• Home
  • Sitemap
  • Contact us
Int J Fire Sci Eng Search

CLOSE


Int J Fire Sci Eng > Volume 37(4); 2023 > Article
Kim: Development of a Photoelectric Smoke Detector with a Built-in Function to Prevent False Alarms Owing to Humidity

Abstract

In this study, a photoelectric smoke detector with a built-in function was developed to prevent false alarms owing to humidity. Smoke-detector sensitivity characteristics were analyzed through room-scale false alarm experiments and by generating humidity conditions measured using a type approval sensitivity tester. The results showed that when humidity was generated, more light scattering occurred as the humidity flowed into the detector chamber along with the smoke, increasing the smoke concentration of the smoke detector. It was found that smoke concentrations increased at a relative humidity of 85 %RH or higher. The findings confirm that false alarms occur owing to high concentrations of humidity in the summer. False alarms that occur owing to high concentrations of water vapor in summer can be prevented by detecting fires using a humidity + smoke concentration fire-detection algorithm that has been quantified through experiments.

1. Introduction

Fire departments are tasked with responding not only to fires but also to a broad spectrum of safety events, upholding the public safety of citizens. The Korean National Fire Agency Statistical Yearbook[1] reveals that in 2021, emergency responses to safety included 200,310 cases of bee extermination and hive removal (37.1%), 95,192 cases of capturing and exterminating dangerous animals (17.7%), 92,866 cases of false alarms (17.2%), 76,428 cases of other safety measures (14.2%), 61,327 cases of confinement accident management (11.4%), 6,222 cases involving electrical and gas safety measures (1.2%), 3,555 cases of water supply and drainage support (0.7%), and 2,772 cases of hazardous icicle removal (0.5%). Notably, responses to false alarms were the third most common, leading to diminished firefighting capacity and unnecessary expenditure of time and resources. The frequency of such false alarms has shown a consistent annual increase, with counts of 20,445 in 2018, 25,020 in 2019, 38,119 in 2020, and 92,866 in 2021. This upward trend is attributed to the increased number of automatic fire detectors installed and the aging of these devices. Various factors, including humidity, dust, detector malfunctions, device aging, and poor adaptability of detectors, contribute to false alarms, which are more prevalent during the summer months when temperatures and humidity rise[2]. In Korea, researchers such as Choi Su-gil[3], Choi Yu-jeong[4], and Seo Byeong-geun[5] have examined the response characteristics of smoke detectors in relation to the conditions that lead to false alarms. However, there has been a lack of active research on developing smoke detectors for practical field use. Reviewing past studies on smoke detector innovation, such as those by Park Seung-hwan[6] and Jeong Jong-jin[7], reveals a focus on creating detectors with multiple sensors and fire detection systems employing Internet of Things (IoT) technology. Such approaches, while technologically advanced, often lead to higher costs per unit for fire detectors. Consequently, many of these devices have not transitioned to practical application, remaining in their project inception stages and failing to be commercialized. Furthermore, some detectors have not even received type approval as they do not meet nationally certified standards, resulting in their research and development efforts being discontinued.
In this study, sensitivity characteristics were investigated through humidity experiments at a room-scale nonfire test site under humidity conditions within a type approval sensitivity tester to identify the circumstances in which false alarms are triggered owing to humidity. The aim of this study was to develop a smoke detector equipped with a humidity sensor that can prevent false alarms by having the detector automatically adjust its fire detection method based on increases in its humidity concentrations based on experimental results. In addition, this study aimed to design and manufacture a photoelectric smoke detector with a built-in humidity sensor without increasing the unit cost using a low-cost humidity sensor such that the detector can be commercialized. Furthermore, this study aimed to resolve the problem of false alarms triggered by humidity, which is an environmental factor that causes smoke-detector false alarms.

2. Experiments

2.1 Humidity experiments at a room-scale nonfire test site

Smoke detectors are installed in areas such as hallways, corridors, stairways, and living rooms within apartment buildings. In this study, a room-scale test site, simulating a one-room residence with an additional multipurpose room, was constructed to perform experiments that replicated humidity conditions typically generated by a coffeepot, in a real one-room residence. Figure 1 illustrates the test site and its specifications, replicating spaces commonly found in apartment buildings. The test space comprised a room dedicated to humidity experiments, measuring 2.5 m by 4 m by 2.5 m, and an adjoining living room designed to allow natural airflow, with dimensions of 7 m by 1.5 m by 2.5 m. A standard kitchen sink was installed on the room's wall at a height of 0.9 m from the floor, from which the coffeepot, the source of humidity, was operated.
An Optical Density Meter (ODM) capable of recording the optical density of the space was installed on the ceiling, 3 m away from the wall, and Analog Smoke Detectors (ASDs) were placed 2.7 m from the wall. These distances were determined based on the point where rising air currents, generated by thermal expansion, moved along the ceiling surface and reached a stable state. Additionally, the smoke detectors were installed 1 m from the exit, adhering to the standard requirement for smoke detectors to be positioned close to exits in rooms with low ceilings/roofs or in narrow spaces.
The experiments were conducted at an indoor temperature of 25 ± 3 ℃ and a relative humidity of 50 ± 10 %RH. The coffeepot, containing 8 L of water, was turned on, and the experiments were conducted over a duration of 30 min to analyze the response characteristics of the smoke detectors to the water vapor generated as the humidity increased.

2.2 Analysis of sensitivity characteristics via humidity generated by type approval sensitivity tester

Differences in smoke concentration characteristics owing to the humidity generated within a type approval sensitivity tester were analyzed. Figure 2 shows the smoke detector sensitivity tester used in smoke-detector operation and nonoperation tests according to Article 19 of the National Fire Agency Notice, "Technical Standards for Detector-Type Approval and Product Inspection."[8] For domestic smoke detectors, an air current is generated within the sensitivity tester at wind speeds of 0.2-0.4 m/s, and filter paper no. 2, which is a single smoke source, is used as the smoke sample to generate smoke within the tester. When the smoke concentration measured using the ODM of the sensitivity tester reaches the concentration level of the test standard, smoke is allowed to flow into the smoke detector, and operation and nonoperation tests are performed to determine whether the detector is suitable. Accordingly, the following experiments were performed to analyze the increase in smoke scattering that occurred during the operation and nonoperation tests owing to the generated humidity within the sensitivity tester, which has a fixed wind speed. Figure 3 shows a photograph of the type approval sensitivity tester used in the experiments. The internal humidity of the tester was controlled by installing a humidifying hose to generate humidity within the tester. Next, smoke detector operation and nonoperation tests were performed according to technical standards for the detector-type approval and product inspection. Quantified masses of combustible material (filter paper no. 2) were observed. The operation and nonoperation concentrations were 15 and 5 %/m, respectively, at a normal relative humidity of 50 %RH within the smoke-detector sensitivity tester. Next, operation and nonoperation tests were performed with identical masses of material at a relative humidity of 80 %RH to analyze the smoke concentration characteristics of the smoke detector. In addition, the differences in characteristics at relative humidity values of 50 %RH and 80 %RH for the two types of nominal operation concentration of 10 %/m of the smoke detectors were analyzed.

3. Experimental Results and Observations

3.1 Results of humidity experiments at the room-scale nonfire test site

Figure 4 shows the data for the humidity generated by the coffeepot, which can occur in everyday life. At 400 s, just before smoke obscuration first occurred in the ODM, the relative humidity of the sink was 86.2 %RH, which exceeded 80 %RH. At 85 %RH, the water vapor from the coffeepot formed a mist that was visible to the naked eye. This phenomenon occurred because the saturation state for the corresponding temperature was reached when the hot water vapor invisibly filled the air and reached the saturation water vapor level. The vapor encountered the cold air and transformed into moisture droplets, and the phase-transformed mist moved along the ceiling surface and gradually dispersed in the air.
The ODM concentration exhibited a rapid increase between 400 and 410 s, as illustrated in Figure 4(a). This increase corresponded to a relative humidity range of 76-80 %RH, a point where mist gradually formed as the water vapor in the air approached saturation, leading to a subsequent sharp increase. Concurrently, as depicted in Figure 4(a), a sustained rise in the smoke concentration, as detected by the analog smoke detector, commenced at 1,000 s from the experiment's start, with the relative humidity being maintained at 88 %RH during this phase. For the ODM, when the smoke concentration reached 26 %/m, numerous water droplets were already dispersed in the air, causing natural airflow past the light-receiving and light-emitting units toward the living room. Furthermore, at 950 s, which coincided with the first peak of the smoke detector, a relative humidity of 85 %RH was recorded. This indicates that the relative humidity influencing the interior of the smoke detector chamber via natural air currents started at 85 %RH, with substantial amounts of visible steam beginning to manifest at this humidity level. As presented in Figure 4(b), at 1,200 s after beginning the experiment, the RHD (humidity meter - detector area) in the detector area was recorded at 89.4 %RH, while the RHD (humidity meter - sink) measured 94.2 %RH, with ODM and ASD values at 20 %/m and 3 %/m, respectively. Additionally, from 1,200 to 1,800 s, a visible vapor mist enveloped the room, and the ODM values oscillated within the range of 15-20 %/m, whereas the smoke concentration varied between 3.4-4 %/m.

3.2 Analysis of sensitivity characteristics by generating humidity using type approval sensitivity tester

Figure 5(a) shows the results for the insertion of identical masses of filter paper at obscuration rates that the sensitivity tester ODM measured as 5 %/m, which is the nonoperation test concentration level of the two types of smoke detectors. Figure 5(a-1) and Figure 5(a-2) show results for the cases in which the relative humidity values were 50 and 80 %RH, respectively. Masses of the filter paper were inserted at an obscuration rate of 5 %/m by the ODM under normal conditions (50 %RH). In the case of the ASD, the initial smoke concentration was 4.5 %/m, and as time passed and the smoke circulated, the concentration decreased to 3.5 %/m until the end of the experiment. However, when a mass was inserted at an obscuration rate of 5 %/m using the ODM after maintaining a relative humidity of 80 %RH, the smoke concentration increased to 15 %/m, which is the operation concentration of the two types of smoke detectors, even though the test level was a nonoperation level.
Figure 5(b) shows the results for the insertion of identical masses of filter paper at an obscuration rate of 10 %/m measured using the sensitivity tester ODM, which is the nominal operation concentration of the two types of smoke detectors, under normal relative humidity conditions (50 %RH) and when the relative humidity was generated (80 %RH). A mass of the filter paper at an obscuration rate of 10 %/m measured using the ODM under normal relative humidity conditions was inserted. In the case of the ASD, the smoke concentration was 9.2 %/m as the smoke concentration initially increased to 9.2 %/m. Subsequently, the smoke concentration gradually decreased until the end of the experiment, as the smoke concentration decreased via circulation because of the wind speed. However, when a high concentration of relative humidity was induced, the ASD yielded the maximum smoke concentration measurement value of 20 %/m, even though a combustible that generated the same amount of smoke was inserted.
Figure 5(c) shows the results for the insertion of identical masses of filter paper at an obscuration rate of 15 %/m determined using the sensitivity tester ODM, which is the operation concentration of the two types of smoke detectors under normal relative humidity conditions (50 %RH) and when the relative humidity was generated (80 %RH). The filter paper no. 2, which had an obscuration rate of 15 %/m measured using the ODM under normal conditions, was inserted, and the ASD initially produced a smoke concentration of 13.9 %/m. Next, a value of 15 %/m was maintained for 5 s, and the smoke concentration continuously decreased throughout the experiment period. However, when a high concentration of relative humidity was generated in the 15 %/m experiments as well, the ASD yielded the maximum smoke concentration measurement value of 20 %/m, even though a combustible that generated the same amount of smoke was inserted.
Based on the experimental results, the smoke concentration measurements were higher at high concentrations of relative humidity than at the normal relative humidity in the 5, 10, and 15 %/m experiments. The smoke sensitivity at the normal concentration of relative humidity showed that the smoke concentration generated by the ODM was reached within 5 s of inserting the combustible. However, when high concentrations of relative humidity were induced, the time until reaching the smoke concentration determined using the ODM was delayed by approximately 20 s. The smoke/aerosol could not reach the upper level of the test device and became diluted because of the high water vapor concentrations, causing a delay until it flowed into the smoke detector chamber. However, the results of the sustained experiments showed that light scattering increased as the water vapor flowed into the smoke detector chamber along with the smoke/aerosol, and high smoke concentrations were maintained. In addition, under normal conditions, the smoke inside the test chamber rapidly diluted and gradually decreased. However, under high concentrations of relative humidity in the 10 and 15 %/m experiments, water droplets flowed into the smoke detector chamber along with the smoke for an extended period, and continuous maximum values were maintained.

4. Design of a Photoelectric Smoke Detector with a Built-in Humidity Sensor

The experiments described in Section 3 demonstrate that high relative humidity concentrations increase the smoke concentrations measured using smoke detectors. Therefore, a smoke detector with a built-in humidity sensor was designed. In the case of the relative humidity, the higher the temperature, the greater the amount of saturated water vapor; that is, the higher the temperature, the greater the amount of water vapor dispersed in the air. Similarly, the lower the temperature, the smaller the amount of water vapor in the air. Many false alarms occur in the summer because the amount of water vapor in the air is greater than in other seasons. The water vapor flows into the detection chambers of photoelectric smoke detectors that use the light-scattering method, which causes light scattering and transmits the resulting fire detection signals to the receiver, generating false alarms. Hence, this study aimed to install a humidity sensor in the detection chamber to analyze the relative humidity in the detector chamber and correct abnormal signals that occur because of increases in relative humidity. The humidity sensors used to develop this approach can be broadly classified as direct and indirect types based on the method by which they measure humidity. The direct type includes dew-point, absorption-based volumetric, and evaporation-based wet and dry bulb thermometers. The indirect type includes electric, capacitive, near-infrared, and diffusion/discharge hygrometers. Because automatic fire detection equipment uses a rated DC voltage of 24 V, the voltage drop consumed by the smoke detector must be minimized. Therefore, electric hygrometers that use low voltage are appropriate for use, and electrical-resistance-type humidity sensors were used from among the indirect measurement methods. The humidity composition within the smoke detector chamber should be clarified to determine the ultimate conditions in which false alarms occur owing to humidity in the smoke detector chamber in advance. The humidity sensor should be a miniaturized type so that it can be installed within the smoke chamber. In this study, the HR202L humidity sensor was used (Figure 6), which is a contact resistance-type sensor typically used for miniaturization. In addition, the humidity sensor has a feature that determines the relative humidity by measuring the temperature and absolute humidity.
Figure 7 shows photographs and a schematic of the photoelectric smoke detector with a built-in humidity sensor, in which the humidity sensor was attached to form the device unit by selecting a location in a space that did not interfere with the infrared sensor of the smoke detector unit. A basic scattered light smoke detection circuit was built via a design of a scattered light smoke detection circuit that used infrared and a photocell to allow for detection. A humidity detection circuit was built by installing an electrical-resistance-type humidity sensor that could measure variations in surface conductivity. The humidity and smoke detector used an MCU to control all IO and was configured such that it could detect fires. It received DC 24 V power and used an internal constant-voltage circuit to generate 3.3 V to drive the IR sensor, which is the light-receiving element. The IR sensor repeatedly turned on and off according to the control signal from the MCU and periodically functioned as a scattered light source. In addition, the frequency of the humidity sensor was applied by the oscillator circuit within the MCU to continuously operate it and record measurements. Fires are assessed based on a comparison operation of the photocell.
Figure 8 shows the fire detection algorithm of the developed humidity + smoke detector. The results of the experiments on false alarms owing to humidity showed that the smoke concentration of the smoke detector was influenced by relative humidities of 85 %RH and higher when the mist appeared. Accordingly, programming was conducted to set the smoke concentration off-set after installation, analyze the initial smoke concentration, and measure the internal temperature such that the smoke detector could determine its settings. After completing the initial settings, the detector was set as a primary condition to detect a fire when a smoke concentration of 15 %/m was maintained for 10 s at a relative humidity of 85 %RH or lower. When a relative humidity of 86 %RH was measured using the humidity sensor in the smoke detector chamber, the fire signal in the photocell was maintained in standby mode, and the humidity sensor in the chamber continuously measured the temperature and relative humidity to prepare for a fire. When the temperature increased by 30 °C from the temperature at the moment when the relative humidity initially exceeded 85 %RH, the detector determined that a fire had occurred, and fire detection occurred via the differential heat detector function. These settings are based on the operation test standards for differential spot-type detectors, which are among the technical standards for type approval product inspections.
In addition, because the relative humidity in the air was reduced by high-temperature heat and flame during fires, the initial smoke-based detection function can be maintained. The algorithm was applied by programming the differential heat detector to operate in parallel such that fires could be detected when they occurred at high concentrations of relative humidity in certain cases.
Additional humidity experiments were conducted in an ISO 9705 standard room using the developed specimen. The experiments were performed at the test site shown in Figure 9 for 600 s using a humidifier and coffeepot to induce a relative humidity of 86 %RH or higher and a wind speed of 0.8 m/s in accordance with previous experiments[3]. The experimental results, as shown in Figure 10, that the photoelectric smoke detectors produced by different manufacturers that were installed on circuits 1, 2, and 3 transmitted fire signals owing to high humidity concentrations. However, the developed detectors with built-in humidity sensors that were installed on circuits 4 and 5 did not transmit fire signals owing to humidity. Thus, false alarms owing to high concentrations of water vapor in summer can be prevented by detecting the fire using a fire detection algorithm that is based on experimentally quantified humidity.

5. Conclusions

In this study, a photoelectric smoke detector with a built-in function was developed to prevent false alarms owing to humidity. Experiments were performed to determine the smoke concentration characteristics of smoke detectors for different relative humidity values. The following conclusions were drawn:
1. Results from experiments conducted at a room-scale false alarm test site, where humidity was generated using a coffeepot, indicated that the smoke concentration measured by an analog smoke detector began to increase at 950 s from the start of the experiment. At this point, the relative humidity was maintained at 85 %RH, a level that can influence the interior of a smoke detector's chamber via natural air currents.
2. Experiments designed to assess the sensitivity characteristics of smoke detectors under varying humidity conditions were performed using a type approval sensitivity tester. These experiments demonstrated that smoke concentration measurements were higher at elevated relative humidity levels compared to normal humidity conditions. The presence of water vapor mixed with smoke/aerosol led to increased light scattering within the smoke detector chamber, resulting in higher smoke concentration readings.
3. A smoke detector equipped with a built-in humidity sensor was developed, and further humidity experiments were conducted in an ISO 9705 standard room using prototypes implementing a specific algorithm. The findings revealed that at a relative humidity of 86 %RH or higher and a wind speed of 0.8 m/s, standard photoelectric smoke detectors erroneously transmitted fire signals due to the high humidity. However, smoke detectors with integrated humidity sensors did not transmit false fire signals under these high humidity conditions.
This study confirmed that actual false alarms are often caused by high humidity concentrations in the summer. Consequently, it is assessed that false alarms due to elevated water vapor levels during summer can be mitigated by using a smoke detector that incorporates a humidity + smoke concentration fire detection algorithm, which has been quantified through experimentation. However, to address some of the numerous false alarms occurring in real-world scenarios, it is considered that this detector can be deployed as a highly reliable development model only after conducting verification experiments across a broader range of false alarm scenarios.

Notes

Conflicts of Interest

The authors declare no conflict of interest.

Acknowledgments

This research was supported by the National Fire Agency of the Republic of Korea (20016764).

Figure 1.
Testbed photographs and dimensions.
KIFSE-91a22c7cf1.jpg
Figure 2.
Type approval equipment for smoke detectors.
KIFSE-91a22c7cf2.jpg
Figure 3.
Test for measuring smoke detector concentration according to humidity composition.
KIFSE-91a22c7cf3.jpg
Figure 4.
Testbed data analysis (coffee pot).
KIFSE-91a22c7cf4.jpg
Figure 5.
Optical density meter readings at smoke concentrations of 5, 10, and 15 %/m.
KIFSE-91a22c7cf5.jpg
Figure 6.
Size and specification of a humidity sensor.
KIFSE-91a22c7cf6.jpg
Figure 7.
Development diagram.
KIFSE-91a22c7cf7.jpg
Figure 8.
Fire detection algorithm.
KIFSE-91a22c7cf8.jpg
Figure 9.
Photographs and specifications of experimental space.
KIFSE-91a22c7cf9.jpg
Figure 10.
Experiment photographs.
KIFSE-91a22c7cf10.jpg

References

1. National Fire Agency 119, "NATIONAL FIRE AGENCY STATISTICAL YEARBOOK" (2022).

2. FPN 119 PLUS MAGAZINE, "Rep. Young-hoon Oh, 99% Malfunction Rate of Automated Fire Reporting Equipment" (2021). https://www.fpn119.co.kr/164763.

3. S. G. Choi, Y. J. Choi and S. K. Kim, “Experimental Research on Operation Characteristics of Smoke Detection Equipment with Respect to Varying Humidity Environments”, Fire Science and Engineering, Vol. 36, No. 1, pp. 23-34 (2022), https://doi.org/10.7731/KIFSE.2ec642f6.
crossref
4. Y. J. Choi, S. G. Choi, Y. J. Nam, J. H. Yu, E. H. Hwang, S. E. Lee and S. K. Kim, “A Study on the Response Characteristics of Fire Detector and Indoor Air Quality Measurement Factor According to UL 268 Cooking Nuisance Test”, Fire Science and Engineering, Vol. 35, No. 1, pp. 67-77 (2021), https://doi.org/10.7731/KIFSE.776dfdaa.
crossref
5. B. K. Seo and S. G. Nam, “Study of the Improvement of False Fire Alarms in Analog Photoelectric Type Smoke Detectors”, Fire Science and Engineering, Vol. 30, No. 5, pp. 108-115 (2016), https://doi.org/10.7731/KIFSE.2016.30.5.108.
crossref
6. S. H. Park, D. H. Kim and S. C. Kim, “Analysis of Unwanted Fire Alarm Signal Pattern of Smoke / Temperature Detector in the IoT-Based Fire Detection System”, Journal of the Korean Society of Safety, Vol. 37, No. 2, pp. 69-75 (2022), https://doi.org/10.14346/JKOSOS.2022.37.2.69.
crossref
7. J. J. Jung and C. T. Kim, “Development of a Fire Smoke Detector that Satisfies the UL 268 (7th Edition) Fire Sensitivity Tests”, Journal of the Korean Society Hazard Mitigation, Vol. 22, No. 2, pp. 101-108 (2022), https://doi.org/10.9798/KOSHAM.2022.22.2.101.
crossref
8. National Fire Agency 119 Notice (No. 2023-51), "Technical Standards for Formal Approval and Product Inspection of Fire Detector" (2023).



ABOUT
BROWSE ARTICLES
EDITORIAL POLICY
AUTHOR INFORMATION
Editorial Office
Room 906, The Korea Science Technology Center The first building, 22, Teheran-ro 7 Gil, Gangnam-gu, Seoul, Republic of Korea
Tel: +82-2-555-2450/+82-2-555-2452    Fax: +82-2-3453-5855    E-mail: kifse@hanmail.net                

Copyright © 2024 by Korean Institute of Fire Science and Engineering.

Developed in M2PI

Close layer
prev next