Beam smoke detector

A beam smoke detector is a smoke detector that sends a beam of light across the protected space to a receiver and gives an alarm when smoke weakens that beam. The sources also call it a projected beam smoke detector or an optical obscuration (beam) smoke detector. It works on the photoelectric light obscuration principle, but the light path spans the area instead of lying inside a small housing. Because detection takes place along the whole path, it is classed as a line detector rather than a spot detector.

Beam detectors are mostly used to protect large open areas and spaces with high ceilings. They can also be mounted below the ceiling of a tall space, so that they still see the smoke when it stops rising before it reaches the roof. According to the SFPE Handbook, the projected-beam detector is the only commercially available smoke detector that senses the attenuation of a light beam. This makes its response easier to relate to the optical density of smoke calculated by fire models than the response of ionisation or light-scattering detectors. The sensitivity is usually set as a total obscuration, typically between 20% and 70% of the beam. Engineers convert this setting into an optical density at alarm, which depends on the length of the beam.

How it works

A beam smoke detector has two main parts: an emitter that produces the light beam and a receiver that measures how much light arrives. The US fire alarm code NFPA 72, as summarised in NUREG-1805, describes the obscuration principle as a light source focused on a photosensitive sensor. When smoke particles enter the light path, part of the light is scattered and part is absorbed, so less light reaches the sensor. The detector processes this loss of signal and gives an alarm when it meets preset criteria. In the projected beam version the same principle is used, but the beam crosses the protected area.

A beam smoke detector: smoke between emitter and receiver weakens the beam; the alarm follows when the received light falls below the setting. Based on Iqbal and Salley (2004), pp. 11-3 and 11-5.

The quantity that matters is therefore the fraction of the original light that still reaches the receiver. Researchers often call the loss of light obscuration, although Schifiliti, Meacham and Custer note that attenuation is the more accurate word. The light may be absorbed, reflected or refracted by the smoke. Percent obscuration is 100 times one minus the ratio of the received light to the light received without smoke. The optical density expresses the same loss on a logarithmic scale, and per metre of path it has the unit m⁻¹.

This principle sets the beam detector apart from the two most common smoke detectors. Ionisation detectors sense smoke through a drop in the electrical conductance of ionised air in a small chamber. Light-scattering photoelectric detectors place the sensor out of the direct beam and respond to light that smoke scatters onto it. Neither type works by light attenuation, so their response cannot be related directly to measured optical density. A scattering detector may, for example, respond at 0.029 m⁻¹ to smoke from a smouldering cotton lamp wick but only at 0.15 m⁻¹ to smoke from a kerosene fire.

Spot detectors, by contrast, cover a limited area each. NUREG-1805 gives up to 84 m² per spot detector, depending on the ceiling surface and the room height. A beam detector senses smoke continuously along its whole path, which is why it is listed among the line detectors. Like other detectors, it is normally part of a wider fire alarm system that warns occupants, calls the fire brigade and can start other building systems. Some designs use coincident detection, in which two or more detectors must operate before the system accepts an alarm.

Where beam detectors are used

NUREG-1805 states that projected beam photoelectric detectors are most often applied for large open area protection and are also useful in areas with high ceilings. In such spaces the actuation time of ceiling-mounted spot smoke detectors may be unacceptably long, and a projected beam detector may then be the preferable means of detection. Full-scale tests in two US Navy aircraft hangars, 15 m and 22 m high, included projected beam smoke detectors alongside spot detectors, flame detectors and sprinklers.

The main physical reason for using beams in tall spaces is smoke stratification. Smoke rises only while the plume is warmer, and therefore lighter, than the surrounding air. When the temperature in the plume falls to that of the surrounding air, the smoke stops rising and stays at that height, whatever the height of the ceiling. This is a concern for low-energy fires and for rooms or volumes with very high ceilings. In one SFPE example, a 1 kW fire of overheated cable insulation in a 7 m high air-conditioned room lifts its smoke only to about 6 m. With detectors on the ceiling only, the design objective may then not be achieved.

Atria are the classic case. Milke reports that the floor-to-ceiling temperature difference in some atria with glazed roofs is of the order of 50 °C, because the sun heats the air under the glazing. Smoke from a growing fire then stratifies below this warm layer. For atrium smoke control, Milke therefore advises a detector other than a ceiling-mounted one, such as a projected beam detector, to start the smoke management system. The beams should be placed below the height of stratification so that they cross the rising plume. Once the smoke management system runs, the warm air layer is extracted and the smoke can reach the ceiling.

Intermediate stratification in an atrium: the smoke stops below a warm layer under the glazed roof, so a beam is placed below that height to intercept the plume. Based on Milke (2002b), pp. 4-304 to 4-305.

Large storage spaces are another application. The SFPE Handbook uses a mattress store of 50 m by 75 m with a 10 m high ceiling as a worked example for a projected beam detector. This places the method in the setting of a warehouse fire, where a large open volume dilutes the early smoke.

Sensitivity and optical density at alarm

A beam detector does not respond to a fixed smoke concentration. It responds when the total loss of light along the beam reaches its setting, and that setting can usually be adjusted. In New Zealand, C/VM2 (second edition, 2025) states that response sensitivities are typically in the range of 20% to 70% total obscuration. This corresponds to a ratio of received light to clear-air light of 0.3 to 0.8. To use such a setting in a fire model, the engineer needs the average optical density of smoke along the beam at the moment of alarm.

If the smoke is spread evenly along the beam, the optical density at alarm follows from the light ratio and the path length. A longer beam passes through more smoke, so the same obscuration setting is reached at a lower optical density per metre. The relation, as given in C/VM2, is:

$$D_{\mathrm{u}}=-\frac{1}{L}{\log }_{10}\left(\frac{I}{I_0}\right)\text{with}\frac{I}{I_0}=\frac{100-\mathrm{O}\mathrm{B}\mathrm{S}}{100}$$

Symbol	Description	Unit
$D_{\mathrm{u}}$	optical density of smoke per unit length at alarm	m−1
$L$	path length of the beam, from emitter to receiver	m
$I$	light intensity at the receiver with smoke present	any consistent unit
$I_0$	light intensity at the receiver without smoke	same unit as $I$
$\mathrm{O}\mathrm{B}\mathrm{S}$	total obscuration setting for alarm	%

The relation assumes a uniform optical density along the whole beam. Where the smoke is not uniform, for example where a narrow plume crosses the beam, the path can be divided into sections. The optical density of each section times its length is then summed over all sections. The logarithm is to base 10, in line with the SFPE definition of optical density. The International Fire Engineering Guidelines (2005) give the same principle and treat the detector as activated once the optical density in the smoke layer exceeds the value at alarm.

As a worked example, C/VM2 takes a beam with a path length of 10 m and an alarm setting of 50% total obscuration. The light ratio is then (100 − 50)/100 = 0.5. The optical density at alarm is −(1/10) × log₁₀(0.5) = 0.030 m⁻¹. For comparison, C/VM2 uses 0.097 m⁻¹ at alarm for spot smoke detectors.

The same calculation shows the effect of the setting and the beam length. For the 10 m beam, a setting of 20% gives about 0.0097 m⁻¹ and a setting of 70% gives about 0.052 m⁻¹. For a 50 m beam at 50%, the optical density at alarm falls to about 0.0060 m⁻¹, one fifth of the value for the 10 m beam. These values follow directly from the relation above, but they hold only for smoke spread evenly along the beam.

Optical density of evenly spread smoke at alarm against the obscuration setting, for three beam lengths. Based on Ministry of Business, Innovation and Employment (2025b), p. 28.

The SFPE Handbook example of the mattress store works in the other direction. The design goal is to detect a flaming fire before about 350 g of polyurethane has burned, with the smoke assumed to be mixed evenly through the 37,500 m³ room. Using a mass optical density of 0.22 m²/g, the smoke gives about 0.002 m⁻¹. Over a 75 m beam, 70.8% of the light still arrives, which is a total obscuration of 29.2%. The detector, adjustable from 20% to 70% in steps of 10%, would therefore have to be set at about 30% or less.

Use in performance-based design

In New Zealand, C/VM2 (second edition, 2025) makes the detection time the first term of the required safe egress time, before notification, pre-travel activity and travel. The same document requires the detection time of an automatic warning system to be found by fire modelling. This can be an algorithm that includes a ceiling jet correlation, or a computational fluid dynamics (CFD) model that solves the velocity, temperature and smoke concentration directly. Detectors must be simulated with the values of the C/VM2 detector table, whatever make and model will actually be installed. For projected beam detectors, that table gives no single number: the optical density at alarm follows from the beam path length and the design obscuration setting.

Two modelling approaches are described in the SFPE Handbook. A zone model calculates one optical density for a uniform hot layer and multiplies it by the path length of the beam. This assumes that the smoke is spread evenly along the beam, which may not be valid. A field model calculates the optical density in many points or segments between emitter and receiver and sums the products of optical density and segment length. The Fire Dynamics Simulator (FDS) follows this second approach. The user sets the positions of emitter and receiver and the total obscuration for alarm, and FDS integrates the obscuration over the path from the soot concentration in each grid cell.

A CFD study of a large atrium building, reported in the SFPE Handbook, shows how a beam can start a smoke control system in practice. The building was 150 m long, 62 m wide and 52 m high, with a central atrium 40 m high. Extract fans with a capacity of 165 m³/s were to be started by a beam detector at first-floor level. The model predicted that this beam would detect a fire on the atrium floor after 65 s, at a heat release rate of 200 kW. The doors that then opened to admit replacement air deflected the plume, and smoke spread across the atrium floor for about 8 minutes.

Limits and uncertainty

The calculation of beam response is simple, but its inputs are uncertain. Smoke measurements depend on the wavelength of the light used. Optical density data are only useful for predicting a beam detector if they were measured at the wavelength of the detector's own light source. Many detectors also use response algorithms instead of simple thresholds, and manufacturers generally do not publish them.

Mixing assumptions are a second source of error. The uniform-path method assumes that the smoke is spread evenly along the beam. In reality a rising plume may cross only a short part of a long beam. Mass optical densities from small-scale tests also introduce error when they are applied to large spaces, and the correlation breaks down for complex fires. Air movement from ventilation systems can dilute the smoke or carry it away, and there are no quantitative methods to estimate these effects.

For these reasons, the SFPE Handbook treats all smoke detector estimates as rough. Because their errors are not generally known, it advises against comparing the estimates with egress time calculations or the time to untenable conditions. It recommends using them to compare how the response of one detector changes when its spacing or location changes. C/VM2 takes a regulatory route instead: it fixes the inputs so that all designs are assessed on the same basis.

Stratification calculations have their own limits. The SFPE method for the height of smoke rise ignores heating and air-conditioning systems and warm or cold walls. A beam placed below the expected stratification height can intercept the plume, but the plume widens only gradually as it rises. Its width is estimated at 25% to 50% of the height above the top of the fuel. The placement of the beam is therefore part of the analysis, not only its sensitivity.

See also

• Aspirating smoke detection
• Detector activation time
• Heat detector
• Flame detector
• Multi-sensor detector

References

• Cox, G., & Kumar, S. (2002). Modeling enclosure fires using CFD. 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. 3-194–3-218). National Fire Protection Association; Society of Fire Protection Engineers.
• 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, Office of Nuclear Reactor Regulation.
• McGrattan, K., Hostikka, S., Floyd, J., McDermott, R., Vanella, M., Mueller, E., & Paul, C. (2026). Fire Dynamics Simulator technical reference guide: Vol. 1. Mathematical model (6th ed., NIST Special Publication 1018-1). National Institute of Standards and Technology. https://doi.org/10.6028/NIST.SP.1018
• Milke, J. A. (2002b). Smoke management in covered malls and atria. 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. 4-292–4-310). National Fire Protection Association; Society of Fire Protection Engineers.
• Ministry of Business, Innovation and Employment. (2025b). Verification Method C/VM2: Framework for fire safety design (2nd ed.).
• 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.
• 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.
• Stroup, D., Lindeman, A., McGrattan, K., Peacock, R., Overholt, K., Joglar, F., LeStrange, S., & Montanez, S. (2016). Verification and validation of selected fire models for nuclear power plant applications: Supplement 1 (NUREG-1824, Supplement 1; EPRI 3002002182). U.S. Nuclear Regulatory Commission, Office of Nuclear Regulatory Research; Electric Power Research Institute.
• Van Rossum, E. (2027). Introduction to Fire Safety Engineering. Fire Safety Academy.