RU2256230C2 - Smoke detection method - Google Patents

Abstract

FIELD: signaling systems and systems for detecting the presence of air-borne smoke particles in environment.

SUBSTANCE: method involves periodical emitting light pulses; receiving reflected light flux by photosensor, wherein light flux is reflected at an angle different from optical radiation axis; measuring charge-discharge cycle durability of photosensor included in RC-circuit in absence of light pulses to be emitted t₁ and that in presence of the light pulses t₂; comparing the difference between t₁ and t₂ values |t₁-t₂|=k with predetermined one and determining smoke absence/presence and operational reliability of optical channel from above comparison.

EFFECT: increased operational reliability due to control operational capability and cleanliness level of optical channel, possibility to turn to digital circuit technology.

4 dwg

Description

The invention relates to automatic alarms, methods for detecting the presence of suspended smoke particles in the environment.

Known smoke detectors (fire detectors, detectors), operating on the principle of periodic emission of light pulses and the subsequent reception by the photosensor of light reflected from smoke particles. The signals (current or voltage) from the photosensors are amplified using various analog amplifying devices (transistor stages, operational amplifiers, etc.) and compared with the reference signal, as a result, a signal is issued indicating the presence or absence of smoke. The emitted and received pulses can be modulated, compared and processed in various ways, contributing to some increase in sensitivity and noise immunity (see Germany, Z. 2361403, RF, Z. 93021357/09, RF, Z. 2037883, RU 2134907, RU 2125739).

These methods do not provide the ability to control the dustiness of the optical channel. When the optical channel is dusty, smoke detection at an early stage of ignition is difficult, and sometimes impossible. The lack of dust control reduces the reliability of the device and is their disadvantage.

Known smoke detector (RF patent 2125739, G 08 B 17/10 from 01/27/99), selected as a prototype, the principle of which is as follows:

- periodically emitted light flux into the optical camera;

- receive reflected signals by a photodetector, the operating point of which is set by applying current (voltage), at an angle different from the optical axis of the emitter;

- reinforce the change in current (voltage) on the photosensor that occurs under the influence of a pulse of the light flux;

- form a signal for detecting smoke, which is the result of a comparison with a reference signal (current, voltage).

In the absence of smoke in the sensitive area of the camera, the photodetector signal is below the threshold value specified by the comparison circuit. The result of comparing the signal from the photodetector with a threshold value in the form of a logical signal is fed to the comparison and storage circuit. When smoke appears, light pulses, reflected from the smoke particles, fall on the photodetector, the signal from which now exceeds the threshold value and from the comparison circuit in the form of a change in the logical signal is fed to the comparison and storage circuit. When this situation repeats four times in a row, a fire signal is issued.

The disadvantages of the prototype should include

- the simultaneous use of both digital and analog circuitry;

- lack of control of the operability of the optical channel during operation. When the prototype is working, the failure of the optical channel (emitter, photodetector, amplifiers) is adequate to give a signal "no alarm" in the presence of smoke, or to give a false alarm "fire";

- lack of control of dustiness of the optical channel.

The objective of the present invention is to increase the reliability of operation by monitoring the health and dustiness of the optical channel and a complete transition to digital circuitry.

The problem is solved by the method of detecting smoke, which consists in the fact that light pulses are periodically emitted and a reflected light is received by the photosensor coming at an angle different from the optical axis of the radiation, characterized in that the charge-discharge cycle of the photosensor, which is an element of the RC circuit, is measured , in the absence of emitting pulses t ₁ and in the emission of light pulses t _2, and compared their difference | t ₂ -t ₁ | = k with predetermined values, which is judged on the presence or absence of smoke and operability STI channel.

Figure 1 presents a block diagram of a device that implements the method. Figure 2 shows the equivalent model of the photodetector 2. Figure 3 shows the timing diagrams of the discharge of the photodetector. Figure 4 shows a schematic diagram of the device of figure 1, which implements the method.

The device of FIG. 1 contains a light pulse source 1, optically coupled to a photodetector 2, which is connected to a logic port of a controller 3, a signaling device 4, the input of which is connected to the output of a controller 3, and the input of a light pulse source 1 is connected to an output of a controller 3.

The device (Fig. 4), in principle, contains a controller 3 - AT90S1200 (DD1), a source of light pulses 1 - an infrared diode of the type AL164B1 (VD1) with a current-limiting resistor R1 = 120 Ohm, a photodetector of the type KDF115, and the signaling device 4 in principle is made in the form a sound element ЗП1 of type ППА1 and LED VD3 of type АЛ307 with current limiting resistance R2 = 500 Ohm, С - storage capacitance of type K-50-35 -100 μF. The smoke detector is connected to a + U power supply with a voltage of 4.5-6 V.

The method for detecting smoke is as follows. Let photodetector 2 be an element of the RC circuit (Fig. 2), in which one or both elements are simultaneously sensitive to the luminous flux F. (R = f (Ф) or C = f (Ф)). Therefore, the duration of the charge (discharge) cycle will also depend on the light flux τ = RC = F (Ф) (Fig. 3). The optical signal emitted by the source of light pulses 1 is reflected from the smoke particles and enters the photodetector 2, while the duration of the charge (discharge) cycle changes.

In real devices, there is an optical connection between the source of light pulses 1 and the photodetector 2, even without smoke. This optical coupling can be used to determine the dustiness and serviceability of the optical channel.

The operation algorithm of the device of figure 1 consists of the following steps. The duration of the cycle t ₁ (Fig. 3) of the charge (discharge) of the photodetector 2 is measured to a certain threshold level U _n in the absence of light flux from the source of light pulses 1, then the source of light pulses 1 is turned on and the duration of the charge (discharge) cycle t _{2 of the} photodetector is measured 2, then find Δt = | t ₁ -t ₂ |. In this case, the influence of destabilizing factors (such as background illumination, temperature, etc.) is subtracted and only the change in the charge Δt time (discharge) caused by the action of the optical emitter remains. Then compare Δt with threshold settings Δt ₁ , Δt ₂ ..Δt _n . For example, if Δt ₂ <Δt <Δt ₃ , then this may correspond to the absence of smoke and the integrity of the optical channel, if Δt ₁ <Δt <Δt ₂ - the optical channel is dusty, if Δt <Δt ₁ - the optical channel is faulty, if Δt> Δt ₄ - the presence of smoke.

In a practical implementation, it is convenient to use the time - number of pulses transformation for measuring t ₁ and t ₂ (similarly to the principle of operation of an analog-to-digital converter with a time-pulse conversion) N1 = t ₁ f _g , N2 = t ₂ f _g , where f _g is the clock pulse generator frequency. In this case, the difference Δt = | t ₁ -t ₂ | is proportional to the difference in the number of pulses k = N1-N2) = f _g | t ₁ -t ₂ |.

Reaching the threshold level U _{p is} controlled by a threshold device that can be performed on logic elements. In this case, the photodetector 2 is directly connected to the logical port. For logic elements with a high input impedance (CMOS - series), the switching threshold from a logical unit to a logical zero, as a rule, is equal to half the supply voltage of the microcircuits. In this case, the number of pulses is measured, during which the voltage at the photodetector 2 from a logical unit reaches a logical zero.

The use of controller 3 with the proposed time-pulsed conversion makes it possible to reduce the number of elements, makes it possible to perform multiple measurements with averaging their results, eliminate systematic errors, automatically control the correct operation of the device, and correct errors caused by the dustiness of the optical camera.

The device of figure 1 works as follows. In the memory of the controller 3, predefined threshold values k1, k2, k3 are recorded. The port of controller 3, to which photodetector 2 is connected, is set to a high state (the photodetector capacitance is charging), then this port of controller 3 is switched as input with a high resistance, and the internal counter (timer) of controller 3 is turned on, the count of which stops when the controller input 3, to which the photodetector 2 is connected, is discharged to the level corresponding to a logical zero, while the logical port m of the controller 3 acts as a threshold device. This state of the counter (N1) is remembered by the controller 3. Then the port of the controller 3, to which the photodetector 2 is connected, is again set to a high state (the capacity of the photodetector 2 is charged), then the controller 3 turns on the light source 1, sets the port of the controller 3 to which it is connected photodetector 2, as an input with a high resistance, simultaneously turns on the internal counter (timer) of controller 3, the count of which stops when the input of controller 3, to which photodetector 2 is connected, is discharged to a level with corresponding to logical zero, this counter state (N2) is stored by controller 3. Controller 3 turns off the light pulse source 1. There is a difference k = N2-N1, which is compared with threshold values k1, k2, k3 by controller 3. If k2 <k <k3, then the smoke detector is operational, the optical channel is dust free, and there is no smoke. If k1 <k <k2, then the controller 3 includes a signaling device 4, which gives a signal characterizing the dustiness of the optical channel. If k <k1, then the controller 3 includes a signaling device 4, which generates a signal indicating a malfunction of the optical channel. If k> k3, then the controller 3 turns on the alarm 4, which gives an alarm indicating the presence of smoke. The signals for each type of alarm are different from each other.

The threshold values k1, k2, k3 can be determined as follows.

Measure the threshold value k = k0 in the absence of smoke and the dust-free optical channel (intrinsic state of the optical channel of the device) with a device that implements the method and determine k2 = 0.7k0. Place the device in a smoke chamber with bright smoke, the optical density of which is, for example, 0.1 dB / m, measure k = k3. Then determine k1 = 0.1k0.

Consider the influence of destabilizing factors. Assume that the RC circuit (Fig. 3) is charged to a voltage of U ₀ . The discharge of such a circuit is carried out according to the well-known law U = U ₀ exp (-t / τ), where τ = RC. The discharge time will be measured to a certain threshold level Uп. It is assumed that the time constant is sufficiently small and the measurement of time intervals t1 and t2 is carried out directly one after another, therefore, a change in the threshold level Uп, background illumination Φ and temperature per measurement cycle can be neglected. With a change in the supply voltage and temperature, U ₀ also changes, as well as Uп (Uп / U ₀ -const) linearly associated with it. Consider how this affects the operation of devices. The time intervals t1 and t2 are determined from the expressions

Uп = U ₀ exp (-t1 / τ1),

Uп = U ₀ exp (-t2 / τ2),

Where τ1 = R1 · C1 and τ2 = R2 · C2 are the time constants of the circuit in the absence and presence of a light flux pulse. By logarithm and simple transformation, we determine the time intervals t1, t2:

t1 = τ1 ln (Uп / U ₀ ),

t2 = τ2 ln (Uп / U ₀ ).

Find Δt

Δt = | t ₁ -t ₂ | = | τ1 ln (Uп / U ₀ ) -τ2 ln (Uп / U ₀ ) | = | τ1-τ2 | ln (Uп / U ₀ ).

Therefore, Δt = | t ₁ -t ₂ | independent of voltage U ₀ .

Consider the effect of background illumination and temperature. Let the resistance R = f (Ф) of the photosensor be sensitive to the light flux. This resistance can be represented as R = R0 + ΔR (Ф ₀ ) + ΔR (Фи) + ΔR (T), where R0 is the resistance of the photodetector in the absence of light, ΔR- (Ф ₀ ) is the change in resistance caused by the influence of background illumination, ΔR (Phi) is the change in resistance when the light flux hits the photodetector from a source of light pulses, ΔR (T) is the change in resistance of the photodetector under the influence of temperature. Using the above calculations, we find Δt:

Δt = ln (Up / U ₀ ) | τ1-τ2 | = ln (Up / U ₀ ) | R1 · C-R2 · C | = ln (Up / U ₀ ) | [R0 + ΔR (Ф ₀ ) + ΔR (T)] · C- [R0 + ΔR (Ф ₀ ) + ΔR (Т) + ΔR (Фи)] · С | = ln (Uп / U ₀ ) ΔR (Фи) · С

Thus; Δt = | t ₁ -t ₂ | independent of background illumination and temperature, and therefore does not reduce the sensitivity of the smoke detector.

The device according to the proposed method is easily implemented on a modern element base (figure 4). In this case, the Atmel microcontroller AT90S1200 (DD1) was used as controller 3, based on field-effect transistors, with a built-in clock generator that does not require external elements and has a frequency f _g , an eight-digit timer-counter, bidirectional logic ports PD and PB that can work as input or output ports. Input impedance over 100 megohms. The source of light pulses 1 is an infrared diode of the AL164V1 (VD1) type with a current-limiting resistor R1 = 120 Ohm, a signaling device 4 - an audio element ZP1 of the PPA1 type and a VD3 LED of the AL307 type with a current-limiting resistance R2 = 500 Ohm, C is a storage capacitance of the K-50- type 35-100 uF. The device is connected to a power supply + U voltage of 4.5-6 V.

The algorithm of the device (figure 4) is as follows. In the memory of the controller 3, predefined threshold values k1, k2, k3 are recorded. The ports of controller 3 PD0, PD5, PB4, PB1 are set as output to high state, the other ports are disabled. When the PD5 port is set to high, the capacitance of photodetector 2 is charged. Then the PD5 port of controller 3 is initialized as an input with a high resistance, and the internal pulse counter of the internal clock of controller 3 is turned on, which stops when the voltage at the photodetector 2 is connected to the PD5 port of controller 3 , will have a level equal to logical zero. This counter state is memorized (N1). Then, the PD5 port of controller 3 is again set to high and the capacitance of photodetector 2 VD2 is charged. The PD5 port of controller 3 is initialized as an input with a high resistance, the PD0 port is turned on as an output in a low state, while current flows through the diode VD1, which is a source of light pulses 1, and the internal pulse counter of the internal clock of controller 3 is turned on, which stops when the voltage at the photodetector 2, connected to the PD5 port of controller 3, will have a level equal to logical zero. This counter state is again remembered (N2) by the controller 3, and the light source 1 - VD1 is turned off. The difference k = N2-N1 is found, which is compared with the threshold values k1, k2, k3 by controller 3. If k2 <k <k3, then the smoke detector works normally, there is no smoke, the camera is dustless. The VD3 LED turns on briefly through the PB4 port, indicating that the fire detector is operating normally. Next, the cycle of operation of the controller 3 is repeated from the beginning. If k1 <k <k2 and this state is repeated many times over a long period of time (for example, for ten minutes), then controller 3 sends short-term pulses in the sound range through port PB1 to control signaling device 4 of ZP1, which characterize the dustiness of the optical channel. If k <k1 and there was no state k1 <k <k2, then the controller 3 through the port РВ1 gives short-term pulses in the audio range, indicating a malfunction of the optical channel. These sound pulses are different from those that show the dustiness of the optical channel. If k> k3, then through the port РВ1 a continuous alarm signal of frequency f2 is issued to the signaling device 4 ЗП1, indicating the presence of smoke.

Thus, in comparison with the prototype, the proposed method allows you to create devices that are completely digital circuitry, to ensure the health and dustiness of the optical channel in the process, which increases the reliability of its operation.

Claims (1)

The method of detecting smoke, which consists in the fact that light pulses are periodically emitted and a light sensor is received by the photosensor, arriving at an angle different from the optical axis of the radiation, characterized in that they measure the duration of the charge-discharge cycle of the photosensor, which is an element of the RC circuit in the absence of emitting pulses t ₁ and when emitting light pulses t ₂ , and compare their difference | t ₁ -t ₂ | = k with the given values, which judge the presence or absence of smoke and the health of the channel.

Images