Modeling and experimental detection of resonance frequency shift of a microwave cavity caused by a small conductive particle

Owing to recent progress in the manufacturing of electrical devices, the detection of metal dust on/in thin films or cloths has become crucial as they are used for separators or insulators. This paper considers the detection of metal particles less than 100 μm in size by monitoring the resonance frequency of a microwave cavity. Metal particles shift the resonance frequency, but the shift is less than the resonance frequency by many orders of magnitude when the cavity consists of an X-band waveguide. This makes it challenging to detect small metal particles in industrial environments. To overcome this difficulty, a new method of detecting small frequency shifts is proposed. First, transmission measurements are performed at a fixed frequency where the resonance peak has the steepest slope. Second, a Mexican hat filter is applied to the measured signal in order to distinguish frequency shifts from background noise and ambient conditions. Experimental detection of a 48 μm stainless steel sphere is demonstrated. The threshold level affects the minimum level of the spheres size and false positive signal. The paper also assesses the ability of available commercial software to accurately predict frequency shifts. The calculated results are compared with experimental data, which should be adjusted for the ambient temperature owing to thermal expansion of the cavity. The corrected experimental shifts agreed with the calculated shifts, demonstrating that the simulation can be used for future studies to improve this method.