Skip to content
From Fire Safety Academy Wiki
Revision as of 19:48, 12 November 2026 by Mark (talk | contribs) (Add SVG figures fse-preaction-sprinkler-system-v1.svg, fse-preaction-sprinkler-system-v2.svg, fse-preaction-sprinkler-system-v3.svg (list-5 visual plan))
Contents
  1. 1Detection and interlock logic
  2. 2Water-filled and gas-filled arrangements
  3. 3Thermal operation and water delivery
  4. 4Waterflow signals and valve supervision
  5. 5Selecting the operating demand
  6. 6Discharge, pressure and assigned area
  7. 7Inspection and evidence of readiness
  8. 8See also
  9. 9Notes
  10. 10References

Preaction sprinkler system is a sprinkler system in which a detection-controlled valve holds water back from piping fitted with closed automatic sprinklers. [1][2] The release sequence depends on whether the system is single-interlock, double-interlock or non-interlock. Detection, sprinkler operation and water arrival can therefore occur at different times. The arrangement limits some unwanted water-admission scenarios but adds release components and, for some sequences, additional delivery delay. [2]

Sprinkler performance combines thermal operation, timely water delivery and an effective discharge pattern. These functions are related, but evidence for one does not establish all the others. The protected hazard and the installed geometry remain part of the design basis. Hydraulic calculations establish delivery under stated assumptions, while the suitability of those assumptions depends on the application and the supporting fire-performance evidence. [3][4] Heat access and water access can be obstructed separately. [5]

Detection and interlock logic

The preaction sprinkler system differs from a dry pipe sprinkler system because a supplemental detection arrangement participates in admitting water. It also differs from a deluge system because its sprinkler outlets remain closed until their own operating elements release. A smoke detector may form part of a specified detection arrangement, but the term preaction does not prescribe one universal detector technology. [1][2]

Preaction sprinkler system: functional relationships. Mechanism and application limits are distinguished.
Preaction sprinkler system: functional relationships. Schematic explanation; dimensions are not installation instructions.

The operating logic is relevant to data centre fire protection and other applications where unwanted discharge is a design concern; the mechanism still needs evaluation against the actual hazard and delay. A dry pipe valve and a preaction release valve do not have identical initiating conditions. Combining their names without describing the interlock can conceal the sequence that determines water delivery. [2][6]

In a single-interlock arrangement, detection admits water before a closed sprinkler necessarily opens; in a double-interlock arrangement, both required events must occur. A non-interlock arrangement permits either event to initiate admission. These distinctions change the consequences of detector operation, accidental sprinkler damage and a developing fire, even where the final sprinkler discharge is hydraulically similar. [2]

The United States nuclear-power methodology NUREG/CR-6850 (2005) explicitly includes preaction detection logic among the conditions to document before crediting suppression; it also records installed head characteristics, spacing and relevant obstructions. That sector-specific analysis does not select a preaction system for another occupancy. It illustrates why a system name alone is insufficient to determine when effective water delivery begins. [7]

Water-filled and gas-filled arrangements

The four basic arrangements described by Fleming are wet pipe, dry pipe, preaction and deluge; each has a different relationship between the discharge outlets, the water and the release mechanism. Their names do not merely describe whether a pipe happens to be wet during an inspection. The sequence by which water becomes available defines the functional distinction. [1]

Four release arrangements. Mechanism and application limits are distinguished.
Four release arrangements. Schematic explanation; dimensions are not installation instructions.

A wet system holds water behind closed automatic sprinklers; when a sprinkler's thermal element releases, water can discharge from that opening without first filling the distribution piping. This arrangement avoids the gas-evacuation delay associated with a dry system. It nevertheless depends on an open supply route and suitable environmental conditions for the water-filled piping. [8][9]

A dry system holds pressurised air or nitrogen downstream of a dry pipe valve; an opened sprinkler releases gas, allowing the valve to trip and admit water. The total delivery time includes both the valve's response and water travel. Residual water can still collect in low points, so the word dry does not imply absence of corrosion or freezing concerns throughout the installation. [8][10][11]

A preaction system uses automatic sprinklers with an additional detection-controlled valve; in the single-interlock arrangement described by the handbook, detection admits water to the piping. Double-interlock arrangements require both detection and sprinkler operation, while non-interlock arrangements permit either condition to initiate admission. The choice changes the balance between unwanted water admission and delivery delay. [2]

A deluge system uses open discharge outlets with a common release valve; once that valve admits water, all connected open outlets form part of the discharge demand. A closed sprinkler system instead discharges only through openings that have operated or otherwise opened. This difference affects both release analysis and water-supply calculations. [1][12]

The installed release chain must be tested as the chain actually used; testing an alarm switch, a detector or a valve independently leaves other functions unexamined. Water delivery and the restoration of any temporarily bypassed interfaces remain separate checks. The handbook describes these distinctions in its treatment of alarm testing and system maintenance. [13][10]

Thermal operation and water delivery

An automatic sprinkler combines a heat-responsive element with a water-distribution device; the element responds to its own temperature, which follows the surrounding gas conditions with a delay. Water then needs an open discharge path and an adequate supply. These are successive physical processes, so activation time and effective water delivery are not interchangeable quantities. [14][15]

Activation is followed by delivery. Mechanism and application limits are distinguished.
Activation is followed by delivery. Schematic explanation; dimensions are not installation instructions.

Temperature rating describes the nominal operating temperature established under specified test conditions; thermal sensitivity describes how quickly the element heats in a moving hot gas stream. A sprinkler can therefore have the same temperature rating as another device while responding differently to a growing fire. Gas velocity, heat conduction and the element's position beneath the ceiling influence that response. [16][11]

The response time index provides a measure of thermal sensitivity; lower values represent faster response under comparable convective heating conditions. The handbook relates this index to a thermal time constant and the square root of gas velocity. Conduction into the sprinkler frame introduces another influence, especially where convection is weak. [17][18]

A wet system contains water up to the closed sprinklers; a dry system contains pressurised gas downstream of a valve, so sprinkler operation first initiates pressure decay and valve release. Water must then travel through the piping. A preaction arrangement adds a detection-controlled release condition whose logic depends on the particular system. [19][14]

Fleming separates dry-system water delivery into valve trip time and water transit time; a quicker release can reduce the first component without eliminating the second. The number and location of open sprinklers affect gas venting and water travel. Consequently, a remote test gives a defined comparison condition, rather than reproducing every possible fire sequence. [11]

The International Fire Engineering Guidelines (IFEG) analysis process likewise distinguishes local activation conditions, notification delays and the effect of suppression; its 2005 framework treats these as inputs to an evolving fire scenario. A detector signal cannot by itself establish that water has reached the burning material. That distinction matters whenever the fire model changes growth behaviour at an assumed intervention time. [20]

Waterflow signals and valve supervision

A waterflow alarm communicates that water is moving through the monitored sprinkler arrangement; depending on the equipment, movement can be detected through an alarm valve, a pressure-operated switch or a vane-type flow switch. The resulting signal can operate local notification and a remote receiving system. The hydraulic event and the subsequent notification are separate stages. [21]

An alarm check valve normally retains water on the sprinkler side; when its clapper lifts during discharge, an alarm passage becomes available. Water or pressure from that passage can operate mechanical or electrical alarm equipment. A retard arrangement limits unwanted signalling from short supply-pressure disturbances. [22]

The alarm passage does not create an additional extinguishing supply; nor does an alarm check valve perform the same purpose as an isolation valve. Its condition influences check and alarm functions within the installed arrangement. The system still requires an available upstream supply and a discharge route through the operating sprinkler. [21][23]

A supervisory signal concerns an abnormal condition that can impair readiness, such as an unsuitable control-valve position; the handbook distinguishes these signals from waterflow fire alarms. The receiving organisation interprets and acts on them differently. This separation avoids treating a warning of reduced protection as evidence that suppression is already operating. [24][25]

Testing through an alarm bypass can exercise the alarm pipe and connected devices; it does not necessarily prove that the main valve opens properly under actual sprinkler discharge. The inspector's test connection produces a defined discharge for testing appropriate waterflow functions. For a dry system, a suitable remote test can also measure valve trip and water delivery. [13][26]

Interconnected systems introduce further consequences; a test signal can initiate building notification, remote response or operation of other equipment. The handbook therefore treats test coordination and restoration of bypassed interfaces as part of the procedure. The final operating state matters as much as the observed signal during the test. [24]

Selecting the operating demand

A design area represents the sprinklers assumed to operate together for a particular hydraulic design method; it is not the whole installed system, nor necessarily the physical area of burning fuel. The selected area connects the fire protection design criteria to a manageable hydraulic demand. Its size and arrangement must remain consistent with the selected hazard and sprinkler method. [27][28]

The hydraulically demanding area is not always the one furthest away on a floor plan; elevation, pipe diameter, branch arrangement and supply routes influence the required pressure. Different candidate areas can therefore govern different aspects of a calculation. Physical distance alone is an unreliable substitute for following the actual water path. [28][29]

The handbook illustrates converting an operating area into a whole number of sprinklers. The count is rounded upwards so that the selected coverage does not fall short of the stated area. [30]

N=⌈AdAs⌉

Symbol Description Unit
N Minimum whole number in this area conversion dimensionless
Ad,As Design area and assigned area per sprinkler m²

The source's rounded example shows the effect of this conversion. Dividing 139.4 m² by an assigned 18.6 m² per sprinkler gives approximately 7.49 sprinklers. Rounding upwards gives eight operating sprinklers, because a fractional outlet cannot represent the assumed arrangement. Their actual combined assigned area is then 148.8 m² using those rounded inputs. [30]

Area shape also affects the pipe network being calculated; the handbook describes extending the area along branch lines and rounding the required number of sprinklers on those lines. This creates a physical arrangement rather than an abstract area total. Pipe flows must then accumulate through that arrangement towards the supply, accounting for pressure-dependent discharge at each sprinkler. [31][29]

A dry system introduces water-delivery delay, during which additional sprinklers may operate; historical NFPA 13 approaches reported by the handbook account for that disadvantage through specified design adjustments. Fast thermal response can support different adjustments only within the permitted application and conditions. Neither concept creates a universal percentage change applicable to every sprinkler or storage system. [31][32]

Storage protection also includes methods specified by a number of operating sprinklers and minimum pressure; those criteria are not interchangeable with a density-area selection. Changing spacing does not automatically permit a reduced minimum pressure for such an application. The calculation must preserve the full tested design basis, including the sprinkler type and storage configuration. [33]

Discharge, pressure and assigned area

Hydraulic calculation starts with the water discharge required at an individual operating sprinkler; in a density-area design, the assigned area and selected application density determine that minimum discharge. The assigned area comes from the spacing method used for the design. It is not an assertion that every point beneath the sprinkler receives an identical local quantity. [27][34]

The discharge coefficient, or K-factor, relates flow to the square root of pressure at the sprinkler; its numerical value depends on the units used for both quantities. A value stated for pressure in bar cannot be inserted unchanged when pressure is in kilopascals. The unit convention is part of the coefficient's definition. [35][36]

q=dAs=Kp,p=(qK)2

Symbol Description Unit
q Required sprinkler flow L/min
d,As Design density and assigned area L/(min·m²); m²
p Sprinkler discharge pressure kPa
K Discharge coefficient in this convention L/(min·√kPa)

Using the rounded metric inputs in Fleming's example, 6.1 L/(min·m²) over 15.6 m² requires 95.2 L/min. With a nominal K-factor of 8.0 L/(min·√kPa), the calculated pressure is approximately 142 kPa. The corresponding bar-based K-factor is 80 L/(min·√bar). The example illustrates the calculation method, rather than selecting a density for a new hazard. [36][35]

Other sprinklers in the operating area may have greater pressure than the limiting sprinkler; their individual discharges then exceed the minimum flow calculated from density and coverage alone. Total demand is the sum of those actual calculated flows, with applicable additional demands considered separately. Simply multiplying nominal density by nominal design area can therefore understate the hydraulic result. [31][37]

A higher K-factor reduces the pressure required for a specified flow; it does not establish a particular spray pattern, response sensitivity or suitability for storage. Those characteristics depend on the complete sprinkler design and its tested application. Hydraulic equivalence is therefore narrower than equivalent fire protection performance. [38][39]

Inspection and evidence of readiness

Inspection, testing and maintenance provide different forms of evidence; visual inspection can identify closed valves, damaged sprinklers, unsuitable storage or missing components. A functional test can demonstrate operation under a defined condition. Maintenance addresses deterioration or defects, while records connect those actions to the installation's intended state. [40]

The handbook treats changes in occupancy, materials, partitions, ceilings and heating as relevant to sprinkler performance; a system can remain mechanically intact while its design assumptions no longer match the protected hazard. An inspection limited to leaks or gauge readings would miss that change. Readiness therefore includes both equipment condition and the continued relevance of the design basis. [41]

An indicating control valve makes its position observable, while electrical supervision can report departure from the intended position; neither proves that every downstream route is unobstructed. Flow testing provides different evidence by exercising a water path. Valve identification and drawings help establish which part of the protection is affected by a particular closure. [23][42]

Deposits on a sprinkler can interfere with heating or mechanical release; corrosion can affect both the operating mechanism and water passage. Inside pipes, scale or foreign material can migrate to smaller fittings and outlets. These mechanisms explain why an apparently pressurised system may still have reduced effective protection. [43]

The purpose of a test determines its interpretation; an alarm test can establish signalling without demonstrating the full water demand. A main-drain comparison can reveal a changed supply condition without reproducing every operating sprinkler. A pump start test exercises starting functions, while a flow test examines delivery capability over selected operating points. [13][44]

The IFEG links commissioning to the performance assumed in a fire engineering evaluation; it also links subsequent management and maintenance to keeping those assumptions valid. This provides a useful framework for reading test records: the observation must answer the performance question being claimed. Successful operation of one component is evidence about that component and the functions actually exercised. [45]

See also

Notes

  1. ↑ 1.0 1.1 1.2 1.3 Fleming (2002), ch. 4-3, pp. 4-72–4-73.
  2. ↑ 2.0 2.1 2.2 2.3 2.4 2.5 Cote (2008), ch. 16.3, pp. 16-35–16-37.
  3. ↑ Fleming (2002), ch. 4-3, pp. 4-72–4-75,4-85–4-87.
  4. ↑ Cote (2008), ch. 16.1, pp. 16-4–16-9.
  5. ↑ Cote (2008), ch. 16.3, pp. 16-45–16-47.
  6. ↑ National Research Council of Canada et al. (2005), pp. 1.7-5–1.7-8.
  7. ↑ Science Applications International Corporation and Sandia National Laboratories (2005), pp. 11-16.
  8. ↑ 8.0 8.1 Cote (2008), ch. 16.3, pp. 16-34–16-35.
  9. ↑ Fleming (2002), ch. 4-3, pp. 4-72.
  10. ↑ 10.0 10.1 Cote (2008), ch. 16.11, pp. 16-211–16-212.
  11. ↑ 11.0 11.1 11.2 Fleming (2002), ch. 4-3, pp. 4-85.
  12. ↑ Cote (2008), ch. 16.9, pp. 16-181–16-182.
  13. ↑ 13.0 13.1 13.2 Cote (2008), ch. 14.4, pp. 14-42–14-43.
  14. ↑ 14.0 14.1 Fleming (2002), ch. 4-3, pp. 4-72–4-73,4-85.
  15. ↑ Cote (2008), ch. 16.1, pp. 16-4–16-6.
  16. ↑ Cote (2008), ch. 16.2, pp. 16-17–16-18.
  17. ↑ Cote (2008), ch. 16.1, pp. 16-4–16-5.
  18. ↑ Cote (2008), ch. 16.6, pp. 16-95–16-96.
  19. ↑ Cote (2008), ch. 16.3, pp. 16-34–16-38.
  20. ↑ National Research Council of Canada et al. (2005), pp. 1.7-4–1.7-8.
  21. ↑ 21.0 21.1 Cote (2008), ch. 14.4, pp. 14-40–14-42.
  22. ↑ Cote (2008), ch. 14.4, pp. 14-41.
  23. ↑ 23.0 23.1 Cote (2008), ch. 16.3, pp. 16-42–16-43.
  24. ↑ 24.0 24.1 Cote (2008), ch. 14.4, pp. 14-43.
  25. ↑ Cote (2008), ch. 16.11, pp. 16-212.
  26. ↑ Cote (2008), ch. 16.11, pp. 16-211.
  27. ↑ 27.0 27.1 Fleming (2002), ch. 4-3, pp. 4-74–4-75.
  28. ↑ 28.0 28.1 Cote (2008), ch. 16.3, pp. 16-47–16-49.
  29. ↑ 29.0 29.1 Fleming (2002), ch. 4-3, pp. 4-74–4-80.
  30. ↑ 30.0 30.1 Cote (2008), ch. 16.3, pp. 16-48.
  31. ↑ 31.0 31.1 31.2 Cote (2008), ch. 16.3, pp. 16-48–16-49.
  32. ↑ Cote (2008), ch. 16.5, pp. 16-85.
  33. ↑ Cote (2008), ch. 16.5, pp. 16-80–16-83.
  34. ↑ Cote (2008), ch. 16.1, pp. 16-6–16-8.
  35. ↑ 35.0 35.1 Cote (2008), ch. 16.2, pp. 16-25–16-26.
  36. ↑ 36.0 36.1 Fleming (2002), ch. 4-3, pp. 4-75.
  37. ↑ Fleming (2002), ch. 4-3, pp. 4-75–4-80.
  38. ↑ Cote (2008), ch. 16.2, pp. 16-23–16-27.
  39. ↑ Cote (2008), ch. 16.5, pp. 16-80.
  40. ↑ Cote (2008), ch. 16.11, pp. 16-205–16-212.
  41. ↑ Cote (2008), ch. 16.11, pp. 16-205.
  42. ↑ Cote (2008), ch. 16.11, pp. 16-209–16-210.
  43. ↑ Cote (2008), ch. 16.11, pp. 16-207–16-212.
  44. ↑ Cote (2008), ch. 16.11, pp. 16-210–16-211,16-220.
  45. ↑ National Research Council of Canada et al. (2005), pp. 1.7-8.

References

  • Cote, A. E. (Ed.). (2008). Fire protection handbook (20th ed.). National Fire Protection Association.
  • Fleming, R. P. (2002). Automatic sprinkler system calculations. In P. J. DiNenno (Ed.), SFPE handbook of fire protection engineering (3rd ed., pp. 4-72–4-87). 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. Australian Building Codes Board.
  • Science Applications International Corporation, & Sandia National Laboratories. (2005). EPRI/NRC-RES fire PRA methodology for nuclear power facilities: Volume 2, detailed methodology (NUREG/CR-6850; EPRI 1011989). Electric Power Research Institute; U.S. Nuclear Regulatory Commission.