Skip to content
Upcoming courses

Fire development stages

Tools
From Fire Safety Academy Wiki
Revision as of 00:33, 2 November 2026 by Jan (talk | contribs) (New FSE wiki article: Fire development stages (written by Kimi Code CLI))
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Contents
  1. 1Initiation and early combustion
  2. 2Growth and fuel-controlled burning
  3. 3Hot-layer development and flashover
  4. 4Fully developed and post-flashover fire
  5. 5Decay and suppression
  6. 6Compartment fires and design scenarios
  7. 7Investigation and evidence
  8. 8Limits and common misunderstandings
  9. 9See also
  10. 10Notes
  11. 11References

Fire development stages are the phases used to describe how a fire grows, reaches developed burning and declines. A common compartment-fire sequence distinguishes growth, fully developed fire and decay. The sequence helps organise changing heat release, smoke and ventilation, but it is not a compulsory path for every fire.[1][2]

The stages are linked by transitions rather than fixed durations. Ignition can be followed by smouldering or flaming, growth can stop before flashover, and developed burning can be limited by fuel or oxygen. Suppression and fuel depletion can produce several forms of decay.[1][3]

Using stages allows different analyses to focus on the relevant period. Detection and escape often depend on early development, whereas structural exposure and firefighting may be dominated by fully developed conditions. A complete assessment nevertheless needs the whole time history.[4][5]

Initiation and early combustion

A scenario begins with an ignition source and a susceptible fuel arrangement. Gas mixtures can ignite through a pilot, liquids can produce an ignitable vapour layer, and solids can require heating and pyrolysis. The resulting combustion can be flaming, smouldering or another mode.[6][7]

Smouldering is slow, flameless surface oxidation in a condensed fuel. It can persist for long periods with limited external signs and can produce toxic products. It may remain the dominant mode or later transition to flaming when gases, heat and oxygen conditions coincide.[8][9]

Early development can include an incubation or establishment phase with slow burning before a simpler growth curve applies. The 2005 International Fire Engineering Guidelines state that this duration cannot be determined reliably and is often omitted in design fires. That simplification is usually conservative because it advances modelled growth.[3]

The first visible flame does not prove persistent fire. Transient ignition can disappear when external heating is removed, while persistent combustion needs continuing fuel production and heat feedback. This distinction affects whether a scenario develops beyond the initiating event.[6]

Growth and fuel-controlled burning

Growth involves increasing heat release, burning area or involvement of additional objects. Heat from flames and hot gases preheats new fuel, while plume transport spreads smoke and energy through the enclosure. Early spread can therefore change both fire size and the location of exposure.[10][11]

A Fuel-controlled fire has sufficient oxygen for current combustion, so fuel-gas production controls heat release. Early local fires commonly have this form because the surrounding enclosure still contains ample oxygen. Fuel type, geometry and heat feedback then dominate development.[12]

A t-squared fire represents growth as proportional to time squared. Common categories use times to 1 MW of 75 seconds for ultrafast, 150 for fast, 300 for medium and 600 for slow growth. These categories provide a simple design-fire family rather than a prediction of natural fire timing.[13][3]

Actual growth can depart from quadratic behaviour. Object-to-object delays, changing combustion efficiency and ignition of different materials can produce steps or plateaus. A simplified curve should be justified by test data, scenario evidence or a deliberately conservative choice.[14][3]

Hot-layer development and flashover

During growth, the plume carries hot gases to the ceiling and feeds an upper layer. The layer deepens and warms, exchanging radiation with lower contents and energy with boundaries. This creates a stratified compartment with hotter gas above cooler gas.[15][16]

Radiation from the hot layer and heated surfaces can preheat many objects simultaneously. Their subsequent pyrolysis and ignition increase heat release and reinforce the feedback. This is one mechanism leading towards widespread compartment involvement.[11][17]

Flashover marks the rapid transition to this widespread involvement. It is associated with upper-layer temperature and heat-flux criteria, but the source ranges are broad and measurement locations differ. A single temperature or fire power should not be used as a universal transition point.[18][17]

Flashover is not inevitable. Active suppression can interrupt growth, and insufficient heat feedback or fuel depletion can prevent the developed stage. The scenario must establish whether the required enclosure and fuel conditions exist.[1][2]

Fully developed and post-flashover fire

A Post-flashover fire has widespread fuel involvement and is commonly ventilation controlled. Abundant pyrolysis gases can be produced while openings restrict oxygen, causing incomplete combustion inside. External flames can then burn these gases after they mix with fresh air.[12][1]

This stage changes the relevant system boundary. Heat released inside the compartment can differ from total heat generated by its fuel, and external surfaces can receive substantial exposure. Internal temperatures and structural heating can remain high for a sustained period.[19][5]

One-zone models often represent developed conditions as a uniform hot compartment, while two-zone models represent earlier stratification. The switch is a simplification of changing physical conditions. Real properties can remain non-uniform even when a one-zone approximation is useful.[20][21]

Fully developed burning does not have to remain at one constant power. Vent changes, fuel distribution and external combustion can produce varying heat release. A steady design plateau is therefore a modelling choice rather than a universal characteristic.[12][14]

Decay and suppression

Decay can start because available fuel becomes depleted or because suppression reduces heat release. Other changes, such as restricted oxygen or altered boundaries, can also weaken combustion. The scenario should identify which mechanism is responsible.[1]

Fuel depletion can be spatially uneven. Some objects can burn out while others ignite later, creating several local peaks within overall decay. Treating the whole compartment as one uniformly depleted fuel package can miss this sequence.[22][10]

Suppression can cool fuel, limit oxygen or interrupt flame chemistry. Early intervention can prevent developed burning altogether, while later intervention can shorten the developed stage. Effectiveness depends on reaching the relevant reaction and fuel regions.[23][1]

Hot materials and smouldering fuel can remain after visible flame declines. Oxygen readmission can then cause reignition, and stored heat can continue to expose structures and occupants. Decay is therefore a stage of changing hazard, not an immediate return to safe conditions.[24][25]

Compartment fires and design scenarios

A Compartment fire supplies the physical setting in which these stages interact. Fuel, plume, layers, openings and boundaries change together, so a stage label must be supported by modelled or observed conditions. The same label can mean different model assumptions in different methods.[16][20]

Schematic design fires often show growth, developed heat release and decay as a simplified curve. The 2005 guidelines identify these as conservative representations for engineering evaluation. They are not intended to reproduce every irregular natural fire.[2]

Smouldering design fires need a separate treatment because they may never enter developed flaming. Their source strength and transition timing can be uncertain, and many scenarios have no smouldering phase. Inserting every stage by default would therefore be inappropriate.[2][3]

For large enclosures, spatial offsets can represent spread between regions instead of simultaneous full-area involvement. The overall curve then emerges from regional development stages. This avoids unrealistic early occupation of the entire floor.[22]

Investigation and evidence

Fire investigation uses stage evidence to reconstruct development. Burn patterns, damaged objects, opening states and the sequence of detected events can help distinguish local growth from developed burning. No single observation proves an entire timeline without its context.[26][27]

Later objects can become new ignition sources, so the first burning item found is not necessarily the original igniter. External flaming can likewise be a result of developed internal combustion rather than the initial event. Source and target roles change as stages progress.[27][19]

Smouldering can leave deep-seated evidence while producing little immediate flame. Central charring in a self-heating body can suggest internal runaway, but other source and preheating evidence still needs evaluation. Stage reconstruction should remain open to alternative sequences.[25]

Time estimates require particular caution. Smouldering-to-flaming transitions have widely varying reported durations, and self-heating time depends strongly on size and criticality. A precise timeline should not be inferred from stage names alone.[9][28]

Limits and common misunderstandings

The stages are not equal intervals and do not have standard durations. A short fire can move quickly through growth, while another can remain in early combustion for a long time. Source and scenario determine each transition.[3][9]

Flashover is not the beginning of fire, and decay is not the immediate end of hazard. The first stage begins with initial combustion, and residual heat and smouldering can persist through decay. These boundaries must be described by conditions, not labels alone.[1][24]

A design curve is not a measurement. It is a selected representation of development intended for analysis. Test data can support it, but the curve still embodies choices about growth, developed burning and decay.[2][29]

Finally, one stage need not occur everywhere simultaneously. Different objects and rooms can be in different stages, especially in large or connected spaces. The overall assessment should preserve this spatial variation.[22][30]

Growth-category times should not be mistaken for stage durations. The 75, 150, 300 and 600 second values in the t-squared family describe time to 1 MW, not time to flashover, developed burning or extinction. A scenario still needs a later-stage model and a fuel or suppression outcome.[13][3]

Early detection depends on the same development history. A slow establishment phase can delay detectable smoke or heat, while a faster growth curve can bring them forward. Omitting that phase changes the predicted response time and should be acknowledged when the result supports alarm or evacuation decisions.[3][4]

The transition from fuel control to ventilation control is another scenario-specific event. It depends on oxygen demand and opening flow rather than on a fixed point in every fire. A stage model should show when and why this limitation begins, especially if unburnt gases and external flaming become important.[12][19]

Spatial development complicates stage labels further. One object can be in decay while another begins to pyrolyse, and one room can be fully developed while the next has only local smoke. A building-wide timeline should therefore be assembled from object and compartment sequences rather than from one global stage label.[22][30]

Test data can anchor parts of this timeline. Open calorimetry can describe an item's heat release, while room tests add layer and ventilation effects. Choosing data from the wrong arrangement can shift both the predicted growth and the onset of developed burning.[14][29]

Incident reconstruction benefits from the same separation. Alarm times, visible flame reports, window failure and external flaming can be compared with candidate stages, but each observation needs its physical meaning. External flames, for example, can indicate developed internal combustion rather than initial ignition.[19][31]

Evidence should also be assigned to the correct spatial region. A room can show developed burning while an adjacent corridor has only smoke transport, and these observations should not be merged into one compartment stage. Mapping evidence to objects and spaces helps preserve the actual sequence.[30][22]

Model stages should be compared with evidence at the same times. A predicted layer average, heat release rate or opening state can be checked against measurements only when their definitions match. This avoids declaring a model correct because two different quantities happen to have similar values.[17][14]

The final timeline should distinguish observed transitions from modelling choices. A flashover criterion, for example, can be selected to switch model zones even when the exact physical transition is uncertain. Recording that distinction prevents a numerical switch from being presented as a directly observed event.[20][17]

Residual conditions belong in the final stage assessment. Stored heat can sustain boundary exposure, and hot porous fuel can react again after oxygen returns. The end of a design heat-release curve is therefore not automatically the end of the safety analysis.[24][32]

Stage criteria should be recorded explicitly in the report. These can include a heat-release threshold, a layer temperature, an opening event or a suppression action, depending on the model. Clear criteria allow another reviewer to reproduce when the analysis changes from one stage representation to another.[20][17]

Where the sequence is uncertain, alternatives can be compared rather than forcing one timeline. Different incubation durations, growth rates, opening times and decay assumptions reveal which conclusions remain stable. This approach also shows which additional observation would most improve the reconstruction.[33]

The selected acceptance criterion should match the stage being evaluated. Detection and escape depend on early changing conditions, while structural resistance often depends on sustained developed exposure. A result that passes at one time should not be presented as passing every relevant stage without showing those comparisons.[30][5]

Stage analysis is most useful when it preserves these distinct questions and the transitions between them. It allows early, developed and residual hazards to be assessed with appropriate data, instead of treating one design point as the entire fire. That makes the sequence and its limitations clear to the reader.[1][2]

See also

Notes

  1. ↑ 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 Van Rossum (2027), §4.11.
  2. ↑ 2.0 2.1 2.2 2.3 2.4 2.5 Australian Building Codes Board et al. (2005), pp. 1.2-29–1.2-30.
  3. ↑ 3.0 3.1 3.2 3.3 3.4 3.5 3.6 3.7 Australian Building Codes Board et al. (2005), p. 2.4-11.
  4. ↑ 4.0 4.1 Australian Building Codes Board et al. (2005), p. 2.5-2.
  5. ↑ 5.0 5.1 5.2 U.S. Nuclear Regulatory Commission (2005), p. F-45.
  6. ↑ 6.0 6.1 Kanury (2002), p. 2-230.
  7. ↑ Australian Building Codes Board et al. (2005), p. 2.4-4.
  8. ↑ Ohlemiller (2002), p. 2-200.
  9. ↑ 9.0 9.1 9.2 Ohlemiller (2002), p. 2-208.
  10. ↑ 10.0 10.1 Quintiere (2002), p. 2-246.
  11. ↑ 11.0 11.1 Van Rossum (2027), §4.3.
  12. ↑ 12.0 12.1 12.2 12.3 Van Rossum (2027), §4.8.
  13. ↑ 13.0 13.1 Van Rossum (2027), §4.5.
  14. ↑ 14.0 14.1 14.2 14.3 Babrauskas (2002), p. 3-1.
  15. ↑ Van Rossum (2027), §6.7.
  16. ↑ 16.0 16.1 Quintiere (2002), p. 3-162.
  17. ↑ 17.0 17.1 17.2 17.3 17.4 Cadorin and Franssen (2003), p. 412.
  18. ↑ Van Rossum (2027), §4.7.
  19. ↑ 19.0 19.1 19.2 19.3 Drysdale (2002), p. 1-95.
  20. ↑ 20.0 20.1 20.2 20.3 Cadorin and Franssen (2003), pp. 400–401.
  21. ↑ Van Rossum (2027), §9.1.
  22. ↑ 22.0 22.1 22.2 22.3 22.4 Australian Building Codes Board et al. (2005), p. 2.6-12.
  23. ↑ Grant (2002), p. 4-152.
  24. ↑ 24.0 24.1 24.2 Ohlemiller (2002), p. 2-206.
  25. ↑ 25.0 25.1 Gray (2002), p. 2-212.
  26. ↑ Kanury (2002), p. 2-229.
  27. ↑ 27.0 27.1 U.S. Nuclear Regulatory Commission (2005), p. F-36.
  28. ↑ Gray (2002), p. 2-213.
  29. ↑ 29.0 29.1 Babrauskas (2002), p. 3-3.
  30. ↑ 30.0 30.1 30.2 30.3 Australian Building Codes Board et al. (2005), p. 1.6-3.
  31. ↑ Australian Building Codes Board et al. (2005), p. 1.6-7.
  32. ↑ Rockett and Milke (2002), p. 1-29.
  33. ↑ Van Rossum (2027), §1.7.

References

  • Australian Building Codes Board, National Research Council of Canada, International Code Council, & Department of Building and Housing New Zealand. (2005). International fire engineering guidelines (2005 ed.). Australian Building Codes Board.
  • Babrauskas, V. (2002). Heat release rates. In P. J. DiNenno (Ed.), SFPE handbook of fire protection engineering (3rd ed.). National Fire Protection Association.
  • Cadorin, J.-F., & Franssen, J.-M. (2003). A tool to design steel elements submitted to compartment fires: OZone V2. Part 1: Pre- and post-flashover compartment fire model. Fire Safety Journal, 38, 395–427.
  • Drysdale, D. D. (2002). Thermochemistry. In P. J. DiNenno (Ed.), SFPE handbook of fire protection engineering (3rd ed.). National Fire Protection Association.
  • Grant, C. C. (2002). Halon design calculations. In P. J. DiNenno (Ed.), SFPE handbook of fire protection engineering (3rd ed.). National Fire Protection Association.
  • Gray, B. (2002). Spontaneous combustion and self-heating. In P. J. DiNenno (Ed.), SFPE handbook of fire protection engineering (3rd ed.). National Fire Protection Association.
  • Kanury, A. M. (2002). Flaming ignition of solid fuels. In P. J. DiNenno (Ed.), SFPE handbook of fire protection engineering (3rd ed.). National Fire Protection Association.
  • Ohlemiller, T. J. (2002). Smoldering combustion. In P. J. DiNenno (Ed.), SFPE handbook of fire protection engineering (3rd ed.). National Fire Protection Association.
  • Quintiere, J. G. (2002). Compartment fire modeling. In P. J. DiNenno (Ed.), SFPE handbook of fire protection engineering (3rd ed.). National Fire Protection Association.
  • Quintiere, J. G. (2002). Surface flame spread. In P. J. DiNenno (Ed.), SFPE handbook of fire protection engineering (3rd ed.). National Fire Protection Association.
  • Rockett, J. A., & Milke, J. A. (2002). Conduction of heat in solids. In P. J. DiNenno (Ed.), SFPE handbook of fire protection engineering (3rd ed.). National Fire Protection Association.
  • U.S. Nuclear Regulatory Commission. (2005). Fire dynamics tools: Quantitative fire hazard analysis methods for the U.S. nuclear regulatory commission fire protection inspection program (NUREG-1805). Author.
  • Van Rossum, E. (2027). Introduction to Fire Safety Engineering. Fire Safety Academy.
Retrieved from "https://fire-safety.academy/w/index.php?title=Fire_development_stages&oldid=669"