Abstract: A kind of large scale numerical control (NC) machine tool condition monitoring system based on AC servo motor encoder, linear or rotary scale and motor current (torque) is proposed. To obtain the state information, the test principles of motor torque, machine tool position and lubrication characteristic are profoundly investigated, and some different sampling methods for open NCs and commercial NCs are suggested. Furthermore, some constant speed tests and lubrication characteristic tests are carried out respectively. The analysis shows that it is feasible and reliable to use the motor current, position, instantaneous velocity, instantaneous acceleration for monitoring the mechanical parts state and evaluating the servo control characteristics, which lays the foundation for large scale numerical control machine tool on-line condition monitoring and fault diagnosis.

Key words: Built-in sensors, Condition monitoring, Fault diagnosis, NC machine tool