Abstract
We have developed a system to obtain both the resonance frequency and resistance from a Quartz Crystal Microbalance(QCM) using a portable vector network analyzer (VNWA). Acoustic loading on a QCM causes a change in admittance,which can be measured by a VNWA. For application as an odor sensor, it is necessary to coat the QCM with a sensing film. Since many sensing films are highly viscous, the noise level is high owing to viscous damping. Thus, we propose a curve fitting method by utilizes all the datapoints within a certain frequency range to obtain the resonance frequency. The scan frequency range, the number of datapoints and the measurement time per datapoint are optimized to achieve minimum noise. Then, the data after curve fitting are applied to a Savitzky–Golay filter with a polynomial approximation within the moving window to further smooth the temporal frequency data, while maintaining the sharpness of the sensor response. All the entire process can be carried out in real time. The Savitzky–Golay filter was found to be able to suppress noise after the optimization of window size optimization. The effectiveness of the system was confirmed by performing an independent t-test.

This publication has 12 references indexed in Scilit: