An experiment was set up at the Oak Ridge National Laboratory (ORNL) to test methods for estimating the cutting forces in real time within machine tools for any spindle speed, force profile, tool type, and cutting conditions. Before cutting, a metrology suite and instrumented tool holder were used to induce magnetic forces during spindle rotation, while on-machine vibrations, magnetic forces, and error motions were measured for various combinations of speeds and forces. A model may then relate the measured accelerations to the forces, such that during cutting, on-machine measured vibrations may be used in the model to estimate the cutting forces in real time. To test this process, the metrology suite and the instrumented tool holder were removed, except that the on-machine accelerometers remained. A dynamometer was then set up on the worktable with a workpiece to independently measure cutting forces during machining. Various cutting passes were performed with an end mill while the dynamometer data and accelerometer data were collected. Even though considerable research has been conducted to estimate cutting forces with accelerometers and measured frequency response functions (FRFs), one main challenge remains: a method must be created to estimate the cutting forces in real-time for any spindle speed, force profile, tool type, and cutting conditions. This dataset can be used to develop or advance such methods for industrial adoption.

An experiment was set up within a machine tool at the National Institute of Standards and Technology (NIST) to test vision-based thermal drift tracking methods. A wireless microscope within a tool holder in the spindle is used to capture videos of image targets attached to the worktable. For each target, one video is captured during spindle rotation orthogonal to the worktable and another video is captured during axis translation orthogonal to the worktable. Data are collected periodically so that the three-dimensional thermal error at each target location may be determined via image analysis.

A linear axis testbed at the Center for Intelligent Maintenance Systems (IMS Center) at the University of Cincinnati was run to failure (the detection of backlash) over one year with periodic data collected from an inertial measurement unit (IMU) on the carriage, two triaxial accelerometers on the ball nut, and the controller. The IMU and its associated data acquisition (DAQ) equipment and software were provided by the National Institute of Standards and Technology (NIST), while the two triaxial accelerometers on the ball nut and its associated DAQ equipment and software were provided by the IMS Center. The linear axis components, including the ball screw, the motor, and the controller were provided by HIWIN Technologies Corp. in collaboration with the IMS Center.Ball screws play a critical role in maintaining the operational accuracy and reproducibility of production systems. Over time, mechanical wear can cause reduced preload levels and eventually backlash development, which leads to non-conforming behavior and a loss of positional accuracy in the ball screw. This dataset can be used to develop methods to track the loss of preload and the emergence of backlash.

In this study, a rail of a linear axis testbed was intentionally degraded to various levels, affecting the carriage motion, which can be deduced with either an inertial measurement unit (IMU) or a laser-based reference system. The rail was degraded over a 10-cm-long region. Thus, the two trucks that move on the rail interact with this region of the rail, changing the translational and angular error motions of the carriage. To mechanically simulate spalling, a handheld grinder was used to wear the surface of the raceway groove of the rail. The degradation zone length was increased incrementally by about 5.4 mm from its nominal state of no degradation (Stage 1) to its final state of significant degradation with a length of about 75 mm (Stage 15). For each stage of degradation, micrometer measurements were taken of the degradation zone, and IMU and laser-based reference data were also collected. For each stage, fifty (50) runs of IMU data were collected bidirectionally at slow (0.02 m/s), moderate (0.1 m/s), and fast (0.5 m/s) axis speeds over a travel range of 322 mm, following the method described in Vogl et al. (https://www.nist.gov/publications/diagnostics-geometric-performance-machine-tool-linear-axes). The axis position from the motor encoder was also collected simultaneously during each motion of the linear axis. Afterwards, ten (10) runs of laser-based reference data were collected bidirectionally at finite positions of travel, specifically every 1 mm between travel positions 1 mm and 321 mm.