Heat release rate - Revision history

Thomas: Improve Heat release rate: sourced explanations, calculations and internal links (Codex)

2026-11-02T17:57:52Z

Improve Heat release rate: sourced explanations, calculations and internal links (Codex)

@@ Line 1: / Line 1: @@
-'''Heat release rate''' is the rate at which combustion reactions produce heat. It expresses fire power, usually in kilowatts or megawatts, and describes how quickly energy is released rather than how much energy is stored in the fuel. Babrauskas identifies it as the essential quantitative characteristic of fire size.<ref name="Babrauskas-3-1">Babrauskas (2002), p. 3-1.</ref><ref name="VR-4.5">Van Rossum (2027), §4.5.</ref>
+'''Heat release rate''' ('''HRR''') is the rate at which combustion reactions produce heat. It expresses fire power, usually in kilowatts or megawatts, and describes how quickly energy is released rather than how much energy is stored in the fuel. Babrauskas identifies it as the essential quantitative characteristic of fire size.<ref name="Babrauskas-3-1">Babrauskas (2002), p. 3-1.</ref><ref name="VR-4.5">Van Rossum (2027), §4.5.</ref>
 Heat release rate changes as ignition, spread, ventilation and fuel depletion alter combustion. It influences flame height, plume flow, gas temperature and the exposure of people and structures. It therefore provides a central input to many fire-dynamics and design calculations.<ref name="Heskestad-2-1">Heskestad (2002), p. 2-1.</ref><ref name="IFEG-2.5-2">Australian Building Codes Board et al. (2005), p. 2.5-2.</ref>
@@ Line 17: / Line 17: @@
 Babrauskas illustrates this variability with Western red cedar tested at 65 kW/m². Effective heat of combustion was roughly 12 MJ/kg during the first part of the test and increased to around 30 MJ/kg during the charring period. A single constant conversion would not represent that whole test.<ref name="Babrauskas-3-1" />
 == Heat release rate per unit area ==
@@ Line 82: / Line 109: @@
 Rounding should reflect uncertain fuel and exposure inputs. Van Rossum recommends two or at most three significant figures for many fire-engineering outcomes and safe-direction rounding for time comparisons. Excess decimal places can suggest confidence that the underlying fire data do not support.<ref name="VR-1.7">Van Rossum (2027), §1.7.</ref>
-== Reporting and comparing heat release ==
+== Interpreting test curves ==
 A heat-release result should state whether it was measured or assumed. A calorimeter produces a measurement under its test conditions, while a design curve represents an analytical choice. Presenting both as an observed fire history would conceal that distinction.<ref name="Babrauskas-3-1" /><ref name="IFEG-1.2-29" />
@@ Line 102: / Line 129: @@
 A peak value can occur at a different time from maximum smoke production or maximum target exposure. Entrainment, thermal storage and changing view factors influence those later outcomes. Heat release rate should therefore be carried through the model as a time-dependent input rather than replaced prematurely by its maximum.<ref name="IFEG-2.5-2" /><ref name="Rockett-Milke-1-29">Rockett and Milke (2002), p. 1-29.</ref>
-Review should also include the measurement chain. Oxygen concentration, flow rate and instrument calibration affect oxygen-consumption results, while specimen and enclosure choices affect their meaning. A precise heat-release number is credible only with those supporting measurements and conditions. The complete chain should be reproducible by another analyst using the same reported data and definitions.<ref name="Drysdale-1-96" /><ref name="Babrauskas-3-3" />
+a traceable report includes the heat-release curve, area definition, combustion-efficiency basis, enclosure and ventilation. It also records whether values are peak, averaged or integrated. These details allow heat release rate to be reused without losing the conditions that give it meaning. This is essential for comparison.<ref name="Babrauskas-3-3" /><ref name="VR-1.7" />
 == See also ==
 * [[Effective heat of combustion]]
 == Notes ==
@@ Line 127: / Line 150: @@
 * 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). Surface flame spread. In P. J. DiNenno (Ed.), ''SFPE handbook of fire protection engineering'' (3rd ed.). National Fire Protection Association.
 * Van Rossum, E. (2027). ''Introduction to Fire Safety Engineering''. Fire Safety Academy.
 [[Category:Fire Dynamics]]

Sanne: Add diagrams fse-heat-release-rate-v1.svg, fse-heat-release-rate-v2.svg, fse-heat-release-rate-v3.svg (level-1 visual plan)

2026-11-02T07:16:15Z

Add diagrams fse-heat-release-rate-v1.svg, fse-heat-release-rate-v2.svg, fse-heat-release-rate-v3.svg (level-1 visual plan)

← Older revision		Revision as of 08:16, 2 November 2026
Line 8:		Line 8:

	Power is energy produced per unit time, with one watt equal to one joule per second. Heat release rate can be integrated over a period to estimate energy released during that period, provided the full time history is known. A peak value alone does not specify total energy.<ref name="VR-4.5" />		Power is energy produced per unit time, with one watt equal to one joule per second. Heat release rate can be integrated over a period to estimate energy released during that period, provided the full time history is known. A peak value alone does not specify total energy.<ref name="VR-4.5" />
			[[File:fse-heat-release-rate-v1.svg\|thumb\|upright=1.5\|alt=Heat-release curve with peak power and shaded energy area over time.\|Heat release rate is power; energy is its time integral. Based on Van Rossum (2027), §4.5.]]

	[[Fire load]] instead describes the chemical energy potentially available from combustible material. A high fire load can sustain a long fire but does not establish its instantaneous power. Heat release rate and fire load answer different questions and use different units.<ref name="VR-4.4">Van Rossum (2027), §4.4.</ref>		[[Fire load]] instead describes the chemical energy potentially available from combustible material. A high fire load can sustain a long fire but does not establish its instantaneous power. Heat release rate and fire load answer different questions and use different units.<ref name="VR-4.4">Van Rossum (2027), §4.4.</ref>
Line 32:		Line 33:

	A [[t-squared fire]] uses quadratic growth, in which heat release rate increases in proportion to the square of elapsed time. Common categories are labelled by the time taken to reach 1 MW: 75 seconds for ultrafast, 150 seconds for fast, 300 seconds for medium and 600 seconds for slow growth in the cited guidance.<ref name="VR-4.5" /><ref name="IFEG-2.4-11">Australian Building Codes Board et al. (2005), p. 2.4-11.</ref>		A [[t-squared fire]] uses quadratic growth, in which heat release rate increases in proportion to the square of elapsed time. Common categories are labelled by the time taken to reach 1 MW: 75 seconds for ultrafast, 150 seconds for fast, 300 seconds for medium and 600 seconds for slow growth in the cited guidance.<ref name="VR-4.5" /><ref name="IFEG-2.4-11">Australian Building Codes Board et al. (2005), p. 2.4-11.</ref>
			[[File:fse-heat-release-rate-v3.svg\|thumb\|upright=1.5\|alt=Times to 1 MW: ultrafast 75, fast 150, medium 300 and slow 600 seconds.\|Characteristic t-squared growth times to 1 MW. Based on Van Rossum (2027), §4.5.]]

	These categories are modelling choices rather than predictions of every real fire. Experiments can show an incubation or establishment phase before quadratic growth, and its duration cannot be determined reliably in the 2005 guidelines. Omitting that phase is identified there as a usually conservative design assumption.<ref name="IFEG-2.4-11" />		These categories are modelling choices rather than predictions of every real fire. Experiments can show an incubation or establishment phase before quadratic growth, and its duration cannot be determined reliably in the 2005 guidelines. Omitting that phase is identified there as a usually conservative design assumption.<ref name="IFEG-2.4-11" />
Line 44:		Line 46:

	Drysdale gives a conversion factor of 13.1 kJ per gram of oxygen consumed for the method. This is different from heat of combustion per gram of fuel. The distinction allows diverse fuels to be compared through their oxygen use rather than their individual mass loss alone.<ref name="Drysdale-1-93">Drysdale (2002), p. 1-93.</ref><ref name="Drysdale-1-96" />		Drysdale gives a conversion factor of 13.1 kJ per gram of oxygen consumed for the method. This is different from heat of combustion per gram of fuel. The distinction allows diverse fuels to be compared through their oxygen use rather than their individual mass loss alone.<ref name="Drysdale-1-93">Drysdale (2002), p. 1-93.</ref><ref name="Drysdale-1-96" />
			[[File:fse-heat-release-rate-v2.svg\|thumb\|upright=1.5\|alt=Fire exhaust flowing to oxygen measurement and heat-release calculation with CO correction.\|Oxygen-consumption calorimetry estimates heat release from oxygen use. Based on Drysdale (2002), pp. 1-95–1-96.]]

	Open-burning calorimeters measure items without a surrounding test room. Room-fire tests instead include hot-layer and surface feedback and possible ventilation limitation. At low heat release rates these can agree, but developed room fires can differ substantially from open burning.<ref name="Babrauskas-3-1" />		Open-burning calorimeters measure items without a surrounding test room. Room-fire tests instead include hot-layer and surface feedback and possible ventilation limitation. At low heat release rates these can agree, but developed room fires can differ substantially from open burning.<ref name="Babrauskas-3-1" />

Jan: New FSE wiki article: Heat release rate (written by Kimi Code CLI)

2026-11-01T23:10:48Z

New FSE wiki article: Heat release rate (written by Kimi Code CLI)

New page

'''Heat release rate''' is the rate at which combustion reactions produce heat. It expresses fire power, usually in kilowatts or megawatts, and describes how quickly energy is released rather than how much energy is stored in the fuel. Babrauskas identifies it as the essential quantitative characteristic of fire size.<ref name="Babrauskas-3-1">Babrauskas (2002), p. 3-1.</ref><ref name="VR-4.5">Van Rossum (2027), §4.5.</ref>

Heat release rate changes as ignition, spread, ventilation and fuel depletion alter combustion. It influences flame height, plume flow, gas temperature and the exposure of people and structures. It therefore provides a central input to many fire-dynamics and design calculations.<ref name="Heskestad-2-1">Heskestad (2002), p. 2-1.</ref><ref name="IFEG-2.5-2">Australian Building Codes Board et al. (2005), p. 2.5-2.</ref>

The quantity is related to mass loss rate and effective heat of combustion, but these need not have a constant relationship. Material decomposition, incomplete combustion and enclosure feedback can change both the rate and the energy released per unit mass.<ref name="Babrauskas-3-1" /><ref name="Drysdale-1-95">Drysdale (2002), p. 1-95.</ref>

== Power, energy and burning rate ==

Power is energy produced per unit time, with one watt equal to one joule per second. Heat release rate can be integrated over a period to estimate energy released during that period, provided the full time history is known. A peak value alone does not specify total energy.<ref name="VR-4.5" />

[[Fire load]] instead describes the chemical energy potentially available from combustible material. A high fire load can sustain a long fire but does not establish its instantaneous power. Heat release rate and fire load answer different questions and use different units.<ref name="VR-4.4">Van Rossum (2027), §4.4.</ref>

[[Burning rate]] is sometimes used informally for heat release rate, but Babrauskas recommends using that term nonquantitatively. Mass loss rate is measured in kilograms per second, while heat release rate is measured in kilowatts. Keeping the terms separate avoids confusing fuel consumption with energy production.<ref name="Babrauskas-3-1" />

[[Heat of combustion]] is the heat released by complete oxidation of a unit quantity of fuel to stable products. Effective heat of combustion accounts for actual burning conditions and can differ from the bomb-calorimeter value. Incomplete combustion means that not all available chemical energy is released.<ref name="Drysdale-1-92">Drysdale (2002), pp. 1-92–1-93.</ref><ref name="Drysdale-1-95" />

Babrauskas illustrates this variability with Western red cedar tested at 65 kW/m². Effective heat of combustion was roughly 12 MJ/kg during the first part of the test and increased to around 30 MJ/kg during the charring period. A single constant conversion would not represent that whole test.<ref name="Babrauskas-3-1" />

== Heat release rate per unit area ==

[[Heat release rate per unit area]] expresses power divided by burning area, commonly in kilowatts per square metre. It describes intensity over the selected area, while total heat release rate describes the whole fire. The area definition must be stated because fuel surface, flame footprint and compartment floor area are not interchangeable.<ref name="VR-4.6">Van Rossum (2027), §4.6.</ref>

A small fire can have high local intensity and low total power, while a larger fire can distribute lower intensity over a greater area. Total heat release rate increases when either area or intensity increases. This relationship helps connect object-scale measurements with design-fire size.<ref name="VR-4.6" />

Van Rossum gives an illustrative developed-building-fire value of 250 kW/m² for a non-storage building. He also gives an average living-room sofa at about 2.5 MW and a car peak of 8.5 MW. These are anchors from the source, not universal values for every sofa, car or occupancy.<ref name="VR-4.5" />

Fire area can change as flame spreads. If intensity remains broadly similar, a larger pyrolysing area can increase total power and strengthen feedback. This is one reason early object-to-object spread matters in fire development.<ref name="Quintiere-2-246">Quintiere (2002), p. 2-246.</ref>

== Growth and design fires ==

A [[Design fire]] specifies a simplified heat-release history or related fire inputs for analysis. It may include growth, developed burning and decay, or a smouldering scenario without a developed flame stage. The curve should represent a defined scenario and purpose.<ref name="IFEG-1.2-29">Australian Building Codes Board et al. (2005), pp. 1.2-29–1.2-30.</ref>

A [[t-squared fire]] uses quadratic growth, in which heat release rate increases in proportion to the square of elapsed time. Common categories are labelled by the time taken to reach 1 MW: 75 seconds for ultrafast, 150 seconds for fast, 300 seconds for medium and 600 seconds for slow growth in the cited guidance.<ref name="VR-4.5" /><ref name="IFEG-2.4-11">Australian Building Codes Board et al. (2005), p. 2.4-11.</ref>

These categories are modelling choices rather than predictions of every real fire. Experiments can show an incubation or establishment phase before quadratic growth, and its duration cannot be determined reliably in the 2005 guidelines. Omitting that phase is identified there as a usually conservative design assumption.<ref name="IFEG-2.4-11" />

Developed burning can be fuel controlled or ventilation controlled. In the first case, fuel-gas production limits power while oxygen is sufficient. In the second, restricted oxygen supply limits combustion and can leave unburnt pyrolysis gases to burn elsewhere.<ref name="VR-4.8">Van Rossum (2027), §4.8.</ref>

Decay can follow fuel depletion or intervention. A design curve may select a decay rate for simplicity, but real decline can be irregular and include smouldering or external burning. Total energy and residual hazard should be assessed over the whole relevant history.<ref name="VR-4.11">Van Rossum (2027), §4.11.</ref><ref name="Ohlemiller-2-206">Ohlemiller (2002), p. 2-206.</ref>

== Oxygen-consumption measurement ==

The [[Oxygen consumption principle]] estimates heat release rate from the oxygen consumed during combustion. Many fuels release approximately similar energy per unit mass of oxygen consumed, allowing measurement of oxygen depletion to be converted into heat release. Carbon monoxide production can require a correction for incomplete combustion.<ref name="Drysdale-1-96">Drysdale (2002), pp. 1-95–1-96.</ref>

Drysdale gives a conversion factor of 13.1 kJ per gram of oxygen consumed for the method. This is different from heat of combustion per gram of fuel. The distinction allows diverse fuels to be compared through their oxygen use rather than their individual mass loss alone.<ref name="Drysdale-1-93">Drysdale (2002), p. 1-93.</ref><ref name="Drysdale-1-96" />

Open-burning calorimeters measure items without a surrounding test room. Room-fire tests instead include hot-layer and surface feedback and possible ventilation limitation. At low heat release rates these can agree, but developed room fires can differ substantially from open burning.<ref name="Babrauskas-3-1" />

Bench-scale instruments such as the cone calorimeter expose a small specimen to controlled radiant heat. They provide material-response curves but do not automatically reproduce full object geometry or enclosure effects. Intermediate and full-scale tests help address those differences when required.<ref name="Babrauskas-3-3">Babrauskas (2002), p. 3-3.</ref>

== Enclosure effects ==

A hot upper layer and heated room surfaces can increase radiant feedback to fuel and raise heat release rate above an open-burning value. Ventilation can simultaneously restrict oxygen and move unburnt gases to other locations. Room effects therefore depend on both thermal feedback and air supply.<ref name="Babrauskas-3-1" /><ref name="Drysdale-1-95" />

Babrauskas reports room augmentation around 20% for some upholstered chairs in fires up to 1 MW, but stronger and less predictable effects for mattresses. Liquid pools require a specific pool model rather than a simple correction. These examples demonstrate that product configuration affects the result.<ref name="Babrauskas-3-2">Babrauskas (2002), p. 3-2.</ref>

Room-fire results cannot be extrapolated directly to rooms with larger ventilation. Open-burning data can be more suitable for well-ventilated enclosures, provided fuel geometry and exposure remain comparable. The source and intended application must be matched explicitly.<ref name="Babrauskas-3-2" />

Heat released inside a compartment is not necessarily all heat generated by fuel originating there. Some volatile gases can escape and burn outside, changing both the measured internal power and the external hazard. System boundaries must therefore be stated when reporting heat release rate.<ref name="Drysdale-1-95" />

== Use in fire models ==

Plume models use heat release rate, often its convective part, to estimate entrainment, temperature and mass flow. Heskestad notes that convective flux is rarely less than 60% to 70% of total heat release in the fire types discussed. The radiative remainder acts in other directions.<ref name="Heskestad-2-2">Heskestad (2002), p. 2-2.</ref><ref name="IFEG-2.5-2" />

Smoke and detection calculations use the resulting transport to estimate when combustion products reach a layer or device. Structural calculations use the fire's heat-release history to determine boundary and member exposure. These applications need consistent timing and fire-area assumptions.<ref name="IFEG-2.5-2" /><ref name="IFEG-2.6-2">Australian Building Codes Board et al. (2005), p. 2.6-2.</ref>

A design heat-release curve can come from an appropriate test, statistical evidence or a simplified growth model. Full-scale data are valuable when they describe the actual fuel source, but even then enclosure and installation differences need review. Similar object names are not sufficient evidence of identical behaviour.<ref name="Babrauskas-3-3" />

For large spaces, spatial development can be represented by dividing the area into regions with ignition time offsets. Summing these regional heat-release histories avoids assuming simultaneous combustion of the entire floor. The offsets must still be supported by spread analysis.<ref name="IFEG-2.6-12">Australian Building Codes Board et al. (2005), p. 2.6-12.</ref>

== Interpretation and uncertainty ==

Peak heat release rate is not total fire energy, average power or fire duration. A short high peak and a long lower plateau can produce very different thermal exposures. The complete curve and the assessment period must be considered.<ref name="VR-4.5" /><ref name="VR-5.1">Van Rossum (2027), §5.1.</ref>

Combustion efficiency adds further uncertainty. Heskestad reports values close to unity for some pool fires but much lower values for polystyrene and a fully involved stack of wood pallets in the cited tests. Total heat release should not automatically be equated with theoretical complete-combustion power.<ref name="Heskestad-2-2" />

Heat release rate per unit area also needs a clear area. Multiplying an intensity by a floor area when the original measurement used burning surface area creates an inconsistent result. Units alone do not catch that definitional error.<ref name="VR-4.6" />

Rounding should reflect uncertain fuel and exposure inputs. Van Rossum recommends two or at most three significant figures for many fire-engineering outcomes and safe-direction rounding for time comparisons. Excess decimal places can suggest confidence that the underlying fire data do not support.<ref name="VR-1.7">Van Rossum (2027), §1.7.</ref>

== Reporting and comparing heat release ==

A heat-release result should state whether it was measured or assumed. A calorimeter produces a measurement under its test conditions, while a design curve represents an analytical choice. Presenting both as an observed fire history would conceal that distinction.<ref name="Babrauskas-3-1" /><ref name="IFEG-1.2-29" />

The system boundary should also be explicit. Results can describe one item, all combustion inside a room, or all reactions caused by fuel originating there. External flaming means these totals need not be identical. A boundary description prevents internal and total fire power from being mixed.<ref name="Drysdale-1-95" />

Time averaging requires similar care. An instantaneous peak, an average over one test interval and an average over the entire fire answer different questions. Comparing them without a common interval can create a false ranking between tests or design scenarios.<ref name="VR-4.5" /><ref name="Babrauskas-3-3" />

For per-unit-area results, the reference area belongs beside every value. A table should not switch between fuel surface and floor area without renaming the quantity. This preserves the distinction between local intensity and compartment-wide energy production.<ref name="VR-4.6" />

Complete-oxidation calculations provide a useful upper reference but not an automatic measurement. Combustion efficiency and effective heat of combustion should be identified if mass loss is converted into power. Where these vary, the conversion needs a time-dependent treatment or direct heat-release measurement.<ref name="Babrauskas-3-1" /><ref name="Heskestad-2-2" />

Convective and radiative portions should likewise be identified when they are used differently. Plume calculations commonly use the convective portion, while target radiation requires the radiative portion. Applying total heat release to both without a stated split can double-count or misassign the energy.<ref name="Heskestad-2-2" /><ref name="VR-9.3">Van Rossum (2027), §9.3.</ref>

The frequently used 70% convective share is a modelling approximation rather than a universal property. Heskestad gives a broader 60% to 70% range for several fire types, and Van Rossum cautions against using the simple split in post-flashover conditions. The selected fraction should therefore be justified for the fire stage.<ref name="Heskestad-2-2" /><ref name="VR-9.3" />

Integrating a heat-release curve gives energy released over its stated interval. This can be compared with fire load to check whether the scenario consumes more or less energy than the inventory contains. Such a check can expose an inconsistent duration or an unreasonably high assumed power.<ref name="VR-4.4" /><ref name="VR-4.5" />

A peak value can occur at a different time from maximum smoke production or maximum target exposure. Entrainment, thermal storage and changing view factors influence those later outcomes. Heat release rate should therefore be carried through the model as a time-dependent input rather than replaced prematurely by its maximum.<ref name="IFEG-2.5-2" /><ref name="Rockett-Milke-1-29">Rockett and Milke (2002), p. 1-29.</ref>

Review should also include the measurement chain. Oxygen concentration, flow rate and instrument calibration affect oxygen-consumption results, while specimen and enclosure choices affect their meaning. A precise heat-release number is credible only with those supporting measurements and conditions. The complete chain should be reproducible by another analyst using the same reported data and definitions.<ref name="Drysdale-1-96" /><ref name="Babrauskas-3-3" />

Test selection should follow the question. Bench-scale data can compare material responses, open full-scale tests can describe objects without room feedback, and room tests can include enclosure effects. Choosing one for convenience alone can leave the most important feedback unmeasured.<ref name="Babrauskas-3-3" />

Finally, a traceable report includes the heat-release curve, area definition, combustion-efficiency basis, enclosure and ventilation. It also records whether values are peak, averaged or integrated. These details allow heat release rate to be reused without losing the conditions that give it meaning. This is essential for comparison.<ref name="Babrauskas-3-3" /><ref name="VR-1.7" />

== See also ==

* [[Effective heat of combustion]]
* [[Flashover criteria]]
* [[Fuel-controlled fire]]
* [[Radiative fraction]]

== Notes ==

<references />

== 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.
* Drysdale, D. D. (2002). Thermochemistry. In P. J. DiNenno (Ed.), ''SFPE handbook of fire protection engineering'' (3rd ed.). National Fire Protection Association.
* Heskestad, G. (2002). Fire plumes, flame height, and air entrainment. 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). Surface flame spread. In P. J. DiNenno (Ed.), ''SFPE handbook of fire protection engineering'' (3rd ed.). National Fire Protection Association.
* Van Rossum, E. (2027). ''Introduction to Fire Safety Engineering''. Fire Safety Academy.

[[Category:Fire Dynamics]]