Available online at www.sciencedirect.

com

Available ScienceDirect
online atonline
Available www.sciencedirect.com
at www.sciencedirect.com

ScienceDirect
Procedia CIRP 00 (2018) 000–000
ScienceDirect www.elsevier.com/locate/procedia

Procedia CIRP 00 (2017)

Procedia 000–000
CIRP 77 (2018) 379–382
www.elsevier.com/locate/procedia

8th CIRP Conference on High Performance Cutting (HPC 2018)

Fundamental analysis 28th CIRPof the cutting

Design edge
Conference, micro-geometry
May 2018, Nantes, France in orthogonal
machining of unidirectional Carbon Fibre Reinforced Plastics (CFRP)
A new methodology to analyze the functional and physical architecture of
existing products for an
Vossassembly oriented product
, Friedrichfamily
Kustera, identification
a, a a
Lukas Seeholzer *, Robert , Frank Grossenbacher Konrad Wegenera
a
Institute of Machine Tools and Manufacturing (IWF), ETH Zurich, Leonhardstrasse 21, 8092 Zurich, Switzerland
Paul Stief *, Jean-Yves Dantan, Alain Etienne, Ali Siadat
* Corresponding author. Tel.: +41 44 6325309; fax: +41 446320731. E-mail address: seeholzer@iwf.mavt.ethz.ch
École Nationale Supérieure d’Arts et Métiers, Arts et Métiers ParisTech, LCFC EA 4495, 4 Rue Augustin Fresnel, Metz 57078, France

*Abstract
Corresponding author. Tel.: +33 3 87 37 54 30; E-mail address: paul.stief@ensam.eu

Machining of carbon fibre reinforced plastics (CFRP) is closely related to strong abrasive tool wear and represents a significant process
challenge due to its anisotropic material structure and brittle fracture behavior. Furthermore, machining of unidirectional CFRP material shows
Abstract
a significant increase in process forces due to tool wear effects, resulting in a changing micro-geometry of the cutting edge. This research work
aims to give a detailed analysis of the tool wear behavior with respect to the five micro-geometry parameters based on an approach of “straight
Inline–ellipse–straight
today’s business environment,
line”, capable thealso
trendtotowards
describemore
the product variety
developing and customization
asymmetric is unbroken.
cutting edge. Therefore,Due to this development,
fundamental orthogonalthe need of
machining
agile and reconfigurable
experiments with varying production systems emerged
fibre orientations and toolto geometries
cope with various products Itand
are conducted. product
can families.
be shown that Tothedesign and optimize
micro-geometry production
changes with
systems as well
increasing as to
cutting choose
length, the optimal
whereby productofmatches,
an increase product
the resulting analysis
friction methods
length are needed.
and spring Indeed,can
back height mostbe of the known
detected, methods
having aim to
a significant
analyze
influencea product or one product
on the process forces. family on the physical level. Different product families, however, may differ largely in terms of the number and
nature
© 2018 ofThe
components. This fact by
Authors. Published impedes
Elsevier anLtd.
efficient
This iscomparison and choice
an open access of appropriate
article under product family
the CC BY-NC-ND licensecombinations for the production
© 2018AThe
system. newAuthors. Published
methodology by Elsevier
is proposed Ltd. existing products in view of their functional and physical architecture. The aim is to cluster
to analyze
(http://creativecommons.org/licenses/by-nc-nd/3.0/)
This products
these is an open inaccess
new article under
assembly the CC
oriented BY-NC-ND
product families license (https://creativecommons.org/licenses/by-nc-nd/4.0/)
for theCommittee
optimization
Peer-review under responsibility of the International Scientific of of
theexisting
8th CIRPassembly lines on
Conference andHigh
the creation of future
Performance reconfigurable
Cutting (HPC
Selectionsystems.
assembly and peer-review
Based on under
Datum responsibility
Flow Chain, of the
the International
physical structure Scientific
of the Committee
products is of the 8th
analyzed. CIRP Conference
Functional on High
subassemblies are Performance
identified, and
2018).
Cutting (HPC 2018).
a functional analysis is performed. Moreover, a hybrid functional and physical architecture graph (HyFPAG) is the output which depicts the
similarity
Keywords:between productplastic;
Fiber reinforced families by edge;
Cutting providing
Wear; design
Machinesupport to both, production system planners and product designers. An illustrative
Tool; Turning
example of a nail-clipper is used to explain the proposed methodology. An industrial case study on two product families of steering columns of
thyssenkrupp Presta France is then carried out to give a first industrial evaluation of the proposed approach.
© 2017 The Authors. Published by Elsevier B.V.
1. Introduction
Peer-review under responsibility of the scientific committee of the 28th CIRP Design Conference
rounding. 2018. HENERICHS [3] conducted various
Subsequently,
orthogonal machining experiments with unidirectional CFRP
Keywords:
Carbon Assembly; Design method;
fibre reinforced Family identification
plastics (CFRP) with their high and detected an asymmetric cutting edge rounding, whereby
specific strength and stiffness properties represent a high its intensity strongly depends on the initial tool geometry and
potential for lightweight constructions, such as those found in the machined fibre orientation. WANG et al. [9] described
1.aerospace applications. Simultaneously, machining of CFRP is
Introduction of the productwear
mechanical rangeasandthecharacteristics
dominant wear mechanism
manufactured in
and/or
still a great challenge due to the distinct tool wear and the machining CFRP, while further wear mechanisms as diffusion
assembled in this system. In this context, the main challenge in
inhomogeneous
Due to the microstructure,
fast development resulting
in inthe andomain
anisotropic
of and dissolution
modelling can beis excluded
and analysis due to
now not only the low
to cope withprocess
single
material behavior. Various researchers [2, 3, 5, 6] showed ANG et al. [9] explained the
communication and an ongoing trend of digitization anda temperatures. Furthermore, W
products, a limited product range or existing product families,
strong correlation
digitalization, between enterprises
manufacturing the wear-induced
are facingcutting edge
important mechanical
but also to beabrasion by the and
able to analyze absence of a stagnation
to compare products tozone in
define
rounding and the occurring machining defects
challenges in today’s market environments: a continuing and process front of the cutting edge. This stagnation zone, which
new product families. It can be observed that classical existingis well
forces. This
tendency paper
towards deals with
reduction a detailed
of product analysis of
development the and
times tool known families
product in metalaremachining,
regrouped inrepresents anclients
function of accumulation of
or features.
wear due to the abrasive carbon fibres, resulting
shortened product lifecycles. In addition, there is an increasing in an workpiece material, acting like a protection for the
However, assembly oriented product families are hardly to find. cutting
increasing cutting edge rounding.
demand of customization, being at the same time in a global edge. According to WANG et al. [9] the relative motion
On the product family level, products differ mainly in two
With respect et al. [7] between the cutting edge and the workpiece is significantly
competition with tocompetitors
the occurring tool wear
all over KOPLEV
the world. This trend, main characteristics: (i) the number of components and (ii) the
conducted orthogonal machining experiments with reduced by the stagnation zone, whereby in accordance to
which is inducing the development from macro to micro type of components (e.g. mechanical, electrical, electronical).
unidirectional CFRP and stated a significant increase of the ARCHARD [1] the intensity of mechanical abrasion decreases.
markets, results in diminished lot sizes due to augmenting Classical methodologies considering mainly single products
process forces as a result of the wear-induced cutting edge In accordance to VOSS [8] tool wear primarily affects a small
product varieties (high-volume to low-volume production) [1]. or solitary, already existing product families analyze the
To cope with this augmenting variety as well as to be able to product structure on a physical level (components level) which
identify
2212-8271 possible
© 2018 The optimization
Authors. Publishedpotentials in This
by Elsevier Ltd. theis an
existing
open access causes difficulties
article under regarding
the CC BY-NC-ND license an efficient definition and
(http://creativecommons.org/licenses/by-nc-nd/3.0/)
production system, it is important to have a precise knowledge comparison of different product families. Addressing this
Peer-review
2212-8271 ©under
2018responsibility
The Authors. of the International
Published Scientific
by Elsevier Committee of the 8th CIRP Conference on High Performance Cutting (HPC 2018)..
Ltd.
This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/)
Selection ©
2212-8271 and peer-review
2017 under
The Authors. responsibility
Published of the
by Elsevier International Scientific Committee of the 8th CIRP Conference on High Performance Cutting
B.V.
(HPC 2018).
Peer-review under responsibility of the scientific committee of the 28th CIRP Design Conference 2018.
10.1016/j.procir.2018.09.040
 380 Lukas Seeholzer et al. / Procedia CIRP 77 (2018) 379–382
2 Lukas Seeholzer/ Procedia CIRP 00 (2018) 000–000
region next to the cutting edge. This so-called micro-geometry
describes the transition between the flank and the rake face.
During CFRP machining, a tool-sided material removal due to
mechanical abrasion occurs, resulting in a distinct change of
the micro-geometry. In order to quantify the corresponding
cutting edge rounding, VOSS [8] proposed a simplified
description of the contact region by five micro-geometry
parameters (lα, lγ, γ*, α*, bc), which will be explained in detail
below. Therefore, it will be shown, that the operative part of
the cutting edge can be sufficiently described by the
simplification «straight line – ellipse – straight line», see
Figure 1. For this purpose the cutting edge is subdivided into
three separate regions according to ZHANG et al. [10].
(a) Real Tool wear (b) Micro-geometry Parameters
new state
γ content as well as the high toughness of the fibre-matrix-
[μm]
worn Region 1: Straight (γ*)
Tool vc Region 2: Ellipse (lα, lγ)
Region 3: Straight (bc, α*)
chipping 1
spring back region 1
region 3 pressing 2 4
4 region 2 5
bc
3 α*
α Exemplary fibre
[μm]
Figure 1: (a) Exemplary new and worn cutting edge profile and (b) evaluation
of worn micro-geometry with five micro-geometry parameters [8]
The chipping region 1 along the rake face is bounded by
the nominal material level ahead of the cutting edge 1 and the
most forward point in cutting velocity direction 2. With
increasing tool wear this region is assumed to stay straight but
with a changing rake angle γ*. The pressing region 2
represents the contact length along the cutting edge rounding
between the most forward point in cutting velocity direction 2
and feed direction 3 respectively. The tool shape in region 2
can sufficiently be described by an ellipse with the semi-major
axes lα and lγ. The spring back region 3 represents the contact
length of the flank face between the most forward point in
feed direction 3 and the profile separation point 4. The latter
defines the transition between the new and the worn profile,
whereby the distance in feed direction between 3 and 4
corresponds to the spring back height bc, which results from
the elastic spring-back of the material. In combination with
α*, the parameter bc defines the friction length rp and thus the
straight in region 3.
2. Experimental Setup and material
To analyse the increasing tool wear in orthogonal
machining CFRP, fundamental machining experiments with
four different tool geometries and two fibre orientations for an
overall cutting length of lg=35 m are conducted on a CNC
lathe Okuma LB15-II. The notation of the fibre orientation
angle correspond to that of HENERICHS et al. [4]. For the
experiments CFRP in form of three 120° angular ring
segments (55 mm width, 5 mm wall thickness, 200 mm
radius) are combined and fixed by modified clamping jaws on
a hydraulic chuck, see Figure 2. To avoid a deflection of the
CFRP ring element due to the high clamping forces, an inner
aluminum ring is used, as shown by HENERICHS et al. [4].
With this test-rig a non-interrupted orthogonal machining
process is achieved, while the fibre orientation, the cutting
speed vc=90 m/min, the feed rate f=0.03 mm/rev and the
width of cut ae=5 mm are constant. To avoid issues due to
powder like CFRP chips a suction unit of the type AL-KO
PowerUnit120/400 is used. In order to enable online
measuring of the process forces a 3D-dynamometer of the
type Kistler 9121 is applied. The unidirectional CFRP
material used for the turning experiments has the specification
M21/34%/UD194/IMA-12k and is widely used in the
aerospace industry [4]. It consists of 41% high performance
matrix material HexPly® M21 and 59% carbon fibres IMA-
12k. In accordance to HENERICHS et al. [4], machining this
material represents a big challenge due to both the high fibre
γ system which cause extensive tool wear.
1
γ* Rotation
Clamping jaw
lα 2 Tool
200mm
lγ 5 120° CFRP ring
segments
3
α orientation
5mm
(a) (b)
Figure 2: (a) segmented ring specimen and (b) schematic illustration of the
test-rig setup for orthogonal cutting experiments
A detailed summary of the physical and mechanical
properties of the CFRP material is presented by HENERICHS et
al. [4]. For the turning experiments, uncoated carbide cutting
inserts of the grade H13A are used. To investigate different
combination of rake and clearance angles, the initial cutting
insert geometry CCMW09T304 (ISO standards) is adjusted
by additional grinding operations.
3. Measurement of tool wear
To analyse the cutting edge rounding as a result of the
increasing tool wear, the orthogonal machining process is
interrupted in regular intervals and the actual wear state of the
cutting edge is analysed by means of optical microscopy using
an Alicona Infinite Focus microscope. In order to reduce the
influence of outlier values, each turning experiment is
conducted twice. In addition to the initial micro-geometry
(lg=0 m) the following wear states are analysed: lg=5 m, lg=10
m, lg=15 m, lg=20 m and lg=35 m. In each wear state the five
micro-geometry parameters (lα, lγ, γ*, α*, bc) are evaluated by
means of a modified Cutting Edge Analyser (CEA) [4]. In
order to determine the spring back height bc, the transition
point 4 between the new and the worn cutting tool is crucial.
To enable a repeatable evaluation of this transition point, the
flank face of the cutting insert is marked by a short pulsed Yb-
laser of the type IPG YLP-HP. For this, the following laser
parameter are used: Power of 14 W, wavelength 1064 nm,
pulse frequency 100 kHz, pulse duration 125 ns and scanner
feed 1200 mm/s. As shown in Figure 3 (a). the marking lines
on the flank face are arranged in form of a grid, where the
distance between the corresponding lines parallel and
perpendicular to the cutting edge are 100 μm and 500 μm
 Lukas Seeholzer et al. / Procedia CIRP 77 (2018) 379–382 381
Lukas Seeholzer/ Procedia CIRP 00 (2018) 000–000 3

respectively. The resulting depth of the laser marks is so small
that they vanish if they get in contact with the spring-back
CFRP-material. Figure 3 (b) shows the shift of the transition
line due to increasing tool wear resulting in an increasing
friction length in region 3.
Figure 4 shows that for both θ=30° and θ=60°, the
dimension of the resulting rake angle γ* is primarily affected
by the initial rake angle γ. Furthermore, Figure 4 implies a
wear-induced change of the parameter γ* with increasing
cutting length. In terms of the tools with varying initial
clearance angles (J, L and M) the wear-induced changes are

Rake face
Detailed view A:
Cutting edge
500 μm
Flank face small. In combination with an overall measuring accuracy of
Rake face
±3° no distinct conclusion about the influence of both the
clearance and rake angle is possible.
The micro-geometrical shape in region 2 is modelled as an
ellipse, described by the two semi-major axes lα and lγ. Figure
100 μm
5 shows the trends of these two parameters for the respective
tool geometries and fibre orientations over the cutting length

Flank face
Rake face of lg=35 m. Fundamentally, it can be stated that all tool
A Cutting edge geometries for both fibre orientations show a similar trend of
the parameters lα and lγ. Starting from the initial circular shape
γ (lα=lγ), the increase of lα is significantly higher than of lγ,
resulting in an increasing asymmetric shape of the cutting
edge in region 2. In the scope of the tested tools and fibre
Flank face orientations, no clear influence of the initial tool geometry to
(a) (b) the resulting trends of lα and lγ is detected. Furthermore, the
α
0μm 50μm 100μm

appropriate trends of lα and lγ are characteristic and could be

Figure 3: (a) Laser marking grid on the flank face and (b) shift of the well approximated by a root function which includes a
transition line due to mechanical wear significant influence of the actual fibre orientation.
Θ=30° Θ=60°
4. Experimental Results
Semi-axis [μm]

Semi-axis [μm]
50 50

In the following the trends of the five micro-geometry 40 40

parameters (lα, lγ, γ*, α*, bc) are discussed in detail with 30 30
respect to the three regions of the cutting edge. The presented 20 20
results include the fibre orientations θ=30° and θ=60° and the
10 10
following four tool geometries: J (γ=20°,α=7°), L
(γ=20°,α=14°), M (γ=20°,α=21°) and H (γ=10°,α=14°). While 0 0
0 10 20 30 40
0 10 20 30 40
the tools J, L and M differ only in terms of the initial l Cutting length [m] Cutting length [m]
tool α
clearance angle, the tool H has a smaller initial rake angle, but lγ region 2
vc
lγ Tool J
lα Tool J
lγ Tool L
lα Tool L
lγ Tool M
lα Tool M
lγ Tool H
lα Tool H
otherwise corresponds to the tool L. Based on the geometric
description of the micro-geometry which is explained in
Figure 5: Trend of lγ and lα with increasing cutting length for different tool
section 1, the wear-induced change of the micro-geometry in geometries in θ=30° and θ=60° material; vc=90 m/min and f=0.03 mm/rot
region 1 is described by the parameter γ*. This parameter
represents the rake angle, which is changed by a wear-induced According to the previous results, Figure 6 shows the
material removal on the rake face. The graphs in Figure 4 trends of the parameters α* and bc, which describe the wear-
show the trend of γ* for the four tool geometries during induced change of the micro-geometry in region 3. Although
machining θ=30° and θ=60° material for an overall cutting the parameter α* primarily depends on the initial clearance
length of lg=35 m. angle, significantly different trends are determined for the
Θ=30° Θ=60°
tools J, L and M. In this context the extrapolated trends of α*,
γ* [°]

γ* [°]

22 22
which are shown in Figure 6 by dotted black lines, imply that
20 20
all clearance angles converge to a single constant clearance
18 18
angle in the worn state of the cutting edge. The fact, that this
16 16
limit value is different for the fibre orientations θ=30° and
14 14
θ=60° indicate a dependency of the convergence behavior on
12 12
10 10
the fibre orientation angle. This fact could be explained by
8 8
different chip formation effects due to a different relative
0 10 20 30 40 0 10 20 30 40 positioning of the fibre with respect to the tool. According to
Cutting length [m] Cutting length [m]
Figure 6, the higher the initial clearance angle, the more
γ* Tool J γ* Tool L γ* Tool M γ* Tool H
distinct is the wear-induced decrease of the parameter α* per
cutting length. Based on the nearly similar trends of the
Figure 4: Trend of γ* with increasing cutting length for different tool
geometries in θ=30° and θ=60° material; vc=90 m/min and f=0.03 mm/rot parameter α* for the tools L and H, the effect of the initial
rake angle is not significant and thus negligible. The trends of
the parameter bc show a crucial influence of the initial
382 Lukas Seeholzer et al. / Procedia CIRP 77 (2018) 379–382
4 Lukas Seeholzer/ Procedia CIRP 00 (2018) 000–000

clearance angle and the fibre orientation, while the influence
of the initial rake angle is approximately neglectable. The tool
J with the smallest initial clearance angle shows the largest
increase of bc and the tool M with the lowest initial clearance
angle the smallest increase. Furthermore, the increase of the
fibre orientation angle from θ=30° to θ=60° results especially
for the tool J with the lowest initial clearance angle in an
overall lower spring back height bc.
Θ=30°
60 region 3 vc
Spring back height bc [μm]
the resulting thrust forces. In contrast, the share of region 3,
defined as the subarea AC has a huge influence and strongly
depends on the initial clearance angle. This subarea AC can be
determined by multiplying the friction length rP and the
cutting width, which corresponds to the constant wall
thickness b. Figure 7 (b) shows the trend of the friction length
with increasing cutting length. Due to the fact that the thrust
forces show nearly comparable trends, it can be concluded
Θ=60° that region 3 is primarily responsible for the increasing thrust
25 60 25
forces during machining CFRP. The evaluation of θ=60° is

Spring back height bc [μm]

bc α*
α* [°]

α* [°]
50 20 50 20 not shown here, but yields an equivalent characteristic.
40 40
15 15
30 30 5. Conclusion
10 10
20 20
5 5 In the scope of this paper, a detailed analysis of the wear-
10 10
induced cutting edge rounding in case of orthogonal
0 0 0 0
0 10 20 30 40 0 10 20 30 40 machining unidirectional CFRP is presented. The trends of the
bc Tool J
Cutting length [m]
bc Tool L bc Tool M
Cutting length [m]
bc Tool H
five micro-geometry parameters (lα, lγ, γ*, α*, bc) are shown
α* Tool J α* Tool L α* Tool M α* Tool H for different fibre orientations and tool geometries.
Additionally, a suitable measurement method is presented to
Figure 6: Trend of α* and bc with increasing cutting length in θ=30° and
θ=60° material; vc=90 m/min and f=0.03 mm/rot
predict the spring back height of the CFRP material and thus
the resulting friction length. Furthermore, the friction length
The results in Figures 4, 5 and 6 show a distinct change of in region 3, which is significantly influenced by the fibre
the micro-geometrical shape with increasing cutting length up orientation and the initial clearance angle is stated as the main
to lg=35 m. Furthermore, the dependencies of the wear- influence factor for the distinct increases of the thrust forces.
induced change on the initial tool geometry and the fibre
orientation are different with respect to the three regions of Acknowledgements
the cutting edge. While the tool wear in regions 1 and 2 is
nearly similar, region 3 significantly depends on both the The authors thank the Commission for Technology and
initial clearance angle and the fibre orientation. Since the Innovation (CTI Project 18309.2 PFIW-IW), the companies
corresponding thrust forces also clearly diverge, a strong Dixi Polytool SA, Heule Werkzeuge AG and Oerlikon
correlation with the wear in region 3 is assumed. In this Surface Solutions AG for the support.
context, Figure 7 (a) shows the trends of the thrust force
exemplary for θ=30° and the tools J, L and M. Basically, the References
smaller the initial clearance angle, the higher the wear-
[1] Archard J. F., Contact and Rubbing of Flat Surfaces. Journal of Applied
induced increase in the corresponding thrust force. This fact Physics, 1953. 24(981).
can be explained by a detailed analysis of the thrust force [2] Chen W.-C., Some experimental investigations in the drilling of carbon
relevant areas of the cutting edge and its influence factors. fiber-reinforced plastic (CFRP) composite laminates. International
Journal of Machine Tools and Manufacture, 1997. 37(8): p. 1097-1108.
1200
Thrust force FZ [N]

1000 [3] Henerichs M., Bohrbearbeitung von CFK unter besonderer

800 Berücksichtigung der Schneidkantenmikrogeometrie. 2015,
600 Eidgenössische Technische Hochschule Zürich (ETH): Zürich.
400 [4] Henerichs M., Voss R., Kuster F., Wegener K., Machining of carbon fiber
200 reinforced plastics: Influence of tool geometry and fiber orientation on
0 the machining forces. CIRP Journal of Manufacturing Science and
5 10 15 20 25 30 35 Technology, 2015. 9: p. 136-145.
(a) Fz, tool J Fz, tool L Fz, tool M Cutting length [m]
[5] Hocheng H., Puw H., On drilling characteristics of fiber-reinforced
600
Friction length rP [μm]

500 thermoset and thermoplastics. International Journal of Machine Tools

400 and Manufacture, 1992. 32(4): p. 583-592.
300 [6] Hocheng H., Tsao C., Comprehensive analysis of delamination in drilling
200 of composite materials with various drill bits. Journal of Materials
100 Processing Technology, 2003. 140(1): p. 335-339.
0 [7] Koplev A., Lystrup A., Vorm T., The cutting process, chips, and cutting
5 10 15 20 25 30 35 forces in machining CFRP. composites, 1983. 14(4): p. 371-376.
(b) rp,
rP tool J rrp,
P tool L rrp,
P tool M Cutting length [m]
[8] Voss R., Fundamentals of Carbon Fibre Reinforced Polymer (CFRP)
Machining. 2017, Eigenössische Technische Hochschule Zürich (ETH):
Figure 7: Trend of rP, lα and the thrust forces with increasing cutting length in Zürich.
θ=30° material; vc=90 m/min and f=0.03 mm/rot [9] Wang X., Kwon P. Y., Sturtevant C., Lantrip J., Tool wear of coated drills
in drilling CFRP. Journal of Manufacturing Processes, 2013. 15(1): p.
127-135.
Since the thrust force primarily depends on the contact area [10] Zhang L. C., Zhang H. J., Wang X. M., A force prediction model for
between the tool and the just machined workpiece surface, the cutting unidirectional fibre-reinforced plastics. Machining Science and
regions 2 and 3 have to be considered. As mentioned above, Technology, 2001. 5(3): p. 293-305.
the influence of the initial clearance angle on the micro-
geometrical shape in region 2 is low and thus also its share to