horizontal surface is termed a horizontal barrier. Horizontal barriers are used in specific sprinkler protection schemes to improve the fire protection of the rack or racks selected. Horizontal barriers are generally constructed of wood and placed at predetermined locations within storage racks. In-rack sprinklers are placed under the horizontal barriers to slow vertical fire propagation and decrease in-rack sprinkler activation time. This usually results in limited ceiling sprinkler activation, which can reduce the hydraulic demand of the sprinkler system (i.e. balancing of the in-rack sprinkler system with the ceiling sprinkler system is not required). This protection scheme is very useful for the protection of challenging rack storage scenarios where the ceiling sprinkler system is not adequate.

Another type of horizontal surface commonly found in storage racks is a solid shelf. Solid shelves, which are also generally constructed of wood, are used to support objects too small or of irregular shape or size to be supported by pallets. Solid shelves are used for storage efficiency and unfortunately block sprinkler spray from reaching the fire location. This helps to promote horizontal fire spread, which can seriously reduce the effectiveness of ceiling sprinkler performance.

The 2016 edition of the National Fire Protection Association’s “Standard for the Installation of Sprinkler Systems” (NFPA 13), includes requirements for the use of horizontal barriers above in-rack sprinklers and for the protection of solid shelves. The basis of these requirements is testing completed in the 1970’s and 1990’s. Much has changed regarding sprinkler technology and the understanding of in-rack fire behavior since then.

The goal of this project was to review the requirements of NFPA 13 and identify knowledge gaps related to sprinkler protection involving horizontal barriers and solid shelves. This included a thorough literature review of fire test data related to horizontal barriers and solid shelves. Data from testing completed from the 1970’s through 2015 was reviewed.

2. LITERATURE REVIEW

The requirements for the use of horizontal barriers in conjunction with in-rack sprinklers as a protection scheme are found in two main areas within NFPA 13:

• Sections 16.3 and 17.3 -Control Mode Density/Area Sprinkler Protection Criteria for Rack Storage Stored Over 25 ft. in Height, and Class I-IV and Plastic Commodities respectively.

• Sections 16.1.2.4 and 17.1.2.9-Alternate Protection, Class I-IV and Plastic Commodities, respectively.

The requirements for the protection of rack storage of Class I-IV commodities first appeared in the 1972 inaugural edition of National Fire Protection Association Standard 231C (NFPA 231C), “Rack Storage of Materials”. These requirements are based upon the findings of the Rack Storage Fire Protection Committee which was organized in August 1967.1 This committee was comprised of a diverse group of members, including insurance, consulting, and manufacturing storage experts. In late 1968, the National Fire Protection Association formed the Rack Storage of Materials Committee, which subsequently became the NFPA 231C code development committee. 2 The autonomous Rack Storage Fire Protection Committee provided data to the NFPA committee for use in preparing the NFPA standard.

1 Schirmer, Chet (1969), Rack Storage Fire Protection Committee Interim Report 2 IBID


JENSEN HUGHES

2.1. Legacy Horizontal Barrier Full-scale Fire Testing

The Rack Storage Fire Protection Committee test program was conducted at the Legacy FM Global Research Center in West Glocester, Rhode Island. Phase 1 of the project explored fire protection of commodity stored within storage racks to a maximum height of 20 feet. Horizontal barriers were not used in this testing.

Phase 2 considered storage heights of 30 feet and higher. A variety of tests were completed with various in-rack sprinkler arrangements, including sprinklers located in the longitudinal flue and at the rack face. In addition, testing with in-rack sprinklers beneath horizontal barriers was conducted.

Approximately 20 full scale tests were completed with a 30-foot storage height (Test Nos. 109-128).3 The test configuration was 6 tiers of rack storage under a 40-foot ceiling. A 25-foot-long ignition array with target arrays, separated by 4 foot aisles on both sides was used.4 Six-inch transverse and longitudinal flue spaces were provided between loads and at the rack uprights. Class II commodity, consisting of 42 inch x 42 inch x 42 inch double tri-wall cardboard boxes with metal liners, was used. This commodity was selected to emulate the commodity normally expected to be stored in “real life” warehouses. 5

A ceiling sprinkler density of 0.30 gpm/ sq. ft. was provided. This criterion was selected based upon hydraulic calculations that showed an Ordinary Hazard pipe schedule system, the predominant sprinkler design approach used for storage protection, was capable of delivering this density when provided with a residual pressure of 100 psi at the base of the sprinkler riser.6

The ceiling sprinklers were K5.6 orifice, 280° F rated, standard response (Response Time Index > 80 (meters-seconds) ½). Sprinkler spacing was 10 ft. x 10 ft. In-rack sprinklers were K5.6 orifice, 160° F rated, standard response. The in-rack sprinklers discharge pressure was 30 psi.

Test Nos. 121, 126, and 127 included horizontal wood barriers, approximately ½ inch thick. The barriers were placed at the 3rd and 6th tiers above the in-rack sprinklers. The barriers extended the entire length and width of the rack obstructing both horizontal and transverse flue spaces. Relevant results of this testing series are shown in Table 1.

A test was considered acceptable if the fire did not:7

• Burn beyond the first row of cartons in the target array.

• Burn to end stack in the ignition array.

• Open all sprinklers within the test facility

3 Englehart, Owen (1976) Rack Storage Test Summary Sheet, Rack Storage Steering Committee, From Schirmer Project Notes 4 IBID.The ignition array was shortened from 50 feet for the 30 ft. tests. 5 IBID 6 Schirmer, Chet (1971), Project Notes. 7 IBID


JENSEN HUGHES

Table 1: 1971-1972 Tests Involving Horizontal Barriers and In-Rack Sprinklers 8

Test No. In-rack Sprinkler Arrangement

Horizontal Barrier Location

No. Ceiling Sprinklers Activated

No. In-Rack Sprinklers Activated

Damage (No. of Boxes)

Results

121 At 3rd and 6th tiers. 8 ft. horizontally in longitudinal flue space. No face sprinklers.

At 3rd and 6th tier 0 2 8 Passed

126 Same Same 0 3 13 Passed

127 Same Same 0 4 21 Passed

128 At the 3rd and 5th tiers. 8 ft. horizontally in longitudinal flue space. No face sprinklers.

No horizontal barriers provided

2 4 48 Failed

Research by Field suggests that if in-rack sprinkler sensitivity is not significantly high, the flame front will move past the in-rack sprinkler prior to the sprinkler’s activation, igniting combustibles at a higher level. 9 Results of Test 128 (without horizontal barriers) confirmed this theory. The addition of horizontal barriers stopped the vertical progression of the fire, concentrating the fire plume energy in the near field of the in-rack sprinkler, which resulted in in-rack sprinkler activation. This occurred despite the low sensitivity of in-rack sprinklers used (standard response). The effectiveness of the horizontal barrier/ in-rack sprinkler protection scheme is confirmed in the data which shows all tests with this arrangement (Test Nos. 121, 126, and 127) were successful with only in-rack sprinklers operating. Although not found in the project documentation, it is logical to theorize that the horizontal barriers were introduced as one solution to the low sensitivity of in-rack sprinklers in the early 1970’s. Test Nos. 117 and 120 included face sprinklers in addition to in-rack sprinklers located within the longitudinal flue spaces at the 3rd and 6th tier of storage. Horizontal barriers were not provided. The results of these tests showed fire control with only in-rack sprinklers (longitudinal flue or face) operating. A comparison of the results of configurations without face sprinklers showed more damage to the commodity in the upper tiers and operation of the ceiling sprinklers compared to configurations with face sprinklers. Acceptable testing results, however were achieved with or without face sprinklers10.

The requirements for the protection of rack storage of plastic commodities first appeared in the 1986 edition of NFPA 231C. These provisions were based primarily upon the full-scale fire testing completed in the Rack Storage Project (RSP) testing series, which was completed from 1973 through 1978. In this series, approximately 16 full -scale fire tests were completed. Storage height was 25 feet or below. The commodity selected was cartoned, unexpanded Group A plastic.11

The NFPA 13 provisions for the protection of plastic commodity over 25 feet includes options for the use of horizontal barriers. The requirements for below 25 feet do not.

8 Englehart, Owen (1976) Rack Storage Test Summary Sheet, Rack Storage Steering Committee, From Schirmer Project Notes 9 Ingason, Haukur(1996) Experimental and Theoretical Study of Rack Storage Fires, Chapter 4 10 Englehart, Owen (1976) Rack Storage Test Summary Sheet, Rack Storage Steering Committee, From Schirmer Project Notes 11 Fire Protection Research Foundation Fire Test Database


JENSEN HUGHES

It appears the requirements for the protection of plastic commodity above 25 feet were extrapolated from the Class I-IV testing series. The requirements are similar apart from a higher ceiling sprinkler density for plastic commodity when the storage height above the top level of in-rack sprinklers exceeds 5 feet (0.45 gpm/ sq. ft. vs. 0.30 gpm/ sq. ft.).

2.2. Alternate Protection Scheme Full-scale Fire testing

An additional option for the protection of rack storage of material was added to the 2016 Edition of NFPA 13.12 This option was intended to address situations where a greater level of protection is required than what is available from the ceiling sprinkler system.13 The Alternate Protection Scheme, as it is called in NFPA 13, utilizes a combination of in-rack sprinklers and horizontal barriers to protect the commodity. A general overview of the requirements for double-row racks is shown in Table 2.

Table 2: Alternate Protection Scheme Details

12 Klaus, Mattew, J. (2016) NFPA 13 Automatic Sprinkler Handbook, and 2016 NFPA 13 Second Draft Report 13 Discussions with Tracy Bellamy, P.E., author of Sections 16.1.2.4 and 17.1.2.9

Commodity Rack Type

Horizontal Barrier

Location

In-rack Sprinkler Location In-rack Sprinkler Details

In-rack Sprinkler Design

Class I-IV and Plastics

Open 12 ft. max vertically

Under the barriers, at each rack upright in the longitudinal flue space and at the face of the rack and at the mid-bay of each rack bay. Maximum linear spacing between sprinklers not greater than 5 ft. at the rack face and 10 ft. in the longitudinal flue space.

Min K8.0, Quick Response

6 sprinklers at minimum flow of 60 gpm. Balancing with the ceiling sprinkler system is not required

Class I-IV and Plastics

Solid Shelf

Every tier vertically

Same Same Same

Barrier Details: Barriers are required to be constructed of minimum 22-gauge sheet metal or 3/8 in. plywood.

Barriers are required to extend to both aisle faces, covering both longitudinal and transvers flue spaces.

Barriers are required to be fitted to within 3 in. of any vertical rack member or other equipment that creates an opening.


JENSEN HUGHES

This approach originated from FM Global sponsored fire testing of motor oil (Class IIIB combustible liquid) in 1 quart plastic containers. Motor oil in plastic containers is a rather newly discovered commodity challenge (1990’s).

The immiscibility of motor oil in water is the biggest challenge when designing sprinkler protection for this commodity. Water discharging onto a burning pool of motor oil will be unable to form a mixture. For water miscible liquids, this mixing increases the water percentage in the mixture causing the flash point and fire point to increase while the heat release rate and heat of combustion are decreased. 14 The motor oil testing was completed at Southwest Research Institute, circa 1990. Three tests in this multi-test program, Test Nos. P29-P31, integrate the use of horizontal barriers to improve the performance of the in-rack sprinklers. In all three tests, only in-rack sprinklers activated.15 See Appendix A for a summary of the test data.

Based upon this testing data, requirements for the protection of Class IIIB combustible liquids were incorporated into the NFPA 30, “Flammable and Combustible Liquids Code”, (NFPA 30 Table 16.5.2.5). FM Global also included a modified version of this protection scheme (Scheme A) in FM Global Loss Prevention Data Sheet (FMDS), 7-29, “Ignitable Liquid Storage in Portable Containers”. Recognizing the merit of this approach, in 2004 FM Global added this option for the protection of Class I-IV and plastic commodities to FMDS 8-9, “Storage of Class 1, 2,3, 4 and Plastic Commodities”. Fire Protection Scheme 8-9A, as it was appropriately renamed in FMDS 8-9, was then integrated into NFPA 13 unchanged. See Table 3 for a comparison of the requirements.

A review of the current NFPA 13 requirements in comparison to the baseline fire tests was conducted and are included in Table 3. Important parameters such as in-rack sprinkler sensitivity, orifice factor, location, and design pressure are the same. Given a maximum of two in-rack sprinklers operated in each fire test, the requirement to calculate 6 or 8 in-rack sprinklers provides a safety factor of 3 or 4 for the designs that were tested.

14 Factory Mutual Research Corporation. July 2014. Ignitable Liquid Storage in Portable Containers. Property Loss Prevention Data sheet 7-29. 15 Nugent, David, P., Schirmer Engineering, (1994) Directory of Fire Test involving Storage of Flammable and Combustible Liquids in Small Containers


JENSEN HUGHES

Table 3: Comparison of Alternate Protection Scheme Requirements

Rack Type

Horizontal Barrier

Location

Ceiling Sprinkler Design

In-rack Sprinkler Location In-rack Sprinkler Details

In-rack Sprinkler Test Results/Design

Requirements

Fire Tests: P-29,30, 31. (20 ft. storage, 30 ft., ceiling)

Open 10 ft. vertically

286 °F, K8.0, Standard response. Not active (positioned only for response)

Tests 29 and 30: Three lines of sprinklers 9 in. above the 2nd tier of storage below the plywood barrier. Sprinkler spacing 4 ft. 3.5 in. on line in the longitudinal flue space and 8 ft. 7 in. along both faces. Face sprinklers located at rack uprights, 3 ft. 5 in. between lines. Test 31: Barriers located at the 2nd and 4th tiers with the same rack sprinkler arrangement.

165° F, K8.0, Quick response

57 psi operating pressure (60.4 gpm). Two sprinklers operated.

NFPA 30 (Table 16.5.2.5)


Not less than 0.20 gpm/ sq. ft. for the most remote 3000 sq.ft

Three lines of sprinklers 6 in. minimum above the tier of storage below the plywood barrier. Sprinkler spacing 4-5 ft. on line in the longitudinal flue space and 8-10 ft. along both faces. Face sprinklers located at rack uprights.

165 ° F, min. K8.0, quick response

Most remote 6 sprinklers (three on two lines) if one barrier installed. Most remote 8 sprinklers (four on two lines) if more than one barrier is installed. Min. operating pressure 50 psi. Balancing with ceiling sprinklers not required.

FM 8-9 ( Section 2.3.4.9)


Designed to protect the highest commodity hazard not protected by the alternate protection option.

Three lines of sprinklers as close as possible to underside of the barrier. Sprinkler spacing 10 ft. maximum on line in the longitudinal flue space and 5 ft. along both faces. All in-rack sprinklers located at rack uprights.

160 ° F, Min K8.0, Quick response

Most remote 6 sprinklers for single-row racks or most remote 8 sprinklers for double or multiple row racks. Minimum flow of 60 gpm. Balancing with the ceiling sprinkler system is not required.

NFPA 13 (Sections 16.1.2.4 and 17.1.2.9)

Same Same Same Same 165 ° F, Min K8.0, Quick response

Same


JENSEN HUGHES

2.3. Contemporary Horizontal Barrier Full-scale Fire Testing

A literature review of contemporary research shows a renaissance in the use of horizontal barrier protection schemes. Work done in 2012-2015 by Reliable Automatic Sprinkler Co., Inc. explored the use of horizontal barriers to protect rack storage of plastic commodities. All tests were conducted at Underwriters Laboratory Inc. large-scale fire test facility in Northbrook, Illinois.

Test No. 1 (August 8, 2012) was completed with double-row rack storage of exposed, expanded Group A plastic. Storage height was 35 feet, beneath a 40-foot ceiling.16 Eight foot aisles were used. One 3/8-inch-thick plywood horizontal barrier was located at the 20-foot elevation. The horizontal barrier was provided with spaces between rack uprights (3 inches wide) including across the longitudinal flue space (See Photograph 1). The ignition location was at the face of the commodity in the aisle space, centered between 2 ceiling and the in-rack sprinklers (See Figure 1).

Photograph 1: Test No. 1 Horizontal Barrier Detail

In-rack sprinklers were provided beneath the barrier. Horizontal spacing was 8 feet, 3 inches on center located in the longitudinal flue space. The in-rack sprinklers selected were K25.2, 212°F, quick response extended coverage pendent. The discharge pressure of the in-rack sprinklers was 30 psi. Face sprinklers were not provided. Ceiling sprinklers were K16.8, 212°F, Early Suppression Fast Response sprinklers spaced at 10 feet x 10 feet. A discharge pressure of 50 psi was used.

This protection scheme was successful. Two of the four in-rack sprinklers operated and one ceiling sprinkler operated. The fire was contained within the ignition array, and maximum one minute average steel temperatures were below 1000 °F. All sprinklers activated within 13 seconds of each other, with the in-rack sprinklers operating first, at 49 seconds (See Figure 2).

The fire traveled a horizontal distance that only spanned two bays. There was minor melting of the target plastic commodity, but ignition did not occur (See Table 4).

16 Steppan, Daniel (2012), Underwriters Laboratories, Inc. Project 12CA35881. Courtesy Reliable Sprinkler


JENSEN HUGHES

Figure 1: Test No. 1 Layout Plan View

Figure 2: Test No. 1 Sprinkler Operating Sequence


JENSEN HUGHES

Table 4: Test 1 Parameters and Results


JENSEN HUGHES

Test No. 2 (October 24, 2013) was completed with double-row rack storage of cartoned, unexpanded Group A plastic. Storage height was 43 feet, beneath a 48-foot ceiling. 17 Standard 4-foot aisles were used. One 3/8-inch-thick plywood horizontal barrier was located at the 30-foot elevation. The barrier covered the entire horizontal plane at the 30-foot elevation (i.e. no gaps at the rack uprights).

In-rack sprinklers were provided beneath the barrier. Horizontal spacing was 8 feet, 3 inches on center located in the longitudinal flue space. The in-rack sprinklers selected were K25.2, 212 °F, quick response extended coverage pendent type. The discharge pressure of the in-rack sprinklers was 30 psi. Face sprinklers were not provided. Ceiling sprinklers were K25.2, 212 °F, quick response, extended coverage type spaced at 14 feet x 14 feet. The ceiling sprinkler design density was 0.70 gpm/ sq. ft. The ignition location was in the transverse flue space, centered between the longitudinal flue space and the aisle, and centered between in-rack sprinklers (See Figure 3).

This protection scheme proved successful. Three of the four in-rack sprinklers operated, the fire was contained within the rack, and maximum one minute average steel temperatures were below 1000 °F. No ceiling sprinklers operated. The three in-rack sprinklers activate nearly simultaneously, at approximately 1:14 minutes after ignition. The fire traveled horizontally only two bays. Damage was limited to the commodity below the horizontal barrier and adjacent to the ignition transverse flue space (See Figure 4 and Table 5).


17 Steppan, Daniel (2013), Underwriters Laboratories, Inc. Project 4786092094. Courtesy Reliable Sprinkler


JENSEN HUGHES


48 ft.

In Rack Sprinkler System, under 3/8

inch plywood horizontal barrier

at the nominal 30 ft. elevation(4 pressurized

sprinklers)

Ignition Location – Two half igniters positioned at base of

commodity as shown, centered on North Transverse Flue Space

30 ft. 10 in. (top of beam)

43 ft. nominal

C L

14 ft. (typ.)14 inches

Steel Beam

dno: did not operate

1:151:151:14dno dno dno


JENSEN HUGHES



JENSEN HUGHES

Test No. 3 (July 16, 2014) was completed with multiple-row rack storage of exposed expanded, Group A plastic. Storage height was 20 feet, beneath a 27-foot ceiling 18. Standard 4 foot aisles were used. One 3/8-inch-thick plywood horizontal barrier was located at the 20-foot elevation. The barrier covered the entire horizontal plane at the 20-foot elevation (i.e. no gaps at the rack uprights).

In-rack sprinklers were provided in the longitudinal and transverse flue spaces beneath the barrier. The in-rack sprinklers were spaced 8 feet, 3 inches on center. The in-rack sprinklers selected were K25.2, 212 °F, quick response extended coverage pendent type. The nominal discharge pressure of the in-rack sprinklers was 30 psi. Ceiling sprinkler were K25.2, 212 °F, quick response, extended coverage type spaced at 14 feet x 14 feet. The ceiling sprinkler design density was 0.70 gpm/ sq. ft. The ignition location was in the transverse flue space, centered between the longitudinal flue space and the aisle (See Figure 5).

This protection scheme proved successful. Five of the fourteen in-rack sprinklers operated, the fire was contained within the rack, and maximum one minute average steel temperatures were below 1000 °F. No ceiling sprinklers operated. The first in-rack sprinkler activated at 47 seconds and the last activated at 8:08. The fire burned from the ignition location to the north and south extremities of the ignition array, but did not ignite the target array (See Figure 6 and Table 6).




JENSEN HUGHES


N

0:59

3:22

0:47

0:47

8:08

dno

dno

dno

dno

dno

dno

dno

dno

dno

dno

dno

dno

dno

dno

dno

dno

dno

dno

dno: did not operate

Ignition Location


JENSEN HUGHES



JENSEN HUGHES

Test No. 4 (August 7, 2015) was a variation of Test No. 2.19 The horizontal barrier was modified by providing openings (3 inches wide) in the longitudinal flue spaces at the rack uprights. In addition, the ignition location was moved to the face of the commodity, centered between 2 boxes and the in-rack sprinklers (See Figure 7).

The results of this test were successful as well. Two of the four in-rack sprinklers operated, the fire was contained within the ignition rack, and maximum one minute average steel temperatures were below 1000 °F. No ceiling sprinklers operated (See Figure 8 and Table 7).




JENSEN HUGHES



JENSEN HUGHES

Table 7: Test No. 4 Parameters and Results


JENSEN HUGHES

Due to aisle ignition location, the activation time of the first sprinkler was much longer than Test No. 1, 4:28 vs 1:14 minutes. Once the first sprinkler operated, sprinkler discharge down the longitudinal flue space drove the fire to the north face of the rack. This resulted in an increase in radiant heat flux to the target array which caused minor surface charring at the 2nd tier of storage (See Photograph 2 and Photograph 3).

The fire traveled horizontally only two bays at the 30-foot elevation. The fire did not travel vertically above the horizontal barriers, despite the barrier openings at the rack uprights. Additional commodity damage occurred below the barrier compared to Test No. 2. This is understandable due to the delay in the activation of the first sprinkler.

Photograph 2: Test No. 2 – East View of Fire with Ignition Within Transverse Flue Space, Early Stages of Fire Development

Photograph 3: Test No. 3 – West View of Rack with Ignition at Rack Face, Early Stages of Fire Development


JENSEN HUGHES

The results of these fire tests show the effectiveness of the horizontal barrier/in-rack sprinkler design. In an open rack without horizontal barriers, as the fire grows, it spreads rapidly through the vertical flue spaces and simultaneously spreads along horizontal channels, growing faster and becoming more intense. As the fire moves vertically through the flue spaces heat flux is applied to the face of the exposed cardboard boxes causing ignition of the cardboard face, preheating the commodity within. Eventually the commodity ignites and the fire spreads horizontally within the storage tier.

When a horizontal barrier is added, the upward traveling fire intersects the horizontal barrier and flames divert horizontally. This causes heating of the tops of the cardboard boxes and ultimately ignition of the commodity within (See Photograph 4).

Photograph 4: Test No. 2 – Fire Movement Under Horizontal Barrier

The in-rack sprinklers become immersed in the flames traveling horizontally which results in prompt operation. At this time, the commodity below may not yet be ignited so the discharge of the in-rack sprinkler pre-wets the commodity and applies water to any burning surfaces. This is shown in Photograph 5. Water is also driven down through the flue spaces which cools the fire plume and pre-wets adjacent commodity.

Photograph 5: Test No. 2 – In-rack Sprinkler Operation Under Horizontal Barrier

This test series demonstrates that the horizontal barrier/ in-rack sprinkler protection scheme is a viable option for the protection of rack storage of materials. The results of Test Nos. 1-3 have been submitted to the NFPA 13 committee and will likely appear in the 2019 edition of the standard.


JENSEN HUGHES

2.4. Solid Shelf Full-scale Fire Testing

The requirements for solid shelves first appeared in the 1973 edition of NFPA 231C. Section 422 (for storage within double-row racks less than or equal to 25 ft. high) required double and multiple-row racks with solid shelves that obstruct both longitudinal and transverse flues to be protected with in-rack sprinklers beneath each shelf. In solid shelf racks, 6 in. transverse flue spaces at rack uprights or between loads and 6 in. longitudinal flue spaces between loads were required. In racks without solid shelves, longitudinal flues spaces were not required if 6 in. transverse flue at rack uprights or between loads were provided. These requirements were commonly interpreted to mean that double-row racks less than or equal to 25 ft. high did not require in-rack sprinkler protection if the solid shelf did not obstruct the longitudinal flue space. This could result in solid shelves 32 sq. ft. in size not requiring in-rack sprinkler protection. These provisions remained unchanged, with the exception that single-row racks were included, until 2002.

The provisions for the protection of solid shelves were rewritten in the 2002 edition of NFPA 13 and remain mostly unchanged in the 2016 edition of NFPA 13. The requirements were generally based upon the requirements found in FMDS 8-9 “Storage of Class 1,2,3,4 and Plastic Commodities” (edition in in effect in approximately 2000). Table 8 and Table 9 provide a comparison of the NFPA 13 and FM Global requirements.


JENSEN HUGHES

Table 8: Comparison of NFPA 13/ FM Global Solid Shelf Sprinkler Location Requirements

Commodity Class Solid Shelf Area

Rack Protection Required

Vertical Spacing

Horizontal Spacing Storage Height

Class I-IV >20 sq. ft. Y 6 ft. max.

(15 ft. max)

10 ft. max.

(10 ft. max. at every other transverse flue space or 8 max at every flue space)

Any Height

(Greater than 15 ft. storage)

Cartoned Plastic Y 6 ft. max.

(15 ft. max)

5 ft. max.

(10 ft. max. at every other transverse flue space or 8 max. at every flue space)

Any Height


Uncartoned Plastic Y 6 ft. max.

(10 ft. max.)

5 ft. max. (10 ft. max. at every other transverse flue space or 8 ft. max. at every flue space)

Any Height


Class I-IV ≥ 64 sq. ft. Y Beneath Every Tier

10 ft. max.

(At every transverse flue space, 8 ft. max.)

Any Height

Cartoned Plastic 5 ft. max.

(At very transverse flue space, 8 ft. max))

Any Height

Uncartoned Plastic Y 5 max.

(At very transverse flue space, 8 ft. max)

Any Height

*(FM Global requirements)


JENSEN HUGHES

Table 9: Comparison of NFPA 13/FM Global Solid Shelf In-rack Sprinkler Design Requirements

Commodity Class

Storage height above top level

of In-rack sprinklers

No of levels installed

Number of in-rack sprinklers flowing

Flow Balancing with Ceiling Sprinkler System

I-III ≤ 10 ft. 1 6 30 gpm

(22 gpm)

Required.

Other option, install face sprinklers under each solid shelf at the rack uprights. In-rack sprinkler design, min. discharge of 60 gpm from eight sprinklers. Ceiling and in-rack sprinklers are not required to be balanced.

2 or more 10 (5 on 2 levels) 30 gpm

(22 gpm)

Same

IV and Plastics

≤ 10 ft. 1 8 30 gpm Same

2 or more 14- 7 (on 2 levels) 30 gpm Same

*(FM Global requirements)

These requirements originated from the work of the 1967 Rack Storage Fire Protection Committee.20 21 Four of the sixty fire tests conducted in this series included racks with solid shelves.22 Two tests, Test Nos. 98 and 147, were analyzed. The other two tests were considered inappropriate for the purposes of this analysis since multiple variables, such as vertical barriers or higher temperature sprinklers, were introduced in each test. Test Nos. 66 and 89 were selected as baseline tests, due to their similarity with the solid shelf tests.

The test configuration used in these tests was rack storage of standard test commodity (Class II, double tri-wall cartons with a metal liner inserted) stored up to 20 feet beneath a 30-foot ceiling. The ignition location was at the junction of the two center racks (i.e. in the transverse flue space offset from the aisle). The ceiling sprinklers used were K5.6 orifice, 165°F rated, standard response. The design densities used for these tests were 0.30 gpm/ sq.ft. in Test 98 and 0.45 gpm/ sq.ft. in Test 147. Storage array details for both tests are shown in Figure 9 with specific details on Test 98 provided in Tables 10 and 11 as well as Figures 10 and 11, with specific details on Test 147 provided in Tables 12 and 13 as well as Figures 12 and 13.23

20 Schirmer, Chet (1969), Rack Storage Fire Protection Committee Interim Report 21 National Fire Protection Association Standard for Rack Storage of Materials, NFPA 231C, 1973 edition 22 Englehart, Owen (1976) Rack Storage Test Summary Sheet, Rack Storage Steering Committee, From Schirmer Project Notes 23 Schirmer, Chet (1969), Rack Storage Fire Protection Committee Interim Report


JENSEN HUGHES

Figure 9: Solid Shelf Storage Array Details

Test 98 had solid shelves below each tier of storage that extended the full length and width of the ignition array (375 sq. ft.). During Test 98 the fire burned horizontally on the first tier to the east rack face then vertically up the face. This continued on each tier allowing the fire to progress both horizontally and vertically through the rack. The fire jumped to and ignited the west face of the target array at approximately 7 minutes. Ceiling bar joist temperatures were 170 °F, which is within the acceptable range.

It is apparent that the ceiling density was insufficient to control the fire when the fire burned vertically up the rack face. Review of the requirements of NFPA 13 for the specific conditions of the test shows that a ceiling sprinkler density of 0.37 gpm/ sq. ft. is required. 24 The sprinkler density provided was 0.30 gpm/ sq. ft., which is approximately 19% less than that required by NFPA 13.

Test 66 (the baseline test) had a better outcome. The fire did not burn beyond the ignition array and ceiling temperatures were within the acceptable range. However, 48 sprinklers activated, 17% less than Test 98.

Test 98 was considered unsuccessful given the amount of fire damage and the number of sprinklers that operated. However, given the inadequate sprinkler density, it is difficult to quantify the effect that the solid shelves had on sprinkler performance.

24 National Fire Protection Association Standard for the Installation of Sprinkler Systems, NFPA 13, 2016 edition


JENSEN HUGHES

Table 10: Test 98 Summary Part 1

Test No. Solid Shelf Size

Flues Space Aisle Sprinkler Density

(gpm/sq. ft.)

First Sprinkler Activation

Time

No. Sprinklers Operated

98 7.5 ft. x 50 ft. (375 sq. ft.)

0 in. 8 ft. Provided-0.30 Required-0.37

4:18 58

66

Baseline Test

None 6 in. Longitudinal and Transverse

8 ft. Provided-0.30 Required-0.37

3:11 48

25


Test No.

Commodity Consumed Main Rack

(%)

Commodity Consumed East Target

(%)

Commodity Consumed

West Target (%)

Max. Ceiling Air

Temperatures (°F)

Bar Joist Steel

Temperature (°F)

98 100 0 18 1140- 1:56 170

66 55 0 0 1630-19:50 180

26 27 28

Figure 10: Test 98 Fire Damage 29

25 Englehart, Owen (1976) Rack Storage Test Summary Sheet, Rack Storage Steering Committee, From Schirmer Project Notes 26 IBID 27 Schirmer, Chet (1969), Rack Storage Fire Protection Committee Interim Report 28 IBID 29 The term conventional rack means a standard 8 ft. aisle is provided for forklift operations. The term automatic rack means a narrower aisle, 4 ft. in width, was provided because automated picking machines require less aisle width.


JENSEN HUGHES

Figure 11: Test 66 Fire Damage Test 147 was provided with 6 in. transverse flue spaces, at the rack uprights and between loads, and 6 in. longitudinal flue spaces. Solid shelves, that were 3-1/2 ft. deep and 7-3/4 ft. wide, were provided between the rack uprights at each tier level, resulting in a solid shelf area of 27 sq. ft.

During Test 147 the fire immediately traveled vertically up the flue spaces to the ceiling. Cartons within the flue spaces exposed to fire ignited, resulting in both vertical and horizontal propagation of the fire. The first sprinkler activated quickly at 1:23, the second sprinkler followed soon after at 1:38. The shielding provided by the solid shelving allowed the fire to move towards the aisle and then up the rack face. The fire burned horizontally thru the ignition array, and jumped the aisle to the east target array. The fire continued to burn through the east target array eventually consuming much of the array. Ceiling bar joist temperatures were 175 °F, which is within the acceptable range.

Review of the requirements of NFPA 13 for the specific conditions of the test shows that a ceiling sprinkler density of 0.44 gpm/ sq. ft. is required. The sprinkler density provided was 0.45 gpm/ sq. ft. Compared to Test 98, the water flux delivered to the storage array had improved success controlling the fire burning up the face of the rack. However, the narrow aisle exposed the target array to increased thermal radiation causing the ignition and subsequent burning of the majority of the east target array.

Test 147 was considered unsuccessful given the fire burning beyond the ignition array and the number of sprinklers that operated. However, the ceiling steel temperatures were within the acceptable range.

Test 89, the baseline test, was successful. Only seven sprinklers operated. The fire did not burn beyond the ignition array and ceiling temperatures were within the acceptable range. Damage drawings for the tests are shown in Figure 12 and Figure 13.


Test No. Solid Shelf Size

Flues Space Aisle Sprinkler Density (gpm/sq.ft.)

First Sprinkler Activation Time

No. Sprinklers Operated

147 3.5 ft. x 7.75 ft.

(27 sq. ft.)

6 in. Longitudinal and Transverse

4 Provided-0.45

Required-0.44

1:23 47

89- Baseline Test

Slave pallets, beneath the pallets (42 in. x 42 in)

6 in. Longitudinal and Transverse

4 Provided-0.45 Required-0.44

2:57 7


JENSEN HUGHES


30 31 32

Figure 12: Test 147 Fire Damage

Figure 13: Test 89 Fire Damage

30 IBID 31 Schirmer, Chet (1969), Rack Storage Fire Protection Committee Interim Report 32 IBID

Test No.

Commodity Consumed Main Rack

(%)

Commodity Consumed East Target

(%)

Commodity Consumed

West Target (%)

Max. Ceiling Air Temperatures

(°F)

Bar Joist Steel Temperature

(°F)

147 91 74 0 1545-6:00 175 89 30 0 0 925- 105


JENSEN HUGHES

3. DISCUSSION

The requirements found in NFPA 13 for the use of horizontal barriers/in-rack sprinklers for the protection of rack storage are based upon testing completed in the 1970’s. Horizontal barriers were tested for commodity stored over 25 feet. The in-rack sprinklers used in the testing were K5.6 orifice, standard response. Review of the test results, however, shows the legacy sprinklers performed well with a minimal number of in-rack sprinklers and no ceiling sprinklers activating. The use of horizontal barriers improved in-rack sprinkler performance and potentially compensated for the thermal lag of the standard response in-rack sprinklers.

The current edition of NFPA 13 requires larger orifice sizes, K ≥ 8.0, and high sensitivity (quick response, RTI ≤ 50 (meters-seconds)1/2) in-rack sprinklers. The performance of these sprinklers will provide superior results to those shown in the successful legacy testing.

A maximum of 4 in-rack sprinklers activated in the legacy fire tests when horizontal barriers were installed. NFPA 13, Sections 16.3.1.3.3 and 17.3.1.15, require the calculation of 6-14 rack sprinklers depending on the number of levels of in-rack sprinklers installed and the commodity classification. A minimum safety factor of 150% is provided by the NFPA 13 requirements, which is consistent with suggested present-day sprinkler design convention. Note this testing included only Class II commodity.

It should be noted that this design approach is not commonly used for ceiling heights below 45 ft. Data from various sources, such as interviews with sprinkler manufacturers and survey work completed by the authors, indicate that only 4% of the respondents surveyed use this approach (See Figure 14). 33 ESFR sprinklers are primarily for these building heights. For very tall warehouse situations (i.e. greater than 45 ft.) horizontal barriers/in-rack sprinklers are used because there is no available ceiling sprinkler only protection approach approved by NFPA 13.

33 Discussions with major sprinkler manufacturers. Survey of the membership of the NFPA 13 Discharge Committee and two leading fire protection engineering firms


JENSEN HUGHES

Figure 14: Percentage of Survey Recipients That Use Horizontal Barrier Scheme

Review of the 2012-2015 test series provides a more accurate indication of the performance of current in-rack sprinklers installed beneath horizontal barriers. Test Nos. 2 and 4 were similar, but important variables were changed, namely ignition location and the extent of horizontal barrier coverage. The ignition location in Test No. 2 was offset in the transverse flues space, in Test No. 4 the ignition location was at the face of the rack. Typically fires ignited at the face initially develop slowly until the fire progresses to the second tier of storage after which the fire moves towards the longitudinal flue and behaves as if it was ignited there. 34 This behavior was confirmed in Test No. 4 which resulted in a delay in operation of the first in-rack sprinkler.

In Test No.2 gaps were not provided in the horizontal barriers around the rack uprights. In Test No. 4 they were provided (3 inches wide). The effect the gaps had on the test results is difficult to determine. Review of the temperatures recorded at the ceiling sprinkler locations closest to the ignition location show minor differences between the two tests. However, due to the lack of a test with the fire in the transverse flue space with openings in the horizontal barrier at the rack uprights, it is not definitive whether fire would be restricted from traveling above the horizontal barrier.

The excellent results of this testing series should lead to new requirements in future editions of NFPA 13 which could open the door to an expanded use of the horizontal barrier/in-rack sprinkler protection scheme.

The technical substantiation of the Alternate Protection Scheme is strong. The requirements found within NFPA 13 correlated well with the results of the fire tests. A maximum of 2 in-rack sprinklers activated in the fire tests. NFPA 13, sections 16.1.2.4.5 and 17.1.2.9.5, require the calculation of 6-8 in-rack sprinklers depending on type of rack and the commodity classification. A minimum safety factor of 300% is provided by the NFPA 13 requirements, which exceeds present day sprinkler design convention.

The requirements for the protection of solid shelves found in the current edition of NFPA 13 are based upon testing performed in the 1970’s. Only four tests were completed, two of which were considered relevant. Test 98 had solid shelves approximately 375 square feet in size placed beneath every tier of storage. Fifty-eight K5.6 orifice, 165 ° F rated standard response sprinklers activated. The sprinkler density was 0.30 gpm/ sq. ft. which is approximately 19% less than that required in NFPA 13 (0.37 gpm/ sq. ft.). The fire burned the entire length and width of the ignition array, and then jumped the east aisle burning through the majority of the target array.

Test 147 had solid shelves approximately 27 square feet in size placed beneath every tier of storage. Storage array and sprinkler parameters were similar to Test 98 except the ceiling density was increased

34 Ingason, Haukar (1996). Experimental and Theoretical Study of Rack Storage Fires, Section 3.

horizontal configuration

4%

other 13%

N/A 83%

horizontal configuration other N/A


JENSEN HUGHES

to 0.45 gpm/ sq. ft., due to the 4-foot aisle space provided. NFPA 13 requires a density of 0.44 gpm/ sq. ft. Slave pallets were provided beneath each pallet load of material. Forty-seven ½ inch orifice, 165 ° F rated standard response sprinklers activated. The fire burned the entire length and width of the ignition array, jumped the east aisle burning through the majority of the target array.

The sprinklers used were standard orifice, K5.6. The current edition of NFPA 13 requires K factors of 8.0-11.2 for the densities used (See Table 14).

Table 14: NFPA 13 K Factor Requirements

Design Density (gpm/sq. ft.)

Minimum Sprinkler K Factor

(gpm/(psi) 1/2

Orifice Size (Inches)

≤ 0.20 5.6 0.5 > 0.20-0.34 ≥ 8.0 ≥ 0.53

> 0.34 ≥ 11.2 ≥ 0.64

The performance improvement of larger K factor sprinklers is pronounced. Work completed in 2001 by Troup and Vincent (FM Global) compared the ability of various K factor sprinklers to protect rack storage of Class II commodity. 35 A storage height of 20 feet, beneath a 30-foot ceiling was used. A sprinkler density of 0.37 gpm/ sq. ft. was provided. The tests with K11.2 sprinklers opened 50% less sprinklers than the tests with K5.6 sprinklers. In addition, commodity damage was reduced by approximately 87% (See Figure 15). It is apparent that a sprinkler system complying with the requirements of the current edition of NFPA 13 would perform much better than the sprinkler systems used in the 1970’s testing.

35 Troup and Vincent Fire Test Performance of K-Factor 25 Control-Mode Extended-Coverage Sprinklers for Storage Occupancies

– NFPA World Safety Congress May 14, 2001


JENSEN HUGHES

Figure 15: Comparison of Orifice Size (K-Factor) to Sprinkler Performance

In addition, Test 147 used 4 foot aisles, which makes comparison between Test 98 and Test 147 challenging. The tests used only Class II commodity. This was appropriate for the 1970’s, today however, plastics are used extensively. NFPA 13 requirements apply to storage greater than 20 feet in height, yet testing was only conducted to 20 feet.

The previous research is based upon the use of control mode density area sprinklers. Data from various sources, such as sprinkler manufacturers and a survey completed in this project, indicate that ESFR or control mode special application (CMSA) sprinklers are specified in the vast majority (67% and 11% respectively) of storage sprinkler systems36 (See Figure 16).

36 Discussion with major sprinkler manufacturers. Survey of the membership of the NFPA 13 Discharge Committee and two leading fire protection engineering firms.

0

5

10

15

20

25

30

35

40

45

0 10 20 30

No

. Sp

rin

kler

s O

per

atin

g/N

o. P

alle

ts

Co

nsu

med

Orfice Size

Orfice Size vs No.Operating Sprinklers

Orfice Size Vs No.Pallets Consumed


JENSEN HUGHES

Figure 16: Sprinkler Type Most Commonly Utilized by Survey Recipients

NFPA 13 requires in-rack sprinkler protection beneath every solid shelf when ESFR or CMSA ceiling sprinklers are used. No fire testing data was found to substantiate these code provisions.

Chet Schirmer, Chairman of the Rack Storage Fire Protection Committee and the engineer in charge of the 1970’s fire testing, concluded regarding the solid shelf testing, that “These tests did not yield sufficient information to develop a comprehensive protection standard for solid shelf racks. Items such as increased ceiling density, use of bulkheads, other rack configurations, and limited shelf length and depth should be considered.” 37

4. GAP ANALYSIS

The fire testing used as a basis for the current requirements for the horizontal barriers/in-rack sprinkler protection requirements found in NFPA 13 was conducted with Class II commodity only. The ability of these protection schemes to protection Class IV and plastic commodities should be explored. In addition, there was limited testing conducted for very tall storage arrays, which is the most common application of this protection approach. More work is needed in this regard.

The solid shelf testing which was conducted in the 1970’s has been used ever since as the basis for NFPA 13 solid shelf protection requirements. This testing is incomplete. Limited scenarios as well as limited types of commodity were tested. The solid shelf arrangements are peculiar and do not represent common shelving situations used today. In addition, the sprinkler technology used is dated, modern sprinklers preform much better.

ESFR sprinklers, the most commonly used storage sprinklers, have not been tested with solid shelves. 38 In general, these sprinklers will outperform the K5.6 sprinklers used in the legacy solid shelf testing. However, the effect the shielding caused by the solid shelves will have on ESFR sprinkler performance is unknown.

Due to the completeness of much of data reviewed, details of in-rack sprinkler location and alignment was not analyzed. More work needs to be completed to examine how this variable affects in-rack sprinkler performance.

37 Schirmer, Chet (1969), Rack Storage Fire Protection Committee Interim Report 38 Do data was found on this topic.

CMDA 15%

ESFR 67%

CMSA 11%

N/A7%

CMDA ESFR CMSA N/A


JENSEN HUGHES

JENSEN HUGHES,

Prepared by:

Garner Palenske, P.E. Garth Ornelas, M.S. Vice President Associate [email protected] [email protected]


JENSEN HUGHES

5. BIBLIOGRAPHY

Beitel, Jesse J. (December 1991). Sprinklered Rack Storage Fire Tests of Motor Oil. Southwest

Research Institute. Billstroem, Michael. (2015). New Storage Protection Options; NFPA 13, 2016 Edition, and Beyond. Oslo.

Reliable Automatic Sprinkler Co., Inc. Chicarello, Peter J., Joan M.A. Troup, and Ronald K. Dean. (May 1986). National Quick Response

Sprinkler Research Project: Large-Scale Fire Test Evaluation of Early Suppression Fast Response (ESFR) Automatic Sprinklers. National Fire Research Foundation. Factory Mutual Research Corporation.

Fleming, Russ. (2006). Solid Shelves in Racks vs. Shelf Storage: Continues Research Remains Vital in

Protecting Life and Property. NFPA Journal January/ February 2006. Francis, Christina F. (2016). New Technology for High-Challenge Warehouses. SUP DET 2016 – New

Technology for High-Challenge Warehouses. Gallagher, Richard, and Michael Gollner. (February 2010). High Challenge Warehouse Case Study

Summary. Zurich Services Corporation. Zurich Risk Engineering. Golinveaux, James. What We Don’t Know About Storage. Tyco Fire and Building Products. Hamins, Anthony, and Kevin McGrattan. (November 1999). Reduced-Scale Experiments to Characterize

the Suppression of Rack-Storage Commodity Fires. Building and Fire Research Laboratory Gaithersburg, MD: National Institute of Standards and Technology.

Ingason, Haukur. (1996). Experimental and Theoretical Study of Rack Storage Fires. Report

LUTVDG/(TVBB-1013). Lund University, Sweden. Institute of Technology. Department of Fire Safety Engineering.

Ingason, Haukur. (1997). In-Rack Fire Plumes. Fire Safety Science – Proceedings of the Fifth

International Symposium, pp. 333-344. Swedish National Testing and Research Institute (SP). Ingason, Haukur. (2001). An Experimental Study of Rack Storage Fires. SP Report 2001: 19. Boras,

Sweden. Swedish National Testing and Research Institute. Ingason, Haukur. (2001). Heat Release Rate of Rack Storage Fires. Interflam 2001: Ninth International

Fire Science and Engineering Conference, Edinburgh Conference Centre, Scotland, 17th-19th September 2001. SP Swedish National Testing and Research Institute, Sweden.

Lee, James L. (May 1984). Early Suppression Fast Response (ESFR) Program Phase I: Determination of

Required Delivered Density (RDD) in Rack Storage Fires of Plastic Commodity. Early Suppression Fast Response (ESFR) Program Technical Report. Factory Mutual Research Corporation.

Multer, Tom. (2014). Storage Protection Using Horizontal Barriers and Large K-Factor, Extended

Coverage In-Rack Sprinklers. SUPDET 2014. Reliable Automatic Sprinkler Co., Inc. National Fire Protection Association. (2016) NFPA 13: Standard For the Installation of Sprinkler Systems.

Quincy, MA. National Fire Protection Association. National Fire Protection Association. (1973). Rack Storage of Materials. NFPA No. 231C. National Fire

Protection Association.


JENSEN HUGHES

National Fire Protection Association. (1985). Technical Committee Documentation. Fall Committee Baltimore, MD 1985. National Fire Protection Association.

National Fire Protection Research Foundation. National Quick Response Sprinkler Research Project:

Preliminary Report of the Group 2 Performance Tests Phase II – ADD Tests. National Fire Protection Research Foundation. Underwriters Laboratories Inc.

Nugent, David P. (2006). Developing Sprinkler System Design Criteria for Flammable and Combustible

Liquid Storage. Aon Risk Consultants. O’Neill, John G. (2008). Integrated ESFR – Mobile High-bay Records Storage Systems Protection.

Suppression and Detection Research and Applications: A Technical Working Conference (SUPDET 2008). The Protection Engineering Group, Inc.

Reliable. (2015). AC460 N-Rack-EC Extended Coverage In-Rack Automatic Sprinkler System. Reliable

Automatic Sprinkler Co., Inc. Retail Fire Research Coalition Committee Report. Report of the Retail Fire Research Coalition. A

partnership of Government Agencies, Retailers, Insurance Companies, Fire Protection Consultants and Testing Laboratories.

Rolf Jensen & Associates. (April 7, 2000). Report on Full-Scale Fire Suppression System Testing for

Protection of records Stored in Shelving 30 Feet High. Rolf Jensen & Associates. Scheffey, Joseph L. & Mealy, Christopher L. Assessment of Hazards of Flammable and Combustible

Liquids in Composite IBCs in Operations Scenarios. Hughes Associates, INC. Scheffey, Joseph L. (October 2006). Protection of Combustible Liquids Stored in Composite Intermediate

Bulk Containers (IBCs) Research Project. The Fire Protection Research Foundation. Sienkiewicz, Seth E. (2010). Large-Scale Fire Testing of Cartoned Plastics. Suppression and Detection

Research and Applications: A Technical Working Conference (SUPDET 2010). Factory Mutual Research Corporation.

Thomas, W.P. (1973). High-Rack Storage Experiments in the U.S.A. NFPA European Fire Conference

Geneva, Switzerland, October 15-17, 1973. Factory Mutual Research Corporation. Troup, Joan M.A. (November 1994). Large-Scale Fire Testing of Rack Stored Group A Plastics in Retail

Operation Scenarios Protected by Extra Large Orifice (ELO) Sprinklers. Factory Mutual Inc. Underwriters Laboratories. (1986). National Quick Response Sprinkler Research Project: Group 2

Performance Tests Phase I – RDD Tests. Underwriters Laboratories Inc., National Fire Protection Research Foundation.

Underwriters Laboratories. (1993). Fact Finding Report on Rack Storage Fire Tests of Flammable and

Combustible Liquids. Underwriters Laboratories Inc. Underwriters Laboratories. (2012). Exposed Expanded Group A Plastics Stored in Racks With a

Horizontal Barrier Protected by K=25.2 EC In-Rack and K=16.8 ESFR Sprinklers. Underwriters Laboratories Inc.

Underwriters Laboratories. (2013). Cartoned Unexpanded Group A Plastic Commodity Stored In Racks

With Continuous Combustible Horizontal Barriers Protected by K=25.2 EC In-Rack and K=25.2 EC Ceiling Sprinklers. Underwriters Laboratories Inc.

Underwriters Laboratories. (2014). Exposed Expanded Group A Plastic Commodity Stored in Multi-row

Racks With Continuous Combustible Horizontal Barriers Protected by K= 25.2 EC In-Rack and K=25.2 EC Ceiling Sprinklers. Underwriters Laboratories Inc.


JENSEN HUGHES

Underwriters Laboratories. (2015). Cartoned Unexpanded Group A Plastic Commodity Stored In Racks

With Combustible Horizontal Barriers Protected by K=25.2 EC In-Rack and K=25.2 EC Ceiling Sprinklers. Underwriters Laboratories Inc.

Underwriters Laboratories. (May 1998). Report of Large Scale Fire Tests. Underwriters Laboratories Inc. Underwriters Laboratories. (Sept 2006). Evaluation of Sprinkler Performance in Protecting Gondola Type

Shelf Storage. The Fire Protection Research Foundation. Wolin, Steve. (2015). The Return of the In-Rack Sprinkler. Fire Protection Engineering Quarter 3 Issue –

The Return of the In-Rack Sprinkler. The Society of Fire Protection Engineers. Wolin, Steve. Protection of Exposed Expanded Plastics Stored in Racks. Code Consultants, Inc You, Hong-Zeng, and Hsiang-Cheng Kung. (1984). Strong Buoyant Plumes of Growing Rack Storage

Fires. Twentieth Symposium (International) on Combustion/ The Combustion Institute. 1984/pp. 1547-1554. Factory Mutual Research Corporation.

Zhou, Liang, Aiping Chen, Xingna Liu, and Fudong Zhang. (2015). The Effectiveness of Horizontal

Barriers in Preventing Fire Spread on Vertical Insulation Panels Made of Polystyrene Foams. Fire Technology, 52, 649-662, 2016.

STORAGE PROTECTION AND HORIZONTAL BARRIERS PAGE A-1 LITERATURE SEARCH / 1GPA16002 November 6, 2017

JENSEN HUGHES

APPENDIX A: ALTERNATE PROTECTION SCHEME TESTING DATA SUMMARY

STORAGE PROTECTION AND HORIZONTAL BARRIERS PAGE A-2 LITERATURE SEARCH / 1GPA16002 November 6, 2017

JENSEN HUGHES

TEST NO. DATE LOCATION SPONSOR COMMODITY

CONTAINER/ CARTON

TYPE

LIQUID VOLUME

AND OTHER

DETAILS

STORAGE HEIGHT

(FT)

CEILING HEIGHT

(FT) STORAGE ARRAY BARRIER

DESCRIPTION IGNITOR

CEILING SPRINKLER

DESCRIPTION

IN-RACK SPRINKLER

DESCRIPTION IN-RACK SPRINKLER

ARRANGEMENT

SPRINKLER ACTIVATION

(Min: Sec) RESULT

P-29 - SWRI FM MOTOR OIL PLASTIC, F TYPE 1 QT. 20.6 30

DRR, 6 IN. LONGITUDINAL FLUE SPACE, OVERALL WIDTH 7 FT. 6 IN. FOUR TIERS OF STORAGE. TEST COMMODITY IN 1ST AND 2ND TIERS, MIDDLE BAY AND HALF OF BOTH END BAYS. DOUBLE TRI-WALL CORRUGATED CARDBOARD CARTONS WITH METAL LINER IN REMAINDER.

3/4 IN. PLYWOOD HOR. BARRIER INSTALLED OVER 2ND TIER. 12 IN. TRANSVERSE FLUE AT RACK UPRIGHTS

8 OZ GASOLINE IN 3 INCH COTTON CELLULOSIC ROLL PLACED AT THE RACK FACE UNDER THE LOWER PALLET CENTER BAY

17/32 IN ORFICE, 286°F, STANDARD RESPONSE (NON-ACTIVE)

17/32 IN ORFICE, 165 °F, RTI- 50, 57 GPM

3 LINES IN-RACK SPRINKLERS 9 IN ABOVE 2ND TIER OF STORAGE BELOW BARRIER. SPACING- 4 FT 3.5 IN. ON LINE AND 3 FT. 5 IN. BETWEEN LINES. FACE SPRINKLERS ON EACH AISLE AT RACK UPRIGHT AT 3 FT. 5 IN. BETWEEN LINES.

TWO FACE SPRINKLERS AT 5:21 AND 5:23

FIRE CONTROLLED

P-30

- SWRI FM MOTOR OIL PLASTIC, F TYPE

2.880 GAL. 20.6 30


3/4 IN. PLYWOOD HOR. BARRIER INSTALLED OVER 2ND TIER. 12 IN. TRANSVERSE FLUE AT RACK UPRIGHTS

8 OZ GASOLINE IN 3 INCH COTTON CELLULOSIC ROLL PLACED IN THE TRANSVERSE FLUE SPACE IN THE CENTER BAY AT THE FIRST TIER

17/32 IN ORFICE, °F, STANDARD RESPONSE (NON-ACTIVE)


3 LINES IN-RACK SPRINKLERS 9 IN ABOVE 2ND TIER OF STORAGE BELOW BARRIER. SPACING- 4 FT. 3.5 IN. ON LINE AND 3 FT. 5 IN. BETWEEN LINES. FACE SPRINKLERS ON EACH AISLE AT RACK UPRIGHT 3 FT. 5 IN. BETWEEN LINES.

LONGITUDINAL FLUE SPRINKLER 0:50

FIRE CONTROLLED

P-31 - SWRI FM MOTOR OIL PLASTIC, F TYPE 1 QT. 20.6 30


3/4 IN. PLYWOOD HOR. BARRIER INSTALLED OVER 2ND TIER AND 4TH TIER. 12 IN. TRANSVERSE FLUE AT RACK UPRIGHTS

8 OZ GASOLINE IN 3 INCH COTTON CELLULOSIC ROLL PLACED UNDER A PALLET ON THE 3RD TIER CENTER BAY

17/32 IN ORFICE, 286 °F, STANDARD RESPONSE (NON-ACTIVE)


3 LINES IN-RACK SPRINKLERS 9 IN. ABOVE 2ND AND 4TH TIER OF STORAGE BELOW BARRIER. SPACING- 4 FT. 3.5 IN. ON LINE AND 3 FT 5 IN. BETWEEN LINES. FACE SPRINKLERS ON EACH AISLE AT RACK UPRIGHT 3 FT. 5 IN. BETWEEN LINES.

TWO FACE SPRINKLERS AT 5:19 AND 5:39

FIRE CONTROLLED

Storage Protection in the Presence of - nfpa.org · test configuration was 6 tiers of rack storage under a 40-foot ceiling. A 25-foot-long ignition array with A 25-foot-long ignition

Documents