Non-dispersed infrared (NDIR) Carbon Dioxide (CO2) Sensor Working Principle

Non-dispersed infrared (NDIR) technology is a traditional way to measure gas concentration and can detect gases in various compound types. A kind of gas (e. g., CO₂, NO₂, CO, CH₄, Hydrocarbon gas, etc.) in the sensor's gas measurement chamber will absorb the energy of infrared rays at the corresponding wavelengths, as shown in Figure 1. Usually, the higher the gas concentration, the more energy is absorbed. Therefore, we can determine the concentration of the target gas by measuring the absorption rate of the infrared beam at the corresponding wavelength. Because the infrared beam emitted from the light source and passing through the gas measurement chamber is composite and non-dispersed, this measurement method is called the NDIR technology. In this article, we take CO2 gas detection as an example to explain the working principle of NDIR gas sensors.

Figure 1. IR absorption for different gases [1]

It is not difficult to see from figure 1 that each gas corresponds to its absorption waveband. If only a specific gas is measured, an optical filter (as shown in figure 2) is usually placed before the thermopile detector to filter out the light components, except for the component that the target gas can absorb. The absorption peak of the CO₂ gas exists around the wavelength of 4.26 μm, which scarcely overlaps with the absorption waveband of the other gases. Therefore, we often use a narrow-band filter with a center frequency of 4.26 μm for the CO₂ NDIR sensor.

Figure 2. NDIR sketch of optical structure

Behind the filter, a thermopile (or pyroelectric) detector is usually used for measuring the narrow-waveband infrared intensity, according to Bill Lambert's Law [2]:

Where I₀ represents the light power intensity emitted to the target gas (unit W/m²), I represents the remaining light power intensity after absorption (unit W/m²), k represents the absorption coefficient of the target gas at its absorption waveband, L represents the equivalent optical path length between the infrared light source and the detector (unit cm), and C represents the target gas concentration (unit mol/dm³). This equation shows that the infrared light intensity in the measured gas decreases exponentially; that is, the higher the measured gas concentration, the absorption coefficient, and the equivalent light path length, the greater the attenuation of the infrared power intensity.

Figure 3. Working principle of thermocouple [3]

Thermopiles are composed of a large number of thermocouples in series. There are two metal materials on the thermocouple. According to the Seebeck Effect, if there is a temperature difference around the two different junctions of two conductors, there is a potential difference between the relatively hot and cold junctions (as shown in figure 3). In NDIR applications, the infrared light passing through the filter is radiated to a set of thermocouples heated to produce a weak thermal potential relative to another reference set, which is positively proportional to the infrared absorption light intensity. A typical MEMS thermopile detector is shown in figure 4.

Figure 4. Sectional view of MEMS thermopilesensor [3]

The thermopile outputs a weak voltage signal, which the signal conditioning circuit needs to amplify and filter. Thermopile generates sensing signals by absorbing thermal radiation, so the environmental temperature fluctuation will inevitably affect the signals. Therefore, the detector integrates a thermistor for temperature reference and compensation. For better performance, an analog-to-digital converter (ADC) digitizes the analog signals, and the firmware embedded in the microprocessor performs digital signal processing (DSP), which includes digital filtering, compensation, etc. At last, the digital communication interface will transfer the concentration readings to the client. The schematic block diagram is shown in figure 5.

Figure 5. System design of NDIR sensor

This paper briefly describes the working principle of the NDIR CO2 sensor. Please refer to other documents from Unitense website to inquire more about field applications. The copyright of this paper belongs to Unitense Innovation. Reproduce is prohibited.

References

[1] D. Popa, et al., “Towards Integrated Mid-Infrared Gas Sensors”, MDPI: Sensors, May, 2019.

[2] L. B. Mendes, "NDIR Gas Sensor for Spatial Monitoring of Carbon Dioxide Concentrations in Naturally Ventilated Livestock Buildings", MDPI: Sensors, May, 2015.

[3] DH Xu, et al., "MEMS-based thermoelectric infrared sensors: A review", Frontiers of Mechanical Engineering, Jun. 2017.