Modeling and Requirements of the Automated Deburring Process

Abstract: An increasing number of robots have been used for the finishing of castings during recent years. This is mainly due to developments to overcome problems that restricted their use during the past decade. The major difficulties have been the need to control reaction forces during metal removing by grinding operations and the lack of tools designed specially for robots. The elimination of these two problems, makes robot cells with high efficiency and good quality products possible. This thesis focuses on requirements of automated deburring process and models necessary to control the process. There may still exist problems that should be solved before these robot cells become feasible. These problems can be divided into two groups. The first group consists of the problems related to the shape of the workpiece. Grinding and deburring is used in a vast domain of products with different types of materials that can take practically any shape. This requires a rather long programming time. Therefore developing methods for simplifying and decreasing the required time for programming jobs are desirable. Improved programming methods increase the flexibility and make the automation economically viable even for small batch production. Improvement of off-line programming software and availability of efficient vision systems has been a great step to achieve this. The second group of problems is related to choice of the grinding methods and equipment. The possibility of choosing among the different available tools such as grinding discs or wheels and rotary files with different sizes and hardness necessitates enough knowledge about the process to find the best solution. This directly affects the production's economics and products' quality. Investigations show that developing a process model for robotic grinding is both possible and desirable. The process model is the basis for selection of the suitable grinding machine (power and speed) and disc (grit size, thickness and hardness). The process model also can be used for selecting an optimized operating point, i.e. establishing the relationships between material removal rate, wheel wear and the process operating data, i.e. feed rate and cut depth.