Elsevier

Computer-Aided Design

Volume 34, Issue 5, 15 April 2002, Pages 391-403
Computer-Aided Design

Compensating for systematic errors in 5-axis NC machining

https://doi.org/10.1016/S0010-4485(01)00111-7Get rights and content

Abstract

The errors introduced during 5-axis machining are higher than the intrinsic repeatability of the machine tool. It is possible to identify such systematic errors and compensate for them, thus achieving higher performance. A group of systematic errors can be compensated for directly in the inverse kinematics equations. Other systematic errors can be combined and compensated for through the total differentials of the inverse kinematics relations. A new general approach on how to compensate for the systematic errors based on the closed loop volumetric error relations is presented. The errors due to the 5-axis toolpath generation in current CAD/CAM and CNC are analyzed in detail. A new strategy for measurement and compensation is presented.

Introduction

Five degrees of freedom are the minimum required to obtain maximum flexibility in tool workpiece orientation. This means that the tool and workpiece can be oriented relative to each other under any angle. To orient two rigid bodies in space relative to each other, 6 degrees of freedom are needed for each body (tool and workpiece) or 12 degrees. However, any common translation and rotation which does not change the relative orientation is permitted reducing the number of degrees by 6. The distance between the bodies is prescribed by the toolpath and allows us to eliminate an additional degree of freedom, resulting in a minimum requirement of 5 (rigid) degrees. If a coordinate system is fixed to each of the five bodies implementing the 5 degrees of freedom in the 5-axis machine tool, there can be six errors per body or 30 errors. The errors due to the relative distances between the five bodies must be added to these. Five rigid bodies can be connected rigidly to each other by nine rigid bars. These additional possible degrees of freedom result in nine more error components. So, from the above, it can be seen that there are 39 independent error components in a 5-axis machine or 18 more than in the case of a 3-axis machine. A new general error model based on the kinematics of the 5-axis machine will be developed. To the above errors an additional component due to the interpolation in the CAD/CAM system and the machine control must be added.

Section snippets

Error models

Ramesh et al. [1] reviewed the current state of research in the field of error compensation for machine tools. The sources of errors and the methods to eliminate errors are reviewed. The review focuses in detail on the error measurements and compensation used in the past. From this review it is clear that the first part in error compensation, i.e. modeling, has been very much neglected in past research. Past research focussed on measurement and compensation and not on modeling. This is

Volumetric error concept

Fig. 1 illustrates the concept of closed loop volumetric error. The Cutter Contact point, or CC point, is the point where the tool envelope touches the theoretical surface to be machined. It is represented in Fig. 1 as O and O′. Point O represents the CC point on the tool side. Point O′ represents the corresponding point on the reference surface. Point O and O′ coincide only in discrete locations (Fig. 5). The difference in the locations O and O′ is not shown in Fig. 1. Due to various error

Machine taxonomy and kinematic chain diagram

A 5-axis machine is a milling machine (Fig. 3) with 5 degrees of freedom: three translatory movements X, Y, Z (in general represented by TTT) and two rotational movements AB, AC or BC (in general RR). The kinematic chain can be represented in a simplified way. Fig. 2 is the representation of the axis configuration in Fig. 3.

Inverse kinematics

The CAM system generates a toolpath in a machine independent coordinate system fixed to the workpiece. The transformation from workpiece to machine coordinates is called the

General mathematical model

The approach used by Srivastava et al. [5] can be simplified by total differentiation of the kinematic model of the machine. The volumetric errors can be expressed analytically as the total differential of the inverse geometry transformation equations. This approach is general and applicable to each type of 5-axis kinematics. The workpiece coordinates differentials dx1, dy1, dz1, dI1, dJ1, dK1 from Eq. (1) can be expressed in function of the machine coordinate differentials dx4, dy4, dz4, dA, dB

Case study of a specific 5-axis machine

The general approach will now be demonstrated for a specific 5-axis machine shown in Fig. 3.

Error measurement

A ball bar approach in the case of a 5-axis machine can be adopted. However, the many possible orientations due to the rotations will require avoidance of interference with the machine and result in many reference point changes. Therefore, an approach based on tuning or reference pieces was chosen.

Each of the five rigid bodies in a 5-axis machine has six error components, three translatory and three rotational errors. Each of the errors of these five bodies will influence each other. The

Conclusion

An overview of the systematic error components and their interrelation, influencing the final accuracy of the part, were outlined in the general case of 5-axis machines in a new approach. There are three distinct components in the systematic errors.

  • Systematic errors which can be compensated directly in the inverse kinematics relations, mainly reference, positional and angular machine axes errors (nine components).

  • Systematic errors which cannot be included in the inverse kinematics, but which

Erik L.J. Bohez is an Associate Professor of Manufacturing Systems Engineering at the Asian Institute of Technology. He graduated in Electromechanical Engineering from the State University of Ghent, Belgium, in 1979. His research interests are CAD/CAM, geometric modeling, 5-axis machining, simulation of metalcutting and CNC.

References (51)

  • C.A. Van Luttervelt et al.

    Symbiosis of modelling and sensing to improve the accuracy of workpieces in small batch machining operations, cutting

    Int J Adv Manuf Technol

    (1999)
  • J.A. Soons et al.

    Modeling the errors of multi-axis machines: a general methodology

    Precision Engineering

    (1992)
  • J.H. Cho et al.

    Volumetric error analysis of a multi-axis machine tool machining a sculptured surface workpiece

    Int J Prod Res

    (1994)
  • Patel AJ, Ehman KF. Volumetric error analysis of a stewart platform-based machine tool. Annals of the CIRP Vol....
  • M. Weck

    Handbook of machine tools

    (1984)
  • Veldhuis SC, Elbestawi MA. A strategy for the compensation of errors in five-axis machining. Annals of the CIRP Vol....
  • McKeown PA, Weck M, Bonse R. Reduction and compensation of thermal errors in machine tools. Annals of the CIRP Vol....
  • K.G. Ahn et al.

    In process modelling and estimation of thermally induced errors of a machine tool during cutting

    Int J Adv Manuf Technol

    (1999)
  • J.S. Chen et al.

    Thermal error modeling for realtime error compensation

    Precision Engineering

    (1997)
  • Heisel U, Richter F, Wurst KH. Thermal behavior of industrial robots and possibilities for error compensation. Annals...
  • Weck M, McKeown PA, Bonse R, Herbst U. Reduction and Compensation of Thermal Errors in Machine Tools. Annals of the...
  • Hatamura Y, Nagao T, Mitsuishi M, Kato KI, Taguchi S, Okumura T, Nakagawa G, Sugishita H. Development of an intelligent...
  • Jedrzejewski J, Modrzycki W. A new approach to modelling thermal behaviour of machine tool under service conditions....
  • Jedrzejewski J, Kaczmarek J, Kowal Z, Winiarski Z. Numerical optimization of thermal behaviour of machine tools. Annals...
  • Balsamo A, Marques D, Sartori S. A method for thermal deformation corrections of CMMs. Annals of the CIRP, Vol....
  • Cited by (89)

    • Estimation of tool orientation contour errors for five-axismachining

      2013, Robotics and Computer-Integrated Manufacturing
      Citation Excerpt :

      The non-Cartesian kinematics of five-axis machining increases the challenges for applying a contour controller. Bohez compensated the systematic errors in five-axis machines based on the closed loop volumetric error relations [14]. Lo designed a coupled proportional–integral–derivative controllers and proposed real-time transformation between the drive-coordinate and workpiece-coordinate bases to show that minimizing the tracking errors of the tool tip in the workpiece coordinates results in better contouring performance [15].

    • Machining error prediction scheme aided smart fixture development in machining of a Ti6Al4V slender part

      2023, Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture
    View all citing articles on Scopus

    Erik L.J. Bohez is an Associate Professor of Manufacturing Systems Engineering at the Asian Institute of Technology. He graduated in Electromechanical Engineering from the State University of Ghent, Belgium, in 1979. His research interests are CAD/CAM, geometric modeling, 5-axis machining, simulation of metalcutting and CNC.

    View full text