A flue gas ultra-low emission dust detection system

Abstract

The present invention discloses a system and method for detecting ultra-low emission dust in flue gas, and belongs to the field of environmental detection. The detection system of the present invention includes a velocity meter, a smoke detection unit, a mass flow meter for detecting air flow, and an industrial computer for data processing, wherein the smoke detection unit includes: a smoke detector, whose air inlet end is connected to a sampling tube and a mass flow meter, and the gas is ionized by releasing high voltage electricity, and dust in the gas is attached to the ions, and the gas concentration is obtained by the current; a light scattering sensor, which is connected to the exhaust end of the smoke detector, and is used to detect the concentration of particles in the discharged mixed gas, and the detected air is transported to the flue through a reflux pipe. The present invention combines the smoke detector with the light scattering sensor to detect particles of different particle sizes, thereby improving the detection accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental monitoring, and more specifically to a flue gas ultra-low emission fine dust detection system. Background Art

[0002] There are many solutions for smoke detection in the existing technology. The concentration detection of particulate matter is already quite accurate. At present, the more important issue is how to achieve accurate detection of fine dust when the amount of smoke discharged from the pipeline is relatively small. The optical dust meter currently used for detection is greatly affected by the particle size of the particulate matter and often cannot accurately reflect the particle concentration of smoke. The accuracy and objectivity of the detection are particularly important. Therefore, how to achieve accurate detection of fine dust under low emission conditions is more important.

[0003] For the related scheme, the Chinese patent discloses a flue particulate matter isokinetic sampling and constant flow control system based on jet modulation and a method for pre-processing other parts of the flue using the above system (application number: 201711059808.7), including: a jet modulation subsystem, a zero gas preparation subsystem, a zero gas control subsystem, a sample gas introduction subsystem, a sample gas mixing subsystem, a sample gas heating subsystem, a sample gas dehumidification subsystem, a particulate matter detection subsystem and an air compressor. An air compression source is used as the power source for multiple jet modulation systems in the system to provide power for the jet modulation system that realizes flow control in each level of functional subsystems, and the negative pressure generated by the jet modulation system is used to suck the sample gas to be tested, and the automatic control of flue gas isokinetic sampling is realized through the jet modulation system (composed of a solenoid valve, a flow meter, a controller, etc.).

[0004] Also disclosed is a particle online monitoring system and method based on the β-ray method and light scattering method (application number: 201510878762.6), which is used for online real-time and accurate measurement of the concentration of particles in flue gas. This scheme combines the advantages of the β-ray method and the light scattering method, and uses the internal reference method to achieve reference calibration of the β-ray method measurement results to the light scattering method measurement results, thereby achieving online real-time and accurate measurement of the concentration of particles in flue gas.

[0005] The above two patent solutions are both for detecting smoke dust. The first solution is to control the humidity of smoke, reduce the influence of temperature, and use constant speed sampling for detection; the second solution is to compare and calibrate the detection of β-ray method and light scattering method to achieve accurate detection. However, when the accurate value is not obtained by itself, it will cause greater error as a benchmark, which has certain defects. Summary of the invention

[0006] 1. Technical problem to be solved by the invention

[0007] The purpose of the present invention is to overcome the deficiency of the prior art that the particle concentration of smoke cannot be accurately reflected, and to provide a smoke ultra-low emission fine dust detection system. The present invention combines the PPS smoke detector with a light scattering sensor to detect particles of different particle sizes, thereby improving the detection accuracy.

[0008] 2. Technical solution

[0009] In order to achieve the above object, the technical solution provided by the present invention is:

[0010] A smoke ultra-low emission dust detection system of the present invention comprises a flow meter, a smoke detection unit, a mass flow meter for detecting air flow, and an industrial computer for data processing, wherein the smoke detection unit comprises:

[0011] The smoke detector, whose air inlet is connected to the sampling tube and the mass flow meter, ionizes the gas by releasing high voltage electricity, and the dust in the gas adheres to the ions, and the gas concentration is obtained by the current;

[0012] The light scattering sensor is connected to the exhaust end of the smoke detector and is used to detect the concentration of particulate matter in the exhausted mixed gas and transport the detected air to the flue through the reflux pipe.

[0013] As a further improvement of the present invention, the air inlet end of the mass flow meter is connected to a solenoid valve S1, and the air inlet end of the solenoid valve S1 is provided with a pressure detector, and the on-off of the solenoid valve S1 is controlled by a feedback signal of the pressure detector.

[0014] As a further improvement of the present invention, a filter is provided on the air intake pipeline of the solenoid valve S1.

[0015] As a further improvement of the present invention, the flow meter is connected to the exhaust end of the solenoid valve S1 through a pipeline for back-blowing the flow velocity probe of the flow meter.

[0016] As a further improvement of the present invention, a three-way pipe is provided on the connecting pipeline between the sampling tube and the smoke detector, the third interface of the three-way pipe is connected to the exhaust end of the solenoid valve S1, and a solenoid valve S2 is provided on the connecting pipeline.

[0017] As a further improvement of the present invention, the mass flow meter calculates the opening flow rate according to the flue gas flow rate signal 4-20mA measured by the flow meter, and the flue gas flow rate v 烟 (m / s) and mass flow meter opening flow Q 计 The algorithm model is:

[0018]

[0019] Flue gas flow rate signal 4~20mA corresponds to flue gas flow rate v烟 (m / s) value is 0~40m / s, then v 烟 =2.5*i-10 where i is the flue gas flow rate output current signal.

[0020] As a further improvement of the present invention, the smoke detector sets the flow rate to L, and the concentration value output by the smoke detector is M pps , the actual particle concentration is m p ' ps , then the conversion relationship is:

[0021] Assume that the output concentration value of the light scattering sensor at a certain moment is M pm The actual converted concentration value is m p ' m , then the conversion relationship is:

[0022] Where: Q 吸 is the actual sampling gas flow rate sucked into the smoke detector; Q 计 Turn on flow for the mass flow meter.

[0023] A method for detecting ultra-low emission fine dust of flue gas of the present invention adopts the above detection system for detection, and comprises the following steps:

[0024] S1. The system is powered on for preheating. The smoke detector is heated to a predetermined temperature and the heating temperature signal is transmitted to the industrial computer.

[0025] S2, start air supply. When the pressure detector detects that the air pressure meets the set value, the solenoid valve S1 is opened, the mass flow meter is turned on, and the opening flow of the mass flow meter is controlled according to the flue gas flow rate detected by the flow meter;

[0026] S3, the gas enters the smoke detector through the mass flow meter, the internal jet of the smoke detector forms a negative pressure, and the smoke in the flue is sucked into the smoke detector through the sampling tube;

[0027] S4. The metal needle in the smoke detector releases high voltage to ionize the gas. The dust particles in the sampled gas attach to the ions and are charged. When the sampled gas flows out, the ammeter in the smoke detector obtains the concentration M of the sampled gas through the measured current. pps ;

[0028] S5. The sampled gas is mixed with air and enters the light scattering sensor to detect the concentration of remaining particles M. pm After detection, the gas is discharged into the flue through the reflux pipe.

[0029] As a further improvement of the present invention, the total particle concentration value M in the sampled gas should be: m = K pps *Mp ' ps +K pm *M′ pm , where K pps and K pm It is the weighting factor during calibration, and the default value is 1.

[0030] As a further improvement of the present invention, step S5 also includes a back-blowing anti-blocking process: closing the valve port of the mass flow meter, opening the solenoid valve S2, and allowing air to enter the sampling tube and the smoke detector for purging.

[0031] The detection and calculation principle of the present invention is as follows: first, the industrial computer collects the flow meter signal and calculates the current flue gas flow rate value; then the industrial control screen outputs the mass flow meter control signal through the fitting algorithm to control the opening flow of the mass flow meter. The smoke detection unit is a device similar to a jet pump, and there is a certain mathematical model between its sampling flow rate and the intake flow rate. In this way, the flow rate when the sampling flow rate of the smoke detection unit is converted to the smoke probe sampling tube is the current smoke flow rate, thereby realizing isokinetic sampling.

[0032] The smoke detection unit first detects the dust mass concentration value of the sampled smoke with a particle size less than or equal to 2.5um, and then outputs the dust mass concentration value of the particle size of 2.5um to 10um detected by the light scattering sensor. However, the light scattering sensor detection value is the dust mass concentration value after the sampled smoke and air are mixed. Therefore, conversion is required to obtain the dust mass concentration value in the sampled smoke.

[0033] 3. Beneficial effects

[0034] Compared with the prior art, the technical solution provided by the present invention has the following beneficial effects:

[0035] (1) The detection system of the present invention adopts a classification method for measuring the particle concentration. When the particle diameter is less than or equal to 2.5 microns, the value measured by the smoke detector is converted. When the particle diameter is greater than 2.5 microns but less than 10 microns, it is measured by a light scattering sensor. The two results are converted and added to obtain the total smoke concentration, thereby improving the detection accuracy.

[0036] (2) The present invention controls the intake flow rate of the mass flow meter through an algorithm, and can realize a constant speed and constant flow detection mode as required, thereby reducing the influence of flow rate on the detection result and helping to improve the accuracy of the detection.

[0037] (3) The present invention extracts smoke through the negative pressure jet principle, which is easy to measure. The back-blowing and anti-blocking structure design enables the equipment to be used for a long time. The structural design is reasonable, the principle is simple, and it is easy to promote and use. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1It is a schematic diagram of the system structure of the present invention.

[0039] Explanation of the numbers in the schematic diagram: 100, filter; 101, pressure detector; 200, mass flow meter; 201, smoke detector; 202, light scattering sensor; 203, flow meter; 300, flue; 301, reflux pipe; 302, sampling tube; 303, flow probe. DETAILED DESCRIPTION

[0040] In order to further understand the content of the present invention, the present invention is described in detail in conjunction with the accompanying drawings and embodiments.

[0041] Example

[0042] Combination Figure 1 A flue gas ultra-low emission dust detection system of this embodiment includes a velocity meter 203, a smoke detection unit and a mass flow meter 200 for detecting air flow, and an industrial computer for data processing.

[0043] The flow meter 203 is used to detect the gas flow rate in the flue. The detection part of the flow meter 203 is the flow probe 303. The flow probe 303 is composed of a Pitot tube and a sensor. The flow probe 303 is fixed on the flue and extended into the flue, and can be used to detect the gas flow rate in the flue.

[0044] The model is established so that the flow meter outputs an analog signal of 4-20 mA, and the corresponding digital value after AD conversion is 4000-20000, corresponding to the actual flue gas flow rate of 0-40 m / s. Suppose a certain instantaneous output analog current signal is i, the corresponding actual flue gas flow rate is v, and the sampled digital value is D 2 , then the conversion relationship between the digital quantity and the corresponding actual flue gas flow rate is: By measuring and calculating the flow rate, isokinetic or constant flow sampling can be achieved.

[0045] According to the analog signal i: 4~20mA corresponds to the actual flue gas flow rate v 烟 : 0~40m / s, the relationship between the two can be established.

[0046] v 烟 =2.5*i-10

[0047] Furthermore, the flow meter 203 is provided with a backflush interface, which is connected to the air inlet pipe, so that backflush can be performed after a period of use to prevent the sensor from being blocked.

[0048] The mass flow meter 200 is used to detect the air flow rate entering from the outside, and can control the flow rate by the valve opening. When used in conjunction with the velocity meter 203, it can achieve constant speed and constant flow sampling. The mass flow meter 200 can use the CS200 model MFC1. One end of the mass flow meter 200 is connected to the air supply pipeline of the external air, and the other end is connected to the air intake pipe at the bottom of the smoke detector 201, which is used to quantitatively transport the external air into the smoke detector 201.

[0049] The smoke detection unit includes a smoke detector 201 and a light scattering sensor 202. The smoke detector 201 is a Pegasor sensor. The pipeline connected to the mass flow meter 200 at the bottom leads to the jet pump, which can generate a jet phenomenon inside. The smoke inlet of the smoke detector 201 is connected to the sampling pipe 302 for obtaining the sampled gas.

[0050] The light scattering sensor 202 is connected to the exhaust end of the smoke detector 201 to detect the concentration of particulate matter in the exhausted mixed gas and transport the detected air to the flue 300 through the return pipe 301 .

[0051] The external air can be compressed air, and a solenoid valve S1 and a pressure detector 101 are set on the pipeline. The pressure in the pipeline is detected by the pressure detector 101, and the feedback signal controls the on and off of the solenoid valve S1. A filter 100 is also set at the air inlet end of the pipeline to filter the air to prevent the presence of particulate matter from affecting the detection result.

[0052] A three-way pipe is provided on the connection pipeline between the sampling tube 302 and the smoke detector 201, and the third interface of the three-way pipe is connected to the exhaust end of the electromagnetic valve S1, and an electromagnetic valve S2 is provided on the connection pipeline. When the valve port of the mass flow meter 200 is closed, external air can enter the sampling tube 302 through the three-way pipe interface for back-blowing, which is used to clean the dust accumulated on the pipeline and prevent it from being blocked. At the same time, the smoke detector 201 can be purged to clean the accumulated smoke.

[0053] In addition, as an implementation method, the solenoid valve S2 can be used to control the backflush pipeline of the flow meter 203 at the same time, that is, the two branches are connected to the intake pipeline using a three-way pipe, and the solenoid valve S2 is set on the intake pipeline. Then, the solenoid valve S2 can be used to control the on and off of the two branches at the same time.

[0054] The industrial computer in the present invention is an overall control system, which is electrically connected to the flow meter 203, the smoke detector 201, and the mass flow meter 200, and is used to receive signals and complete corresponding signal processing to generate corresponding instructions. The solenoid valve S1 and the pressure detector 101 are connected to the industrial computer and controlled by the industrial computer.

[0055] In order to achieve isokinetic sampling, the mass flow meter 200 and the velocity meter 203 need to cooperate with each other. The mass flow meter 200 is used to precisely control the gas and to open the flow rate according to the flue gas velocity signal measured by the velocity meter 203. In this embodiment, the flue gas velocity signal 4-20mA is used to calculate the opening flow rate. The flue gas velocity v 烟 (m / s) and mass flow meter opening flow Q 计 The algorithm model is:

[0056]

[0057] Flue gas flow rate signal 4~20mA corresponds to flue gas flow rate v 烟 (m / s) value is 0~40m / s, then v 烟 =2.5*i-10 where i is the flue gas flow rate output current signal.

[0058] The internal flow rate of the smoke detector 201 is set to L, that is, the rated flow rate is L. When the air flow rate Q 计 When the value increases, the corresponding sampling gas flow rate will decrease, thus changing the flow rate and achieving isokinetic sampling. For the range of flue gas flow rate signal 4-20mA, this example derives the relationship between it and the flow rate through design and calculation, and the isokinetic sampling result is more accurate.

[0059] The smoke detector 201 sets the flow rate to L, and the concentration value output by the smoke detector is M pps , the actual particle concentration in the sampled gas is M p ' ps , then the conversion relationship is:

[0060] Assume that the output concentration value of the light scattering sensor at a certain moment is M pm The actual conversion corresponds to the concentration value in the sampled gas: M p ' m , then the conversion relationship is:

[0061] Where: Q 吸 is the actual sampling gas flow rate sucked into the smoke detector; Q 计 Turn on flow for the mass flow meter.

[0062] The above calculation formula adds the two detection values ​​to obtain the total concentration value. In conventional detection devices, most of them are detected separately, resulting in inaccurate results. In the present invention, the particles in the flue gas are classified and detected. When the particle concentration diameter is less than or equal to 2.5 microns, the smoke detector PPS is used to detect and convert the actual concentration value. When the particle diameter is greater than 2.5 microns and less than 10 microns, it is measured by a light scattering sensor and converted to obtain the actual concentration value. Finally, the particle concentration values ​​of the two are added to obtain a more accurate flue gas concentration.

[0063] During use, since the device itself may have certain errors, the device can be calibrated before use. Use a fixed flow rate of simulated flue gas to circulate in the flue, and then use the detection system to detect and convert the corresponding values. The corresponding weighting coefficients are set for both concentration values, and the weighting coefficient value can be obtained through multiple measurements.

[0064] The total particle concentration value M in the sampled gas should be: M = K pps *M′ pps +K pm *M′ pm , where K pps and K pm is the weighting factor during calibration.

[0065] If the overall device and flow rate detection are both relatively accurate, the weighted coefficient is 1, which is directly the sum of the two. Therefore, the weighted coefficient can be set to a default value of 1.

[0066] The present invention also provides a method for detecting ultra-low emission fine dust in flue gas, which uses the above detection system for detection, and includes the following steps:

[0067] S1. The system is powered on for preheating. The smoke detector 201 is heated to a predetermined temperature and the heating temperature signal is transmitted to the industrial computer. When working, the smoke detector 201 is first used to preheat the internal system. The industrial computer continuously sends signals to read the heating temperature information. When the smoke detector 201 is heated to a predetermined temperature, the corresponding signal is output to the industrial computer.

[0068] S2. Start air supply. When the pressure detector 101 detects that the air pressure meets the set value, the solenoid valve S1 is opened, the mass flow meter 200 is turned on, and the opening flow of the mass flow meter 200 is controlled according to the flue gas flow rate detected by the flow meter 203.

[0069] During the heating period, the back-blowing function is turned off, and the solenoid valve S1 is controlled by the external pressure. When the set pressure value is met, the solenoid valve S1 is opened, otherwise it is closed. After the solenoid valve S1 is opened, the mass flow meter 200 is turned on.

[0070] Calculate the opening flow rate when the flue gas flow rate signal is 4-20mA, and the flue gas flow rate v 烟 (m / s) and mass flow meter opening flow Q 计 The relationship is:

[0071]

[0072] Where v 烟 =2.5*i-10; i is the flue gas flow rate output current signal.

[0073] S3, external gas enters the smoke detector 201 through the mass flow meter 200, and the internal jet of the smoke detector 201 forms a negative pressure, sucking the smoke in the flue 300 into the smoke detector 201 through the sampling tube 302.

[0074] S4. The metal needle in the smoke detector 201 releases high-voltage charge to form an electric field, ionizing the gas. The dust particles in the sampled gas attach to the ions and are charged. When the sampled gas flows out, the ammeter in the smoke detector obtains the concentration M of the sampled gas through the measured current. pps .

[0075] The smoke detector 201 is connected to the mass flow meter 200. When external compressed air or zero gas passes through the flow meter 200 and enters the smoke detector 201, the air is ionized and then mixed with the sampling gas. The dust particles in the sampling gas are charged. After the mixed gas passes through the external electric field, the remaining ions attached to the particles are absorbed by the electric field, leaving only the ion charges attached to the particles. When the gas flows out, the resulting current reflects the concentration of the sampling gas.

[0076] The actual particle concentration in the sampled gas is M p ' ps , then the conversion relationship is:

[0077] Assume that the output concentration value of the light scattering sensor at a certain moment is M pm The actual conversion corresponds to the concentration value in the sampled gas: M p ' m , then the conversion relationship is:

[0078] Q 吸 , Q 计 are the smoke detector sampling flow rate and mass flowmeter output flow rate, Q 吸 It can be calculated based on the current flue gas flow rate and the cross-sectional area of ​​the sampling port.

[0079] Q 计 With Q 吸 The sum is the same as the total flow of the smoke detector.

[0080] Then the total particle concentration M in the sampled gas can be calculated as: M = K pps *M′ pps +K pm *M′ pm , where K pps and K pm It is the weighting factor during calibration, and the default value is 1.

[0081] S5: The sampled gas is mixed with air and enters the light scattering sensor 202 to detect the concentration of remaining particles M. pm After detection, the gas is discharged into the flue 300 through the reflux pipe 301.

[0082] After being used for a period of time, a lot of dust may accumulate inside, and then back-blowing can be performed to prevent blockage. The back-blowing anti-blocking process is as follows: close the valve port of the mass flow meter 200, open the solenoid valve S2, and allow air to enter the sampling tube 302 and the smoke detector 201 for purging.

[0083] The present invention adopts a classification method for measuring the particle concentration. When the particle diameter is less than or equal to 2.5 microns, the value measured by the smoke detector is converted. When the particle diameter is greater than 2.5 microns but less than 10 microns, it is measured by a light scattering sensor. The two results are converted and added to obtain the total smoke concentration, thereby improving the detection accuracy.

[0084] The present invention and its embodiments are described schematically above, and the description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. Therefore, if a person skilled in the art is inspired by it and designs a structural method and an embodiment similar to the technical solution without creativity without departing from the purpose of the invention, they shall all fall within the protection scope of the present invention.

Claims

1. A flue gas ultra-low emission dust detection system, comprising a flow meter (203), a smoke detection unit and a mass flow meter (200) for detecting air flow, and an industrial computer for data processing, characterized in that: The detection part of the flow meter (203) is a flow probe (303), and the smoke detection unit comprises: The smoke detector (201) has an air inlet end connected to a sampling tube (302) and a mass flow meter (200), and releases high voltage electricity to ionize the gas, so that fine dust in the gas adheres to the ions, and the gas concentration is obtained through the current; A light scattering sensor (202) connected to the exhaust end of the smoke detector (201) for detecting the concentration of particulate matter in the exhausted mixed gas and conveying the detected air to the flue (300) through the return pipe (301); The air inlet end of the mass flow meter (200) is connected to a solenoid valve S1, and the air inlet end of the solenoid valve S1 is provided with a pressure detector (101), and the on-off of the solenoid valve S1 is controlled by a feedback signal of the pressure detector (101); The flow meter (203) is connected to the exhaust end of the electromagnetic valve S1 through a pipeline, and is used to backflush the flow probe (303) of the flow meter (203); A filter (100) is provided on the air intake pipeline of the solenoid valve S1.

2. The ultra-low emission dust detection system for flue gas according to claim 1, characterized in that: A three-way pipe is provided on the connecting pipeline between the sampling tube (302) and the smoke detector (201), the third interface of the three-way pipe is connected to the exhaust end of the solenoid valve S1, and a solenoid valve S2 is provided on the connecting pipeline.

3. The ultra-low emission dust detection system for flue gas according to claim 1, characterized in that: The mass flow meter (200) calculates the opening flow rate according to the flue gas flow rate signal 4-20 mA measured by the flow meter (203), and the flue gas flow rate v 烟 (m / s) and mass flow meter opening flow Q 计 The algorithm model is: Flue gas flow rate signal 4~20mA corresponds to flue gas flow rate v 烟 (m / s) value is 0~40m / s, then v 烟 =2.5*i-10 where i is the flue gas flow rate output current signal.

4. The ultra-low emission dust detection system for flue gas according to claim 1, characterized in that: The smoke detector (201) sets the flow rate to L, and the concentration value output by the smoke detector is M pps , the actual particle concentration is M′ pps , then the conversion relationship is: ; Assume that the output concentration value of the light scattering sensor at a certain moment is M pm The actual converted concentration value is M′ pm , then the conversion relationship is: ; Where: Q 吸 is the actual sampling gas flow rate sucked into the smoke detector; Q 计 Turn on flow for the mass flow meter.

5. A method for detecting ultra-low emission fine dust in flue gas, using the detection system according to any one of claims 1 to 4 for detection, characterized in that: The following steps are involved: S1, the system is powered on for preheating, the interior of the smoke detector (201) is heated to a predetermined temperature, and a heating temperature signal is transmitted to the industrial computer; S2, start air supply. When the pressure detector (101) detects that the air pressure meets the set value, the solenoid valve S1 opens. The mass flow meter (200) is turned on, and the opening flow rate of the mass flow meter (200) is controlled according to the flue gas flow rate detected by the flow meter (203); S3, the gas enters the smoke detector (201) through the mass flow meter (200), and the internal jet of the smoke detector (201) forms a negative pressure, so that the smoke in the flue (300) is sucked into the smoke detector (201) through the sampling tube (302); S4. The metal needle in the smoke detector (201) releases high voltage to ionize the gas. The dust particles in the sampled gas are attached to ions and charged. When the sampled gas flows out, the ammeter in the smoke detector obtains the concentration M of the sampled gas by measuring the current. pps ; S5. The sampled gas is mixed with air and enters the light scattering sensor (202) to detect the concentration of remaining particles M. pm After detection, the gas is discharged into the flue (300) through the reflux pipe (301).

6. A method for detecting ultra-low emission dust in flue gas according to claim 5, characterized in that: The total particle concentration value M in the sampled gas should be: M = K pps *M′ pps +K pm *M′ pm , where K pps and K pm It is the weighting factor during calibration, and the default value is 1.

7. A method for detecting ultra-low emission dust in flue gas according to claim 5, characterized in that: After step S5, a back-blowing anti-blocking process is also included: closing the valve port of the mass flow meter (200), opening the solenoid valve S2, and allowing air to enter the sampling tube (302) and the smoke detector (201) for purging.

Images