三軸數控工具機之即時體積誤差補償系統之研製

Abstract

數控工具機的位置回授控制均採用位移感測器(光學尺或編碼器)的單感測系統，但是其位移感測器之量測軸線永遠無法和切削刀具在加工空間的運動軸線同軸，非同軸所產生之誤差為眾所周知之阿貝原理，而各軸移動時具有無法避免之基本六自由度幾何誤差(三線性三角度)，其中偏角誤差更會因阿貝原理而放大，因此阿貝誤差是工具機體積誤差來源中主要成因。本研究中自行研發體積誤差補償系統為多感測回饋補償系統(Multi-sensor feed-back compensation system)，由現有NC控制器之單感測系統與多角度感測器所組成，可即時量測運動平台之偏角誤差與阿貝偏位量，動態補償工具機體積誤差。本系統可建於PC-base控制器內，使得數控工具機具有內藏式體積誤差補償系統功能。量測系統利用實驗室自行開發之自動視準儀與滾動度感測器，組成一套可同時量測三角度(Pitch, Yaw, Roll)之三角度感測器，配合電路板處理將電壓訊號傳輸至PC-based控制器，再配合控制器內根據不同機台之結構配置、位置回授方式，進而推導體積誤差通式，最後進行實驗驗證之。實驗結果可知，以旋轉編碼器做位置回授，其定位誤差主要由阿貝誤差造成，經補償後，其定位精度可有效提升80%以上，工具機精度可大幅改善。

The position feedback control of NC machine tool controllers mostly employs displacement sensor (such as optical scale or encoder) as a single sensor system. Distance between axis of sensor and axis of cutting could lead to massive enlargement of the positioning error of each axis which is known as the Abbe principle. The movement of linear stage inherently has position and orientation errors in six degree of freedom, which are three linear errors (positioning error, horizontal straightness and vertical straightness) and three rotational angles (pitch, yaw and roll). As a result, Abbe error is considered to be the major cause for volumetric error of machine tool. In this study, a multi-sensor feed-back compensation system which consists of three angular sensors and NC controller is developed to greatly enhance the machine tool accuracy by detecting angular errors in real-time and corresponding Abbe offsets so as to compensate for the Abbe errors dynamically. It can also be embedded in NC controllers to improve the accuracy of machine tool. In the measurement system, autocollimator and roll are deployed to monitor the angular errors of machine tool by constantly converting the measuring signals into Pc-based controller(with embedded volumetric error model) Experiments show that the positioning error of an investigated NC machine tool Fis mainly affected by Abbe error. With the proposed volumetric error compensation system, the positioning accuracy can be significantly improved by more than 80%.