Chapter 5 Experimental Design, Materials and Procedures
5.3.2 Acoustic Emission (AE) Sensor Calibration
AE sensors are designed and manufactured to extract and record a set of parameters from measured AE signals. The characterisation of these signals depends on the type of sensor commercially available. Therefore, verifying AE signal parameters recorded by the AE sensor is very important. This is achieved by carrying out checks in form of calibration on the sensor and other devices to be used in the AE test.
Results of calibration are then compared with published results and manufacturer‘s specifications. ASTM-E976 is the standard guide for determining the reproducibility of AE sensor while ASTM-E1106 is the standard method for primary calibration of AE sensors [194].
An example of a published calibration certificate for the AE sensor used in this study is shown in Figure 5.10.
Figure 5.10: Published AE sensor (VS900-M) calibration certificate [194].
The success of any AE measurement depends on the selection of the correct sensor.
This is because the sensors are the starting point in the measurement chain. They are attached behind the specimen surface to detect dynamic motion resulting from the AE events and then convert the motion to voltage-time signals that are analysed and interpreted in the measurement. The types and features of the sensor control the characteristics of the obtained signal. Hence, the repeatability and success of the measurement depends on the resulting electrical signals. Sensors are categorized based on the mechanism of their transduction [194], e.g. laser interferometer, displacement and capacitive sensors, etc. The sensors used in this study are piezoelectric sensors because they utilize piezoelectric elements for transduction. The element (made of a special ceramics called zirconate titanate (PZT) [194]) is shown in Figure 5.11. The surface of the sensor is attached to the surface of the specimen so that the dynamic surface motion propagates into the piezoelectric element which in turn generates an output voltage signals that are processed by the acquisition system. The AE devices (sensors, preamplifiers and two-channel acquisition system) were supplied by Vallen AE Company, Germany [194]. They were chosen because Ferrer et al [172, 173, 187] have successfully used Vallen AE devices to study abrasion-corrosion of
AISI 304L austenitic stainless steel in acidified saline solution using jet impingement apparatus.
Figure 5.11: Illustration of the components of piezoelectric sensor [168].
During the calibration, the sensor was coupled to the X65 carbon steel sample by means of vacuum grease and connected through a BNC connector to the pre-amplifier which is then connected to the Vallen AMSY 6 acquisition system. The PC with the acquisition and analysis software is thereafter connected to the acquisition system through a USB 2.0 port. The set-up is schematically illustrated in Figure 5.12.
Figure 5.12: Schematic illustration of AE sensor calibration set-up.
The procedure involves pencil lead breaks. It was performed by breaking a 2H lead of 0.3 mm diameter and length of from its tip by pressing it against the surface of the sample as shown in Figure 5.12. This generated an intense AE signal that is similar to natural AE source that the sensor can detect. The generated signal was processed, analysed and compared with manufacturer‘s published calibration certificate which shown in Figure 5.10. The signal generated by the pencil lead break was processed and analysed in time-domain (with sampling rate of 2.5 MHz), frequency domain using Fast Fourier Transform (FFT) and frequency-time domain using Wavelet Transform (WT). The results are shown in Figure 5.13.
Figure 5.13: Lead pencil break signal results in (a) time-domain, (b) frequency- domain and (c) frequency-time domain.
The time history displayed in Figure 5.13(a) indicates that sensor adequately responded to the pencil lead break which created an AE burst with maximum amplitude of 54 mV and signal duration of less than 300 às. The frequency-domain (Figure 5.13(b)) of the signal obtained by performing Fast Fourier Transform (FFT) of the AE signal shows a good response of the sensor with frequency range of 100 kHz to 900 kHz and resonant (peak) frequency of 150 kHz. The response conforms to the manufacturer‘s specifications (wideband frequency 100 kHz to 900 kHz) as shown in Figure 5.10. This means that the sensor can detect AE events occurring between 100 kHz and 900 kHz and will be suitable for erosion-corrosion measurements [173]. This is important because the sensitivity of AE sensor to corrosion, erosion and/or erosion- corrosion can be increased by matching the response frequency of the sensor to the frequency range of corrosion, erosion and/or erosion-corrosion events.
It is also important to note that the sensor frequency response (especially the resonant frequency) can change with different calibration procedures, material, distance from source, etc due to attenuation (reduction of signal strength in form of amplitude) and the wave behaviour of the AE in different materials [194]. For example, ASTM E976, steady state, face to face excitation calibration of the sensor in similar materials may give higher resonant frequency. For the sensor calibrated in this study (VS900-M), the resonant frequency can go up to 350 kHz [194] when steady state, face to face excitation calibration is used but the wideband frequency range (100 kHz to 900 kHz) will still be the same as the transient pencil lead break calibration.