Controller design and implementation of six-degree-of-freedom magnetically levitated positioning system with high precision

Abstract

This paper presents the controller design and implementation of a high-precision six-degree-of-freedom (6-DOF) magnetically levitated (maglev) positioner. This high-precision positioning system consists of a novel concentrated-field magnet matrix and a triangular single moving part that carries three three-phase permanent-magnet planar-levitation-motor armatures. Since only a single levitated moving part, namely the platen, generates all required fine and coarse motions, this positioning system is reliable and potentially low-cost. The three planar levitation motors based on the Lorentz force law not only produce the vertical force to levitate the triangular platen but also control the platen's position and orientation in the horizontal plane. The main contribution of this paper is that all 6-DOF motions are solely controlled by magnetic forces without any other means to support the platen's weight against gravity, and the most suitable controller of the magnetic levitation system was designed and implemented. The platen can be regarded as a pure mass, and the spring and damping effects are neglected except for the vertical directions. Single-input single-output digital lead-lag controllers were designed and implemented on a digital signal processor. This 6-DOF fully magnetically levitated positioner has a total mass of 5.91kg and exhibits a 120 × 120 mm maximum travel range. The position resolution of 20 nm and position noise of 10-nm root mean square are demonstrated. The positioner has sub-microradian angular resolution in about the x, y, and z-axes. A maximum velocity of 24.8 mm/s in y is achieved. This single-moving-part maglev positioner structure is highly suitable for semiconductor manufacturing applications such as wafer steppers. Several experimental motion profiles are presented to demonstrate the maglev stage's capability of accurately tracking any planar and three-dimensional paths. © IMechE 2008.