See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/277896186

DEVELOPMENT OF A CUTTING TOOL FOR COMPOSITES WITH

THERMOPLASTIC MATRIX

Article  in  MM Science Journal · October 2013

DOI: 10.17973/MMSJ.2013_10_201312

CITATIONS READS

7 408

3 authors:

Petr Mašek Pavel Zeman

Czech Technical University in Prague Czech Technical University in Prague
9 PUBLICATIONS   22 CITATIONS    33 PUBLICATIONS   146 CITATIONS   

SEE PROFILE SEE PROFILE

Petr Kolar
Czech Technical University in Prague
18 PUBLICATIONS   231 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Centre of Competence – Manufacturing Technology View project

Next Generation Cutting Tools for Manufacturing Advanced Lightweight Materials View project

All content following this page was uploaded by Pavel Zeman on 23 July 2015.

The user has requested enhancement of the downloaded file.

 There are many variants of fibres in these two main groups and
DEVELOPMENT OF other groups such as aramide, basalt and natural fibres. Two mostly

A CUTTING TOOL FOR used types of fibres are carbon fibres and glass fibres. Carbon fibres
are tougher and conduct heat better than glass fibres. The type of
COMPOSITES WITH reinforcement also influences the behaviour of the composite during
milling. The fibres are very hard and abrasive. They cause rapid wear
THERMOPLASTIC MATRIX of the cutting tool. The content and orientation of fibres in composites
influence tool life time and surface quality (delamination, burrs
Petr Masek, Pavel Zeman, Petr Kolar creation) as well [Sheikh Ahmad 2009]. There are three main types
Czech Technical University in Prague of reinforcement in the case of fabric: unidirectional, multidirectional
Research Center of Manufacturing Technology and woven reinforcement [Ehrenstein 2009].
Prague, Czech Republic This paper focuses on the milling of FRTC materials. Milling of
composite materials is a branch which is primarily focused on
e-mail: [email protected] trimming of flat or curved panels as parts of planes, cars and
Fibre reinforced thermoplastic composites (FRTC) are becoming an industrial and energetic devices. Other milling operations are
increasingly important material for a wide range of applications. Thus, milling of internal shapes and holes and slotting. The trimming
the demand for high performance tools for machining of this material operation of flat panels was chosen in this paper because it is the
is increasing. The main goal of the presented research results is to most representative technology of the milling of FRTC materials. The
decrease cutting costs during trimming operations of thin plates and main task was to improve the surface quality and productivity and
to increase the quality of machined composite materials by suitable decrease costs which are connected with this type of machining.
shape, geometry and material of the cutting tool and by suitable Some papers have been focused on milling of composite materials.
cutting conditions. Statistical methods were used in order to obtain The results presented in these papers affected the design of our
the control factors (cutting conditions) which are the most important experiment. The up milling was identified as better than down milling
for machining FRTC. Subsequently, an optimal adjustment of cutting for better surface roughness [König 1985] but the situation is different
conditions was identified. The paper is a brief summary of the cutting for the delamination of composite. Down milling was slightly better
tool development. It includes important parts of tests and their results as in Coligan and Ramulu paper [Colligan 1991]. This paper was also
well as some interesting findings implicit in experiments and measuring. focused on the orientation of the fibres in the composite and various
types of delamination. Cutting speed and feed rate was identified
Keywords: as relevant for testing on the basis of the Davim’s experiments. The
thermoplastic composite, cutting tool, cutting conditions, feed rate was more significant than the cutting velocity but both were
development, milling statistically significant. The delamination and cutting forces were
smaller when the feed rate was smaller [Davim 2004]. The depth
1. Introduction of cut could be also relevant for testing. The surface roughness is
Machining of composite materials is nowadays an issue which influenced by depth of cut [Sheikh Ahmad 2009].
includes mainly cutting tool lifetime and surface quality of machined The tests were divided into two stages. The first one was oriented
parts. This development was focused on the milling of FRTC to the test of standard commercially available tools which were
materials. Different applications for using these materials need different recommended by producers for machining of polymeric composites.
composition of composites materials. Composite materials usually Six tools were chosen from the wide portfolio of cutting tools on the
consist of a polymeric matrix and fibre reinforcement. A thermoplastic market as suitable representatives (Fig. 1). The second stage focused
matrix of composite materials with fibre reinforcement is rather on testing new cutting tools which were proposed on the basis of the
different from thermoset polymers. Thermosetting matrices are more test in the first stage.
brittle than thermoplastic and during machining create chips like
dust [Peters 1998]. Exhausting is necessary if this type of composite is
machined. Thermoplastics are softer and more ductile. Small chips
are created during machining. The amount of dust is minimal and
consists mainly from debris of broken fragile fibres. Temperature
must be carefully observed. If the temperature of machining exceeds
the glass transition temperature for a given polymeric matrix, the
composite loses its mechanical properties. Polymeric materials do
not conduct heat very well. Heat concentration in the cutting area
occurs during machining at high cutting velocities. The matrix
became soft and molten. The molten chips can adhere to each other,
to the material and also to the cutting tool. It leads to a decrease in A B C D E F
the surface quality of the composite and a decrease in the cutting
tool lifetime [Quadrini 2007]. Figure 1. Six chosen cutting tools.

Identification of Number of Diameter Clearance Rake Helix

Material of cutting edge
cutting tool teeth [mm] angle angle angle
A 2 8 PCD 17.84 5.06 5.09
B 2 8 PCD 17.16 –15.58 5.13
C 2 12 PCD 10.09 0.37 2.53
D 6 12 Carbide with diamond coating 11.28 20.72 16.03
E 9x2 12 Carbide with diamond coating 8.68 12 20.17
F 2x2 12 Carbide with Ti28 coating 22.34 21.64 17.83
Table 1. Parameters of tested cutting tools.

DEVELOPMENT OF A CUTTING TOOL FOR COMPOSITES WITH THERMOPLASTIC MATRIX | October | 2013 | 422/423
2. Test of standard commercially available tools ability of this method is decreased by this. It was not possible to apply
Commercial tools which were chosen for testing in the first stage the S/N ratio method in regard to small differences between results in
are standard tools recommended by different companies for the the experiment runs for all tools. But there was an obvious difference
trimming operation. The tools were compared in terms of cutting between the results for all tools (Fig. 3). The higher positivity of the
forces, surface roughness and burr creation. The Taguchi design of rake angle caused elimination of burr creation.
experiment (DOE) was used [Taguchi 1987]. It allows a reduction
in the number of necessary experiments in the test by consolidating
some less important effects of interactions with the main effects.
The control factors for this experiment were cutting velocity, feed
per tooth, radial depth of cut, milling strategy and surface ply
orientation. Two interactions were chosen on the basis of preliminary
tests. The first one was interaction between the feed rate per tooth
and cutting velocity. The second one was interaction between the
feed per tooth and radial depth of cut. They were chosen on the
basis of the ANOVA test results and heredity of significance. This
means that if the control factors are very statistically significant, it is
possible that their interaction will be also statistically significant. The
DOE is in Tab. 1.

Figure 3. Direct comparison of burr area.

D E
Run ID A (ft) B (vc) AB C (ae) AC (strat.) (ply orient.)
1 1 1 1 1 1 1 1 It is possible to deduce significance of the cutting tool and its
2 1 1 1 2 2 2 2 geometry in the case of burr creation from the previous diagram.
3 1 2 2 1 1 2 2 The values of data could be averaged in order to unambiguously
4 1 2 2 2 2 1 1 determine the suitability for cutting-off of burrs. We can see that
5 2 1 2 1 2 1 2
cutting tool “F” has the most suitable geometry. This cutting tool has
a very sharp geometry. It has a very positive geometry of rake and
6 2 1 2 2 1 2 1
clearance angle, more than 20° and the helix angle is 18°. This might
7 2 2 1 1 2 2 1
be the most superior choice for cutting tool which creates clear edges
8 2 2 1 2 1 1 2
without burr.
Table 2. Taguchi DOE. It is possible to say after comparison of Fig. 3 with Table 3 that
the rake angle together with the clearance angle may be the most
The DOE needs two levels of control factors and chosen important factor for removing burr. This is true for cutting tool “F”,
interactions. Levels for each control factor were chosen as shown “D” and “A”. The slight positivity of the rake angle and helix angle
in Table 2. are probably of benefit. Cutting tool “E”, which also has high values
of the rake angle and helix angle, is limited by a high number of
Unit Symbol Levels teeth. Flutes of the cutting tool are not sufficient for removing chips.
Parameters/Factors It could be the reason for the deterioration of cutting-off of the top
Level 1 Level 2 and bottom burr on the specimen. Cutting tool “B” has a rake angle
feed per tooth mm ft 0.05 0.1 equalling 0. Its results are similar to cutting tool “E” in the case of
burr. Cutting tool “B” has the most unsuitable geometry. Its negative
cutting speed m/min vc 150 300 geometry of the rake angle, small helix angle and the fact that the
depth of cut mm ae 1 3
helix only has a sense in one direction (compressed helix) causes the
creation of large burr on the bottom edge of the composite.
strategy – – Up milling Down milling The surface of composites differs from the surface which is created
after machining metals. While metals are homogenous, composites
ply orientation – – PW PX
consist of fibers which are very hard and brittle and a matrix which is
Table 3. Machining parameters and their levels. rather yielding. It is possible to observe small holes on the machined
surface after milling. These holes were formed by pulling out of the
The polyphenol sulfide (PPS) with woven carbon fibres reinforcement bonds of the fibers or destroying these bonds. A photograph of these
was used as tested material. The test Material had 5-harness satin holes was taken by a confocal microscope (see Fig. 4).
weave structure. Two different orientations of surface plies PW and
PX were cut (Fig. 2).

Figure 2. Diagrams of surface ply orientation PW and PX and specimen of

PPS/C on the right.

2.1 Results and discussion for commercial tools

The statistical comparison fails for measuring of burrs. The method
of measuring used is significantly influenced by the place of cutting Figure 4. Holes on the machined surface from confocal microscope,
on the specimen and the definition of the photograph. The resolution zoomed 480x.
 The holes are not bigger than 70 μm in diameter. The holes are
impossible to see by the naked eye but they can have a significant
influence on the evaluating of surface roughness. The stylus tip on
the pick-up of Surtronic 3+ has a radius of the tip 5 μm. It is possible
that the stylus tip may go down the hole. Some of these holes are
almost 30 μm deep and evaluating of Rz surface roughness may be
probably influenced by this [Masek 2012].
The statistical evaluation of surface roughness was made for both
tools. The differences between results in runs were too small and
this was the reason why the S/N ratio failed (Fig. 5). It is not possible
to decide which factor was more significant than others. It was also
impossible to decide how beneficial the chipbreaker was in the case
of surface roughness. The differences between cutting tools were
smaller then differences between runs and both were smaller than
the measuring device uncertainty of measurement.

Figure 5. Comparison of the Rz parameter for all tools in each run of experiment.

Figure 7. An example of S/N ratio evaluation for resultant cutting force

– cutting tool “C”.

forces less than expected. Low velocities can be recommended as

suitable for trimming. Orientation of fibres was almost insignificant
for most cutting tools.
On the basis of the first block of test, the most effective parts of
Figure 6. Results of maximal resultant forces for all tools and all runs. cutting tools were determined. A compressive helix was effective in
the case of burr reduction. The upper plies are compressed into the
The smallest resultant forces were gained by cutting tool “F” as can composite. Created burrs are then easier to cut by the next tooth.
be seen from Fig. 6. The resultant forces were up to 60 N for low The best cutting tools in terms of small cutting forces and burrs had
feed rates and up to 80 N for high feed rates. The highest resultant a very positive rake and clearance angle. The longest lifetime of
forces were obtained during milling by cutting tool “A”, where the the cutting tool can be achieved only with diamond coating on the
cutting forces were at a level from 150 to 190 N for low feed rates carbide substrate or with PCD materials. Tools with PCD were not
and from 190 to 250 N for higher feed rates, i.e. almost 3 times conspicuously worn and the diamond coated tool only slightly after
higher values of the resultant force. Again, the most suitable and completion of the first block.
the least suitable cutting tools were determined, this time for the
maximal peaks of the resultant force. This result corresponds to 3. Proposal of new tools
the results of burr evaluation. Other cutting tools deserve deeper Two prototype tools were developed using information from
analysis because their rank of suitability was not definitive for all runs preliminary tests of commercial milling tools. The main difference in
of this experiment. It is possible to estimate from the diagram that the tool design is cutting edge material and geometry. Both cutting
generally, cutting tools “E” and “B” were less suitable and tools “D” tools had a compressive helix. One was cemented carbide with
and “A” were more suitable, as in the case of burr evaluation. diamond coating and the second one was carbide tool with soldered
S/N ratio was made in order to identify the most significant control
factors. The most significant control factor was feed per tooth for
most cutting tools. Another result was only for “D”. It had radial depth
of cut as the most significant factor probably due to a high number
of teeth. It is recommended to machine with low feeds. All tested
tools reached smaller forces for low level of feed per tooth in the test.
Radial dept of cut and milling strategy were similarly significant. The
optimal milling strategy was up milling. Cutting velocity influenced Figure 8. New developed cutting tools DEMO1 (left), DEMO2 (right).

Identification of Number of Diameter Material of Clearance Rake Helix

cutting tool teeth [mm] cutting edge angle angle angle
DEMO1 5x2 8 Carbide with diamond coating 13 20 20
DEMO2 4x2 8 PCD 16 3 16

Table 4. Parameters of new cutting tools.

DEVELOPMENT OF A CUTTING TOOL FOR COMPOSITES WITH THERMOPLASTIC MATRIX | October | 2013 | 424/425
PCD tips. A maximum number of teeth was designed for the diameter The cutting tool with soldered PCD tips had the best abrasion
of 12 mm in order to meet the requirements of high positivity of cutting resistance. It is one of the hardest materials used for cutting. The
angles and helix for maximizing cutting tool productivity. wear of cutting tool DEMO2 is less than 70 μm on the flank after
The rake angle for the PCD cutting tool was only 3°. The reason for 30 minutes of machining. The cutting tool with diamond coating
this were difficulties in production. PCD tips are made as flat plates. had a less rapid process wearing off. The flank wear increased
Higher positivity of the rake angle can be reached only by sloping almost linearly for the diamond coated tool whereas the flank wear
the PCD tip in the carbide bed. That means that the rake angle was of the PCD tool almost did not rise after 10 minutes of machining.
variable along the whole length of the helix and 3° of the rake angle A titanium coating of cutting tool “F” was worn rapidly. The flank
was only in the cutting plane. wear of this tool was more than twice as much for new tools after
Both cutting tools were tested on high performance material ten minutes of machining. The intensive wearslowed down after
PEEK/C. PEEK is a polymer matrix characterized by higher toughness then the titanium coating disappeared and the cutting edge began
and higher melting point and glass transition temperature than PPS. abrading. This situation was observed also for DEMO1 but first after
The reinforcement was multidirectional and carbon fibres were AS4. 20 minutes of machining (Fig. 11).
Criteria of testing were mainly burr creation and tool life time.

3.1 Results and discussion for new tools

The burr area was measured for both the new cutting tools and the
best commercial tool “F” for comparison. The results of the test are
in Fig. 9. The cutting tool DEMO2 created big burrs along the whole
length of the specimen as is shown in the same figure. The burrs
were bigger for small feed per tooth than for high in the case of this
tool. Tool DEMO2 had a small rake angle and it could be the cause
of the bad results in terms of the burr creation. The results of the new
carbide tool with diamond coating and commercial “F” are almost
the same. Both tools had a very positive cutting edge, which was
beneficial for successful cutting–off of burrs.
Figure 11. Abrasion wear of DEMO1 cutting tool.

4. Conclusions
The test of commercial cutting tools were made in order to find the
best cutting shape geometry and cutting material for machining
FRTC materials. The best choice is a very positive rake and clearance
angle, compressive helix and very hard cutting material such as PCD
or diamond coating on a carbide substrate.
Cutting forces were very small during trimming of FRTC materials.
If the cutting forces are very small, the differences between various
cutting conditions may be also very small in the statistical evaluation.
This was caused by vibrations which distorted all measurement,
namely measurement of multitooth cutting tools. It made the results
of evaluated S/N ratios slightly confusing for weak control factors.
Surface roughness was influenced by many factors. Nevertheless,
Figure 9. Creation of burrs – Comparison of the new cutting tools measurements showed that surface roughness Rz was smaller
and cutting tool “F“. than 6 μm for all tools. Unfortunately, the differences between
measurements were too small and the S/N ratio could not be evaluated.
The next comparison was focused on the tool life time tests. The The burr-less machined edge of the composite was obtained only
testing was made under the same cutting conditions and these were: when the tool had high positivity of the rake and clearance angle.
Feed per tooth 0.1 mm, cutting velocity 300 m/min, radial depth of It is easy to reach a high positive angle on carbide tools but they only
cut 3 mm, down milling, and the surface ply orientation PW. have limited tool lifetime. PCD tools have much longer lifetime but
there is an issue with the small positivity of the rake angle. The burrs
were bigger for PCD tools.
Results from testing of standard commercially available tools were
used for the design of two new specific tools. Both the new designed
tools have double helix design for minimizing burr size. The first one
has diamond tips and 4 teeth. The tool geometry creates medium
burr because of a less positive rake angle (the tip itself has a rake
angle of 0°; the positive rake angle is achieved just by declining the
tip in the tool body). The tool life time is very long because of the
application of PCD. The cutting productivity is very high because of
the high number of teeth.
The other one is made from cemented carbide with a diamond-
like coating. The ground geometry generates small burr because of
non-changing positive cutting edge geometry. The tool life time is
approximately 10 times lower than in the case of PCD. The carbide
cutting edge with the diamond-like coating has lower performance
Figure 10. Tool lifetime tests – Comparison of the new cutting tools than the diamond cutting edge. The cutting productivity is very high
and cutting tool “F“. because of the high number of teeth.
Acknowledgements [Masek 2012] Masek, P., Zeman, P. and Kolar, P., Technology
The authors would like to thank the European Community for funding optimization of PPS/C composite milling using Taguchi method,
the large-scale EU project „FibreChain – Integrated Process Chain MATAR2012, Prague, September 2012, pp. 1-8
for Automated and Flexible Production of Fibre-Reinforced Plastic [Peters 1998] Peters, S. T., Handbook of Composites, London:
Products“ as part of the European Community’s Seventh Framework Chapman and Hall, 1998.
Programme under grant agreement no. 263385. [Quadrini 2007] Quadrini, F., Squeo, E. A., and Tagliaferri, V.,
Machining of glass fiber reinforced polyamide, Rome, Italy: eXpress
References Polymer letters Vol. 1, No.12, 2007.
[Coligan 1991] Colligan, K. and Ramulu, M., Delamination [Sheikh-Ahmad 2009] Sheikh-Ahmad, J. Y., Machining of polymer
in surface plies of graphite/epoxy caused by the edge trimming composites, New York: Springer Science + Business Media, 2009.
process, Processing and manufacturing of composite materials – e-ISBN 978-0-387-68619-6.
ASME, Washington: PED-CP.49/MD-Vol. 27, 1991. [Taguchi 1987] Taguchi, G. and Konichi, S., Taguchi methods,
[Davim 2004] Davim, P. J., Reis, P. and Antonio, C. C., A study Orthogonal arrays and linear graphs, tools quality engeneering,
on milling of galss fiber reinforced plastics manufactured by hand- MI: ASI Press, 1987.
lay up using statistical analysis (ANOVA), Porto, Portugal: Elsevier
– Composite structures 64, 2004, pp. 493-500. Contacts
[Ehrenstein 2009] Ehrenstein, G. W., Polymer Composites (in Czech), Ing. Petr Masek, Research Center of Manufacturing Technology
Prague: Scientia, 2009. ISBN 978-80-86960-29-6. Czech Technical University in Prague, Faculty of Mechanical Engineering
[König 1985] König, W., et. al, Machining of fibre reinforced plastics, Horska 3, Prague 2, 128 00, Czech Republic
CIRP Annals – Manufacturing Technology, Vol. 34, No. 2, pp. 537-548, tel.: +420 221 990 945, +420 224 359 348
1985. e-mail: [email protected], www.rcmt.cvut.cz

DEVELOPMENT OF A CUTTING TOOL FOR COMPOSITES WITH THERMOPLASTIC MATRIX | October | 2013 | 426/427

View publication stats