Skip to content
From Fire Safety Academy Wiki
Peter (talk | contribs)
Add SVG figures fse-video-smoke-detection-v1.svg, fse-video-smoke-detection-v2.svg, fse-video-smoke-detection-v3.svg (list-4 visual plan)
Peter (talk | contribs)
Wiki review: corrected 9 errors: citation (6), markup (3) (Claude Opus 5.5)
 
(One intermediate revision by one other user not shown)
Line 1: Line 1:
'''Video smoke detection''' identifies smoke by analysing changes in camera images. The Fire Protection Handbook describes software that selects changing image regions and evaluates characteristics associated with smoke. Unlike a point [[Smoke detector|smoke detector]], this method does not require the smoke to enter a local sensing chamber. Its coverage instead depends on the camera view and the analysis applied to that view.<ref name="F-02ae6bb73d">Cote et al. (2008), chapter 14.2, pp. 14-19, 14-22–14-23.</ref>
'''Video smoke detection''' identifies smoke by analysing changes in camera images. The Fire Protection Handbook describes software that selects changing image regions and evaluates characteristics associated with smoke. Unlike a point [[Smoke detector|smoke detector]], this method does not require the smoke to enter a local sensing chamber. Its coverage instead depends on the camera view and the analysis applied to that view.<ref name="F-02ae6bb73d">Cote et al. (2008), chapter 14.2, pp. 14-19, 14-22–14-23.</ref>
[[File:fse-video-smoke-detection-v1.svg|thumb|upright=1.5|alt=Show the dependence on the monitored scene.|Show the dependence on the monitored scene. Based on Cote et al. (2008), chapter 14.2, pp. 14-19, 14-22–14-23; Cote et al. (2008), chapter 14.2, pp. 14-22–14-23, 14-26; Schifiliti et al. (2002), pp. 4-2–4-3.]]
[[File:fse-video-smoke-detection-v1.svg|thumb|upright=1.5|alt=A camera views a scene with rising smoke and a masked region; the frames pass to scene analysis and then to an alarm decision|Video smoke detection depends on the monitored scene: camera frames, with masked regions excluded, are analysed before an alarm decision is made.]]


The technique differs from a [[Flame detector|flame detector]] analysing radiant signatures and from [[Aspirating smoke detection|aspirating smoke detection]] transporting air through pipework. It can be relevant to large spaces where smoke transport delays point detection. [[Atrium smoke control]] and [[Tunnel fire|tunnel fire]] protection nevertheless require application-specific assessment rather than an assumption that an available camera provides suitable coverage.<ref name="F-f7251c191b">Cote et al. (2008), chapter 14.2, pp. 14-22–14-23, 14-26.</ref><ref name="S-4be2477a1d">Schifiliti et al. (2002), pp. 4-2–4-3.</ref>
The technique differs from a [[Flame detector|flame detector]] analysing radiant signatures and from [[Aspirating smoke detection|aspirating smoke detection]] transporting air through pipework. It can be relevant to large spaces where smoke transport delays point detection. [[Atrium smoke control]] and [[Tunnel fire|tunnel fire]] protection nevertheless require application-specific assessment rather than an assumption that an available camera provides suitable coverage.<ref name="F-f7251c191b">Cote et al. (2008), chapter 14.2, pp. 14-22–14-23, 14-26.</ref><ref name="S-4be2477a1d">Schifiliti et al. (2002), pp. 4-2–4-3.</ref>
Line 11: Line 11:
The monitored view can include masks excluding regions associated with expected nuisance sources. Such masking changes the area actually assessed by the software. The useful coverage is therefore the combination of camera position, visible scene and configured analysis. Obstructions or changes to the scene can alter that coverage without producing an obvious camera failure.<ref name="F-b52f18b04d" />
The monitored view can include masks excluding regions associated with expected nuisance sources. Such masking changes the area actually assessed by the software. The useful coverage is therefore the combination of camera position, visible scene and configured analysis. Obstructions or changes to the scene can alter that coverage without producing an obvious camera failure.<ref name="F-b52f18b04d" />


The image can also help an operator interpret an event. The handbook identifies this as an additional source of information for emergency management. However, the presence of a view and the automatic decision to raise an alarm are separate functions. Their contribution depends on the intended response and the people receiving the information.<ref name="F-5ea64bae68">Cote et al. (2008), chapter 14.2, pp. 14-23.</ref>
The image can also help an operator interpret an event. The handbook identifies this as an additional source of information for emergency management. However, the presence of a view and the automatic decision to raise an alarm are separate functions. Their contribution depends on the intended response and the people receiving the information.<ref name="F-5ea64bae68">Cote et al. (2008), chapter 14.2, p. 14-23.</ref>


Large open facilities are among the applications discussed because smoke can be visible before it reaches a distant point detector. That advantage remains conditional on suitable views and a detectable image signature. The engineering approach still compares plausible fires, nuisance conditions and response objectives. A camera field of view does not provide a universal measure of detection time.<ref name="F-b52f18b04d" /><ref name="S-4be2477a1d" />
Large open facilities are among the applications discussed because smoke can be visible before it reaches a distant point detector. That advantage remains conditional on suitable views and a detectable image signature. The engineering approach still compares plausible fires, nuisance conditions and response objectives. A camera field of view does not provide a universal measure of detection time.<ref name="F-b52f18b04d" /><ref name="S-4be2477a1d" />


== Fire signatures and detector selection ==
== Fire signatures and detector selection ==
A fire signature is a measurable effect associated with combustion. Heat, suspended smoke particles and radiant energy provide different signatures, and detector technologies respond to different parts of that information. A detector can therefore be highly sensitive to one type of fire and comparatively slow for another. Selection begins with the fires and environmental conditions expected in the protected space.<ref name="S-fd76472e02">Schifiliti et al. (2002), pp. 4-3.</ref><ref name="F-e5e5ba2507">Cote et al. (2008), chapter 14.2, pp. 14-15.</ref>
A fire signature is a measurable effect associated with combustion. Heat, suspended smoke particles and radiant energy provide different signatures, and detector technologies respond to different parts of that information. A detector can therefore be highly sensitive to one type of fire and comparatively slow for another. Selection begins with the fires and environmental conditions expected in the protected space.<ref name="S-fd76472e02">Schifiliti et al. (2002), p. 4-3.</ref><ref name="F-e5e5ba2507">Cote et al. (2008), chapter 14.2, p. 14-15.</ref>
[[File:fse-video-smoke-detection-v2.svg|thumb|upright=1.5|alt=Explain the mechanism supporting video smoke detection.|Explain the mechanism supporting video smoke detection. Based on Schifiliti et al. (2002), pp. 4-3; Cote et al. (2008), chapter 14.2, pp. 14-15; Cote et al. (2008), chapter 14.2, pp. 14-15–14-19.]]
[[File:fse-video-smoke-detection-v2.svg|thumb|upright=1.5|alt=Heat, smoke particles and radiant energy as fire signatures feeding sensor selection, with environmental interference also influencing the choice|Selecting a sensing principle depends on the expected fire signatures and on environmental interference; no detector ranking applies independently of the fire and environment.]]


Flaming and smouldering fires illustrate this distinction. A rapidly flaming fire can produce substantial heat and a strong radiation signal. A smouldering fire can produce smoke without a visible flame and with relatively little heat. A detection strategy based only on heat or visible flame may consequently respond differently from one based on smoke. The relevant comparison is response to the selected scenario, rather than an abstract ranking of all detectors.<ref name="F-b20170c94c">Cote et al. (2008), chapter 14.2, pp. 14-15–14-19.</ref>
Flaming and smouldering fires illustrate this distinction. A rapidly flaming fire can produce substantial heat and a strong radiation signal. A smouldering fire can produce smoke without a visible flame and with relatively little heat. A detection strategy based only on heat or visible flame may consequently respond differently from one based on smoke. The relevant comparison is response to the selected scenario, rather than an abstract ranking of all detectors.<ref name="F-b20170c94c">Cote et al. (2008), chapter 14.2, pp. 14-15–14-19.</ref>
Line 41: Line 41:
Different detection methods impose different transport conditions. An aspirating system draws sampled air through pipework, introducing transport time and possible dilution between sampling ports. A projected beam measures smoke along an optical path rather than at a single point. Video detection analyses an image and therefore does not require smoke to enter a local sensing chamber. Each arrangement introduces its own limitations concerning coverage, obstruction and interpretation.<ref name="F-724337824e" />
Different detection methods impose different transport conditions. An aspirating system draws sampled air through pipework, introducing transport time and possible dilution between sampling ports. A projected beam measures smoke along an optical path rather than at a single point. Video detection analyses an image and therefore does not require smoke to enter a local sensing chamber. Each arrangement introduces its own limitations concerning coverage, obstruction and interpretation.<ref name="F-724337824e" />


Duct detection has a different purpose from open-area detection. Smoke in a duct may be diluted by air returning from other rooms, and it may not reach the detector when the fan is stopped. The handbook consequently distinguishes control of smoke circulation from protection of an occupied space itself. Coverage is assessed against the function being claimed, including the relevant operating and non-operating states of the ventilation system.<ref name="F-97c29afec9">Cote et al. (2008), chapter 14.2, pp. 14-27.</ref>
Duct detection has a different purpose from open-area detection. Smoke in a duct may be diluted by air returning from other rooms, and it may not reach the detector when the fan is stopped. The handbook consequently distinguishes control of smoke circulation from protection of an occupied space itself. Coverage is assessed against the function being claimed, including the relevant operating and non-operating states of the ventilation system.<ref name="F-97c29afec9">Cote et al. (2008), chapter 14.2, p. 14-27.</ref>


== Heat response and the meaning of spacing ==
== Heat response and the meaning of spacing ==
Line 50: Line 50:
The handbook distinguishes the test basis of heat-detector spacing from the conventional nominal spacing used for smoke detectors. Listed heat-detector spacing relates to a defined fire-test arrangement and response comparison. It does not mean that a fire of any size will be detected within a fixed time anywhere in the resulting area. Ceiling height, ambient conditions, fire growth and the detector's thermal response still influence the result.<ref name="S-21ffe36703" /><ref name="F-97c29afec9" />
The handbook distinguishes the test basis of heat-detector spacing from the conventional nominal spacing used for smoke detectors. Listed heat-detector spacing relates to a defined fire-test arrangement and response comparison. It does not mean that a fire of any size will be detected within a fixed time anywhere in the resulting area. Ceiling height, ambient conditions, fire growth and the detector's thermal response still influence the result.<ref name="S-21ffe36703" /><ref name="F-97c29afec9" />


Performance assessment therefore combines the fire environment with the sensing element's response. A model can estimate gas temperature and velocity near a detector and then estimate heating of the element. The assumptions must match the type of fire and enclosure being represented. The handbook cautions that a plume-based approach developed for flaming fires cannot simply be transferred to low-energy smouldering scenarios.<ref name="F-97c29afec9" /><ref name="S-6654fefff7">Schifiliti et al. (2002), pp. 4-4.</ref>
Performance assessment therefore combines the fire environment with the sensing element's response. A model can estimate gas temperature and velocity near a detector and then estimate heating of the element. The assumptions must match the type of fire and enclosure being represented. The handbook cautions that a plume-based approach developed for flaming fires cannot simply be transferred to low-energy smouldering scenarios.<ref name="F-97c29afec9" /><ref name="S-6654fefff7">Schifiliti et al. (2002), p. 4-4.</ref>


== Protection objectives and response time ==
== Protection objectives and response time ==
Line 63: Line 63:
== Inspection, testing and continuing performance ==
== Inspection, testing and continuing performance ==
Inspection, testing and maintenance answer different questions about the installation. Inspection can reveal damage, obstruction, unsuitable surroundings or changes in the protected space. Functional testing demonstrates an intended response under a stated test condition. Maintenance addresses defects and deterioration so that the installation can continue to perform its specified functions. The handbook treats these activities as part of the original protection design.<ref name="F-c183541efa">Cote et al. (2008), chapter 14.5, pp. 14-53–14-54, 14-61–14-62.</ref>
Inspection, testing and maintenance answer different questions about the installation. Inspection can reveal damage, obstruction, unsuitable surroundings or changes in the protected space. Functional testing demonstrates an intended response under a stated test condition. Maintenance addresses defects and deterioration so that the installation can continue to perform its specified functions. The handbook treats these activities as part of the original protection design.<ref name="F-c183541efa">Cote et al. (2008), chapter 14.5, pp. 14-53–14-54, 14-61–14-62.</ref>
[[File:fse-video-smoke-detection-v3.svg|thumb|upright=1.5|alt=Distinguish complementary evidence needed to assess video smoke detection.|Distinguish complementary evidence needed to assess video smoke detection. Based on Cote et al. (2008), chapter 14.5, pp. 14-53–14-54, 14-61–14-62; Cote et al. (2008), chapter 14.5, pp. 14-61–14-62; Cote et al. (2008), chapter 14.2, pp. 14-28.]]
[[File:fse-video-smoke-detection-v3.svg|thumb|upright=1.5|alt=Physical inspection, functional test, integrated response and recorded conditions (camera view, scene and masking), each with what it establishes and does not prove, combined into an assessment within recorded conditions that is reviewed after change|Complementary evidence for assessing video smoke detection, and what each kind of evidence does not prove.]]


Acceptance testing checks the installed arrangement against its intended sequence of operation. Changes to hardware, wiring or software can require reacceptance testing because they can affect previously demonstrated functions. The scope of the assessment depends on the change and its consequences. Records make the original arrangement, subsequent changes and observed performance available for later comparison.<ref name="F-d70f9bf924">Cote et al. (2008), chapter 14.5, pp. 14-61–14-62.</ref>
Acceptance testing checks the installed arrangement against its intended sequence of operation. Changes to hardware, wiring or software can require reacceptance testing because they can affect previously demonstrated functions. The scope of the assessment depends on the change and its consequences. Records make the original arrangement, subsequent changes and observed performance available for later comparison.<ref name="F-d70f9bf924">Cote et al. (2008), chapter 14.5, pp. 14-61–14-62.</ref>


Electrical supervision cannot identify every physical impairment. A communicating smoke detector may have an obstructed smoke-entry path, and an interface may remain electrically normal while its mechanical connection is ineffective. The handbook specifically identifies painting, covering and changed airflows as reasons for examining detectors physically. Test planning therefore includes the real stimulus and the downstream response wherever those are necessary to demonstrate the required function.<ref name="F-527d72e812">Cote et al. (2008), chapter 14.2, pp. 14-28.</ref><ref name="F-d70f9bf924" />
Electrical supervision cannot identify every physical impairment. A communicating smoke detector may have an obstructed smoke-entry path, and an interface may remain electrically normal while its mechanical connection is ineffective. The handbook specifically identifies painting, covering and changed airflows as reasons for examining detectors physically. Test planning therefore includes the real stimulus and the downstream response wherever those are necessary to demonstrate the required function.<ref name="F-527d72e812">Cote et al. (2008), chapter 14.2, p. 14-28.</ref><ref name="F-d70f9bf924" />


The Australian part of the International Fire Engineering Guidelines, published in 2005, connects commissioning with management and maintenance throughout building use. It also identifies alterations as a possible challenge to the original engineering assumptions. A changed layout or operating procedure can invalidate an earlier assessment without causing an electrical fault. Continuing performance consequently involves reviewing both equipment condition and the circumstances in which the equipment is expected to work.<ref name="I-abe1241982">National Research Council of Canada et al. (2005), §0.3.2.4–0.3.2.8; pp. 0.3-6–0.3-7.</ref>
The Australian part of the International Fire Engineering Guidelines, published in 2005, connects commissioning with management and maintenance throughout building use. It also identifies alterations as a possible challenge to the original engineering assumptions. A changed layout or operating procedure can invalidate an earlier assessment without causing an electrical fault. Continuing performance consequently involves reviewing both equipment condition and the circumstances in which the equipment is expected to work.<ref name="I-abe1241982">National Research Council of Canada et al. (2005), §0.3.2.4–0.3.2.8; pp. 0.3-6–0.3-7.</ref>

Latest revision as of 00:34, 16 November 2026

Video smoke detection identifies smoke by analysing changes in camera images. The Fire Protection Handbook describes software that selects changing image regions and evaluates characteristics associated with smoke. Unlike a point smoke detector, this method does not require the smoke to enter a local sensing chamber. Its coverage instead depends on the camera view and the analysis applied to that view.[1]

A camera views a scene with rising smoke and a masked region; the frames pass to scene analysis and then to an alarm decision
Video smoke detection depends on the monitored scene: camera frames, with masked regions excluded, are analysed before an alarm decision is made.

The technique differs from a flame detector analysing radiant signatures and from aspirating smoke detection transporting air through pipework. It can be relevant to large spaces where smoke transport delays point detection. Atrium smoke control and tunnel fire protection nevertheless require application-specific assessment rather than an assumption that an available camera provides suitable coverage.[2][3]

A photoelectric smoke detector measures particle effects within a defined optical arrangement. Video detection interprets a scene, bringing different questions about illumination, obstruction and unwanted image changes.[4]

Sensing principle and application

The handbook's visible-light video example requires an illuminated space. Images are digitised and areas showing changes receive further processing. The software attempts to distinguish smoke characteristics from other changes in the scene. An ordinary surveillance image therefore does not itself constitute a validated smoke-detection function.[5]

The monitored view can include masks excluding regions associated with expected nuisance sources. Such masking changes the area actually assessed by the software. The useful coverage is therefore the combination of camera position, visible scene and configured analysis. Obstructions or changes to the scene can alter that coverage without producing an obvious camera failure.[5]

The image can also help an operator interpret an event. The handbook identifies this as an additional source of information for emergency management. However, the presence of a view and the automatic decision to raise an alarm are separate functions. Their contribution depends on the intended response and the people receiving the information.[6]

Large open facilities are among the applications discussed because smoke can be visible before it reaches a distant point detector. That advantage remains conditional on suitable views and a detectable image signature. The engineering approach still compares plausible fires, nuisance conditions and response objectives. A camera field of view does not provide a universal measure of detection time.[5][3]

Fire signatures and detector selection

A fire signature is a measurable effect associated with combustion. Heat, suspended smoke particles and radiant energy provide different signatures, and detector technologies respond to different parts of that information. A detector can therefore be highly sensitive to one type of fire and comparatively slow for another. Selection begins with the fires and environmental conditions expected in the protected space.[7][8]

Heat, smoke particles and radiant energy as fire signatures feeding sensor selection, with environmental interference also influencing the choice
Selecting a sensing principle depends on the expected fire signatures and on environmental interference; no detector ranking applies independently of the fire and environment.

Flaming and smouldering fires illustrate this distinction. A rapidly flaming fire can produce substantial heat and a strong radiation signal. A smouldering fire can produce smoke without a visible flame and with relatively little heat. A detection strategy based only on heat or visible flame may consequently respond differently from one based on smoke. The relevant comparison is response to the selected scenario, rather than an abstract ranking of all detectors.[9]

Non-fire processes can resemble the signature that a detector is intended to recognise. Airborne dust, process aerosols and normal temperature changes can therefore influence selection and location. Increasing sensitivity can shorten response to a fire while increasing response to unwanted stimuli. The handbook describes selection as a balance between useful sensitivity and discrimination under the expected environmental conditions.[10]

A combined detector does not necessarily treat every sensor in the same way. Some combinations contain independently operating sensing elements. Multicriteria equipment instead combines information through a processing algorithm before deciding whether to signal. The resulting response belongs to that particular sensing and processing arrangement. The presence of several sensors does not establish a universal response time or remove the need to examine nuisance conditions.[6]

Smoke sensing and signal interpretation

Ionisation and photoelectric detectors measure different effects of suspended particles. An ionisation chamber maintains a small electrical current through ionised air. Smoke particles reduce the mobility of charged species and change the chamber's conductivity. A photoelectric detector instead measures the effect of particles on light. Its optical arrangement can respond to scattered light or to attenuation along a beam.[11]

In a light-scattering detector, the source and receiver are arranged so that the receiver normally sees little direct source light. Particles entering the sensing region scatter some light towards the receiver. In an obscuration detector, smoke reduces the amount of light reaching the receiver along an established path. These are distinct optical arrangements, even though both are described as photoelectric detection.[12]

The handbook associates ionisation response with particle number and photoelectric response with optical properties. It reports different responses to flaming and smouldering test fires, while also explaining that distance and smoke development influence the comparison. Particle properties and transport therefore matter alongside nominal detector sensitivity. A claim about one tested fire cannot establish which technology will respond first in every building scenario.[13]

Analogue reporting and addressability add different capabilities. Analogue information describes a measured condition beyond a simple normal-or-alarm state. Addressability identifies the individual reporting device. Processing can use this information for sensitivity management, diagnostics and locating an event. Compensation for gradual contamination nevertheless cannot demonstrate that smoke reaches the sensing chamber. Physical inspection and functional testing remain separate parts of performance assessment.[14]

Transport, location and coverage

A detector responds to conditions at its sensing location, which may differ from conditions beside the fire. A plume carries hot combustion products upwards and entrains surrounding air. The resulting mixture becomes cooler and more dilute as it travels. Beneath a ceiling, the flow can then spread horizontally before reaching a detector. Ceiling height and horizontal separation therefore influence the available signal.[15][16]

Beams, partitions and other obstructions can redirect or restrict that movement. Building ventilation can also transport smoke or bathe a detector in relatively clean supply air. For low-energy fires, the airflow produced by building services can be especially relevant to where smoke travels. A regular detector grid is consequently a layout arrangement, not a direct measurement of response time.[16]

Different detection methods impose different transport conditions. An aspirating system draws sampled air through pipework, introducing transport time and possible dilution between sampling ports. A projected beam measures smoke along an optical path rather than at a single point. Video detection analyses an image and therefore does not require smoke to enter a local sensing chamber. Each arrangement introduces its own limitations concerning coverage, obstruction and interpretation.[4]

Duct detection has a different purpose from open-area detection. Smoke in a duct may be diluted by air returning from other rooms, and it may not reach the detector when the fan is stopped. The handbook consequently distinguishes control of smoke circulation from protection of an occupied space itself. Coverage is assessed against the function being claimed, including the relevant operating and non-operating states of the ventilation system.[17]

Heat response and the meaning of spacing

Heat detectors respond through a sensing element whose temperature changes as heat is transferred to it. A fixed-temperature device signals when that element reaches its operating condition. The surrounding gas can already be hotter because the element takes time to heat. This difference is thermal lag, and it explains why a rated temperature alone does not determine activation time.[18]

A rate-of-rise detector responds to a sufficiently rapid temperature increase. A rate-compensation detector has a different purpose: it reduces the effect of thermal lag on operation near a specified ambient temperature. The two descriptions are therefore not interchangeable. A combination detector may include both rate-of-rise and fixed-temperature functions so that different heating histories can produce operation.[19]

The handbook distinguishes the test basis of heat-detector spacing from the conventional nominal spacing used for smoke detectors. Listed heat-detector spacing relates to a defined fire-test arrangement and response comparison. It does not mean that a fire of any size will be detected within a fixed time anywhere in the resulting area. Ceiling height, ambient conditions, fire growth and the detector's thermal response still influence the result.[15][17]

Performance assessment therefore combines the fire environment with the sensing element's response. A model can estimate gas temperature and velocity near a detector and then estimate heating of the element. The assumptions must match the type of fire and enclosure being represented. The handbook cautions that a plume-based approach developed for flaming fires cannot simply be transferred to low-energy smouldering scenarios.[17][20]

Protection objectives and response time

A detection and alarm design begins with the outcome that the protection is intended to achieve. Life safety, property protection and continuity of a building's activities can require different responses to the same fire. A warning intended to support evacuation has to arrive while useful escape time remains. Protection of sensitive equipment may instead demand detection before a small fire produces unacceptable contamination. These objectives influence the devices, processing and response arrangements chosen for the installation.[3]

The response is a sequence rather than a single event. Products of combustion must reach an appropriate sensor, the system must recognise its input, and the intended action must follow. A person receiving a warning may then need time to interpret it and begin moving. For an automatic protective function, the corresponding delay includes operation of the connected equipment. Treating detector operation as the end of the process therefore omits part of the protection mechanism.[3][21]

The SFPE design approach separates delays associated with transport from delays associated with processing and subsequent response. Transport depends on the fire, distance and enclosure geometry. Processing delays depend on the selected system and its configuration. This distinction helps identify whether an improvement requires a different sensor location, different processing or a different response arrangement.[3]

A single favourable fire scenario does not establish performance for every possible fire. Different fuels, locations and growth histories can change which signature reaches the system first. Assessment therefore compares plausible scenarios and the consequences of delayed or unsuccessful operation. The resulting performance claim belongs to the complete arrangement and its stated assumptions, rather than to an isolated component.[3]

Inspection, testing and continuing performance

Inspection, testing and maintenance answer different questions about the installation. Inspection can reveal damage, obstruction, unsuitable surroundings or changes in the protected space. Functional testing demonstrates an intended response under a stated test condition. Maintenance addresses defects and deterioration so that the installation can continue to perform its specified functions. The handbook treats these activities as part of the original protection design.[22]

Physical inspection, functional test, integrated response and recorded conditions (camera view, scene and masking), each with what it establishes and does not prove, combined into an assessment within recorded conditions that is reviewed after change
Complementary evidence for assessing video smoke detection, and what each kind of evidence does not prove.

Acceptance testing checks the installed arrangement against its intended sequence of operation. Changes to hardware, wiring or software can require reacceptance testing because they can affect previously demonstrated functions. The scope of the assessment depends on the change and its consequences. Records make the original arrangement, subsequent changes and observed performance available for later comparison.[23]

Electrical supervision cannot identify every physical impairment. A communicating smoke detector may have an obstructed smoke-entry path, and an interface may remain electrically normal while its mechanical connection is ineffective. The handbook specifically identifies painting, covering and changed airflows as reasons for examining detectors physically. Test planning therefore includes the real stimulus and the downstream response wherever those are necessary to demonstrate the required function.[24][23]

The Australian part of the International Fire Engineering Guidelines, published in 2005, connects commissioning with management and maintenance throughout building use. It also identifies alterations as a possible challenge to the original engineering assumptions. A changed layout or operating procedure can invalidate an earlier assessment without causing an electrical fault. Continuing performance consequently involves reviewing both equipment condition and the circumstances in which the equipment is expected to work.[25]

Warning, recognition and movement

A warning reaches its life-safety objective through the response of the people who receive it. Hearing a sound or seeing a light is only the first part of that process. People must recognise its significance, interpret the situation and decide what action to take. Movement towards safety follows those earlier stages rather than occurring automatically at the instant of detection.[21]

Van Rossum distinguishes awareness, decision-making and the actions preceding purposeful movement. Behaviour depends on perceived danger, previous experience, training, roles and the conduct of other people. The same audible signal can therefore produce different responses in different occupancies. A calculation that assigns no time to interpretation removes a part of the evacuation process described in the source.[26]

Information content also matters. A simple alarm pattern identifies an emergency condition but carries little detail about its location or the required response. A voice message can convey additional instructions, provided that the message is understandable. The Fire Protection Handbook consequently treats audibility and intelligibility as separate performance questions. Greater sound output alone does not establish that a listener understands spoken instructions.[27]

The intended audience influences the notification arrangement. Public-mode notification addresses occupants generally, while private-mode arrangements address trained people responsible for emergency actions. The handbook uses hospitals and prisons as examples where staff may mediate the response. Such arrangements depend on the defined management and assistance strategy. They cannot be evaluated solely by checking whether a warning appliance operates electrically.[28][29]

See also

Notes

  1. ↑ Cote et al. (2008), chapter 14.2, pp. 14-19, 14-22–14-23.
  2. ↑ Cote et al. (2008), chapter 14.2, pp. 14-22–14-23, 14-26.
  3. ↑ 3.0 3.1 3.2 3.3 3.4 3.5 Schifiliti et al. (2002), pp. 4-2–4-3.
  4. ↑ 4.0 4.1 Cote et al. (2008), chapter 14.2, pp. 14-21–14-23.
  5. ↑ 5.0 5.1 5.2 Cote et al. (2008), chapter 14.2, pp. 14-22–14-23.
  6. ↑ 6.0 6.1 Cote et al. (2008), chapter 14.2, p. 14-23.
  7. ↑ Schifiliti et al. (2002), p. 4-3.
  8. ↑ Cote et al. (2008), chapter 14.2, p. 14-15.
  9. ↑ Cote et al. (2008), chapter 14.2, pp. 14-15–14-19.
  10. ↑ Cote et al. (2008), chapter 14.2, pp. 14-15, 14-26.
  11. ↑ Cote et al. (2008), chapter 14.2, pp. 14-19–14-21.
  12. ↑ Cote et al. (2008), chapter 14.2, pp. 14-21–14-22.
  13. ↑ Cote et al. (2008), chapter 14.2, pp. 14-19–14-22, 14-27.
  14. ↑ Cote et al. (2008), chapter 14.2, pp. 14-19–14-20, 14-28.
  15. ↑ 15.0 15.1 Schifiliti et al. (2002), pp. 4-3–4-4.
  16. ↑ 16.0 16.1 Cote et al. (2008), chapter 14.2, pp. 14-26–14-27.
  17. ↑ 17.0 17.1 17.2 Cote et al. (2008), chapter 14.2, p. 14-27.
  18. ↑ Cote et al. (2008), chapter 14.2, pp. 14-16–14-17.
  19. ↑ Cote et al. (2008), chapter 14.2, pp. 14-17–14-19.
  20. ↑ Schifiliti et al. (2002), p. 4-4.
  21. ↑ 21.0 21.1 Van Rossum (2027), §7.2–7.3.
  22. ↑ Cote et al. (2008), chapter 14.5, pp. 14-53–14-54, 14-61–14-62.
  23. ↑ 23.0 23.1 Cote et al. (2008), chapter 14.5, pp. 14-61–14-62.
  24. ↑ Cote et al. (2008), chapter 14.2, p. 14-28.
  25. ↑ National Research Council of Canada et al. (2005), §0.3.2.4–0.3.2.8; pp. 0.3-6–0.3-7.
  26. ↑ Van Rossum (2027), §7.3.
  27. ↑ Cote et al. (2008), chapter 14.3, pp. 14-29–14-32.
  28. ↑ Cote et al. (2008), chapter 14.3, pp. 14-29–14-30.
  29. ↑ National Research Council of Canada et al. (2005), §0.3.2.6; p. 0.3-6.

References

  • Cote, A. E., Grant, C. C., Hall, J. R., Jr., & Solomon, R. E. (Eds.). (2008). Fire Protection Handbook (20th ed.). 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 (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., chapter 4-1, pp. 4-1–4-43). National Fire Protection Association.
  • Van Rossum, E. (2027). Introduction to Fire Safety Engineering. Fire Safety Academy.