*5.3. Data Acquisition*

A low-cost piezoelectric diaphragm (PZT), model 7BB–35–3, from Murata Electronics, was attached to the workpiece holder with a cyanoacrylate of medium viscosity, from Tekbond. As in Viera et al. [29], the glue was evenly distributed over the entire diaphragm surface, thus maintaining a thickness of less than 1 mm. The piezoelectric diaphragm consists of (1) a bottom electrode (brass) of 35 mm diameter × 0.3 mm thickness (which is glued to the holder); (2) a piezoelectric ceramic of 25 mm diameter × 0.23 mm thickness (active element), and 3) an upper electrode (silver) of 23 mm diameter.

The study conducted by Freitas and Baptista [18] showed that the Murata piezoelectric diaphragms have a similar frequency response, determining that for most applications, the size of the diaphragm does not influence the results. Thus, the 35 mm model was chosen to validate the proposed approach. Other diaphragm models with different sizes and thicknesses can be tested in future studies. The piezoelectric diaphragm was glued to the center of the workpiece holder, thus positioning it as close as possible to the acoustic source (workpiece surface). Finally, the PZT diaphragm was covered with

silicone in order to protect it from the coolant fluid. The PZT transducer was exclusively used in the passive configuration operating under *d*<sup>31</sup> mode; in other words, the transducer was not excited by external sources; its function was to convert the acoustic waves generated during the process into electric voltage through the piezoelectric effect.

For comparison, an acoustic emission (AE) sensor was also fastened to the workpiece holder by a screw. The sensor was connected to a signal unit, model DM – 42, from Sensis manufacturer, which has an input–output gain of 3. The AE sensor consists of a metal housing of 30 mm × 20 mm × 20 mm in length, width, and height, respectively. Inside the metal housing is the piezoelectric ceramic (active element), as well as a filter and backing material, which reduces the resonance of the sensor. The AE sensor was screwed at a distance of 25 mm from the right end of the holder and at the same height as the PZT diaphragm.

The level of saturation and sensitivity of the two sensors was verified before the grinding tests. Both sensors were attached to the workpiece holder in order to reproduce an industrial application, where changes in machine cycles would be unacceptable. An oscilloscope, model DL850, manufactured by Yokogawa, collected the AE and PZT raw signals at a sample rate of 2 MS/s. The PZT output was directly connected to the oscilloscope. On the other hand, the AE output was first connected to the signal unit, which improves the signal conditions. The aliasing effect was prevented by an anti-aliasing filter, inbuilt in the oscilloscope data acquisition board, with a 65-dB cutoff centered at 80% of the sampling frequency. The experimental setup is shown in Figure 2.

**Figure 2.** Experimental setup.
