Title :
Methods for detecting errors in numerically controlled machining of sculptured surfaces
Author :
Jerard, Robert B. ; Drysdale, Robert L. ; Hauck, Kenneth ; Schaudt, Barry ; Magewick, John
Author_Institution :
Dept. of Mech. Eng., New Hampshire Univ., Durham, NH, USA
Abstract :
The numerically controlled (NC) machining of sculptured surfaces is an error-prone process often requiring several attempts before an error-free NC program is produced. A technique designed to move the NC proof process into software is described. The system outputs a color-coded graphics display of the machined surface that shows out-of-tolerance areas. To gain efficiency, surface curvature and cutting-tool size are used as inputs to a surface discretization algorithm, which guarantees that a user-defined level of simulation accuracy is achieved. The simulation time grows linearly in both desired accuracy and in the number of tool movements. In typical test cases, NC programs for complex automotive body panels were simulated and verified in CPU times that ranged between 5 and 30 minutes.<>
Keywords :
automobile industry; computerised numerical control; digital simulation; engineering graphics; program verification; CPU times; NC proof process; color-coded graphics display; complex automotive body panels; cutting-tool size; error detection methods; error-free NC program; error-prone process; machined surface; numerically controlled machining; out-of-tolerance areas; sculptured surfaces; simulation accuracy; simulation time; surface curvature; surface discretization algorithm; tool movements; user-defined level; Application software; Automotive engineering; CADCAM; Computer aided manufacturing; Computer errors; Error correction; Machining; Mathematical model; Solid modeling; Testing;
Journal_Title :
Computer Graphics and Applications, IEEE
