Surface roughness parameters determination model in machining with the use of design and visualization technologies

A virtual reality machine shop environment was developed for the determination of critical quantitative and qualitative machining process parameters. Material removal and milling machine axes kinematics are simulated in real time. The G-Code program serves as the input to the system. A graphical model for the calculation of quantitative data affecting the machined surface roughness was developed. In the model, the machined surface topomorphy is derived as a cloud of points, retrieved from the visualization system Z buffer. The current study is focused on the verification of the quantitative data acquired by the system. A numerical model experimentally verified in the past was integrated to the system to directly evaluate results accuracy. The model accuracy was also verified with results determined in cutting experiments. The calculated surface roughness values were found to be in agreement with the values determined from the numerical model and the experiments. mentation of production plans in the virtual environment, aiming at errors detection in the executed operations. Kim et al. (2000) presented a cutting forces determination model for milling processes of sculptured surfaces with ball-end cutters. In order to determine chip thickness from the intersection area between the cutter and the workpiece, a graphical method based on the Z-map was developed. Roth et al. (2003) presented a cutting forces determination model for milling processes in multi axis machines. The model is based on the Z buffer and exploits the system graphics card. 3 VIRTUAL ENVIRONMENT FOR MACHINING PROCESSES SIMULATION A virtual environment was developed using commercial software tools for the realistic visualization of machining processes. The structure of the virtual environment is shown in Figure 1. A complete machine shop is being visualized and the functional characteristics of a three axes CNC milling machine are being simulated. In Figure 2 the virtual environment for machining processes simulation is presented. The virtual environment has the following functional characteristics:  Objects behavior is as realistic as possible.  User has the ability to interact with all the objects in the virtual environment.  User selects cutter from the cutters table and it is installed automatically in the CNC machine spindle.  User selects workpiece from the workpiece table and it is installed automatically in the CNC machine worktable.  During machining process simulation, CNC machine axes and cutter animate in a realistic way.  Workpiece material removal is being visualized in a realistic way during machining process simulation.  Machining process information like the G code command simulated in the CNC machine, feed, spindle speed and cutter path are being visualized in a special table.  User can pause, stop or restart the machining process.  When the machining process is completed, quantitative data and graph charts related to the machined surface roughness are being visualized, according to user preferences. 4 VIRTUAL ENVIRONMENT SOFTWARE MODULES The following modules of the environment were developed and integrated in the employed Virtual Reality platform for the implementation of the machining processes simulation environment:  Geometrical models visualization: Models developed in CAD like the machine shop environment and the machines, dynamic geometry models like the workpiece, 3D text and graphs, virtual models visualization characteristics, like color and texture.  Geometry models hierarchy, constraints, interaction attributes.  CAM system integration.  Cutter path determination according to the G code program.  CNC machine axes and cutter animation.  Workpiece material removal visualization during machining processes simulation. Figure 1. Structure of the virtual environment for machining processes simulation Figure 2. Virtual Environment for machining processes simulation  Quantitative data determination.  Verification checks during machining processes simulation. The current study focuses on the quantitative data capabilities of the system and the process for their verification. 5 QUANTITATIVE DATA DETERMINATION A model was developed for surface roughness quantitative data determination. In the model, parameters such as the cutting speed, feed, cutting depth, cutter diameter, height, cutter type, number of teeth and cutting edges geometry that contribute in surface roughness formation are considered. Parameters that contribute in surface roughness formation, such as the cutter material, quality and type of the cutter, cutter wear, quality of jigs, fixtures, the use of lubricant, vibrations in the machining process are not considered. 5.1 Determination of machined surface topomorphy The model for the determination of the machined surface topomorphy was implemented in a three dimensional graphics environment developed in OpenGL (Fig. 3). The simulated motion of the cutter and the cutting conditions are obtained directly from the G code file. The sweep surface produced by the cutting edges is being determined. Cutting edges shape is defined from the outer edge profile of each cutting edge (Fig. 4). Each cutting edge is approximated by equal elementary sections that are straight lines (Fig. 5). The produced sweep surface from each cutting edge is approximated by triangles. Therefore there are overlapping triangles, since part of each cutting edge sweep is being overlapped by the next cutting edge sweep or the next cutter pass sweep. If the cutter sweep surface is projected from its down side, the final machined surface topomorphy is derived, since the overlapped triangles are not visible in this projection, due to the hidden line algorithm, that projects to the user only the geometry visible in each point of view. This final machined surface topomorphy is derived from this projection in the form of cloud of points. The coordinates for the cloud of points are determined. The pixels within workpiece limits visualize the machined surface. These pixels are converted into X and Y coordinates in the graphics environment coordinate system. Z coordinate is derived from the visualization system Z buffer. The coordinates for the cloud of points describing the machined surface topomorphy is exported in a text file and used to calculate quantitative parameters for the machined surface roughness in the virtual environment. 5.2 Calculation and visualization of surface roughness parameters in a virtual environment A table was developed for surface roughness quantitative data visualization (Fig. 6). On the upper part of the table the measurement plane on the workpiece machined surface is defined. On the lower part of the table the selected measurement plane topomorphy and the respective surface roughness parameters values are being visualized. For the determination of the measurement plane, a handler was developed (Fig. 7). The handler defines a plane vertical to the machined surface in which topomorphy will be determined and surface roughness parameters will be calculated from this topomorphy. System user defines the position of each handler end. Handler ends define surface roughness measurement plane limits. For surface roughness parameters determination, cloud points near the vertical measurement plane are retrieved from the file. The user has the ability to create a machining process report in the form of a text file (Fig. 8). In the machining process report file, surface roughness data for a region around the measurement plane are being stored and can be acquired for further use. This process can be repeated in any other region of the machined surface. 6 MODEL VERIFICATION In this section the process for the verification of the developed surface roughness quantitative data determination model is described. The process is shown in Figure 9. The results acquired by the model were verified with the experimentally verified machining simulation numerical model MSN (Milling Simulation by Needles), which was integrated in the OpenGL software in order to provide direct control on model results, when implementing experiments is not possible. Moreover, quantitatively and qualitatively model results were verified with experimental data. 6.1 Verification of the machined surface topomorphy with the MSN model The accuracy of the developed model for machined surface topomorphy determination is influenced by the following parameters:  Cutter model discretization density.  Cutter trajectory discretization density.  Hidden lines algorithm employed by the system. Figure 3. OpenGL machining processes simulation environment for machined surface topomorphy determination (a) Edge on a ball end mill (b) Six cutting edg-