Depth of dressing optimization in CBN wheels of different friabilities using acoustic emission (AE) technique

Grinding is a manufacturing process that has the objective of granting the workpiece a high-quality surface and is located at the end of the sequence of machining processes. During grinding operation, the abrasive grains of the wheel surface are worn and the pores are filled with debris. This phenomenon makes the cutting tool less efficient to remove material and sometimes improper to be used if a process to correct the cutting surface is not applied to the tool. Dressing is defined as a conditioning process which gives shape to the wheel and has the purpose of improving its capacity to remove material. In this context, this work proposes the monitoring of the dressing process of CBN wheels through acoustic emission technique (AE) along with the processing of digital signals. Dressing tests were done in a cylindrical grinder with two types of CBN wheels and the surface after the process was evaluated through micrographs. The AE signals were acquired with a 2 MHz sampling rate. In sequence, statistics such as RMS (root mean square) and counts were applied to the sample signals and analyses in the frequency domain were done to select frequency bands that are more related to the dressing process. The results show that the counts' analysis applied to the signals filtered in the selected bands is effective to detect the best moment to stop the dressing process.