Stephen J. Ludwick

Design of a rotary fast tool servo for ophthalmic lens fabrication

Abstract We have developed a novel fast tool servo and associated prototype diamond- turning machine for the production of plastic spectacle lenses. Our fast tool servo carries the cutting tool on a rotary arm and, thus, on a circular path, as opposed to straight line paths in conventional designs. The actuator, sensors, and bearings are standard elements that together allow experimentally demonstrated 500 m/s2 instantaneous accelerations at the tool tip over a 3-cm range of cutting depth. We also describe in this paper new approaches we have developed for toolpath generation and calibration. The paper also presents associated control algorithms, because the controller must supply very high dynamic stiffness to the tool servo axis at multiples of the spindle frequency. This stiffness is achieved by means of repetitive control techniques. The new fast tool servo is shown to have great promise for machining asymmetric surfaces with large amplitude asymmetries.

Modeling and control of a six degree-of-freedom magnetic/fluidic motion control stage

In this paper we describe a six degree of freedom positioning stage for fine motion control. The stage achieves a 6 /spl sigma/ positioning noise of 0.3 nm at a 1 Hz measuring bandwidth and a controller bandwidth of 5 Hz. Total travel is within a cube of 100 /spl mu/m. The single moving element is immersed in oil, forming squeeze film dampers between itself and the frame. This design results in a highly overdamped and vibration resistant system. Twelve electromagnets provide the forces necessary to suspend and servo the platen while six capacitance probes sense the position. We perform all of the controls digitally, using a PC-based digital signal processing board. The controller performs two functions that are essential to achieving 0.3 nm positioning resolution. First, it uses a detailed model of the stage in a feedback linearization scheme that linearizes and decouples the degrees of freedom. Second, it uses a combination of a digital filter and an estimator to reduce the effect of measurement noise by about two orders of magnitude. The end result is a stage that is suitable for positioning a sample with better than atomic resolution. Envisioned applications of the stage include producing the scanning motions required in scanned probe microscopy or as a motion control stage for integrated circuit metrology.

Development of 2.737 Mechatronics at MIT

We describe the development of the curriculum and associated laboratory facilities and exercises for the undergraduate course 2.737 Mechatronics in the Mechanical Engineering Department at the Massachusetts Institute of Technology. The course is centered on laboratory experiences which teach integration across the mechanical, electrical, and control engineering disciplines. Within this broad focus, the labs emphasize the application of feedback design in a variety of contexts as a means to motivate the broader mechatronic design issues. The labs utilize PC-based hardware and software for rapid control prototyping on a target digital signal processing (DSP) board. This DSP board is programmed through a commercially available block-diagram-based graphical environment, in order to eliminate the need for low-level programming and thereby allow students to concentrate on the higher-level challenges posed in each lab exercise. A students lab performance is evaluated primarily through a one-on-one interview with a member of the teaching staff in conjunction with a conventional written lab report.

Nonlinear Control Algorithm for Improving Settling Time in Systems With Friction

A nonlinear control algorithm that greatly reduces settling time in precision instruments with rolling element bearings is proposed. Reductions of 80.5%-87.4% in settling time were achieved when settling to within 3-100 nm of the commanded position. Final settling of such systems is typically impacted by the nonlinearity in the pre-rolling friction regime, which manifests as a hysteretic stiffness. Consequently, the integral term in the controller can take a long time to respond. In this paper, a nonlinear integral action settling algorithm is presented. The nonlinear integral gain takes the form of a Dahl friction model. Since the integral gain mimics hysteretic stiffness, the output of the integral control term is instantaneously set to a large value after each direction change, greatly improving settling response. A nearly first-order error dynamic results, which has a user-definable time constant. Before the algorithm can be implemented, the Coulomb friction and initial contact stiffness in the Dahl model must be experimentally determined for the stage. A sensitivity study is performed on the initial contact stiffness, which was found in other works to dictate the stability of the algorithm.

Determination of a Dynamic Grinding Model

In precision abrasive machining, it is important to control process variables such as the material removal rate, normal force and power input, as these factors influence surface finish, dimensional precision, and material damage. In this research, a linear grinding process model, with enhancements over past models, is developed relating normal force to material removal rates. Two experimental procedures for the determination of the grinding models parameters are presented. Simulations are performed to validate the grinding model. The determined model is found to be a valid representation of the grinding process that should prove useful in adaptive control with real-time parameter estimation.

A Test Setup for Evaluation of Harmonic Distortions in Precision Inertial Sensors

Steadily improving performance of inertial sensors necessitates significant enhancement of the methods and equipment used for their evaluation. As the nonlinearity of sensors decreases and gets close to that of the exciters, new challenges arise. One of them, addressed in this research, is a superposition of errors caused by the nonlinearity of tested devices with nonlinear distortions of excitation employed for experimental evaluation. This can lead to a cancellation, at least partial, of the effects of both imperfections and underestimation of the actual distortions of the evaluated sensors. We implement and analyze several system architectures and evaluate components of applicable motion generation systems from the viewpoint of satisfying the relevant, often conflicting requirements posed by the evaluation of high performance inertial sensors. Robust mechanical integration of the guidance, actuation, and measurement functions emerges as a key factor for achieving the needed quality of generated test patterns. We find precision air bearing stages, such as ABL1500 series (Aerotech) most suitable for implementing the needed experimental setup. We propose an architecture with two reciprocating stages, implement and evaluate its core components, and illustrate its performance with experimental results.

Design and Implementation of a Tube Wall Thickness Measurement System

This paper describes the design and implementation of a gage for measuring the wall thickness of hot rolled steel tubing at an actual production facility. The design is based upon a statistical analysis of the tube samples, and upon the economic realities of mill operating conditions. Issues relating to time constraints, uneven sample preparation, a mill operating environment, and differences between gage operators are addressed. The gage records more data from the tubes than could be taken manually, and therefore allows the accompanying software to apply unique algorithms for the calculation of hexing and eccentricity in each sample. All results are recorded and stored in a database for continuous use in statistical process control.