Damage Assessment of Gearbox Operating in High Noisy Environment Using Waveform Streaming Approach

Threshold and timing parameters are two fundamental factors that control arrival times and waveform signatures collected in an acoustic emission test. While this approach works successfully in many applications, recording the complete, continuous waveforms emitted from an AE test independently of threshold value is more appropriate for some specific applications. Applications such as those with high background noise or complex structures with numerous boundaries and interfaces are better served by continuously streaming with duration in the order of hundreds-of-ms instead of the traditional hundreds-ofs-long AE waveforms. State-of-the-art AE data-acquisition hardware enables this “AE Waveform Streaming” approach at high sampling rates throughout a traditional AE test. These waveforms can subsequently be analyzed in real time to extract traditional AE features in the time and frequency domain. In this paper, the application of waveform streaming will be shown on detecting corrosion pitting on gear teeth under a high background noise environment. The tests were performed using an FZG machine to generate pitting corrosion on gear teeth under fatigue. In these tests, a recently designed “integrated” AE-vibration sensor was used for collecting highand lowfrequency data while monitoring the condition of the gears. Recording the complete, continuous waveforms provided flexibility during post-processed data analysis, which can be implemented in real time during consecutive tests. The paper also presents the impact of the waveform streaming approach in terms of understanding threshold effects on waveform signatures, in the ability to set and analyze different digital filters, and in the extraction of low frequency vibration information. Keyword: Waveform streaming, threshold independent, gearbox, frequency analysis, real time analysis