Abstract
Tool wear is a critical factor in milling operations, as increasing tool wear can affect product quality and contribute to increased cutting forces and spindle bearing degradation. This dataset contains vibration measurements acquired for the development and evaluation of an on-device tool condition monitoring system based on TinyML. The data were collected using an Arduino Nano 33 BLE Sense Rev2 development board, equipped with a BMI270 MEMS accelerometer and based on an Arm Cortex-M4 processor. The sensor node was magnetically attached to the spindle housing to capture vibration signals during milling operations. Milling experiments were performed using both a sharp end mill and a visibly worn end mill under nine different process configurations, providing data representing distinct tool conditions and machining conditions.
