False alarms have become increasingly rare in industrial areas over the past 15 years. Where there was previously no adequate fire detection, fires are now often detected very early on – with practically no false alarms. The following article explains why this is the case, and what role air sampling smoke detectors play in this development.
When fires cause significant property damage or even endanger human lives, it is usually simply the result of an accepted risk, combined with the ever-present possibility of such damage occurring. Operational downtime, negative impacts on delivery capability or business capacity, and loss of customers are all financial risks that must be accounted for when planning fire protection. Modern fire protection technologies allow operators to set appropriate protection objectives practically without compromise—even in very difficult environmental circumstances.
Previously, in the absence of other alternatives, heat detectors were often used in very dusty areas, even though these detectors must “wait out” a great deal of valuable time before an incipient fire has reached a detectable size. Standard and test fires contributed to the fact that even a large number of professionals severely underestimated the potential time advantages of very early fire detection.
Whereas only a few seconds can be gained when it comes to detecting quick-burning test fires, about two-thirds of real, damage-causing fires can be identified minutes or even hours faster. This head start necessarily translates to drastic reductions in the extent of the damage—often meaning the difference between total losses and minor damage. Rather than combating open flames, the statistical majority of cases involve merely resolving an electrical issue, such as a short circuit. Rather than requiring great quantities of extinguishing water, problems can be solved by merely cutting power to an area, thereby withdrawing the energy the incipient fire would need for maintenance and further development. There are no statistics documenting the damage successfully prevented through such very early countermeasures.
Even as recently as twenty years ago, experts considered a smoke detector "good" if it was set to such a low degree of sensitivity that it only reacted at the very limits of the prescribed tolerance ranges when exposed to standard test fires. This low level of sensitivity was presumably the best known method of preventing false alarms; heat detectors were the only known method of improving the situation. Numerous technical developments since that time (though a few of them first appeared on the market early on) have ensured that smoke detectors are now significantly less prone to false alarms despite having become exponentially more sensitive:
a) Dual detector dependency (Type A or B):
Fire alarm conditions are only triggered if two linked automatic detectors send out alarm signals. This method is primarily used to control automatic extinguishing systems. However, its effectiveness is limited when it comes to deceptive phenomena if the two alarms used in the dual detection system employ the same detection principles.
b) Fire pattern recognition:
Smoke detectors cannot reliably determine whether detected particles are smoke particles or merely interference factors. Highly effective algorithms for evaluating fire patterns instead analyse whether the signal characteristics of an event correspond to those of a fire situation and/or a non-fire situation.
c) Multi-criteria evaluation:
Combining different fire detection criteria can provide additional security against false alarms. How effective such a system is depends on both the sensors in use and their wiring. Using detection techniques of differing sensitivity levels can mean that the alarm is only as effective as its least sensitive component.
New approaches are currently being tested in which aerosols are analysed using various wavelengths and scattering angles, resulting in a system that can not only differentiate between false alarm situations and fires, but can even detect which type of deceptive phenomenon is present or what material is burning.
a) Drift compensation:
A great deal of the progress that has been made in general fire alarm technology in terms of preventing false alarms can be attributed to the implementation of drift compensation, wherein the fire alarm threshold is adapted to the change in quiescent level (through air pollution or other background noise within the area of use) within the framework of normative specified limits. The goal of such an approach is to provide consistent detection quality, so that the same quantity of smoke is always necessary in order to trigger an alarm (absolute fire detection). With older detectors that do not have drift compensation, on the other hand, detection quality is inevitably affected by changes in background noise.
b) Noise ratios:
Combining high-quality detectors and complex algorithms makes it possible to produce fire detectors with large signal-to-noise ratios, meaning the noise signal power is very low compared to the power of the desired signal. Such fire detectors are thus both highly sensitive and highly secure against false alarms.
c) Physical dust filtration:
In areas prone to a great deal of dust accumulation, air filters can be used to remove the majority of larger dust particles, while allowing smaller smoke particles to pass through unhindered. This method is restricted to air sampling fire detectors.
d) Collection effect:
Air sampling smoke detectors make use of the effect that smoke gases begin distributing heavily throughout the room even early on in the incipient fire stage. Each individual air sampling point must fulfil at least the same standards of fire detection as a conventional, point-type smoke detector. With every additional air sampling point that takes in smoke, the smoke density in the detector unit increases. Many types of dust, on the other hand, sink at a higher speed, and thus only reach the air sampling point(s) in the direct vicinity of where they were released.
Implementing one or more of the technical measures listed above is very expensive in point-type fire detectors, as the highly complex technology must be operated at each detection point separately. Air sampling smoke detectors represent a cost-effective option in this regard, as they use a central system that can cover dozens of detection points.
Besides the benefits of centralised technology and intake air filtration, air sampling smoke detectors offer a number of other advantages for applications in critical environmental conditions, such as: