Micromanufacturing and Nanotechnology

4 Soft Geometrical Error Compensation Methods Using Laser Interferometer (p. 63-64)

K. K. Tan and S. N. Huang
National University of Singapore, Singapore

4.1 Introduction

Micromachining is a key technology in the important fields of microimprinting, scanning microscopy, microlithography and automated alignment. A large proportion of these machines require accurate positioning of tools or probes with respect to a workpiece. Thus, much challenges behind the metrology is concerned with the accurate measurement of absolute positions and the subsequent reduction of the geometrical errors associated with this position. Compensation for geometrical errors in machines has been applied to Co-ordinate Measuring Machines (CMMs) and machine tools to minimise the relative position errors between the end-effector of the machine and the workpiece (Hocken 1980).

Given an adequate machine design, a large proportion of these errors are completely repeatable and reproducible, and therefore amenable to modelling and compensation. While widespread incorporation of error compensation in machine tools has remained to be seen, the application in CMMs is tremendous and today, it is difficult to find a CMM manufacturer who does not use error compensation in one form or another (Hocken 1980). The development in error compensation is well documented by Evans (Evans 1989). Early compensation methods utilised mechanical correctors, in the form of leadscrew correctors, cams and reference straightedges. Maudslay and Donking, for example, used leadscrew correction to compensate for the errors in their scales producing machine. Compensation via mechanical correction, however, inevitably increases the complexity of the physical machine. Furthermore, mechanical corrections rapidly cease to be effective due to mechanical wear and tear. The corrective components have to be serviced or replaced on a regular basis, all of which contribute to higher machine downtime and costs. The evolution of control systems from mechanical and pneumatic-based subsystems to micromachining-based systems has opened up a wide range of new and exciting possibilities. Many operations which used to be the result of complex linkages of levers, cams, bailing wires and optical sensors can now be carried out efficiently with program codes residing in the memory of a standard computer. Soft compensation schemes thus blossomed in the 1970s. The first implementation was on a Moore N.5 CMM, a pioneering piece of work which earned Hocken the CIRP Taylor Medal in 1977. Since then, there has been an explosion of interests in soft compensation of CMM and micromachining system errors with new methods developed and implemented (Hocken et. al. 1977; Love and Scarr 1973; Kunzmann and Waldele 1984). Common to all these works and more is a model of the machine errors, which is either implicitly or explicitly used in the compensator. A common issue with these approaches is the challenge to have an adequately accurate model, which is also amenable to practical use.

4.2 Overview of Geometrical Error Calibration

Error modeling typically begins with a calibration of the errors at selected points within the operational space of the machine. For a 3D working space, the resultant geometrical errors in positioning may be decomposed into 21 underlying components. For the XY table with zero tool offsets, the error sources reduce to 7 components, including two linear errors, two angular errors, two straightness errors and the orthogonal error between the X- and Y-axis. These errors may be measured accurately using an independent metrology system such as a laser interferometer which can typically measure linear displacement to an accuracy of 1nm and angular displacement to an accuracy of 0.002 arcsec. These errors are subsequently cumulated using the overall error model to yield the overall positional error. A 3D machine has a volumetric error model described by (Zhang et al. 1985).