터치트리거프로브를 이용한 다축 공작기계의 위치오차 측정

Abstract

본 연구에서는 다축 공작기계의 기하오차 중 공작기계의 조립정확도로 기인하는 위치오차를 터치트리거프로브와 기준구를 이용하여 단시간에 규명 할 수 있는 일반적인 측정방법을 제안한다. 먼저, 다축 공작기계의 기준좌표계에서 4축 및 5축 공작기계의 모든 위치오차를 정의하고, 공작기계를 강체구조로 가정한 뒤 동차변환행렬을 이용하여 기준좌표계 원점에서부터 회전축에 장착된 기준구의 위치 및 공구축에 장착된 터치트리거프로브의 위치를 각각 모델링한다. 이 모델링 과정으로 터치트리거프로브로 측정된 기준구 중심점은 이미 정의한 다축 공작기계의 모든 위치오차 요소로 수식적으로 표현되며, 이 수식을 이용하여 모든 위치오차 요소를 분리할 수 있는 측정수식이 유도할 수 있다. 이 측정방법에서 4축 공작기계의 경우, 공작물을 높이방향으로 이동시킬 수 없기 때문에 높이가 다른 두 기준구 또는 하나의 기준구의 높이를 조절하여 사용하며, 5축 공작기계는 틸팅회전축을 이용하여 기준구의 높이방향을 조절이 가능하므로 하나의 기준구를 사용한다. 특히 수평형 5축 공작기계의 경우, 틸팅회전축의 각도에 따라 터치트리거프로브의 접근성에 제한이 발생하므로 수평형 공작기계에 적용이 가능한 측정방법을 별도로 개발하였다. 제안된 측정방법을 검증하기위해 4축 및 5축 공작기계를 대상으로 측정실험을 진행하였다. 수직형 및 수평형 4축 공작기계를 대상으로 유도한 측정수식을 이용하여 4축 공작기계의 모든 위치오차를 분리시킬 수 있었다. 측정결과는 정밀 직각미러와 정전용량형 변위센서를 이용한 직각도오차 측정결과 및 제어기 보정을 이용한 보정실험 전/후의 위치오차 측정결과를 비교하여 각각 검증하였다. 또한, 수직형 5축 공작기계를 대상으로 수직형 및 수평형 측정방법을 각각 적용하여 각 측정방법으로 분리된 위치오차를 비교하여 각 측정방법을 검증하였고, 터치트리거프로브와 기준구 사이에서 발생하는 오차벡터 수식을 제어기에 입력하여 보정 전/후의 위치오차 결과를 비교하여 위치오차 결과를 검증하였다.; The demand for workpieces with high geometric complexity and the increase in machining productivity have led to an increase in the use of multi-axis machine tools. One of the major sources of error affecting the accuracy of multi-axis machine tools is location error. The location errors consist of squareness errors and offset errors and these errors depend on how accurately the machine tool was assembled. It is difficult to measure location errors since multiple axes must be controlled simultaneously, and the errors can only be measured on fully assembled machines. However, most measurement methods for location errors are focused on measuring the location errors of the rotary axes, and they require changing several measurement setups. The objective of this study is to develop a measurement method for all location errors using a touch trigger probe and sphere artifacts with a measurement setup. The measurement method was developed using a rigid-body kinematics model and a homogeneous transformation matrix (HTM). The equations for separating all location errors of a 4-axis and a 5-axis machine tool are induced repectively. Since a 4-axis machine tool cannot adjust the height direction of the workpiece, the measurement method was developed using two sphere artifacts with different heights or using a sphere artifact by adjusting the height. However, the measurement method for a 5-axis machine tool uses a sphere artifact because of a tilting rotary axis of a 5-axis machine tool. The verification experiments were performed on the location errors of a vertical and horizontal 4-axis machine tools, respectively. Total 7 location errors of each 4-axis machine tool were fully separated. The measurement results of a vertical 4-axis machine tool were verified by comparing the squareness errors using a precision square mirror and the maximum difference between the results was about 18.3 . Also, the error results of a horizontal 4-axis machine tool were verified by comparing the results before and after NC compensation. Through the compensation, the maximum squareness and offset errors decreased from -17.3 ard and -7.2 m to 6.9 rad and 3.1 m respectively. The reliability of the location error results was confirmed by the uncertainty analysis. It was also confirmed that the height difference between the sphere aritifacts was the most significant factor to the location error measurement results. In case of a 5-axis machine tool, two measurement methods were developed depending on the type of machine tools. The accessibility of the touch trigger probe is limited at some tilting rotary positions due to the structural characteristics of a horizontal 5-axis machine tool. The measurement method for a horizontal 5-axis machine tool(2 rev. method) is to measure at the tilting rotary position 0 deg. and any tilting rotary position accessible by the touch trigger probe(+a or a deg.) by assuming the setup error of the sphere artifact can be minimized. On the other hand, the measurement method of the vertical 5-axis machine tool(3 rev. method) is to measure at the tilting rotary position of 0 and a deg.. To verify the measurement methods, measurement experiments were conducted on a vertical 5-axis machine tool. The maximum standard deviations between the location error results calculated by each method were 5.0 rad in squareness error and 4.3 m in offset error. The location error results were verified by comparing the results before and after NC compensation. Through the compensation, the maximum squarness and offset errors decreased from 89.5 rad and 28.4 m to -11.2 ard and 13.1 m, respectively. The reliability of the position error results was confirmed through the uncertainty analysis and it was also confirmed that the tilting angle was the most significant factor to the location error measurement results of 2 rev. method. Consequently, the proposed methods can identify all location errors without expertise skills in the measurement setup and reduce the loss of machining time due to minimizing the number of measurement setups.