Development of a Servo System for Linear X-Y Table Based on DSP Controller Ming-Hung Tsai, Ying-Shieh Kung*, Member IEEE, Chung-Chun Huang Department of Electrical Engineering, Southern Taiwan University of Technology No.1, Nan-Tai Street, Yung-Kang, Tainan County, 710, TAIWAN *e-mail:kung@mail.stut.edu.tw Abstract- This paper presents a servo system for linear X-Y table, which is driven by two permanent magnetic linear synchronous motors (PMLSMs), based on a TMS320F2812 DSP controller. In this proposed system, the fully digital servo controller for each PMLSM and the trajectory motion controller for X-Y table are all implemented by a DSP chip. That is, the detection of quadrature encoder pulse (QEP), coordinate transformation, vector control, current loop controller, position loop controller, PWM output of each PMLSM drive system, and the trajectory motions for X-Y table, such as, motions of point-to- point, circle, line etc, are all performed in this DSP chip. To increase the control performances of the linear X-Y table, the fuzzy controller has been employed as their position controller at each linear servo motor drive. The TMS320F2812 DSP adopted herein has fast computational capability and excellent peripheral function for two sets of motor drives, so that it is very suitable to develop a compact servo control system of the X-Y table. At last, an experimental system which includes linear X-Y table, DSP- based servo control board, inverters, rectifiers have been set up and some experimental results have been demonstrated the effectiveness of the proposed control system. I. INTRODUCTION Owing to the advantages of the superior power density, high performance in motion control - fast speed and better accuracy, permanent magnet linear synchronous motors (PMLSM) have gradually used in many automation control fields as an actuators [1-4], including computer-controlled machining tools, X-Y driving devices, robots, semiconductor manufacturing equipment, transport propulsion and levitation, high-speed injection molding machines, DC solenoid, etc. But in industrial applications, there are many uncertainties, such as system parameter uncertainty, external load disturbance, friction force, unmodeled uncertainty, always diminish the performance quality of the pre-design of the motor driving system. To cope with this problem, in recent years, many intelligent control techniques [5-7], such as fuzzy control, neural networks control etc., have been developed and applied to the position control of servo motor drives to obtain high operating performance. With rapid developments of the system-on-chip, the high-performance digital signal processor (DSP) has become a popular area of research in the field of the digital control [8-10] for ac drives because they exhibit high-speed performance, and combine peripheral circuits, memory and an optimized CPU structure on a single chip. In particular, a new-generation DSP controller TMS320F28x [11] produced by Texas Instrument, which provides the advantages of high speed (150MIPS), up to 128Kx16 flash, two sets (total 12 lines) of PWM outputs, two sets (total 4 lines) of QEP inputs, a 16-channels 12-bit A/D converter (200ns conversion time) and a 56-bits GPIO. Therefore, it is not only suitable two-axis motor control, but also the complex control algorithm, such as fuzzy control or neural network control, applied to servo-system to improve the dynamic performance become possible. Accordingly, in this study, a TMS320F2812 DSP controller with two sets of current vector control, SVPWM (Space Vector Pulse Width Modulation) scheme and fuzzy control has been developed to build a fully digital motion controller for drive of linear X-Y table. All software implemented in DSP is coded in C language. II. THEORETIC DESCRIPTION The architecture of the proposed DSP-based servo system for a linear X-Y table is shown in Fig. 1, in which the trajectory motion control, the current vector and the position control of the linear X-Y table are all implemented in a TMS320F2812 DSP chip. The modeling of each PMLSM and the fuzzy control algorithm will be introduced as follows: A. Mathematical modeling of PMLSM The mathematical model of a typical PMLSM is described, in two-axis d-q synchronous rotating reference frame, as follows [6] d d q p d q d d s d v L i x L L i L R dt di 1 + τ π + − = & (1) q q p q f q q s d p q d q v L 1 x L i L R i x L L dt di + τ λ π − − τ π − = & & (2) where v d and, v q are the d and q axis voltages; i d and, i q , are the d and q axis currents, R s is the phase winding resistance; L d and, L q are the d and q axis inductance; p x and, dx p /dt are the translator displacement and velocity; f λ is the permanent magnet flux linkage, and τ is the pole pitch. The developed electromagnetic thrust force is q f d q d e i i L L F ) ) (( 2 3 λ τ π + − = (3) The current control of a PMLSM drive is based on vector control, such that if i d is controlled to zero in Fig.1, the PMLSM is decoupled. Therefore, (3) can be simplified to q t q f e i K i 2 3 F ∆ λ τ π = (4) 1-4244-0726-5/06/$20.00 '2006 IEEE 2907