應用於定位系統之相位對數位轉換器

Abstract

近年來由於半導體產業與電子產業的興起，精密定位量測產品大量應用於半導體產業與電子產業的製程與測試設備中，隨著製程技術的日益精進，更小的線寬與更精密的製程技術不斷的被開發出來，甚至縮小尺寸到奈米等級，因此精密定位量測技術也必須推陳出新。典型的精密定位量測產品為旋轉編碼器與光學尺，依其應用場合的不同而有不同的技術要求。在傳統加工為主的機械產業，其技術要求是能在污染與高溫環境下仍能精確定位的定位量測系統；而在現代之電子與半導體產業，為配合高速驅動的特性定位量測產品自然被要求具備高精度與高速位移量測的需求。 目前工業上大量應用的定位量測儀器，除雷射干涉儀之類高精度位移量測系統外，概分為線性編碼器與旋轉編碼器兩類；編碼器應用感測器通過週期性柵極所產生的等距週期性弦波訊號來量測移動距離，量得的相位即對應於移動距離，基於感測的物理量的不同，主要可分為應用光柵原理的光學尺與磁柵原理的磁性尺等兩類。光學尺具有精度高的優點因而普遍應用於電子與半導體產業，不過由於光學物理特性與機械組裝等因素使得其精度仍然無法滿足現代之電子與半導體產業的需求，因此電子細分割技術不斷的被開發出來，它藉由分割感測器量得的相位而達到增進編碼器的量測精度與解析度的目的，且其型態也隨應用場合而有所不同；一部份利用類比/數位轉換器將編碼器感測器所感測到的訊號轉成數位訊號之後再依不同的理論達到細分割的目的，但是這樣的方法往往需要的週邊電子零件相當多而不利於縮小電路板的面積，雖然其細分割倍率可以達到數千倍卻因為定位系統精度之限制而失去意義，若要將其電路開發成ASIC又因為其電路複雜而使得開發成本過高；另一部份的電子細分割方法雖然其細分割倍率不高卻已足夠滿足現今定位系統的需求，而且其電路結構簡單適合開發成ASIC，不僅能降低編碼器的成本、縮小編碼器讀頭的體積，而且還能提高其抗雜訊的能力、增進定位系統穩定度。 本文基於這樣的趨勢，針對目前普遍使用的光學式編碼器製作一細分割倍率40倍的ASIC，而且還包含光學感測元件所需的訊號擷取電路以進一步提高其抗雜訊的能力；不僅如此，本文也提出一相位差調整方法以很簡單的方法補償由於編碼器組裝等問題造成的誤差；此ASIC符合目前的技術趨勢並在UMC 0.5µm 2P2M CMOS製程完成製作與測試。

Recently, because of the development of semiconductor and electronic industry, precise measurement and positioning products have been used on large scale in the process and testing equipment of electronic and semiconductor industry. Process techniques are changing; smaller line width and more precise process have been developed even scale down to nano-meter. Therefore precise measurement and positioning techniques must be improved. The typical measurement and positioning products are rotary encoder and optical scale. They require various techniques in different application. In traditional manufacture industry, the required techniques are the measurement and positioning products can work in high pollution and high temperature environment. However, in modern electronic and semiconductor industry, they need high precision and high-speed system for high driving ability. The main measurement and positioning apparatus, besides the lacer interferometer that is more high precision displacement measurement system, can be classified into two types, linear and rotary encoder. The encoder always employs sensor passing through periodic and equal distance grating and then generates periodic quadrature scaling signals for displacement measurement. The phase is relative to the movement. According to the sensor type of encoder, it can be classified into two types, optical and magnetic encoder. The optical encoder is popular in modern electronic and semiconductor industry because of the high positioning accuracy. However, its accuracy is still restricted by optical properties and mechanical assembly. To improve encoder accuracy or resolution, electronic interpolation technique had been developed. It improves encoder accuracy by subdivision the phase of quadrature scaling signals. Some of them utilizes analog-to-digital converter to digitize input quadrature scaling signals and then achieve interpolation via different algorithm. This type can achieve high-fold interpolation even ~thousands but need lots of additional component. Therefore it is hard to minimize the circuit size and it cost high to integrate into a chip because of the complex circuits. Other interpolation methods satisfy the requirement of modern electronic and semiconductor industry although its interpolation-fold is not high. In additional, its structure is simple and suitable for integrating into a chip. Not only cost down the encoder but minimize the encoder size, integrating the interpolation circuit into a chip can also improve anti-noise ability and positioning system stabilities. According to the trends, an ASIC with conditioning circuit for optical sensor and a 40-fold interpolation circuit has been fabricated and measured in UMC 0.5µm 2P2M CMOS process. Integrating the conditional circuit for optical sensor into the ASIC improve the anti-noise ability furthermore. In addition, a phase difference adjustment for quadrature signals is also proposed in the thesis to improve assembly tolerance.