Cone calorimeter

Cone calorimeter is a laboratory instrument that measures the heat released by a small burning specimen using oxygen consumption. A cone-shaped electric heater exposes the specimen to controlled radiation, while an exhaust system collects its combustion products. The instrument can also measure ignition, mass loss and smoke production during the same experiment.

The oxygen consumption principle relates heat release to the mass of oxygen consumed during combustion. The heater provides the exposure that challenges the material; its radiation is distinct from the chemical heat released by burning. This distinction allows the apparatus to investigate how a material responds to different imposed heating conditions.

Cone results describe a particular specimen, mounting arrangement and exposure. They support comparisons between products and provide input for fire calculations, but do not directly establish the behaviour of a complete room. Sample thickness, backing, exposed edges and the surrounding fire environment can change the result. Interpretation therefore depends on the test record as well as the reported peak value.

What the instrument measures

Heat release rate describes the rate of chemical energy production by combustion. The cone calorimeter determines this quantity from exhaust flow and oxygen measurements, rather than from the temperature increase of the specimen. A load cell independently records the changing specimen mass. Together, these measurements distinguish the rate of material loss from the rate of energy release.

The heat release rate per unit area expresses measured power relative to the exposed specimen area. Reporting this quantity makes specimens of different exposed areas easier to compare. However, normalisation does not remove effects of specimen size, thickness or mounting. A small specimen can receive different flame feedback from a large burning surface, even when both results use the same units.

The ratio of heat release rate to mass loss rate gives an effective heat of combustion for the conditions measured. This quantity need not remain constant throughout a test. Babrauskas describes changing behaviour during the burning and subsequent charring of wood. Consequently, multiplying an entire mass-loss record by one assumed combustion value can conceal changes in the burning process.

Optical smoke measurement records how strongly the exhaust obscures a light beam. Gravimetric soot measurement instead collects a known fraction of the exhaust on a filter and determines its mass. These are different measurements, although both concern combustion products. Additional gas analysers can provide information about carbon monoxide, carbon dioxide or other species when suitable sampling arrangements are installed.

Heater, specimen and exhaust

The cone heater surrounds an opening through which the flame plume can rise. Its shape promotes reasonably uniform exposure across the specimen while allowing combustion products to enter the hood. The heater temperature is controlled so that changes in specimen burning do not simply become uncontrolled changes in imposed radiation. This separates the intended exposure from unwanted heating of nearby apparatus components.

The heater imposes radiation while the exhaust measurements determine chemical heat release. Based on Babrauskas (2002b), pp. 3-64–3-67.

In the apparatus described by Babrauskas, standard specimens are nominally 100 mm by 100 mm. Standard testing uses a horizontal specimen with its exposed surface facing upwards. The chapter reserves vertical orientation for special investigations, including some studies of flame regions and surface temperatures. These descriptions concern the methods discussed in the 2002 handbook, rather than an assertion about every later standard edition.

The exhaust must capture the products without creating an excessive disturbance near the specimen. Too little extraction permits products to escape the hood, undermining the gas balance used to determine heat release. Excessive extraction can alter the ignition process through airflow effects. The handbook describes a standard exhaust setting of 24 L/s for the apparatus and reports comparative ignition experiments at other settings.

An oxygen analyser measures the sampled exhaust, while flow instrumentation measures the quantity of gas passing through the duct. Because the analyser responds to oxygen partial pressure, pressure variation requires compensation. Sampling arrangements, moisture removal and instrument stability also affect the result. The apparatus therefore depends on a controlled measurement system, rather than a single oxygen reading taken above a flame.

The nominal heater setting and the heat absorbed by a burning specimen are different quantities. A heat-flux meter held cool during calibration does not reproduce the surface temperature of a hot sample. Babrauskas discusses convective losses from heated specimens despite the predominantly radiant imposed exposure. This distinction matters when test results are interpreted through a heat-transfer model rather than simply compared under the same apparatus settings.

Specimen preparation changes the result

The specimen backing determines how easily heat escapes through its rear surface. For a sufficiently thick specimen, the rear surface has little influence before ignition because appreciable heating has not reached it. Thinner specimens can respond strongly to their substrate. Babrauskas therefore discusses representative backing materials and distinguishes the finished product thickness from an arbitrary thickness chosen for convenient testing.

Exposed edges can produce burning that differs from the behaviour of a small area within a large product. Heat and combustible gases can move through these edges, while local deformation changes the exposed surface. Aluminium foil around the sides and bottom limits unwanted material movement. An edge frame can restrain the sample or reduce excessive edge burning, but also introduces possible heat losses.

Materials that swell, curl or melt present further complications. A retaining grid can prevent an expanding surface from moving into an unsuitable position, although its own presence may affect the measurement. Horizontal mounting helps retain molten material that might leave a vertical holder. The test report therefore records the frame, grid and orientation alongside the specimen identity and exposure.

These mounting details are part of the experimental boundary conditions. A result obtained with a restraining grid cannot automatically describe an installation where the material deforms freely. Similarly, an insulating backing cannot automatically represent a product attached to a heat-conducting support. The relevant comparison concerns the complete tested arrangement, including the ways it receives heat and loses material.

Heat release equations and calibration

For many common fuels, heat release is approximately proportional to oxygen consumption. The US nuclear-power-plant guidance NUREG-1805 gives 13,100 kJ per kilogram of oxygen consumed. It describes this approximation as accurate within about ±5% for most hydrocarbon fuels. The relationship below expresses that principle; the oxygen consumption rate itself requires the exhaust-flow and composition measurements.

Q̇ ≈ E_O₂ ṁ_{O₂,consumed}

Division by exposed area expresses the result as a heat release rate per unit area. The separate mass-loss measurement permits calculation of an effective combustion value, provided corresponding quantities are compared. These relationships concern chemical heat release and do not add the heater power to the measured burning rate.

q̇″ = Q̇/A; ΔH_c,eff = Q̇/ṁ_fuel

Symbol	Description	Unit
Q̇	Chemical heat release rate	kW
EO₂	Energy released per mass of oxygen consumed	kJ/kg
ṁO₂,consumed	Oxygen consumption rate	kg/s
A	Exposed specimen area	m²
q̇″	Heat release rate per unit area	kW/m²
ṁfuel	Specimen mass loss rate	kg/s
ΔHc,eff	Effective heat of combustion	kJ/kg

The heater and the heat-release measurement require separate calibrations. A heat-flux meter placed at the specimen position checks the imposed radiation. A burner supplied with a measured flow of high-purity methane checks the heat-release measurement. Babrauskas uses a methane net heat of combustion of 50 MJ/kg for this calibration procedure.

For illustration, a selected methane flow of 0.10 g/s corresponds to 5.0 kW when multiplied by that combustion value. This calculated input is a calibration illustration, not a reported specimen test result. Agreement between expected burner output and measured heat release checks the calorimetry system. It does not independently establish that the specimen receives the intended radiant exposure, which requires the separate flux calibration.

Ignition measurements and a worked exposure example

The time to ignition recorded in the handbook's test procedure is the time to sustained flaming. A spark provides a pilot close to the region where combustible gases emerge. Its purpose is ignition without imposing a substantial additional local heat source on the specimen. Tests without this pilot address a different ignition condition and must identify that difference in their records.

Changing imposed radiation changes the heating history, while backing, thickness and losses influence the surface response. Thus an ignition time has meaning only with its exposure and specimen conditions. Studies addressing critical heat flux for ignition require consideration of ignition behaviour across imposed exposures. A single successful ignition experiment establishes neither an exposure-independent ignition time nor the lowest flux capable of producing sustained flaming.

The handbook reports ignition after 76 s for a horizontal, 13 mm polymethylmethacrylate specimen exposed to 35 kW/m². The exhaust setting was 24 L/s, and the table describes experimental scatter of approximately ±10%. These details identify one experimental condition. They do not define a universal ignition time for polymethylmethacrylate or for every product containing it.

Incident power and energy calculated for the nominal specimen area under the reported PMMA exposure. Based on Babrauskas (2002b), pp. 3-63 and 3-69.

Using the full nominal 100 mm square as an illustrative exposed area gives 0.010 m². Multiplying this area by the reported irradiance gives an incident power of 0.35 kW. Over the reported 76 s before ignition, the corresponding incident energy is 26.6 kJ, or approximately 27 kJ. These calculations assume constant irradiance and exposure across the full nominal area.

This incident energy is not the energy stored in the specimen at ignition. Reflection and heat losses separate incident energy from absorbed and retained energy. It is also unrelated to the subsequent total chemical energy released by combustion. The example demonstrates why radiation exposure, ignition time and heat release are separate quantities despite appearing together in one test report.

Reporting and uncertainty

A useful report preserves the time history as well as selected summary values. The handbook lists the specimen's initial and final masses, exposed area, irradiance, exhaust flow and mounting details. It also records ignition and flameout times, peak heat release, total heat release and average values over specified periods. In the procedure described there, a complete test includes three individual runs.

The meaning of a cone result depends on its specimen, exposure and reporting conditions. Based on Babrauskas (2002b), pp. 3-70–3-78.

An average heat release rate requires a defined averaging interval. Comparing averages over the entire burning periods of two specimens can give a misleading impression when those periods differ. Babrauskas discusses fixed intervals after ignition so that results describe comparable parts of fire development. A peak, a fixed-period average and the total released energy answer different questions about the specimen's behaviour.

Calibration addresses systematic measurement errors, while repeated testing reveals variation between results. Repeatability concerns results under closely repeated conditions; reproducibility includes differences between laboratories. The handbook reports interlaboratory investigations for ignition, heat release, combustion energy and smoke measurements. Their differing uncertainties show why a single precision statement cannot describe every output of the instrument.

The smoke measurement system includes a reference light signal as well as the beam passing through the exhaust. The reference helps distinguish changing laser output from attenuation caused by smoke. Neutral-density filters provide a separate optical calibration. A well-calibrated heat-release channel therefore does not establish the accuracy of the smoke channel, which relies on its own instruments and checks.

Gas measurements also depend on the sampling arrangement. Carbon monoxide and carbon dioxide analysers can share the oxygen sampling line described in the handbook. Some other gases require a separate heated sampling system and heated soot filter. The presence of an oxygen analyser alone therefore does not provide a complete description of toxic combustion products or their production rates.

The stopping point of a test also affects the interpretation of its data. The handbook describes termination criteria involving cessation of combustion, very small continuing mass loss, or a maximum test duration. These practical rules establish a consistent procedure but do not make whole-test averages comparable when burning durations differ. Stating the averaging period preserves the distinction between procedural termination and a defined performance measure.

Using results beyond the apparatus

Cone calorimetry provides controlled data within fire testing, but it does not reproduce every feature of a developing compartment fire. A room can radiate additional heat back to the fuel and increase material decomposition. Its openings can also limit oxygen supply, changing combustion and the gases leaving the room. Both effects complicate direct transfer of a small open-air test to a complete enclosure.

The standard apparatus described by Babrauskas burns specimens in room air. Controlled-atmosphere versions allow research at different oxygen concentrations, but these are distinct experimental arrangements. Their results require their own atmosphere and flow conditions to be reported. Consequently, a conventional cone result does not by itself characterise burning in an oxygen-depleted fire compartment.

Scale also changes the radiation received from the specimen's own flames. The handbook describes size effects for both horizontal and vertical burning surfaces and discusses separate intermediate-scale and room tests. Selecting data for an engineering calculation therefore includes establishing whether the tested product and exposure represent the intended application. A numerical heat-release curve remains tied to the conditions that generated it.

See also

• Heat flux
• Single Burning Item test

References

• Babrauskas, V. (2002a). Heat release rates. In P. J. DiNenno (Ed.), SFPE handbook of fire protection engineering (3rd ed., chapter 3-1, pp. 3-1–3-37). National Fire Protection Association.
• Babrauskas, V. (2002b). The cone calorimeter. In P. J. DiNenno (Ed.), SFPE handbook of fire protection engineering (3rd ed., chapter 3-3, pp. 3-63–3-81). National Fire Protection Association.
• Iqbal, N., & Salley, M. H. (2004). Fire dynamics tools (FDTs): Quantitative fire hazard analysis methods for the U.S. Nuclear Regulatory Commission fire protection inspection program (NUREG-1805). U.S. Nuclear Regulatory Commission.
• Tewarson, A. (2002). Generation of heat and chemical compounds in fires. In P. J. DiNenno (Ed.), SFPE handbook of fire protection engineering (3rd ed., Chapter 3-4). National Fire Protection Association.