How do you optimize acoustic emission parameters for different machine types and applications?

1 Frequency range

One of the key parameters of AE is the frequency range of the signals, which depends on the type and size of the source, the material properties, and the propagation path. Generally, higher frequencies correspond to smaller and more localized sources, such as cracks or impacts, while lower frequencies indicate larger and more distributed sources, such as friction or wear. Therefore, you need to select a frequency range that matches the expected source characteristics and minimizes the background noise and attenuation. For example, if you are monitoring a gearbox, you may want to use a frequency range of 100-300 kHz, as this can capture the tooth meshing and bearing defects, while avoiding the low-frequency vibration and lubrication noise.

2 Sensitivity and threshold

Another important parameter of AE is the sensitivity of the sensors, which determines the amplitude of the signals that are detected and amplified. The sensitivity should be high enough to capture the relevant AE signals, but not too high to saturate the system or pick up unwanted noise. Similarly, the threshold is the minimum amplitude level that triggers the data acquisition and analysis. The threshold should be low enough to avoid missing significant AE events, but not too low to generate false alarms or overload the system. For example, if you are monitoring a pipeline, you may want to use a sensitivity of 40 dB and a threshold of 35 dB, as this can detect the AE signals from leaks or corrosion, while filtering out the ambient noise and pressure fluctuations.

3 Filters and processing

A third parameter of AE is the use of filters and processing techniques to enhance the signal quality and extract useful information. Filters can be applied to remove unwanted frequency components or noise from the signals, such as high-pass, low-pass, band-pass, or notch filters. Processing techniques can be used to analyze the signals in different domains, such as time, frequency, amplitude, or energy, and to extract features that indicate the source type, location, severity, or evolution. For example, if you are monitoring a bridge, you may want to use a band-pass filter of 50-300 kHz and a processing technique called b-value analysis, as this can identify the AE signals from cracks or fatigue, and measure their activity and growth rate.

4 Best practices and challenges

To optimize AE parameters for different machine types and applications, certain best practices must be followed and obstacles must be overcome. These best practices include conducting a baseline measurement to establish normal AE behavior and noise level, performing a calibration test to verify sensor performance and system settings, using multiple sensors with appropriate mounting techniques to ensure good signal coverage and coupling, and applying proper data acquisition and analysis software and hardware to handle the high data rate and complexity of AE signals. Challenges include dealing with environmental noise and interference that can mask or distort the AE signals, distinguishing between different AE sources that may have similar or overlapping frequency ranges or features, and interpreting the AE results to correlate them with the physical condition and damage of the machine or structure.