Search

The paper describes a new method for a finite element based pseudo-simulation of Numerically Controlled (NC) machining (material removal) processes. Industrial machining of a component usually results in warping or distortion due to the re-establishment of equilibrium in the retained part along with the relief of the insitu residual stresses in the removed part. In many cases, these distortions can be so large that the part may no longer be able to serve its designated functionality. Considering that the machining process is fundamentally a material removal process, a new method based on an automated removal of finite elements in the cutting area has been developed in the finite element analysis (FEA) software MSC.Marc to conduct pseudo-simulation of the NC machining process., A number of key software enhancements have been made to facilitate the pseudo-simulation of the NC machining process. First, a seamless interface has been developed to import APT/CL data generated by CAD/CAM systems. Then, the cutting paths have been generated based on information in the APT/CL files and used for the automatic detection of the intersection between the cutter and the finite element mesh. With each incremental motion of the cutter, the FEA solver detects all the elements that are located within the cutting path. Such elements are then deactivated in a step-by-step manner that is consistent with the actual machining process. In order to improve the fidelity of the cut area, local adaptive mesh refinement in the vicinity of the cutting tool is undertaken. This enables relatively coarser meshes away from the cut area and provides more accurate representation of the actual volume that is removed. As demonstrated by an industrial example, the enhanced software features in MSC.Marc have made it possible to practically and efficiently analyze complex machining processes of 3D production parts and provide an elegant tool for predicting distortions in large structures due to the relief of insitu residual stresses. © 2004 American Institute of Physics [ABSTRACT FROM AUTHOR]

This paper discusses various computational tools developed for the numerical simulation of welding processes. The tools can be broadly classified into two groups: Process related tools that facilitate simulation accuracy; Speed and ease-of-use related tools that facilitate simulation and operational efficiency. Process related tools that are discussed include the weld heat flux specification, the weld path and weld fillers that are created in conjunction with the moving heat source. Some of the issues associated with the weld heat flux computation are: correct choice of weld flux expression depending on the type of welding process; an appropriate scaling of heat input for 2D finite element simulations and an accurate numerical integration of the weld heat flux expression. Some of the issues associated with the weld path computation are: Options necessary for accurate and convenient specification of the weld path and arc orientation and the calculation of the local coordinate system of the moving source. Some of the issues associated with the weld filler element computation are: techniques available for the treatment of weld fillers in an analysis; boundary conditions for filler elements and multi-pass welds. Speed and ease-of-use related tools that are highlighted include parallel welding computations using the domain decomposition technique, local adaptive meshing schemes, 2D to 3D state variable transfer and global — local structural zooming. This paper describes several algorithms that have been recently developed and implemented in the commercial, non-linear finite element software MSC.Marc. © 2004 American Institute of Physics [ABSTRACT FROM AUTHOR]