Linear detectors – high level of fire safety for facilities.

Linear detectors – high level of fire safety of objects.

Linear smoke detectors are increasingly in demand on the Russian fire safety market. Such popularity is quite justified. With the release of the «Technical Regulations on Fire Safety Requirements», this is one of the few types of fire detectors that are allowed to protect objects up to 21 m high (in accordance with the set of rules SP 5.13130.2009). Earlier detection of fire by a linear detector compared to point smoke detectors in real conditions is noted, as well as high detection efficiency of both white (from smoldering wood, textiles) and black smoke (from burning plastic, rubber, cable insulation, etc.).Linear smoke detectors are indispensable for protecting production facilities, warehouses, hangars, tunnels, gyms, etc., where the installation of point detectors is difficult, and sometimes even impossible or impractical for economic reasons. The article, offered to the attention of readers, examines this type of detector from the point of view of the operating principle, design and its features, and also compares the detection efficiency of linear and point smoke detectors.

Operating principle and design options

Fig. 1. Operating principle of a two-component linear detector The operating principle of a linear detector can be understood from Fig. 1. On opposite walls of the protected room, under the ceiling, there are a receiver and a transmitter of an infrared signal. The IR range is used to reduce interference from artificial and natural lighting, and a pulse signal is used to reduce current consumption. The transmitter signal is recorded by the receiver. In the event of a fire, smoke rises to the ceiling and «spreads» along it, gradually increasing the area filled with it. The passage of transmitter signals through a smoke-filled environment is accompanied by their attenuation. In the receiver, the ratio of the current signal level to the signal level corresponding to an optically transparent environment is calculated. As soon as the ratio reaches the set threshold, a «Fire» signal is generated, which is transmitted via a loop to the control panel (CP).
Today, there are two main design options for linear detectors. These are two-position or two-component detectors and single-position or one-component detectors. In a two-position detector, the receiver and transmitter are made in separate housings and are installed on opposite walls during installation, as shown in Fig. 1.
Single-position detectors are more modern and consist of a transceiver unit and a passive reflector. The operating principle of a two-component detector was described above. The operating principle of a single-component linear detector differs from a two-component detector only in that the pulse signal passes through the monitored area twice: from the transceiver to the reflector and back (Fig. 2).

Design features
Due to their design features, two-position detectors have a number of disadvantages.
High-quality linear detectors use optical systems with fairly narrow directional patterns, which leads to certain difficulties during adjustment and operation. The result of adjusting a linear detector is finding such positions of the receiver and transmitter that provide the maximum transmitted signal. For two-position detectors, this process is especially labor-intensive. During operation, changing the position of the receiver or transmitter leads to a deviation of the directional pattern and, as a consequence, to a decrease in the signal level and the formation of a false alarm — there is only one way out in this situation — readjustment. Accordingly, the receiver and transmitter of a two-component detector can only be installed on capital structures.
Also, for a two-component detector, it is necessary to ensure a stable transmitter signal level over the entire range of operating temperatures and supply voltages, since a decrease in the transmitter signal level will lead to the formation of a false «Fire» signal. To ensure the operation of two-component detectors at different ranges, it is usually necessary to use several transmitter signal levels and adjust the receiver gain, which creates additional difficulties during setup and adjustment. Another significant drawback is the need to connect both the transmitter and the receiver to a power source, which leads to significant cable consumption, usually exceeding the distance between the receiver and the transmitter. In addition, when installing several linear detectors in parallel in one room, it is necessary to prevent signals from neighboring transmitters from reaching the receiver. Some manufacturers in this case recommend installing receivers and transmitters in a staggered pattern, which leads to additional installation work and an increase in cable consumption. Moreover, installation of this part of the loop is usually difficult due to high ceilings or the need to perform hidden wiring.Almost all of these disadvantages are absent in single-position linear smoke detectors (Fig. 2). The placement of the receiver and transmitter in one block provides the ability to automatically select the signal level measurement range during adjustment, automatic adjustment of the transmitter radiation level and the receiver gain depending on the range of the monitored zone.
The reflector of such detectors is passive, i.e. it does not require power supply and adjustment. This feature allows to reduce cable consumption, labor intensity of installation and adjustment several times. The reflector itself is a cataphot consisting of a large number of prisms, the structure of which provides signal reflection in the direction of the source. The reflector allows installation on non-permanent and even vibrating structures. Modern single-position detectors allow changing the reflector position within ±10°. At large angles, there is a decrease in the level of the reflected signal due to a decrease in the equivalent area of ​​the reflector. In addition, modern single-position detectors have a time selection of signals, which allows using one reflector for two or three detectors (if they are located close to each other). This can be especially important if the detectors generate a signal to control the building automation, fire extinguishing system, type 5 warning system.

Sensitivity and its control
The sensitivity of a linear detector is determined similarly to that of a point optical detector, but is characterized by the value of the optical density of the medium for the set maximum range at which the detector is triggered. The requirements for such detectors are defined in Part 4.9 «Linear optical-electronic fire and smoke detectors» of GOST R 53325-2009. According to the specified requirements, the sensitivity of a linear detector should be in the range from 0.4 dB to 5.2 dB. In the technical documentation for a detector, its sensitivity can be specified as a percentage of signal attenuation, which is especially typical for Western manufacturers. In general, one unit can be converted to another using the following formula; a signal decrease of ±10% corresponds to an attenuation of L dB:
L = 10lg[100/(100 ±12; ±10%)] dB (1)
Table 1 shows an example of calculation using formula (1).
Table 1

Modern linear detectors have several sensitivity thresholds and compensation for dusty optics, which allows taking into account operating conditions, eliminating false alarms and reducing maintenance costs.
When the limit of the automatic compensation range is reached, such detectors generate a separate signal «Maintenance», indicating the need for maintenance (Fig. 3).

Fig. 3 Optical system dust compensation Even today, in the budget detector segment, you can find linear detectors without automatic compensation for optics dust. When dust accumulates during operation, such detectors are subject to false alarms, which require more frequent maintenance to eliminate. As a result, the initial savings on equipment are offset by more frequent maintenance.
Western single-position detectors of the latest generation have adaptive thresholds to eliminate false alarms caused by an increase in optical density in the monitored room during working hours (see Fig. 4).

Fig. 4 Adaptive threshold Unlike the fixed threshold, with the adaptive threshold, slow changes in the optical density of the environment during the day are compensated within the specified limits. Thus, in one of the widely known linear detectors, in addition to four fixed sensitivity levels of 25%, 30%, 40%, 50% attenuation, there are two adaptive levels of 30–50% and 40–50%. When setting the adaptive threshold, for example, 30–50%, the sensitivity will actually be maintained at a level of 30% and there will be no need to roughen it to 50% to eliminate false alarms during working hours.
To test a linear detector, it is sufficient to attenuate the signal by the response threshold. As a rule, this is done using a special filter with a certain transparency value (attenuator), which is installed in front of the optical system of the transmitter or receiver. Such a filter usually has a periodic structure, for example, in the form of dots on a transparent material or holes in an opaque material, the diameter of which is significantly smaller than the dimensions of the optical system of the receiver and transmitter (Fig. 5). The ratio of the opaque area of ​​the filter to the total area determines the percentage of attenuation introduced.

Fig. 5. Example of a test attenuator To test the sensitivity of a two-component linear detector, it is sufficient to have two filters for each sensitivity level. For example, to test a response threshold of 30%, you can use two filters with attenuation of 25% and 35%. These filters are the simplest devices and are usually included in the kit of high-quality Western-made linear detectors. Optical filters provide a complete check of the operability of the linear detector during operation.
To test a single-component detector, you can also use optical filters of the appropriate size, which are installed in front of the transceiver or in front of the reflector. However, in a single-component linear detector, it is more convenient to introduce signal attenuation by “shading” a certain area of ​​the reflector (Fig. 6).

Fig. 6 Reflector shading A question often arises: why is it necessary to cover more than half of the reflector area to simulate signal attenuation by 30%, and about 3/4 of the area for 50%? There is no error here, since in a single-component linear detector, unlike a two-component one, the signal passes the controlled area twice: from the transceiver to the reflector and back. Accordingly, with real smoke, weakening the signal by 3 dB (by 50%), the signal will return to the transceiver weakened by 6 dB (by 75%). A simple calculation for a reflector without a scale: for example, the set sensitivity level is 30%, when the signal is weakened by 30%, 70% of the signal will reach the reflector, i.e. 0.7 of the original level, and on the way back there will also be 0.7 of the signal reflected from the reflector, and in total 0.7 x 0.7 = 0.49, or 49% will return, the attenuation will be 1 — 0.49 = 0.51, i.e. 51%. This effect shows that the potential sensitivity of a single-component linear detector is twice as high as that of a two-component one, and in reality, when setting the same sensitivity, the noise immunity is also higher due to the threshold being doubled.

Detection efficiency
Incorrect testing of a linear smoke detector, even by experienced installers, leads to false conclusions about its lower sensitivity compared to a point optical-electronic detector. Indeed, if a point detector is tested using cigarette smoke, which is blown into the optical chamber and creates unrealistic smoke concentrations in it, the detector is quickly activated, but with similar smoke filling of the linear detector's light filter, no such reaction is observed. Such testing cannot demonstrate the operability of either a linear or a point detector, since smoke filling of an insignificant volume of the room near the detectors does not even remotely reproduce the physical processes accompanying a real fire.
To fully understand this issue, it is necessary to compare the reaction of the detectors to a real fire source. Therefore, it is worth turning to fire tests. The methodology for their implementation is described in the European standard EN 54 for smoke point detectors in Part 7 and linear detectors in Part 12, as well as in our GOST R 53325-2009 in Appendix H «Fire tests of fire detectors». It is worth noting that in Europe, fire tests are mandatory for the certification of detectors, and in Russia only «when putting into production, changing the design or electrical circuit diagram of detectors.» We will not describe the methodology itself in detail, but will only dwell on its main points.

Fig. 7 TP-3 fire source There are six types of test fires: TP-1 – open burning of wood, TP-2 – smoldering wood, TP-3 – smoldering cotton, TP-4 – burning polyurethane, TP-5 – burning heptane and TP-6 ​​– burning alcohol.
Smoke detectors are tested on four test fires TP-2, TP-3, TP-4, TP-5. Each test source not only consists of a specific material, but also has a very specific configuration and dimensions.
For example, the TP-3 hearth consists of approximately 90 cotton wicks, 800 mm long and weighing approximately 3 g each, attached to a wire ring 100 mm in diameter, suspended on a tripod (see Fig. 7). The ends of the wicks, bundled together, are ignited with an open flame, then the flame is blown out until smoldering, accompanied by a glow, appears.
The tests are carried out in a room 9–11 m long, 6–8 m wide and 3.8–4.2 m high, in the center of which a test fire source is located on the floor. The tested point detectors are located on the ceiling along a circle at a distance of 3 m from its center in a 60° sector (see Fig. 8). An optical density meter of the environment m (dB/m), a radioisotope meter of the concentration of combustion products Y (relative units) and a temperature meter T (°C) are also installed here. Two tested linear detectors are located symmetrically, and their optical axes are at a distance of 2.5 m from the center of the room.

Fig. 8 Dimensions of the room and the layout of the detectors. For each type of test source, the boundary values ​​of optical density m, temperature T and concentration of combustion products Y are established.
To be able to compare, it is necessary to evaluate the sensitivity of linear and point detectors in the same units. The sensitivity of a linear detector is defined in absolute attenuation units, and the sensitivity of a point detector is specified in specific units, i.e. the attenuation value at a distance of one meter. In accordance with Part 4.7 «Point optical-electronic smoke fire detectors» of GOST R 53325-2009, the sensitivity of point detectors should be within 0.05–0.2 dB/m. To convert the absolute attenuation value into specific units of optical density of the medium, it must be divided by the length of the zone in meters. Accordingly, the requirements for the sensitivity of a linear smoke detector from 0.4 dB to 5.2 dB with uniform smoke filling of a 10-meter zone correspond to a specific optical density in the range from 0.04 dB/m to 0.52 dB/m, and with a zone length of 100 m — in the range from 0.004 dB/m to 0.052 dB/m. Theoretically, with constant sensitivity, the efficiency of a linear detector increases with an increase in the length of the protected zone.

These results clearly show that the 6500 linear detector has no dependence of sensitivity on the type of smoke. It responds equally well to both light smoke emitted by smoldering wood and textile materials, and to black smoke emitted by burning plastic, cable insulation, rubber products, bitumen materials, etc. For comparison, Table 3 shows the test results of smoke point optical-electronic detectors. These tests were conducted at different times, as a result of which there are differences in the rates of increase in the optical density of the medium, the concentration of suspended particles and temperature.

Thus, even with relatively low ceilings (4 m) and a small optical beam length (5 m), the linear detector is activated at lower levels of specific optical density of the environment compared to point optoelectronic detectors. Moreover, if for a point detector the test conditions correspond to the operating conditions at most facilities with minor deviations, then for linear detectors these conditions are the most unfavorable for its operation. With an increase in the length of the protected zone with a fixed sensitivity level in absolute attenuation units, the linear detector will be activated, accordingly, at lower values ​​of specific optical density. With an increase in the height of the room, the advantages are further enhanced, since smoke dispersion at a high altitude affects the linear detector to a lesser extent than a conventional point detector.
It is not surprising that the sensitivity of point detectors in these tests does not meet the GOST requirement of 0.05–0.2 dB/m. According to EN54, the sensitivity limit for a point smoke detector during fire tests is 2 dB/m. Since for a point detector, the aerodynamic resistance of the smoke inlet of the fire detector has an effect in real conditions. An unsuccessful design of the smoke inlet and smoke chamber of the fire detector, a relatively small area of ​​the smoke inlet compared to the internal volume of the detector can lead to a decrease in sensitivity in real conditions by more than 10 times. To one degree or another, this effect is manifested in any point smoke detector with a smoke chamber and with structural elements for protection against dust. In a linear smoke detector, this effect is completely absent, since smoke enters the controlled area without overcoming any obstacles.

Conclusion
Modern linear smoke detectors, if correctly installed and configured, provide a high level of fire protection. They are more effective than point smoke detectors in facilities with extended zones and high ceilings. They are highly effective in detecting virtually any type of fire with various types of smoke: from smoldering wood and textiles to burning plastic, rubber, bitumen, cable insulation, which ensures their universal application. Using a linear detector of a single-component design, compared to a two-component one, reduces the amount of installation work, cable consumption and adjustment time several times.