Researches On Fault Detection Method Of Micro Loudspeaker And Its Application

Micro loudspeaker’s fault that caused by improper operation in the assembly process directly affects the quality of the products. Loudspeaker abnormal tone is detected by human ear monitoring. The related references have made study of loudspeaker automatic detection online. In order to facilitate the application, digitization, networked, miniaturization of electronic products develop faster and faster, micro loudspeaker have a wide range of applications. But the references did not give special researches of automatic detection of micro loudspeaker online. Micro loudspeaker has low power, small size, so micro loudspeaker and loudspeaker have many differences in performance. Researching for online detection method and application for micro loudspeakers of these characteristics is the industry’s urgent needs.This paper analyzed various fault reasons of micro loudspeaker, and proposed two fault detection method of micro loudspeaker. One fault detection method is based on the Euclidean distance of the high frequency region. A sweep signal stimulates micro loudspeaker to produce a sound signal. The microphone changes the sound signal into an electrical signal and sent to capture card that can change the analog signal into a digital signal. The digital signal is transmitted to a computer for analysis. A one-dimensional digital signal is converted to a two-dimensional image signal by short-time Fourier transform. This paper calculates the Euclidean distance of the high frequency region between the detected image and the standard template image to identify the fault micro loudspeaker and fault severity by observation and analysis of the image. Another detection method is based on the time-domain and frequency-domain analysis. Natural frequency signal stimulates micro loudspeaker to produce a sound signal. A digital signal obtained by the microphone and the capture card and transmitted to a computer for analysis. The digital signal in the time-domain is converted to the signal in the frequency-domain, by the Fourier transform function After the high-pass filtering of the frequency-domain and then to return to the time domain, and then the signal measured in the time-domain by peak to detect the fault micro loudspeaker and fault severity.In order to address the impact of the production site ambient noise on the signal acquisition, the paper finished the preliminary design of the muffler box for micro loudspeaker. In order to complete the validation of the detection methods, this paper has been further improved the muffler box on the basis of laboratory homemade muffler box and designed a board to fix the micro loudspeaker.Software system of two fault detection method has developed by VC++6.0, and this paper completes hardware and software integration.The experiments show that the two different fault detection methods can well detect the fault micro loudspeaker and the severity of the fault, on this basis, fault detection method of the time-domain and the frequency-domain analysis can identify small sound, bottoming and loose particles.