Integrated solid modeller based solutions for machining

https://doi.org/10.1016/S0010-4485(00)00042-7Get rights and content

Abstract

Most existing machining simulation software is geometric in nature, and ignores the physical aspects of the process. This paper reports research progress towards developing a comprehensive physical machining process simulation program based on a solid modelling kernel. Both flat and ball end tools are modelled, and multi-axis motion is supported. A Finite Element Analysis model is used to predict both tool and part deflection. A special sweep representation is used for Boolean part model updating. The system can be used for prediction as well as for intelligent factory floor control. Several examples of completed work are included to illustrate developed techniques. A discussion section presents computation time data, and outlines how parallel processing could be utilised.

Introduction

Solid modellers are well established for Computer Aided Design (CAD) and Computer Aided Engineering (CAE) of mechanical parts. Common applications include part modelling, assembly modelling, and Finite Element Analysis (FEA). Computer Aided Manufacturing (CAM) software is also available, but generally is limited to geometric tool path planning. Process simulation at the geometric level is available, but the physical aspects of the process are ignored. Machining success, however, depends on consideration of cutting forces, torques, part and tool deflection, chatter, tool breakage, and tool wear.

This paper reports research progress towards developing a comprehensive physical machining process simulation program for use in process planning applications, as well as factory floor monitoring and control. The system is based on extensions to the ACIS solid modelling kernel. To use the system, an industry standard Cutter Location Data (CL-DATA) file is first created using commercial CAM tool path planning software. This file and a solid model representing the shape of the part before machining are presented as inputs to the simulation program. The user interactively enters machining process constraints such as maximum force, deflection, etc. During simulation, each tool path motion is subdivided into short steps along the path, and angular steps of spindle rotation. The granularity of the steps, and the complexity of process model, are chosen to suit the particular application. At each simulation step, the geometric immersion of the cutting tool edges with the part is calculated. This information is used by the machining process simulator to calculate physical machining process quantities such as cutting force, torque, tool deflection and finished part surface error.

Models for both flat end and ball nose tools are implemented, and multi-axis motion is supported. Flexible tools are modelled as cantilevers. Flexible parts are created as solid models, are meshed using the ACIS cellular topology husk, and are solved using a specially written dynamic FEA program. The mesh is periodically updated as material is machined away from the part. A stitched sweep representation is then used to generate the solid necessary for Boolean subtraction and part model updating.

The machining feed rate is adjusted to maximise the material removal rate (MRR) subject to imposed constraints such as tooth chip load and cutting force. Expected spindle power is estimated, and is used for online tool wear monitoring. Special departure paths are pre-calculated so that, in the event of sudden tooth breakage, undesirable dwell marks can be minimised by smoothly moving the tool clear of the part. This monitoring and control information is passed to a supervisory Computer Numerical Control (CNC) computer in the form of special comment tags. The supervisory computer compares expected values with online sensor readings and modifies the tool motion as required.

The remainder of the paper is organised as follows. Section 2 reviews major contributions from previous work. Section 3 describes a proposed machining process planning, simulation, and control architecture. The next two sections present examples to illustrate the major contributions of the paper. Section 4 illustrates a pocket machining simulation integrated with online monitoring and control. Section 5 illustrates offline, solid modeller based simulation for side milling of complex shaped, flexible parts. A brief discussion appears in Section 6, followed by a final summary in Section 7.

Section snippets

Literature review

Numerous applications involving solid modeller based preprocessing of Numerical Control (NC) programs, and integration with factory floor control, have appeared in the literature. A chronological overview of previous major work is provided in this section.

Early research by Voelcker and Requicha [1] was purely geometric in nature. The intent was to verify that the tool path would produce the desired part shape without gouging. Later work using the z buffer technique was presented by Van Hook [2]

Machining system architecture

In modern practice, solid modeller based CAD/CAM/CAE software is widely used for mechanical design, analysis, and manufacturing. Each part is designed as a solid model. FEA is possibly performed. Tool path planning software is then used to prepare a CL-DATA file for machining. A geometric verification and optimization software step may be included prior to the file being post processed into a machine tool specific part program. For this paper, the machine tools used conformed to the EIA1

Geometric simulation

As stated earlier, commercially available tool path planning software is used to generate the CL-DATA file. This file, along with an in-process stock model of the part prior to machining, is input to the geometric simulator. The geometric simulator then determines the immersion geometry experienced by the tool teeth as the part is machined. This provides the information necessary for machining process simulation. After immersion calculations for the tool path are complete, a Boolean subtraction

Side milling of flexible parts

In applications such as turbine blade machining, the part shape is thin and flexible. This causes significant part surface dimensional error due to dynamic interaction. In this section, a methodology for predicting the part and tool dynamic response in multi-axis side milling is discussed. The cutting tool is modelled as a cylindrical cantilever. Because of the complex shape, a solid model of the part is used. The solid is automatically meshed to create an ACIS cellular topology based finite

Discussion

An important issue for complex machining process simulation is the computation time required. The work herein was programmed for a Sun Sparc Ultra 1/140 workstation with 96 MB RAM. Runs were executed during periods when no other applications are running.

For the 2 1/2 D spindle power simulation, execution time was 50 s. This included static deflection cutting force prediction [46]. The actual machining time was 2 min. On a larger part test, execution time was 10 min, and actual machining time was 20 

Summary

This paper has presented two applications that integrate solid modelling with milling process simulation. The first application illustrated geometric and milling process simulation combined with online monitoring and control for a 2 1/2 D pocket milling application. It was shown that the tool immersion could be determined by classifying a semi-circular arc with respect to the part surface. Cutting forces, torque, and power could then be calculated using a static milling process model. To

Acknowledgements

The authors wish to thank the Natural Sciences and Engineering Research Council (NSERC) of Canada for financial support. The reviewers provided many helpful comments, and additional references.

Allan D. Spence received his BMath degree in Applied Mathematics from the University of Waterloo in 1984. He received his MASc degree in Mechanical Engineering, again from the University of Waterloo, in 1986. He received his PhD in Mechanical Engineering from The University of British Columbia in 1992. In 1994 he joined the Department of Mechanical Engineering at McMaster University, and was promoted to Associate Professor in 1999. He has industrial experience in mechanism design, vacuum and

References (56)

  • X. Liu

    Five-axis NC cylindrical milling of sculptured surfaces

    Computer Aided Design

    (1995)
  • H.B. Voelcker et al.

    Geometric modeling of mechanical parts and processes

    IEEE Computer

    (1997)
  • T. Van Hook

    Real-time shaded NC milling display

    Computer Graphics, Proceedings ACM SIGGRAPH

    (1986)
  • J.P. Menon et al.

    Advanced NC verification via massively parallel raycasting

    Manufacturing Review

    (1993)
  • I.T. Chappel

    The use of vectors to simulate material removed by numerically controlled milling

    Computer Aided Design

    (1993)
  • M.E. Merchant

    Mechanics of the cutting process I

    Journal of Applied Physics

    (1945)
  • M.E. Merchant

    Mechanics of the cutting process II

    Journal of Applied Physics

    (1945)
  • E.H. Lee et al.

    Theory of plasticity applied to the problem of machining

    Journal of Applied Mechanics

    (1951)
  • M.E. Martelloti

    An analysis of the milling process

    Transactions of the ASME

    (1941)
  • M.E. Martelloti

    An analysis of the milling process. Part II: Down milling

    Transactions of the ASME

    (1945)
  • S. Smith et al.

    An overview of modeling and simulation of the milling process

    Transactions of the ASME, Journal of Engineering for Industry

    (1991)
  • Milling handbook of high-efficiency metal cutting

    (1980)
  • P. Bertok et al.

    A system for monitoring the machining operation by referring to a predicted cutting torque pattern

    Annals of the CIRP

    (1983)
  • CA CGTech, Irvine.http://www.cgtech.com/optipath.htm,...
  • J. Tlusty et al.

    Dynamics of cutting forces in end milling

    Annals of the CIRP

    (1975)
  • L.A. Kendell et al.

    Intelligent supervisory control prototype for machining systems

    (1988)
  • B.K. Fussel et al.

    Computer generated CNC machining feedrates

    (1992)
  • A.D. Spence et al.

    A solid modeler based milling process simulation and planning system

    Transactions of the ASME, Journal of Engineering for Industry

    (1994)
  • Cited by (61)

    • Frame-sliced voxel representation: An accurate and memory-efficient modeling method for workpiece geometry in machining simulation

      2017, CAD Computer Aided Design
      Citation Excerpt :

      Section 6 provides quantitative comparison results to demonstrate the improvements the FSV-rep can bring over other modeling methods used in machining simulation. Solid modeling based machining simulation typically uses the boundary representation (B-rep) modeling scheme [3–5]. B-rep is a way of modeling an object by using surfaces and curves to represent the boundary of the object.

    • Mechanistic modeling of five-axis machining with a general end mill considering cutter runout

      2015, International Journal of Machine Tools and Manufacture
      Citation Excerpt :

      Comparatively, solid modeling methods can improve the calculation accuracy. Spence et al. [16] identified CWE boundaries using a CSG based process simulation system to predict cutting forces. Larue and Altintas [17] extracted the cutter immersion angles for flank milling by intersecting the cutting tool with the workpiece in ACIS solid modeling environment.

    View all citing articles on Scopus

    Allan D. Spence received his BMath degree in Applied Mathematics from the University of Waterloo in 1984. He received his MASc degree in Mechanical Engineering, again from the University of Waterloo, in 1986. He received his PhD in Mechanical Engineering from The University of British Columbia in 1992. In 1994 he joined the Department of Mechanical Engineering at McMaster University, and was promoted to Associate Professor in 1999. He has industrial experience in mechanism design, vacuum and injection mould making, and computer-aided design. His current research program concentrates in computer-aided design and manufacturing, with particular emphasis on manufacturing process simulation and coordinate metrology. Dr Spence is a member of the Professional Engineers of Ontario, and is currently Chapter Chairman for Hamilton Chapter 42 of the Society of Manufacturing Engineers (SME).

    Farid Abrari received his MEng (1994) and PhD (1998) degrees from McMaster University. His research interests are in application of FEA to manufacturing engineering problems. Dr Abrari is currently employed as a research engineer at McMaster University.

    M.A. Elbestawi received his MEng (1976) and PhD (1980) degrees from McMaster University. He joined Ontario Hydro in 1979 as a Research Engineer, and was subsequently promoted to Unit Head specializing in structural dynamics and condition monitoring of power plant equipment. In 1986 he joined the Department of Mechanical Engineering at McMaster, where he is currently Professor and Chair of the Department. Dr Elbestawi currently holds the NSERC Industrial Research Chair in Precision Machining. His research interests are in Manufacturing Engineering, specifically machine tools, metal cutting, and computer-aided manufacturing. He is a Fellow of the American Society of Mechanical Engineers (ASME) a Fellow of the Canadian Society for Mechanical Engineering (CSME), an active member of CIRP, a senior member of the Society of Manufacturing Engineers (SME), and a member of the Professional Engineers of Ontario.

    View full text