Transcript
Early detection of fire phenomena characteristics
Application and experience of multi-sensor detectors AUTHOR: DIPL.-ING., DIPL.-KFM. WALDEMAR OLLIK ❏ Heat detectors monitor temperature differences and ❏ Ionisation detectors register smallest smoke particles and invisible aerosols.
F I R E P R OT E C T I O N
Modern multi-sensor detectors, also known as multi-criteria fire detectors, involve a combination of optical detectors, thermal detectors, ionisation smoke detectors or a gas sensor and are able to detect interference factors. The detectors may also contain an integrated flashing light, audible or voice alarm.
Figure 1: Fire sensitivity test of smoke detectors
The author of this article, Dipl.-Ing., Dipl.-Kfm. Waldemar Ollik, is Product Manager Fire at Novar GmbH, Neuss, Germany. Contact: waldemar.ollik @honeywell.com
1 Introduction Modern fire detection systems not only apply increasingly intelligent methods to detect fires but also imitate defined processes automatically and provide emergency services on the ground with important information. Early and reliable fire detection using appropriate detectors has top priority. The detection criteria for automatic detectors are determined by the combustion products produced by the fire.
The purpose of fire detection systems is to detect a fire in its earliest stage and activate the alarm. From its initial phase, a fire is subject to various conditions, which have different effects on its development. The interaction of fire cause, combustible material and air supply can produce a smouldering fire, an open fire or a combination of the two. The combustion process generates temperatures, fire gases and smoke. The latter can take a variety of forms, from invisible aerosols to black sooty smoke, depending on the combustible material and the progression of the fire. Fire detectors must be able to detect these different fire phenomena characteristics of the fire development quickly and reliably. A number of different detection technologies are applied. They include the following:
❏ Manual call points This article looks at point detectors and their development.
2 An overview of the types, construction and special features of multi-sensor detectors The most popular types of automatic point fire detectors can roughly be divided into two groups: smoke detectors and heat detectors. They include conventional detectors that monitor limit values, intelligent bus-powered detectors and multicriteria fire detectors with several integrated sensors and intelligent measurement value processing. Limit monitors have a preset threshold for the detection limit. When this threshold is exceeded, an alarm signal is sent to the control centre. Intelligent fire detectors usually involve processor-based measurement value processing and send the measured values or the status to the control centre via a bus protocol. The most popular smoke detectors operate on either a scattered light principle or an ionisation principle. They involve different physical processes and consequently also respond to different aerosols.
2.1 Mono-criterion detectors Detection methods of different detector types: ❏ Optical detectors detect visible smoke particles generated by the fire,
18 s+s report international 1 / 2008
❏ Point detectors ❏ Flame detectors ❏ Line type smoke detectors ❏ Line type heat detectors ❏ Aspirating smoke detection systems
2.1.1 Heat detectors Heat detectors detect the rise in temperature that is caused by a fire. In practice, heat detectors are used
in areas that cannot be monitored by smoke detectors. This type of detector can be divided into two groups that use different evaluation methods:
Measurement chamber Isolator Reference chamber
+Electrode (pin)
Figure 2: Ionisation smoke detector
Ionisation compound
Ions
Heat detectors with a fixed temperature function activate the alarm when a pre-set temperature has been exceeded. The response temperature must be significantly higher than the normal ambient temperature to prevent the detector from responding to temperature rises caused by normal heating or direct sunlight. Heat detectors with differential response also evaluate the rate of temperature rise in addition to its maximum value. 2.1.1.1 Rate-of-rise detectors: Advantages: Rate-of-rise detectors are used where operating conditions cause small or slow temperature fluctuations. They are therefore suitable for areas where smoke or similar aerosols may be present for operational reasons but which are also at risk of open and rapidly spreading fires in the event of an alarm. Disadvantages: In the event of a fire, heat detectors respond only if temperatures rise very quickly or if the static response temperature has been exceeded. Heat detectors are unable to detect glowing or smouldering fires, for example. 2.1.1.2 Maximum temperature detectors: Advantages: These detectors are used in areas where temperatures fluctuate greatly and where the exceeding of a temperature threshold is to be interpreted as an alarm. Disadvantages: If the measured temperature exceeds a specified value for a certain length of time, the alarm signal is initiated.
Aerosol particles
Ion deposits
2.1.2 Ionisation detectors The air in the sensor chamber is ionised by a mildly radioactive compound, the alpha emitter americium 241 with an activity of less than 5 k Bq. A voltage is applied to the pin electrode to produce a defined current flow. Smallest fire aerosols attach to the ions and reduce the current flow. The changed signal is evaluated for the alarm decision. Ionisation smoke detectors can identify both dark and light aerosols effectively. They are also able to detect invisible fire aerosols. Their disadvantages include the fact that the measured value is highly dependent on airflow and that the system requires a radioactive ionisation source, typically a radionuclide. 2.1.3 Optical smoke detectors Optical smoke detectors using a scattered light principle consist of a transmitter LED and a receiver photo diode. The diodes are arranged at a defined angle to each other and optically separated by a screen. In the event of a fire, part of the light beam from the transmitter LED is scattered diffusely onto the receiver when fire aerosol particles enter the detector. The scattered light causes an increase in the signal at the receiver. The incoming signal is evaluated. Optical detectors respond particularly well to visible smoke and are able to detect a fire reliably even at higher wind speeds. The detector is especially suitable for the detection of smouldering fires, light smoke, open fires such as plastic fires or smoke-emitting liquid fires.
α radiation
The disadvantage of optical detectors is their sensitivity to fog or moisture, which promote the formation of droplets. Insects entering the detector can trigger false alarms. Optical sensors can only detect visible aerosols. The advantages and disadvantages of fire detectors with a single sensor led to demands for a multi-criteria fire detector relatively early on. However, this only became feasible as a result of the rapid development of microprocessor technology in the components industry and the introduction of SMD technology in the production sector.
Abbildungen 4 (links) und 5 (unten)
2.2 Multi-sensor detectors 2.2.1 OTI detectors One of the first multi-sensor detectors was the OTI multi-sensor detector, which contained three different sensors. It was introduced at the same time as the OT multi-sensor detector at the Security Fair in 1990. The OTI multi-sensor detector consists of an optical smoke detector, a heat detector and an ionisation detector (OTI = optical, thermal and ionisation sensors). The conditions of the individual sensors are recorded and compared to each other using a complex algorithm before an alarm decision is made. This idea
Smoke particles
Figure 3: Optical smoke detector Individual display and test diode
Scattered light
Receiver opt. part
s+s report international 1 / 2008 19
Transmitter opt. part
F I R E P R OT E C T I O N
❏ Heat detectors with a fixed temperature function ❏ Heat detectors with differential response and a fixed temperature function
Figure 4: OTI multisensor detector
Optical sensor chamber Heat detector
Receiver opt. part Isolator Reference chamber
Transmitter opt. part
Ionisation chamber
was borne from the fact that the wider the range of information collected the more reliable the decision as to whether a fire has actually developed would be.
F I R E P R OT E C T I O N
2.2.2 O2T detectors Another milestone in the development of multi-sensor detectors is the O2T detector, which was launched in 2000. Like OTI detectors, O2T detectors contain three different sensors: two optical smoke detectors (O2) and a thermal sensor (T). The advantage of O2T detectors compared to conventional scattered light detectors is their two-angle technology. All popular scattered light detectors use infrared light with a wavelength range between 800 nm and 1 μm. The light is emitted by an IR transmitter diode and the scattered light is measured by a receiver that has been adjusted to this wavelength. Scattered light detectors basically differ in the construction of their external casing and the selection of the scattering angle. Most scattered light detectors operate within a Figure 5: O2T multisensor detector
Transmitter
Receiver
20 s+s report international 1 / 2008
scattering angle range between 90° and 180°. This range is referred to as forward scattering while the range between 0° and 90° is referred to as backward scattering. The closer the angle gets to 180°, the higher the wanted signal. A large scattering angle produces a higher signal yield for light smoke, however, the signal yield for dark smoke is lower by a factor of ten. Scattering angles of less than 90°, i.e. backward scattering, reduce the total signal yield. However, the difference between dark and light smoke is now considerably less than a factor of ten. Accordingly, the differences between light and dark types of smoke are getting smaller. Particles, dust or salt crystals generate a significant backscattering signal in detectors using backward scattering. This phenomenon causes problems with these detectors and may result in false alarms. Most optical smoke detectors use a scattering angle over 90°. For this reason, modern scattered light detectors are not able to detect dark aerosols, such as those genera-
Smoke particles
Transmitter
ted by liquid fires, e.g. diesel, oil, petrol and heavy or medium-weight hydrocarbon compounds, as effectively as light aerosols, such as those generated by a glowing or smouldering fire involving wood or cotton, for example. Scattered light detectors are very sensitive to light smoke, e.g. smouldering fires, to ensure that the alarm is activated in time. As a result, light aerosols, such as steam, cigarette smoke, vapours released by operating processes, dust, process-based aerosols, exhaust gases and hot fat vapours, can trigger a false alarm. Whatever the chosen scattering angle, it is impossible to distinguish between different types of smoke on the basis of a singular signal alone. However, this distinction is necessary if the false alarm rate is to be improved. Conventional methods offer no solution since the detector has to be adjusted to the aerosol that is the most difficult to detect. The O2T multi-sensor detector solves this problem by using two scattered light signals that operate at different angles. The lights are arranged so that one scattered light path is very effective at detecting light aerosols and the other is very effective at detecting dark aerosols. The different types of smoke can be distinguished by correlating the measured results of each path. The type of smoke can then be determined by a microprocessor that is integrated in the detector. The result is practically constant sensitivity in the presence of different aerosols and much lower false alarm rates since the resultant sensitivity is much lower than that of conventional scattered light detectors. The application of two-angle technology in O2T detectors makes it possible to identify certain sources of false alarms, such as steam or dust and vapours from operating processes, and clearly distinguish them from smoke. Interference factors can be further reduced by the applied algorithm. Known interference factors can even be eliminated completely by
Backward scattering
Smoke particles
Smoke particles Transmitter
Transmitter
Receiver
Transmitter
parameterisation. These features are not available with conventional scattered light detectors, even those fitted with an additional thermal component (OT detector), because of their "optical one-dimensionality". 2.2.3 OTG detectors Almost all fires generate the invisible and odourless toxic gas carbon monoxide (CO) during their initial phase. Smoke poisoning is the most common injury among fire victims.
Receiver
night, and most fire victims suffocate in their sleep. Every examination of fire victims has so far shown that carbon monoxide was the primary cause of death. In addition to smoke and thermal sensors, OTG multi-sensor detectors also contain an integrated CO detector. Through the early detection of fire gases, the OTG detector is able to detect a fire before it becomes visible. The system can also activate the alarm when it detects a lifethreatening concentration of odourless carbon monoxide.
95% of all fire victims are injured during the smouldering phase of the fire. This risk is particularly high at
Transmitter
Figure 6 (far left): Greatly improved signal detection for light aerosols due to the large amount of forward scattering Figure 7 (left): Greatly improved signal detection for dark aerosols due to the large amount of backward scattering
2.2.4 OTblue detectors Instead of an infrared light source, this type of detector uses a blue diode that emits very short wave light. The much shorter wavelength makes it possible to detect much smaller particles that are invisible to the human eye.
F I R E P R OT E C T I O N
Forward scattering
OTblue detectors can be used in areas that are currently monitored by ionisation detectors. They are able to detect liquid fires, open wood fires, invisible aerosols and even particles that previously could only be detected by ionisation detectors. Compared to ionisation detectors, they are far less susceptible to interFigure 8: Detection by O2T multisensor detectors during a test with different test fires
Test fire: Steam
Test fire: n-Heptane
Test fire: Cotton
In contrast to conventional scattered light detectors, O2T detectors do not trigger an alarm when exposed to steam.
They also detect dark smoke reliably and activate the alarm earlier than other detectors.
O2T detectors detect fires emitting light smoke much earlier and more reliably than conventional scattered light detectors.
s+s report international 1 / 2008 21
Figure 9: OTblue detector
Paraffin oil mist => Large particles – Similar response to blue and infrared light – Slightly more sensitive to infrared light
Smouldering cotton fire => Small particles – Significant differences in response – Three times more sensitive to blue light – This ratio becomes even more evident with even smaller particles
ference factors, such as airflow and moisture and do not require a radioactive source. The following example illustrates the advantages of optical smoke detectors with a blue LED over those with conventional infrared LEDs.
F I R E P R OT E C T I O N
3 The development of multi-criteria fire detectors compared to mono-criterion detectors The development of multi-criteria fire detectors in Germany can be illustrated on the basis of figures from Novar GmbH. The figures refer only to intelligent bus-powered fire detectors. Conventional limit monitors are not included in the statistics. Figure 11: Worldwide distribution of multi-criteria fire detectors compared to optical detectors (Novar GmbH) Red = Multicriteria fire detectors Blue = Optical detectors
22 s+s report international 1 / 2008
Figure 10: Advantages of optical smoke detectors with a blue diode compared to those with conventional infrared LEDs This section compares multi-criteria fire detectors to optical detectors since they represent the most popular type among mono-criterion detectors. Figures 11 and 12 show the steady rise in market share of multi-criteria fire detectors in Germany and internationally. Between 2002 and 2006, their share rose from approximately 24% to approximately 32%. The level of distribution on the German market (figure 12) is much higher. The share of multi-criteria fire detectors increased from approximately 38% to approximately 57%
between 2002 and 2007. The market share is not just higher at present but also continues to show higher growth rates. The differences in the market shares of multi-criteria fire detectors in Germany and abroad (figure 11) are partially due to the fact that different and less stringent standards apply abroad. For example, many countries do not require the fire detection systems to be linked to the fire brigade. Higher costs are another important factor that makes mono-criterion detectors the preferred option.
Technical progress in microprocessor technology with regard to performance and cost has led to the integration of microprocessors in fire detectors. As a result, different sensors could be combined in a single multi-criteria fire detector for the first time. This also led to the introduction of intelligent measurement value processing in point fire detectors. The resulting advantages compared to conventional detectors include the following: ❏ Interference impulses can be suppressed or filtered out by pre-filtering raw measured values using different methods ❏ Processor-controlled decentralised signal processing ❏ Alarm decisions based on intelligent algorithms ❏ Pattern recognition of fire phenomena characteristics ❏ Automatic detector adjustment to various ambient conditions ❏ Tracking of response values ❏ Optimum self-diagnostics of electronic and sensor systems ❏ Early detection of sensor contamination ––> prevention of false alarms ❏ Automatic maintenance requests
❏ Parameterability and hence optimum sensor adjustment to ambient conditions ❏ Time-controlled switching between different parameter sets depending on operating conditions ❏ Sensors can be evaluated individually or collectively ❏ Virtually constant sensitivity with regard to detecting different combustible materials ❏ Partial suppression of interference factors The advantages of multi-criteria fire detectors are demonstrated by early fire protection and a significant reduction in false alarms.
this positive development. The improved reliability of fire detection systems can be divided into different categories: ❏ Equipment and installation standards or specifications (DIN EN 54 equipment standard, DIN VDE 0833 system development, DIN 14675 Planning – Installation – Maintenance) ❏ Higher available technology of the overall system ❏ More professional system planning and installation ❏ High-quality maintenance tasks increase the reliability of the system over its entire operating time
Is there any supporting evidence for this claim? Unfortunately, there are no statistical evaluations that permit an assessment of multi-criteria fire detectors with regard to the development of false alarms.
Areas of application for O2T detectors
According to fire brigade statistics on false alarms and nuisance alarms, the number of false alarms has remained constant or has even fallen to some extent despite a continuous increase in the application of fire detection systems (source: "False Alarm Statistics on Fire Detection Systems", Jürgen Weiß, VdS conference on 9 December 2004). Multi-criteria fire detectors have certainly played an important role in
❏ Factories with high ceilings and dust or aerosol formation during production processes and machine usage, e.g. forklift trucks (figure 13) ❏ Big events with changing stage equipment and different activities, e.g. use of fog machines and smoke development (figure 14) ❏ Industrial kitchens with steam and cooking vapours.
5 Practical examples
F I R E P R OT E C T I O N
4 Advantages of multicriteria fire detectors
Typical areas of application for O2T detectors include:
Figure 12: Distribution of multi-criteria fire detectors compared to optical detectors in Germany (Novar GmbH) Red = Multicriteria fire detectors Blue = Optical detectors
s+s report international 1 / 2008 23
F I R E P R OT E C T I O N
Figure 14: Big events often involve changing stage equipment and the use of fog machines and smoke development ❏ Hospitals or nursing homes where developing gases must be detected as early as possible because bedridden or disabled patients may not be able to react quickly (figure 15). ❏ Underground garages that require an alarm for excessive gas concentrations due to high parking density or inadequate or defective ventilation systems. Figure 13: Factories with high ceilings and dust or aerosol formation during production processes and machine usage, e.g. forklift trucks
Each object is subject to specific interference factors that can cause conventional scattered light detectors to trigger a false alarm, e.g. steam from paper rolls in printing plants, paper mills, shower cubicles in hotel rooms and microparticles from humidifiers in museums. Typical areas of application for OTG detectors include:
6 Potential of future developments The introduction of multi-criteria fire detectors was an important step towards even more reliable fire detection with point detectors. We know that fires not only generate smoke, heat and flames but also produce fire gases. Since fire ga-
ses are produced during the early stage of the fire before smoke and temperatures develop, there is a potential here to detect fires immediately after ignition and consequently gain even more valuable fire fighting time. The first systems with gas sensors are already available, e.g. OTG multi-criteria fire detectors with electrochemical cells that detect carbon monoxide (CO) as described above. Other gases can be measured with the appropriate sensors but the result is only the selective detection of individual fire gases. Fires are based on certain fire gas mixtures, or fire gas patterns, which contain different gases in different concentrations. The aim of a multi-gas sensor system, the electronic nose, is to clearly identify these gases. There are many cases where nature shows us how it is done: ❏ Humans can sense and distinguish odours. People are able to smell danger such as fire without having to see it. ❏ The Black Fire Beetle (Melanophila acuminata) has a natural multicriteria fire detector. This beetle is able to smell guaiacol over large distances and heads towards the location of the fire using IR sensors.
Figure 15: In hospitals and nursing homes the formation of gas development has to be detected as early as possible because of long reaction times
Learning to smell fire gases is one of many challenges facing the development of future fire detectors.
24 s+s report international 1 / 2008
