Author: Christoph K ä mmerer, field application engineer, ADI company

Modern smoke alarm – improving detection performance and safety

Today’s buildings are equipped with many different sensors to make daily life more convenient and provide safety protection. In addition to environmental sensors and smart home applications (such as power and heating regulation), safety related sensors also play an important role in safety protection. This includes smoke alarms. Smoke alarms are indispensable and required by law, but many smoke alarms on the market are not suitable for use in kitchens or bathrooms because cooking smoke or other steam will increase the risk of false alarm. We cannot underestimate false positives, which will induce users to turn off smoke alarms and incur high costs due to unnecessary fire deployment.

However, the lack of smoke alarms in bathrooms and kitchens is a serious problem because it increases the risk of fire, especially in kitchens. The kitchen in modern apartments is usually integrated with the living room, so it faces greater risks. Modern buildings use a large number of synthetic building materials, which will spread rapidly in case of fire. Therefore, in order to accurately implement fire detection, smoke alarm network must be accurately deployed.

Many global standards require new tests to detect different types of smoke to meet these new requirements. The standards in different regions are slightly different: en standard is adopted in Europe, UL standard is adopted in North America, and ISO standard is adopted internationally. In the latest version (UL 268: 7th edition and UL 217: 8th Edition) released in June 2021, UL added a new test, namely hamburger smoke interference alarm test. In this test, it must be possible to distinguish between the smoke concentration produced by hamburger patties and that produced by flammable polyurethane. This test helps to reduce the false positive rate of the kitchen. This paper will introduce this test and discuss how to design a new detector to pass this new test.

Hamburger smoke interference alarm test focusing on UL

This hamburger smoke interference test is designed to replicate real cooking smoke. The concept of hamburger smoke interference alarm test is very simple, but even modern smoke alarm faces a challenge: hamburger patties need to be fried for a certain time. During this process, it will be checked to confirm whether the smoke alarm is triggered by rising smoke (starting to rise within a certain time limit). This is also a standardized test, so all smoke alarms can be tested under exactly the same conditions. Take the measured value of light reduction rate as the reference. In this test, a light source with a beam diameter of 10 to 15 cm is placed at a distance of about 2 meters. A steam lamp with a wavelength of 589 nm was used as the light source. Smoke between the steam lamp and the detector will block the light. Figure 1 shows the principle and schematic diagram of the reference measurement setting.

Figure 1. Schematic diagram of reference system based on UL standard.

The degree to which the light beam is obscured by smoke is compared with the reference signal in a smoke-free environment. The smoke density and smoke concentration can be obtained according to the dimming rate. For the same particles, the higher the dimming rate, the higher the concentration. Of course, the dimming rate varies not only with the concentration, but also with the particle type. According to the scattering cross section, there are obvious differences among different particle types.

Dimming time is also closely related to alarm triggering. Therefore, according to this standard, an alarm will be triggered after a certain time limit is reached in the reference system or the masking limit is reached. Therefore, the hamburger interference alarm test stipulates that during the frying of hamburger patty, the alarm shall not be issued until the dimming rate reaches more than 1.5% / ft.

In the second phase of the test, polyurethane is ignited to simulate real objects such as armchairs. The smoke alarm must distinguish the difference between the two and trigger the alarm when the dimming rate reaches 5% / ft.

This test is very challenging because it is difficult to distinguish between real fire smoke and cooking smoke. However, this test is only one of many tests defined in UL 217 and UL 268. Multiple identical smoke alarms must also be used to pass this test to eliminate random results and ensure that the detector has a wide range of high mass density.

How did the smoke alarm pass this test

Most modern smoke alarms use the photoelectric working principle. In the hamburger smoke interference test, the light beam is reflected by particles after being emitted. Scattering depends on particle type, particle concentration and scattering angle. The smoke alarm determines whether to trigger the alarm according to the scattering signal.

In order to pass the hamburger smoke interference alarm test, the detector must have a high signal-to-noise ratio to distinguish hamburger smoke from other types of smoke.

Analog Devices’ ADPD188BI integrated opTIcal sensor module equips smoke alarm manufacturers with the technology that can pass this difficult test. Figure 2 demonstrates the working principle of the ADPD188BI.

ADI’s adpd188bi integrated optical sensor module helps smoke alarm manufacturers’ products pass this rigorous test. Figure 2 shows how adpd188bi works.

Figure 2. Working principle of adpd188bi.

The shell of the new smoke detection integrated module contains two transmitter LEDs – the blue LED has a wavelength of 470 nm and the infrared LED has a wavelength of 850 nm. Both transmitters are located in the chamber on the left. On the right side of the housing is a photodiode and an analog front end. The LED emits a light beam, and the smoke particles reflect the light beam back to the photodiode. In addition, an LED driver is integrated, which is switched by an internal time slot. Through these time slots, the user can adjust the timing of the whole front end without constantly rewriting the registers.

The analog front-end includes a current voltage converter and an ambient light analog filter. The latter consists of a band-pass filter for detecting constant ambient light and an integrator for detecting variable ambient light (such as light emitted by fluorescent lamps). After that, the integrated analog-to-digital converter converts the voltage into a digital signal.

Adpd188bi smoke sensor module has high integration density and has many advantages. Since only a few external components are required, the whole system is easier to calibrate. Through two-color light wavelength detection, it not only supports the measurement of a single wavelength, but also supports the analysis of ratio composition, so as to further reduce the false alarm rate. In addition, this module has smaller volume and lower power consumption than the traditional detector. The working power consumption of infrared LED is ~ 5 μ W/Hz。 By fully integrating LEDs and photodiodes into the analog front end, smoke alarm manufacturers can provide an overall module solution.

The high integration of adpd188bi module is related to the “success or failure” of hamburger smoke interference test. Under fixed current, the luminous intensity of different LEDs usually varies greatly, so smoke alarm calibration was carried out by smoke alarm manufacturers in the past. Calibrating the slope and offset between LED luminous intensity and current can ensure that all LEDs maintain the same performance. Because the LED and the whole signal path are integrated into adpd188bi, ADI will pre calibrate the sensor module, which can reduce the difference between devices. Since smoke alarm manufacturers can use pre calibrated modules, the system design is simplified.

The calibration method used by Adi company is to directly calibrate the slope and offset of LED. Therefore, the adpd188bi is placed under the reflector, and the reflected light is measured by the integrated photodiode. The slope and offset can be determined for each adpd188bi, and the calibration coefficient is stored in the nonvolatile memory (i.e. eFuse register) of the chip. By reading these coefficients, the difference between chips can be reduced as much as possible. This means that the alarm threshold can be set more accurately in the algorithm, so as to reduce the false alarm rate and finally pass the UL test.

Figure 3 shows the test results of adpd188bi in a standardized UL test environment – once Hamburg smoke (left) and once flammable polyurethane smoke (right).

Figure 3. Apdpd188bi UL test results: Hamburg smoke and flammable polyurethane smoke were detected in a smoke-free room.

The figure shows the test results over time (x-axis) in two cases, and the y-axis on the left represents the adpd188bi signal. Expressed as power transmission ratio, it describes the relationship between the working power of LED and the received power of photodiode. The power transmission ratio formula is as follows:

This parameter allows different modules to compare with each other. The y-axis on the right represents the dimming rate in% / ft. The green curve is the UL reference beam, and the blue and purple curves are the blue and infrared signals emitted by adpd188bi respectively.

As shown in Figure 3, the two signal curves sent by adpd188bi are very different in the two scenarios, indicating that the sensor can clearly distinguish the two types of smoke. One difference is that the signal changes with time. It can be seen that polyurethane reaches the alarm threshold after 220 seconds, and the time length is 1 / 4 of Hamburg smoke reaching the alarm threshold (after more than 1000 seconds). A critical level can be detected after 4 minutes of polyurethane combustion.

As shown in the figure, the change of particle concentration can also be clearly distinguished and recorded by using the high signal-to-noise ratio of the sensor. For example, in the test of flammable polyurethane, the slope increases suddenly. See the red mark in Figure 3.

In addition, adpd188bi measures two wavelengths, and the ratio of the two is another parameter, which can be used to calibrate the reliable algorithm used to detect hamburger smoke, so as to judge whether it has passed the hamburger smoke test.


Why is the new integrated optical smoke detection module an important turning point

The newly launched hamburger smoke interference test is difficult to pass because the smoke particles produced when frying hamburger patties are not different from normal smoke particles. Therefore, the smoke sensor needs to have a high signal-to-noise ratio to distinguish the smoke generated by hamburger patty from other types of smoke. In this process, the low difference between sensors plays a decisive role. Only when the measurement is completed reliably and passed the test; In order to reduce the false alarm of smoke sensor in the final application. The new integrated optical smoke detection module adpd188bi of ADI company is a highly sensitive integrated sensor module. It not only has high signal-to-noise ratio and supports two-color detection, but also can fully reduce the differences between devices, thus simplifying the design and algorithm development.

Introduction to the author

Christoph K ä mmerer has been working for ADI in Germany since February 2015. He graduated from Erlangen Nuremberg University with a master’s degree in physics in 2014. Then he worked as a process development intern at ADI company in Limerick. After the completion of the trainee program in December 2016, Christoph officially became a field application engineer of ADI company, specializing in emerging application fields.

Leave a Reply

Your email address will not be published. Required fields are marked *