Abstract: In recent years, due to intensified global competitiveness in die machining, further reduction of lead time, lower costs and higher accuracy are in demand. 5-axis machining is receiving attention as a means to solving these issues. However, differing to the conventional 3-axis machining, complex thinking tasks such as determining tool orientation while considering interference are required in the process design phase prior to machining, which not only places pressure on even those experienced workers, but also consumes a significantly longer amount of time than that required for the actual machining itself. Accordingly, with the aim of shortening the overall lead time from machining process design through to machining itself, we have developed a support system which makes highly efficient indexing 5-axis machining process design possible in a short time even for inexperienced workers. This report presents the algorithms contrived and calculation test results.
Abstract: As the feed rate of cutting tools is influenced by the quality of NC programs, this causes two problems when cutting freeform surfaces using NC machine tools, one is the feed rate does not reach the specified value, and the other is the quality of the machined surface decreases due to the large acceleration and deceleration of the control axes. One reason for these problems is that the method to generate cutter location data does not take into consideration the sampling time of the NC controller. To improve these problems, the approximation of offset curves with a series of circular arcs has generally been proposed. But conventionally, circular arc interpolation algorithms are complicated and not rational. So, in this study, we proposed two simple NC program generation methods. (1) Approximate the offset form with circular arcs based on the location of the center of curvature. (2) Make the distance between the cutter location points of the NC data an integral multiple of the “feed rate command F × sampling time T”. We also confirm the effectiveness of the proposed methods by simulations and experiments.
Abstract: While it is important that evaluation of thickness of a product made with press working to evaluate a mold, currently, there is not an appropriate system which can measure thickness of a free curved plate. The study deals with a development of an automatic measurement and evaluation system for thickness of free curved plates, called Orthros. The system measures thickness of a plate with laser displacement gauges and an industrial robot. While the system has been capable for measurement of a relatively simple shaped workpiece, it had collisions between laser-path and workpiece for that of a complicated shape, owing to a geometrical limit about measuring posture of the gauges. In this article, a new method to detect collision and to avoid collision by changing measuring posture on the basis of CAD data is proposed. In the method, not only the geometrical characteristic of the laser-path but also the measuring error characteristic of the gauges depending on the measuring posture are considered and continuous measuring path having change in posture to avoid collision is devised. From the experiment result, it is confirmed that the proposed method has an ability to measure a thickness of a workpiece with complicated shape.
Abstract: We present a framework for reuse of digital mock-up models. Our idea is to find reusable models in the archives by shape matching. Since the reusable models are already designed by operators in the past projects, we can easily create various types of data by using existing models. Indeed, we demonstrate this framework is efficient for surface simplification and CAE mesh generation problems.
Abstract: Reverse engineering systems are used to construct mathematical models of physical models such as clay model based on measurement data. In this study, we proposed a reverse engineering method which can construct high quality surface data automatically. This method consists of the following steps; The first globally and regionally smooths measured data based on the target shape by fitting quadric surface to measurement data. The second defines quadric surfaces and converts measurement points into 3D lattice points to obtain uniform measurement data density. As the positions of measurement data are converted from coordinate values into 3D lattice points, it is easier to find neighboring points and clarify neighboring relations between surfaces. The third acquires segment measurement data based on maximum curvatures and normals at each point. The last defines NURBS surfaces for each segment using the least square method to average positional errors. In order to validate the effectiveness of the proposed method, we developed a reverse engineering system and constructed mathematical models through basic experiments using clay car model measurement data.
Abstract: Computer simulation of the cutting force change in a milling process is important for realizing more stable, precise, and efficient metal cutting. In the latest cutting force simulation method, a cutter engagement feature representing a contact area between the milling cutter and the workpiece must be extracted for each small motion of the cutter. In this paper, a new algorithm is proposed for computing a cutter engagement feature based on the dexel representation of the workpiece. Most prior studies compute the feature based on the cutter removal volume of the workpiece. Differently from them, our technology to propose computes the cutter engagement feature based on the contact analysis between the cutter and dexels representing the workpiece shape. An experimental system is implemented and some computational experiments are performed. Our system can achieve the total cutter engagement feature extraction of a complex 3-axis NC milling process in a few minutes.
Abstract: This report describes a two-dimensional monitoring system for milling machines using small-diameter tools. Small-diameter tools are used for high-speed milling, and they are indispensable for high-quality and high-productivity manufacturing. However, tool breakage occasionally occurs, and this becomes a serious problem in automated production. This study aims to develop a system for monitoring the cutting state that can prevent tool breakage. The proposed system consists of two charge-coupled device (CCD) cameras, an image processing device, a man-machine controller (MMC), and a machining center with an open computer numeric (CNC). This monitoring system is connected to the machining center by Ethernet. It enables the precise measurement of tool deflection during high-speed milling. In an experiment, we applied this system to the end milling of a steel plate under different cutting conditions, and we examined the relationship between tool deflection and the cutting conditions. In addition, we applied this monitoring system to measure tool wear, and we examined the relationship between tool deflection and tool wear. It was found that the proposed system enabled in-process monitoring of the cutting state and tool wear.
Abstract: Detection of a tool fracture is necessary to ensure cutting accuracy and to avoid a tool breakage because tool fracture is one of the significant prediction signals of the tool breakage. For monitoring the tool condition, generally additional sensors are used. However using these sensors causes high cost and increase of failure rate. In this paper, a novel sensor-less detection method of tool fracture in drilling process is proposed on the basis of a disturbance observer theory. It is applied to the x-y stage of the machine tool. The proposed method requires no external sensor because it uses only the servo information of the spindle control system. Since structures of normal drills with two floats are symmetrical with respect to a point, theoretically the cutting force in the x and y directions does not work. When the drill is fractured, its structure becomes asymmetry so that unbalanced forces would exert in the x and y directions at intervals of the spindle speed. Therefore, it is possible to detect a tool fracture by the frequency analysis of estimated disturbance force with a wavelet transform. The experimental results show that the proposed method is available for detection of the small tool fracture effectively.
Abstract: Thermal shrink fit chucks are widely used in high performance machining where excellent concentricity and high torque transmission are required. It was reported that in those milling operations, severe damage of tools, workpieces, and also machine tools occurs due to an extraction of the milling tool out of the shrink fit chuck during the process. Although, theoretically the interference fit assembly should withstand certain process forces, milling tools are apparently pulled out under special process conditions. The resulting increase of the cutting depth often leads to tool overload and breakage. So far, the phenomenon of tool extraction could not be explained. This paper presents an experimental approach of the investigation of the phenomenon of axial tool extraction. Therefore, a unique type of test rig for main spindles and tool interfaces is used. Experimental investigations on dynamic force and torque combinations leading to tool extraction are described. Results show, that the holding force is not only affected by geometrical parameters of the shrink fit chuck, but also by the applied dynamic load.
Abstract: In this study, we examined influencing factors on the identification accuracy and improving method of the identification accuracy. As a result, the influence of the condition of the dynamic characteristic values is large and these values should correspond to the tangential and vertical directions. The influence of treating dynamic characteristic parameters of the same plural joints is also large, and then the parameters of the same plural joints should have the same value. In addition, the tangential dynamic characteristics should be identified from the vibration mode which has large tangential relative displacement between the joint interfaces. Similarly, the vertical one should be identified from the vibration mode which has large vertical relative displacement between the joint interfaces. Finally, the identification accuracy can be improved greatly by considering these influences.