Prestandautvärdering av infraröda detektorer, optisk fiberöverföring och digital filtrering för högtemperaturgnistdetektering

This thesis investigates methods for improving the performance of infrared spark detectors operating in high-temperature industrial environments. Two main approaches were explored: evaluating alternative infrared-sensitive materials to replace traditional lead sulfide (PbS) sensors, and assessing the feasibility of transmitting infrared signals through optical fibres combined with digital filtering. First, the performance of lead selenide (PbSe) and indium arsenide (InAs) detectors was compared to that of PbS through controlled experiments. Metrics such as signal-tonoise ratio (SNR) and relative voltage amplitude changes were evaluated across different temperatures and modulation frequencies, to see at which frequency the responsivity of the materials starts to decrease. The results showed that PbSe consistently matched PbS in terms of SNR and surpassed it at higher modulation frequencies, indicating a higher maximum bandwidth. This suggests PbSe is better suited for detecting faster-moving ignition particles in dynamic industrial environments, making it a strong candidate for future use. InAs exhibited a weaker signal strength and lower signal-to-noise ratio, making it less suitable for the intended application. Second, the feasibility of using Fluoroindate (InF3) optical fibres for remote infrared sensing was studied. Measurements demonstrated that although signal attenuation occurs, it is possible to transmit infrared radiation effectively through a 1-meter fibre. To mitigate noise introduced during transmission, both a simplified Kalman filter and a second-order Butterworth low-pass filter were implemented in MATLAB. Results showed that both filters notably improved SNR without introducing significant computational cost.