The cause of cross-sensitivities in NDIR gas analyzers is an overlapping of the absorption bands of the affected gases. By skillful selection of absorption bands of the involved gases these cross-sensitivities can at best be avoided, but often at least significantly reduced. By internal compensation mechanisms, remaining cross-sensitivities are also usually largely eliminated.
What this means is shown in the following example of the simultaneous measurement of methane (CH4), ethane (C2H6) and propane (C3H8). Figure 1 shows that the absorption bands of the three gases overlap strongly in the range around 3.4µm wavelength. Optical measurement channels in this range cause strong cross-sensitivities between all gases, which can only be reduced unsatisfactorily by calibration and mathematical compensation algorithms.
In the range between 7.5µm and 8µm, methane shows a further strong absorption band in which only insignificant absorptions of the other two gases are present. However, a relevant overlap with the absorption bands of water exists. By compensating this cross-sensitivity, a good measurement of methane is possible. By including this measurement, ethane (C2H6) and propane (C3H8) in the range around 3.4µm can now also be detected well. Thus, all three gases can be measured with high accuracy. The necessary multi-parameter calibration is provided ex works at m-u-t and the complex mathematical compensation algorithms are executed directly in the gas sensor, so that the user receives calibrated measurement values immediately.
m-u-t is highly experienced in selecting optimal wavelength combinations and developing algorithms for complex cross-sensitivity compensation. The associated multi-parameter calibration is fully automated at m-u-t, ensuring reproducible measurement characteristics of the series-produced multi-gas sensors.
m-u-t detectors are characterized by a highly innovative measurement technology and sophisticated electronics. All sensors have a modular design, whereby electronic modules are usually independent embedded systems with their own microcontrollers. Thus, each module can be optimized for its respective task and programmed for specific applications. A superordinate processor unit combines the data of all connected modules, executes the complex compensation algorithms based on the stored calibration data and communicates with the customer's system. RS232 and RS422 are standard interfaces. Further interfaces such as USB, RS485, Ethernet, CAN (optional WLAN, Bluetooth or GSM) or other fieldbus interfaces can be integrated into m-u-t gas sensors easily and cost-effectively.
By using powerful infrared emitters based on microelectromechanical systems (MEMS), sensitive infrared detectors and highly sensitive electronics, we achieve the best possible signal-to-noise ratio. The path taken by the infrared radiation through the sample chamber is deflected up to nine times. This multi-reflection design extends the available distance for gas absorption and improves the sensitivity accordingly, without significantly increasing the size. The fact that only minimal signal loss occurs on the way through this optical arrangement is a challenge and at the same time one of our core competencies.
Gas sensors are very sensitive to changes in environmental conditions. Additionally, multigas sensors are often subject to cross sensitivities of the monitored gases. Both can be reduced by calibrating the integrated compensation mechanisms to such an extent as to enable the sensors to function with the required properties under all conditions occurring in real operation. In m-u-t's efficient environmental laboratories the conditions of the customer's applications are simulated by systematically varying all relevant parameters. This multi-parameter calibration in combination with the corresponding compensation algorithms results in directly calibrated measurement values, which are immediately available for the customer.
The electronic measuring signal of infrared detectors is, as with any highly sensitive measuring principle, influenced by noise, which limits the resolution and sensitivity of the gas sensor. Noise as well as external disturbances are electronically minimized by special measuring methods. The necessary intelligence is integrated in a dedicated microcontroller, coordinating the measuring process and controlling all components involved.
