Available online at www.sciencedirect.
com
ScienceDirect
Procedia CIRP 46 (2016) 147 – 150
7th HPC 2016 – CIRP Conference on High Performance Cutting
Design and Optimisation of an Electromagnetic Linear Guide
for Ultra-Precision High Performance Cutting
B. Denkena, D. Dahlmann and R. Krueger*
Institute of Production Engineering and Machine Tools, Leibniz Universität Hannover, An der Universität 2, D-30823 Garbsen, Germany
* Corresponding author. Tel.: +49-511-762-4840; fax: +49-511-762-5115. E-mail address: [email protected]
Abstract
Ultra-precision machining is rarely used in the production industry due to high costs as a consequence of disproportionally long primary and
secondary processing times. In this context, the implementation of innovative machine technologies presents a suitable approach to increase
productivity and reduce manufacturing costs. This paper introduces the implementation of an electromagnetic linear guide within a two-axis
positioning stage for ultra-precision and micro machining. Using analytical models and FEM simulations, an optimised design for the guide’s
structure and magnet configuration is developed with regard to the intended application in ultra-precision high performance cutting. The new
guide system provides frictionless operation for rapid and precise feed movements. Stiffness and damping of the electromagnetic guide can be
adjusted to current process requirements. Fine positioning of the levitating carriage within the air gap enables an increase of the overall position
accuracy.
© 2016
© 2016TheTheAuthors.
Authors.Published
Published byby Elsevier
Elsevier B.VB.V.
This is an open access article under the CC BY-NC-ND license
Peer-review under responsibility of the International Scientific Committee of 7th HPC 2016 in the person of the Conference Chair Prof.
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Matthias Putz.
Peer-review under responsibility of the International Scientific Committee of 7th HPC 2016 in the person of the Conference Chair
Prof. Matthias Putz
Keywords: High performance cutting; Magnetic guide; Ultra-precision machining
1. Introduction ultra-precision machining. In this context, electromagnetic
guides provide the necessary capabilities to improve the
Ultra-precision machining acts as a key technology for the performance of ultra-precision machining processes. As
manufacture of optical components. Established production mechatronic components, active magnetic guides function as
techniques for precision machining are suitable for the a combined actuator and sensor system. Stiffness and
generation of micro-structured surfaces and freeform surfaces damping can be adjusted to current process requirements
with optical properties. The area of application for precision within physical limits. The active damping of disturbance
machined parts covers astronomy, metrology, medical forces (such as process forces, breakdown torques or
devices, automotive components and optical industries [1]. unbalances of rotating components) allows for an increased
However, available ultra-precision machining processes cutting performance without sacrificing surface quality or
are generally characterised by low productivity. On the one process stability. Positioning of the levitating carriage in 5
hand, limitations in productivity result from low applicable DOF enable the compensation of production and mounting
feed rates. An increase in feed velocity causes increased tolerances in order to increase overall accuracy of the guide
dynamic disturbances and deviations compromising the system. Furthermore, monitoring of the magnets’ coil currents
optical quality of the machined surface. On the other hand, and air gaps enables identification of forces affecting the
time-consuming manual alignment of workpiece and cutting carriage. Thus, sensory properties of electromagnetic guides
tool leads to long secondary processing times. can be used for process monitoring and simplified workpiece
The implementation of innovative machine technologies and tool setup.
presents a suitable approach to increase the productivity of
2212-8271 © 2016 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the International Scientific Committee of 7th HPC 2016 in the person of the Conference Chair
Prof. Matthias Putz
doi:10.1016/j.procir.2016.02.056
148 B. Denkena et al. / Procedia CIRP 46 (2016) 147 – 150
Known implementations of electromagnetic levitation 2.2. Design considerations
technology in machine tools focus on high performance
cutting [2] [3] or workpiece positioning for non-mechanical The novel electromagnetic guide requires a design
processing [4]. Existing electromagnetic guides for high approach which incorporates two fundamental aspects: First,
performance cutting operations show insufficient accuracy for functional requirements of magnetic guides have to be
ultra-precision machining; magnetic guides for precision considered; second, principles of precision machine design
application do not provide the necessary stiffness for have to be taken into account. Imperative design points are the
machining processes. Thus, development of a novel magnet arrangement, magnet design and placement of the
electromagnetic guide is required in order to exploit the feed drive system. The previous design points affect the
potential of electromagnetic levitation technology for high- structural design of the guide. Furthermore, the choice of
precision application. construction material determines thermal and mechanical
properties of the guide’s carriage and frame.
Nomenclature A methodology for the conceptual design of an
electromagnetic guide for use in ultra-precision machining is
A magnet surface presented in [5]. Using the proposed approach, a concept for a
B flux density novel electromagnetic ultra-precision linear guide was
F magnet force developed (Fig. 1). Key considerations revolve around
I electric current functional independence and a minimum error budget.
n number of turns Accordingly, the design features a differential magnet
G air gap arrangement for independent horizontal and vertical
) magnetic flux positioning. If possible, active components are mounted on
μ0 magnetic constant the guide’s frame to reduce cable drag as a source of non-
linear friction. A recessed mounting position of the feed drive
achieves minimal breakdown torques. High resolution
2. Conceptual design of a prototypical electromagnetic
measuring systems are integrated to monitor the carriage’s six
linear guide for ultra-precision machining
DOF. Granite as the main construction material for the
guide’s carriage and frame provides thermal stability and high
Electromagnetic levitation guides distinguish themselves in
damping.
several ways from conventional guide systems. The most
noticeable difference is the reversed bearing force direction.
Electromagnets generate only pulling forces; bidirectional
force application requires a pair of two opposing
electromagnets. Thus, simple substitution of guide
components in conventional machine tools is usually not
possible. Instead, integration of an electromagnetic guide
within an existing machine structure requires a redesign of the
surrounding modules.
2.1. Requirements specification
A fly-cutting process with increased cutting performance
serves as the reference process for the novel guide system. Fig. 1: Concept for an electromagnetic ultra-precision linear guide:
High performance machining of the planar reference surface (a) Front view; (b) Top view
with optical quality presents the main objective. Table 1
shows the technical specification for the magnetic guide. 3. Electromagnet design and optimisation
Table 1. Requirements specification for the prototypical guide system
The magnetic guide’s overall properties mainly result from
the mechanical and electrical configuration of the
Specification Value
electromagnets. Framework conditions for electromagnet
Feed rate [mm/min] 3000 design are predominantly set by the requirements regarding
Acceleration [m/s²] 9.81 the magnet forces. Calculation of magnet forces necessitates
Travel range [mm] 100 detailed knowledge of the electromagnet’s geometry and the
Straightness (over travel range) [μm] 0.16 non-linear properties of the core material. For the application
Position accuracy [μm] <1
at hand, an analytical model of the magnetic circuit provided
an approximation of magnet forces for an initial design.
Resolution of air gap measurement system [nm] 1
Finally, accurate results were obtained using the finite
Position noise [nm] < 10 element method. Optimisation of magnet parameters was
Bearing stiffness [N/μm] 200 achieved using parametric finite element simulation in
Load capacity [kg] 50 conjunction with a multi-objective genetic algorithm.
B. Denkena et al. / Procedia CIRP 46 (2016) 147 – 150 149
3.1. Analytical calculation circuit’s flux density, the resulting magnet forces and the
inductivity of the coil.
An approximation of magnet forces for the initial design
was made by analytical calculation in consideration of
feasible simplifications. The required magnet surface was
estimated using Maxwell’s pulling force formula for the iron-
air transition
B2 A
F . (1)
2μ0
An E-shaped core with three poles was chosen to realise a
compact design. Laminated steel as the material for the
magnet’s core and back iron reduces eddy currents for
increased force dynamics.
The electromagnet works with a nominal flux density of up
to B = 1.0 T; therefore, magnetic saturation of the core
material does not occur at rated operation. Without saturation
the iron core can be neglected in the magnetic circuit. Fig. 2: Initial magnet design: (a) Exploded view; (b) Simplified model for
Therefore, a closed solution for the magnetic flux can be finite element analysis; (c) Simulated flux density distribution
obtained as deduced in [6]:
3.3. Parameter optimisation
AP0 nI
) . (2) The electromagnet design presents a highly complex task
4G due to interdependencies of magnet parameters. The design
process can be described as a multi-objective optimisation
Further, the magnetic flux can be calculated with the problem. Optimisation was performed at two stages for a
generally valid expression for symmetrical magnets practical magnet design approach. First, the magnet’s
geometry was optimised using parametric simulation in
BA ANSYS Workbench. The geometry of the inner and outer
) ³ BdA |
A/ 2
2
. (3)
poles was altered for an even flux density in the magnet core.
At the second stage, optimal electrical parameters were
identified to achieve the required force dynamics.
Thus, equations (2) and (3) yield a correlation between Force dynamics are primarily limited by the coil’s
electrical excitation and force generating induction which is inductivity as well as the performance of the used current
valid for B 1.0 T: amplifier. Current amplification is achieved with a digital
servo drive with output control by pulse width modulation
P0 nI (PWM). In comparison to analogue amplifiers digital drives
B . (4)
2G offer lower power dissipation and dynamic parametrisation
options. The possibility of dynamic parametrisation is
Accordingly, the pulling force formula (1) can be modified especially helpful for applications with alternating load
with equation (4) to determine the coil parameters for the conditions. Then again, pulse width modulation of the output
initial design: signal leads to a current ripple which results in unintentional
magnet force deviations. Consequently a jitter is induced at
standstill.
AP0 (nI )²
F . (5) The digital current amplifier’s DC link voltage determines
8G ² current injection in the coil depending on the coil’s
momentary inductivity. A low coil inductivity is required to
3.2. FEM simulation achieve sufficient magnet forces for a wide frequency range.
At the same time, coil inductivity and DC link voltage
The analytical results were verified by numeric calculation significantly affect the current ripple. High inductivity in
using the finite element method. The initial magnet design combination with low DC link voltage are preferable for
delivered a FEM model for magnetostatic simulation in minimum current ripple. However, this condition is not
ANSYS Workbench (Fig. 2). The model exploits the compatible with dynamic force requirements.
magnet’s two symmetry planes to enable a fast and effective In order to find a solution for the conflicting objectives a
calculation; rounded corners and mounting bores were multi-objective genetic algorithm [7] was applied. The
neglected for an efficient surface mesh generation. The finite optimisation goal for the genetic algorithm was the
elements analysis provided accurate results on the magnetic minimisation of the field coil’s inductivity for a set magnet
150 B. Denkena et al. / Procedia CIRP 46 (2016) 147 – 150
force at nominal flux density. Secondary objectives include
minimum resistance of the field coil which minimizes the
electrical losses. Coil parameters (such as number of turns and
coil current) and nominal air gap were varied to produce
populations. Individuals were evaluated with regard to force
dynamics and electrical losses.
3.4. Final electromagnet design
For set magnet core and coil dimensions, the magnet’s
electrical time constant and electrical losses correlate with the
magnet force. The optimisation results demonstrate that
magnet configurations with low coil inductivity feature a low
current sensitivity. Hence, only minor differences in force
deviations due to current ripple were identified regardless of
Fig. 4: (a) Carriage of the electromagnetic guide; (b) Two-axis position stage
the coil’s inductivity. Thus, coil parameters like the number of
turns and the nominal coil current were adjusted to achieve
5. Summary and conclusion
minimum coil inductivity and maximum force dynamics.
Table 2 presents the technical specification of the final
This work focuses on the design and optimisation of
magnet design. Fig. 3 shows the assembly of the final magnet
electromagnetic actuators as the core components of active
design.
magnet guides. An initial magnet design was developed based
on an analytical model of the iron circuit. Analytical results
were verified by finite element analysis. A parametric FEM
simulation was set up and coupled with a multi-objective
genetic algorithm in order to optimise magnet geometry and
electrical parameters. The final magnet design was evaluated
with regard to current amplification and force dynamics.
The optimised magnet design was incorporated in a novel
electromagnet guide system. Implementation of the new guide
within a two-axis-positioning system for ultra-precision
machining was considered.
Fig. 3: (a) Final electromagnet assembly; (b) Bottom view of the core unit Acknowledgements
Table 2. Technical specification of the optimised magnet design The authors thank the German Research Foundation for
funding the research group FOR 1845 “Ultra-Precision High
Specification Value Performance Cutting”.
Pole surface [cm²] 8.25
Air gap [mm] 0.2 References
Number of turns 50
[1] Riemer O. Advances in Ultra Precision Manufacturing. Proc. JSPE, 2011.
Coil inductivity (@ 1.0 T) [mH] 3.95 [2] Denkena B, Kallage F, Ruskowski M, Popp K. Machine Tool with Active
Coil resistance [ȍ] 0.084 Magnetic Guides. Annals of the CIRP, Volume 53, Issue 1, 2004, p. 333-
336.
Coil current (nominal/maximum) [A] 6.2/12
[3] Denkena B, Guemmer O, Floeter F. Evaluation of electromagnetic guides
Flux density (nominal/maximum) [T] 1.0/1.6 in machine tools. Annals of the CIRP, Volume 63, Issue 1, 2014, p. 357-
Magnet force (nominal/maximum) [N] 300/700 360.
[4] Laro, D. Mechatronic Design of an Electromagnetically Levitated Linear
Dynamic magnet force (@ 1000 Hz) [N] 100 Position System using novel Multi-DOF Acuators. Dissertion, Delft
University of Technology, 2009, p. 20-26.
[5] Denkena B, Dahlmann D, Krueger R. Electromagnetic Levitation Guide
4. Ultra-precision two-axis positioning stage
for Use in Ultra-Precision Milling Centres. Procedia CIRP 37, 2015,
p. 199-204.
The new design was incorporated within a two-axis [6] Ruskowski M. Aufbau und Regelung aktiver Magnetführungen.
positioning stage for ultra-precision machining (Fig. 4). The Dissertion, Leibniz Universität Hannover, 2004, p 21-34.
development of the positioning stage focused on a minimum [7] Konak A, Coit DW. Smith A.E. Multi-objective optimization using
genetic algorithms: A tutorial. Reliability Engineering and Safety
error budget. A box-in-a-box design minimises cumulative
Systems, Volume 91, 2006, p. 992-1007.
errors. Granite is used as the main construction material for
the positioning stage to lessen the effects of thermal
expansion.
