Skip to content
From Fire Safety Academy Wiki
Revision as of 03:08, 8 November 2026 by Peter (talk | contribs) (FSE wiki: Correct NFPA 555 six-level hierarchy and qualify open-burning flashover example after source factcheck (written by Claude Opus 5.5))
Contents
  1. 1Definition and principle
  2. 2Development and standard test methods
  3. 3Apparatus and measurements
  4. 4Calculating the heat release rate
  5. 5Typical results
  6. 6Use in regulation and design
  7. 7Limitations and uncertainty
  8. 8See also
  9. 9Notes
  10. 10References

A furniture calorimeter is a large-scale fire test apparatus in which a complete item, such as a chair, sofa or mattress, burns in the open on a weighing platform beneath a collection hood.[1] It measures the heat release rate of the item from the oxygen that the fire takes out of the air. The method relies on the oxygen consumption principle: about 13.1 MJ of heat is released per kilogram of oxygen consumed, to within about 5 %.[2] Vytenis Babrauskas and co-workers developed the first furniture calorimeter in 1982 at the National Bureau of Standards (NBS), now the National Institute of Standards and Technology (NIST).[2]

In the United States, ASTM standards apply the method to upholstered seating (ASTM E1537), mattresses (ASTM E1590) and stacked chairs.[2][1] Since its 1990 revision, California Technical Bulletin 133 has allowed testing in a furniture calorimeter instead of a burn room, with a pass/fail limit of 80 kW.[3] The data also serve as input to compartment fire models and smoke transport models.[1] Because the item burns in the open, the test leaves out the radiant heating from a hot smoke layer and hot room surfaces. Once a fire becomes large, that heating can raise the burning rate of some items strongly.[4]

Definition and principle

In the SFPE Handbook, Janssens describes furniture calorimeters as devices that measure the heat release rate of objects such as chairs, sofas and mattresses. The specimen stands on a load cell platform beneath a hood and exhaust system.[1] In the 2008 edition of the Fire Protection Handbook, Madrzykowski and Stroup call them open calorimeters, because the item burns freely rather than inside a test room.[5] Hirschler describes furniture calorimeter tests as an added way of assessing the heat release of full-scale items, with the product on a load cell under a hood.[6] The term thus names a test configuration rather than one particular standard.

The heat release rate is the rate at which the combustion reactions of a burning item produce heat, expressed in kilowatts.[4] Babrauskas calls it the single most important variable describing fire hazard, because it puts a number on the question of how big a fire is. Knowing the full-scale heat release rate of an object makes a separate inquiry into its flame spread unnecessary. Slow or incomplete flame spread shows up directly as a low or slowly growing curve.[2]

The calorimeter does not measure heat directly. Thornton showed in 1917 that many organic liquids and gases release a nearly constant amount of heat per unit mass of oxygen consumed. Huggett found the same for organic solids, with an average of 13.1 kJ per gram of oxygen, accurate to within ±5 % with very few exceptions.[7] The heat of combustion of the fuel itself can vary widely, but the method does not require it to be known or constant.[2] Measuring the oxygen removed from the air is therefore enough to determine the heat released.[7]

Development and standard test methods

Before 1982 the heat release rate of burning items could not be discussed, because no measuring tools existed. In that year Babrauskas and co-workers at NIST, then still the NBS, developed the furniture calorimeter. It allowed simple, low-cost testing of whole furniture articles under realistic burning conditions, with smoke and toxic gases measured as well.[2] At about the same time, Heskestad developed an open-burning calorimeter at the Factory Mutual Research Corporation.[4] Other open calorimeters were built by NORDTEST, Underwriters Laboratories, FM Global Research and Statens Provningsanstalt in Sweden.[5]

A European standard based on furniture calorimeter testing followed.[2] This is the Nordic method NT FIRE 032 for the burning behaviour of upholstered furniture, published by NORDTEST in 1987 and in a second edition in 1991.[8][9] United States standards then developed the method further, in ASTM E1537 for upholstered furniture and ASTM E1590 for mattresses.[2] ASTM E1822 covers stacked chairs, which became a concern after the 1978 Cathedral Hill Hotel fire in San Francisco.[3] By 2008, furniture calorimeters were available in many test laboratories in the United States and abroad.[2]

The first large-scale fire test for upholstered furniture to gain wide acceptance was California Technical Bulletin 133 (T.B. 133). It originally burned a real item or a full-scale mock-up inside a room, ignited with 90 g (3.2 oz) of balled-up newspaper.[3] After a NIST research programme, the method was revised in 1990. A gas burner replaced the newspaper to improve reproducibility, and the main pass/fail criterion became a heat release rate of 80 kW. Testing in the hood of a large-scale furniture calorimeter was added as an option.[3]

In 1992 the test became mandatory in California for certain public occupancies, though not for residential use. ASTM International adopted it as ASTM E1537.[3] The California mattress test T.B. 129 uses the same features with a burner beside the edge of the specimen and a limit of 100 kW; it became ASTM E1590.[3]

In the equivalence table of the 2008 Fire Protection Handbook, ASTM E1537 corresponds to NFPA 266 and UL 1056, and ASTM E1590 to NFPA 267 and UL 1895. All four NFPA and UL documents are marked as withdrawn.[10] UL 1975, a test for foamed plastics used for decorative purposes, is a standard version of a furniture calorimeter with a 340 g wood crib as its ignition source.[6] By 2008 the NFPA Technical Committee on Fire Tests was developing NFPA 289, a furniture calorimeter test for all kinds of individual fuel packages.[6][10]

In the United States, the open-flame regulation of the Consumer Product Safety Commission (CPSC) for mattress flammability took effect on 1 July 2007. Its test, based on NIST research, applies gas burners of 18 kW to the top and 9 kW to the side of the mattress to simulate burning bedclothes.[3] The CPSC allows testing in an open calorimeter or in a burn room of 3.05 × 3.66 m (10 × 12 ft). The permitted peak is 200 kW, and the limit for total heat released in the first 10 minutes was lowered from 25 MJ to 15 MJ.[11] Radiant heating from the upper part of a room greatly increases the burning of a mattress, so the two options are not equivalent. Babrauskas (2008b) expected manufacturers to choose the open calorimeter.[11]

Apparatus and measurements

In a furniture calorimeter the specimen rests on a load cell platform beneath a hood and exhaust system. Instruments in the exhaust duct measure the flow rate, the concentrations of oxygen and other gases, and smoke obscuration.[1] The hood is usually also instrumented for the temperature and velocity of the exhaust gases and for carbon dioxide and carbon monoxide.[12] The NBS furniture calorimeter also measured mass loss, smoke concentration and heat flux.[5] The ASTM furniture standards also have room versions, with the item on a load cell near a corner of a room 2.4 m × 3.7 m and 2.4 m high.[13][6]

The calculation needs at least two measurements: the flow rate of the combustion products in the exhaust and a product gas concentration in that flow.[14] All combustion products must be collected, and the gases must be sampled far enough downstream to be well mixed. The inflow of air then does not have to be measured.[15] Some large-scale calorimeters can safely handle fires of tens of megawatts.[14] By 2008, FM Global, Underwriters Laboratories and Southwest Research Institute had enlarged their calorimetry laboratories to test larger commodities.[5]

The NIST tests for the World Trade Center investigation show the scale such apparatus can reach. Single office workstations burned under a 9 m by 12 m hood with a capacity of 10 MW, with unrestricted air supply.[16] Eight bi-directional probes with thermocouples measured the flow in the 2 m exhaust duct. A paramagnetic analyser measured oxygen, and an infrared analyser measured carbon monoxide and carbon dioxide, which show how complete the combustion is.[17] Four load cells of 900 kg capacity recorded the mass of the item, giving the mass loss rate and the effective heat of combustion.[17]

Burning an item inside a room vented to the hood includes the effect of the room. A lack of oxygen in the room can, however, prevent complete combustion. The calorimeter also cannot distinguish burning inside the room from burning of gases outside it.[12] Burning in the open with excess oxygen allows complete combustion and gives a better measure of the total heat release rate.[5]

Calculating the heat release rate

The heat release rate follows from an oxygen balance over the hood. It equals the heat released per kilogram of oxygen multiplied by the mass of oxygen removed per second. That mass is the oxygen carried in with the air minus the oxygen leaving in the exhaust gas.[15]

Q˙=E(m˙aYO2a−m˙eYO2e)

In practice the equation is rewritten, because oxygen analysers read mole fractions on a dried gas sample and flow measurements give volume rather than mass.[15] The working forms are given under the oxygen consumption principle. When only oxygen is measured, they are accurate to within about ±10 %, provided combustion is complete.[18]

For design work, a measured curve is often approximated by a t-squared fire, whose heat release rate grows with the square of the time since a virtual origin. The growth coefficient follows from the growth time, the time at which the fire would reach 1055 kW.[19] The virtual time is the approximate moment at which the test curve began to follow the power law. The energy released up to a given time is the area under the curve.[20]

Q˙=α(t−tv)2,α=1055tg2,Qtot=α(t−tv)33

The heat release rate needed for flashover in a room can be estimated from the area and height of its ventilation opening. The health care chapter of the handbook gives this expression for it.[21]

Q˙fo≥750Ah

Symbols
Symbol Description Unit
Q˙ heat release rate kW
E heat released per mass of oxygen consumed, about 13,100 kJ/kg
m˙a, m˙e mass flow rate of combustion air and of exhaust gas kg/s
YO2a, YO2e mass fraction of oxygen in the combustion air (0.232 in dry air) and in the exhaust gas –
α fire growth coefficient kW/s²
t, tv time since the start of the test, virtual time of origin s
tg growth time, the time to reach 1055 kW s
Qtot energy released since the virtual origin kJ
Q˙fo heat release rate needed for flashover kW
A, h area and height of the ventilation opening m², m

The values of E and the oxygen mass fraction are from Janssens (2002).[15] The t-squared form describes only the growth phase. It does not predict when the item stops following the law or when its fuel runs out.[20]

Worked example. Schifiliti et al. list NBS furniture calorimeter Test 38, a sofa of 51.5 kg (specimen F32). Its growth time is 100 s, its growth coefficient 0.1055 kW/s², its virtual time 50 s and its peak 3000 kW.[22] The coefficient checks out as 1055/100² = 0.1055 kW/s². The sofa then reaches 1000 kW 97.4 s after the virtual origin, between the ultra-fast (75 s to 1 MW) and fast (150 s) design growth rates.[23]

The standard ISO 9705 room has a doorway 0.8 m wide and 2.0 m high.[24] The opening area is 1.6 m² and the square root of its height is 1.414, so flashover needs 750 × 1.6 × 1.414 = 1697 kW. This agrees with the 1.5–1.75 MW that Babrauskas gives for a room ventilated through one normal-sized door.[4]

The time to reach 1697 kW is the square root of 1697/0.1055, or 126.8 s after the virtual origin. Adding the virtual time gives 176.8 s after the start of the test. If the open-burning curve applied unchanged in the ISO 9705 room, this simple calculation would put the flashover threshold at about 177 s after the test started. The peak of 3000 kW would follow at 50 + 168.6 = 218.6 s, with about 169 MJ released during the fitted growth phase since the virtual origin. This fits the handbook statement that a medium-sized sofa can reach 2 to 3 MW within 3 to 5 minutes.[25] In a real room the developing hot layer increases the burning rate of items before flashover, so the time could be shorter.[26]

Typical results

Upholstered furniture normally has the highest heat release rate of all the goods in a normal home.[27] A medium-sized sofa can reach 2,000 to 3,000 kW only 3 to 5 minutes after ignition, and about 5,000 kW is common for large or sectional sofas. Less than 2 MW is enough for flashover in a room with a normal-sized door opening.[25] Introduction to Fire Safety Engineering uses 2.5 MW as a reference value for an average living-room sofa.[23]

In NIST tests, chair F21 with polyurethane foam and polyolefin fabric reached about 2 MW. With cotton fabric, as in chair F24, the peak fell to 700 kW.[28] More examples are discussed under upholstered furniture fire.

Schifiliti et al. collected the NBS furniture calorimeter tests in a table of t-squared fits for the design of detection systems.[22] The selection below shows the range of behaviour.

Selected NBS furniture calorimeter tests fitted with t-squared growth[22]
Test Item (mass) Growth time (s) α (kW/s²) Virtual time (s) Peak heat release rate (kW)
19, later stage Chair F21 (28.15 kg) 50 0.4220 190 2000
22 Chair F24 (28.3 kg) 350 0.0086 400 700
27 Chair F29 (14.0 kg) 100 0.1055 70 1850
37 Loveseat F31 (40.40 kg) 80 0.1648 100 2750
38 Sofa F32 (51.5 kg) 100 0.1055 50 3000
41 Plywood wardrobe with fabrics (36.0 kg) 40 0.6594 40 6000
56 Wood-frame chair with latex foam cushions (11.2 kg) 500 0.0042 50 85
67, later growth Mattress and box spring (62.36 kg) 350 0.0086 400 500

The data suggest a range of 50 to 500 s for the growth time. Only a few fast fires had growth times below 50 s, and three slow fires had values above 500 s.[20] Some items never grew: one group chair with a metal frame and foam cushion never exceeded 50 kW.[22] Chairs that do not support fire spread over their surface normally show full-scale peaks of 20 to 100 kW.[29]

Idealised NIST curves in the health care chapter of the handbook give peaks of 2000 kW for a plywood wardrobe and 250–300 kW for chairs. For beds they give 2720 kW with latex foam, 1620 kW with polyurethane, 970 kW with cotton and polyester, and 30 kW with fire-retardant cotton.[30] Office workstations burned under the NIST 10 MW hood reached net peaks of 3.82 to 7.72 MW after subtracting the ignition burner.[31]

Use in regulation and design

For health care, detention and similar occupancies, Babrauskas (2008b) states that T.B. 133-complying furniture should be used, as described in the occupancy chapters of NFPA 101, the Life Safety Code. In the consumer market there was essentially no demand for such furniture.[25] T.B. 133 limits the peak to less than 80 kW, which presents a negligible fire hazard in almost any circumstance.[28]

For a health care occupancy, NFPA 101, as described in the 2008 edition of the Fire Protection Handbook, suggests a maximum of 250 kW for upholstered furniture, mattresses and wardrobes.[21] The heat release rate needed for flashover in a patient room can then be compared with values from the NIST furniture calorimeter.[21] In United States film and television studios, as described by Grill (2008), foam plastic scenery and props are limited to 100 kW when tested to UL 1975.[32]

In fire engineering design, the primary use of furniture calorimeter data is as input to compartment fire models and smoke transport models.[1] Most fire models are driven by a heat release rate curve entered by the user. In fire reconstructions, an exemplar of the first item ignited may be tested if its construction is known.[26] Budnick et al. list furniture calorimeter data as one of three kinds of laboratory data, beside the cone calorimeter and ISO 9705 room burns. The data must correspond to the application or the correlation in which they are used.[33] The t-squared fits above can be used to choose a growth rate for detector spacing, for example for a warehouse storing sofas like the one in Test 38.[20][34]

Several guidance documents rank furniture calorimeter data among their sources of design data. The 2008 handbook describes six NFPA 555 techniques in order of preference. They begin with a full compartment fire test, then tests on individual full-scale items and large-scale mock-ups. Small-scale tests of representative composites and individual materials follow; quantitative or relative estimates from test results come last.[6][35] The International Fire Engineering Guidelines (2005) rank, for the rate of fire growth, full-scale experiments first, then furniture calorimeter data, statistical data and finally t-squared fires.[36] In Australia, the Fire Safety Verification Method data sheets (2020) list furniture calorimeter tests among the sources for the growth stage of a design fire.[37]

Because full-scale testing costs much more effort than testing small specimens, bench-scale predictions have been sought.[25] The European research programme Combustion Behavior of Upholstered Furniture (CBUF), completed in 1995, produced prediction equations based on cone calorimeter data.[25] Its predictions were checked against peaks, total heat and times to peak measured in the furniture calorimeter.[38] The intermediate-scale calorimeter tests 1.0 m by 1.0 m specimens, but its data are not full scale either and need further analysis before use in fire modelling.[24]

Limitations and uncertainty

The main limitation is the absence of a room. At very low heat release rates, burning in a room and in the open give the same result. After flashover, room values can differ drastically from open burning, mainly through radiant heat from the hot gas layer and hot surfaces.[4] For upholstered chairs the room effect stays at about 20 % up to a 1 MW fire, because chairs largely see their own surfaces.[4][39]

For mattresses it appears at much lower values. In the CBUF tests a drastic room effect occurred in the ISO 9705 room above about 300 kW of open-air heat release.[40] One mattress peaked at 313 kW in the furniture calorimeter and at 1700 kW in the room.[41]

A heat release rate measured in a room cannot be extrapolated to rooms with larger ventilation, whereas open-burning data could apply to such well-ventilated rooms.[39] For items that support fire spread, the measured value is only a lower bound of what may occur in a real fire. Such fires can ignite other items in the room that were not part of the test.[29]

A furniture test is a system test, not a materials test. With the same materials, a smaller chair can pass T.B. 133 while a larger one fails.[3] Data therefore describe the specific item tested, and a similar item may perform very differently.[24] The ignition source matters too: an item that hardly responds to one source may show a runaway fire with a slightly larger one.[11]

The measurement itself carries an uncertainty of several percent. Calorimetric measurements involve dozens of independent measurements. For the Single Burning Item test and the room corner test, relative expanded uncertainties of 0.07 to 0.14 have been estimated. The oxygen concentration, the heat of combustion factor and the mass flow rate were the main sources.[42] For the NIST workstation tests, an analysis of 78 inputs gave a relative standard uncertainty of 0.061.[43] Errors grow when much carbon monoxide or soot is produced, or when the products include species such as hydrogen chloride.[18]

See also

Notes

  1. ↑ 1.0 1.1 1.2 1.3 1.4 1.5 Janssens (2002), p. 3-57.
  2. ↑ 2.0 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 Babrauskas (2008b), p. 6-119.
  3. ↑ 3.0 3.1 3.2 3.3 3.4 3.5 3.6 3.7 Babrauskas (2008b), p. 6-116.
  4. ↑ 4.0 4.1 4.2 4.3 4.4 4.5 Babrauskas (2002b), p. 3-1.
  5. ↑ 5.0 5.1 5.2 5.3 5.4 Madrzykowski and Stroup (2008), p. 2-41.
  6. ↑ 6.0 6.1 6.2 6.3 6.4 Hirschler (2008a), p. 6-50.
  7. ↑ 7.0 7.1 Janssens (2002), p. 3-45.
  8. ↑ Madrzykowski and Stroup (2008), p. 2-47.
  9. ↑ Babrauskas (2008b), p. 6-128.
  10. ↑ 10.0 10.1 Hirschler (2008a), p. 6-38.
  11. ↑ 11.0 11.1 11.2 Babrauskas (2008b), p. 6-117.
  12. ↑ 12.0 12.1 Madrzykowski and Stroup (2008), p. 2-40.
  13. ↑ Hirschler (2008a), p. 6-49.
  14. ↑ 14.0 14.1 Madrzykowski and Stroup (2008), p. 2-37.
  15. ↑ 15.0 15.1 15.2 15.3 Janssens (2002), p. 3-46.
  16. ↑ Ohlemiller et al. (2005), p. 9.
  17. ↑ 17.0 17.1 Ohlemiller et al. (2005), p. 10.
  18. ↑ 18.0 18.1 Janssens (2002), p. 3-47.
  19. ↑ Schifiliti et al. (2002), p. 4-10.
  20. ↑ 20.0 20.1 20.2 20.3 Schifiliti et al. (2002), p. 4-14.
  21. ↑ 21.0 21.1 21.2 O'Connor (2008), p. 20-144.
  22. ↑ 22.0 22.1 22.2 22.3 Schifiliti et al. (2002), p. 4-13.
  23. ↑ 23.0 23.1 Van Rossum (2027), §4.5.
  24. ↑ 24.0 24.1 24.2 Babrauskas (2002b), p. 3-3.
  25. ↑ 25.0 25.1 25.2 25.3 25.4 Babrauskas (2008b), p. 6-120.
  26. ↑ 26.0 26.1 Custer et al. (2008), p. 3-117.
  27. ↑ Babrauskas (2002b), p. 3-28.
  28. ↑ 28.0 28.1 Babrauskas (2002b), p. 3-29.
  29. ↑ 29.0 29.1 Babrauskas (2008b), p. 6-121.
  30. ↑ O'Connor (2008), p. 20-145.
  31. ↑ Ohlemiller et al. (2005), p. xxxi.
  32. ↑ Grill (2008), p. 20-214.
  33. ↑ Budnick et al. (2008), p. 3-147.
  34. ↑ Schifiliti et al. (2002), p. 4-17.
  35. ↑ Hirschler (2008a), p. 6-51.
  36. ↑ National Research Council of Canada et al. (2005), p. 2.4-10.
  37. ↑ Australian Building Codes Board (2020a), Data Sheet B1, p. 10.
  38. ↑ Babrauskas (2008b), pp. 6-122–6-123.
  39. ↑ 39.0 39.1 Babrauskas (2002b), p. 3-2.
  40. ↑ Babrauskas (2002b), p. 3-21.
  41. ↑ Babrauskas (2002b), p. 3-22.
  42. ↑ Hamins and McGrattan (2007), p. 3-2.
  43. ↑ Ohlemiller et al. (2005), p. 53.

References

  • Australian Building Codes Board. (2020a). Data sheets: Fire Safety Verification Method (Handbook annex, Print version 1.1).
  • Babrauskas, V. (2002b). Heat release rates. In P. J. DiNenno, D. Drysdale, C. L. Beyler, W. D. Walton, R. L. P. Custer, J. R. Hall Jr., & J. M. Watts Jr. (Eds.), SFPE handbook of fire protection engineering (3rd ed., pp. 3-1–3-37). National Fire Protection Association; Society of Fire Protection Engineers.
  • Babrauskas, V. (2008b). Upholstered furniture and mattresses. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 6-103–6-128). National Fire Protection Association.
  • Budnick, E. K., Hunt, S. P., & Wright, M. T. (2008). Closed form enclosure fire calculations. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 3-145–3-166). National Fire Protection Association.
  • Custer, R. L. P., Sheppard, D. T., & Wood, C. B. (2008). Applying models to fire protection engineering problems and fire investigations. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 3-111–3-120). National Fire Protection Association.
  • Grill, R. A. (2008). Motion picture and television studios and soundstages. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 20-211–20-215). National Fire Protection Association.
  • Hamins, A., & McGrattan, K. (2007). Verification and validation of selected fire models for nuclear power plant applications: Vol. 2. Experimental uncertainty (NUREG-1824; EPRI 1011999). U.S. Nuclear Regulatory Commission, Office of Nuclear Regulatory Research; Electric Power Research Institute.
  • Hirschler, M. M. (2008a). Concepts and protocols of fire testing. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 6-35–6-60). National Fire Protection Association.
  • Janssens, M. (2002). Calorimetry. In P. J. DiNenno, D. Drysdale, C. L. Beyler, W. D. Walton, R. L. P. Custer, J. R. Hall Jr., & J. M. Watts Jr. (Eds.), SFPE handbook of fire protection engineering (3rd ed., pp. 3-38–3-62). National Fire Protection Association; Society of Fire Protection Engineers.
  • Madrzykowski, D., & Stroup, D. W. (2008). Flammability hazard of materials. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 2-31–2-48). National Fire Protection Association.
  • National Research Council of Canada, International Code Council, Department of Building and Housing, & Australian Building Codes Board. (2005). International fire engineering guidelines (2005 ed.). Australian Building Codes Board.
  • O'Connor, D. J. (2008). Health care occupancies. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 20-139–20-155). National Fire Protection Association.
  • Ohlemiller, T. J., Mulholland, G. W., Maranghides, A., Filliben, J. J., & Gann, R. G. (2005). Fire tests of single office workstations (NIST NCSTAR 1-5C). National Institute of Standards and Technology.
  • Schifiliti, R. P., Meacham, B. J., & Custer, R. L. P. (2002). Design of detection systems. In P. J. DiNenno, D. Drysdale, C. L. Beyler, W. D. Walton, R. L. P. Custer, J. R. Hall Jr., & J. M. Watts Jr. (Eds.), SFPE handbook of fire protection engineering (3rd ed., pp. 4-1–4-43). National Fire Protection Association; Society of Fire Protection Engineers.
  • Van Rossum, E. (2027). Introduction to Fire Safety Engineering. Fire Safety Academy.