ARTICLE IN PRESS

International Journal of Machine Tools & Manufacture 48 (2008) 715–720

www.elsevier.com/locate/ijmactool

Short Communication

Effects of high speed in the drilling of glass fibre reinforced plastic:

Evaluation of the delamination factor
J. Campos Rubioa,, A.M. Abraoa, P.E. Fariaa, A. Esteves Correiab, J. Paulo Davimc
a
Department of Mechanical Engineering, Federal University of Minas Gerais Av. Antônio Carlos 6627 – Pampulha,
Belo Horizonte MG, CEP: 31.270 901, Brazil
b
Department of Mechanical Engineering and Industrial Management, School of Technology, Polytechnic Institute of Viseu - IPV Campus de Refeses,
3504 510 Viseu, Portugal
c
Department of Mechanical Engineering, University of Aveiro, Campus Santiago, 3810 193 Aveiro, Portugal
Received 16 January 2007; received in revised form 18 October 2007; accepted 23 October 2007
Available online 1 November 2007

Abstract

High speed machining (HSM) is an outstanding technology capable of improving productivity and lowering production costs in
manufacturing companies. Drilling is probably the machining process most widely applied to composite materials; nevertheless, the
damage induced by this operation may reduce drastically the component performance. This work employs HSM to realize high
performance drilling of glass fibre reinforced plastics (GFRP) with reduced damage. In order to establish the damage level, digital
analysis is used to assess delamination. A comparison between the conventional (Fd) and adjusted (Fda) delamination factor is presented.
The experimental results indicate that the use of HSM is suitable for drilling GFRP ensuring low damage levels.
r 2007 Elsevier Ltd. All rights reserved.

Keywords: Delamination; Polymer-matrix composites; Drilling; HSM

1. Introduction that they provide a unique balance of chemical and

mechanical properties combined with wide processing
The use of fibre reinforced plastic (FRP) composites in versatility [4]. Within reinforcing materials, glass fibres
the automobile and aerospace industries has increased are the most frequently used in structural constructions
considerably in recent years due to their particular because of their specific strength properties [3].
properties, such as high specific stiffness and strength, Drilling is probably the machining process most broadly
high damping, good corrosive resistance and low thermal applied to composites since components made out of
expansion [1]. Compared to metal systems, FRP of lighter composite materials are usually near net shaped, thus
weight possess low radar and thermal signatures and resist requiring holes for assembly integration. Delamination is a
to degradation caused by chemicals, biological agents and major problem associated with drilling fibre reinforced
even to nuclear radiation, providing a comparable level of composite materials and, in addition to reducing the
protection [2]. structural integrity of the material, it also leads to poor
Composite materials are constituted of two phases: the assembly tolerances and has the potential for long-term
matrix, which is continuous and surrounds the other phase, performance deterioration. Two delamination mechanisms
often called the dispersed (or reinforcing) phase [3]. Epoxy associated with drilling FRP are known as peel-up at the
resins are widely used as matrix in many fibre reinforced drill entrance and push-out at the drill exit [1]. The key for
composites; they are a class of thermoset materials of solving the problem lies in reducing the thrust force when
particular interest to structural engineers owing to the fact drilling.
Several techniques have been employed to measure
Corresponding author. Tel.: +55 31 3499 5138; fax: +55 31 34433783. delamination after drilling composites, such as optical
E-mail address: [email protected] (J. Campos Rubio). microscopy and scanning and digital photography.

0890-6955/$ - see front matter r 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ijmachtools.2007.10.015
 ARTICLE IN PRESS
716 J. Campos Rubio et al. / International Journal of Machine Tools & Manufacture 48 (2008) 715–720

Generally speaking, a quantitative evaluation is required in of the hole, which increases together with the thickness of
order to assess the effect of the principal cutting parameters the fibre bundle for the same edge position angle.
and of the geometry of the drill [5]. The delamination Nevertheless, few works have been conducted using
factor has been widely broadly to characterize the level of HSM and obtaining a relationship between the various
damage on the work material at the entrance and exit of controllable parameters and their influence on the quality
the drill. The delamination factor (Fd) may be calculated of hole. Owing to the fact that HSM leads to lower cutting
from the ratio of the maximum diameter (Dmax) of the forces and energy consumption together with higher
delamination zone to the drill diameter (D0), as indicated in removal rates, less damage in the work piece is expected.
Inoue et al. [9] investigated the relationship between the
Dmax internal damage around the drilled hole and the number of
Fd ¼ . (1)
D0 holes produced. Glass fibre reinforced epoxy resin used in
printed circuit board was tested as work material. The
Alternatively, the ratio of the delaminated area to the hole
holes were drilled under the following conditions: spindle
area may also been used. Davim et al. [5] used digital image
speeds from 20 000 to 80 000 rpm and feed rates from 10 to
analysis to evaluate the damage area after drilling
50 mm/rev using 1 mm diameter drills. The results showed
composite laminates. In this case, the adjusted delamina-
that drilling at high feed rates a large number of holes are
tion factor (Fda) is calculated from Eq. (2), in which the
produced with uniform quality, whereas when drilling at
first part of Eq. (2) represents the size of the crack
low feed rates a small number of holes with high quality are
contribution (conventional delamination factor—Fd) and
produced. Finally, the authors concluded that the ratio of
the second part represents the damage area contribution:
the radius of the drill to the width of the yarn is an
Dmax Amax important factor to estimate the tool life of small diameter
F da ¼ a þb , (2) drills. In the case of drills with a radius larger than the
D0 A0
width of the yarn, high wear rates were recorded.
where Amax is the area related to the maximum diameter of In order to reduce tool wear when drilling composite
the delamination zone (Dmax) and A0 is the area of the materials, Kao [10] investigated the tribological properties
nominal hole (D0). The parameters a and b denote weights of coated drills against glass fibre reinforced epoxy resin.
to the parts of Eq. (2). b is considered as the ratio of the The results indicated that coating microdrills with 5%
damage area (Ad) to the area corresponding to Dmax minus MoS2–Cr, a two-fold increase in tool life was observed
the nominal hole area (A0Amax). compared to uncoated drills. In addition to that, the
The influence of factors such as tool geometry and quality of the holes was not impaired when high speed
machining parameters on delamination have been studied drilling through holes at 100 000 rpm.
by several researchers. Nevertheless, few authors have In the present work, HSM was employed to high
approached both tool geometry and high speed machining performance drilling of glass/epoxi composite material
(HSM) when drilling composites, more specifically glass with minimal damage.
fibre composites. Davim et al. [6] compared the influence of
different drill geometries on the delamination of hand lay-
up glass fibre reinforced plastic (GFRP) laminates. A 2. Experimental procedure
toolmaker’s microscope was used to evaluate the damage
as the machining parameters were altered. The delamina- Drilling experiments were conducted on a machining
tion factor was considered as the ratio of maximum centre with 11 kW spindle power and a maximum spindle
diameter in the damage zone to the drill diameter. The speed of 10 000 rpm. An aerostatic headstock by Air
results indicated that the damage increases with both Turbine Tools (Boca Raton, USA) with 40 000 rpm
cutting speed and feed rate. maximum rotational speed was used for the HSM tests.
Arul et al. [7] induced axial vibrations in the feed The laminates were produced by the hand lay-up technique
direction and studied the influence of the vibration and were made up of epoxy matrix reinforced with 50%
frequency and amplitude when drilling GFRP composites. weight of woven glass fibre with an orientation of 0/901.
In order to establish the extent of the defects caused by Ten alternating layers of glass fibre were used resulting in a
drilling, specimens were examined using an optical micro-
scope coupled to an image analyser. The delamination
factor was obtained from the ratio between the maximum
Table 1
diameters in the damage zone to the drill diameter.
Identification and properties of the matrix and reinforcing materials
Aoyama et al. [8] investigated the damage after drilling
holes of small diameter in printed wiring boards and Epoxy matrix Araldite M
concluded that delamination (which leads to ion migration) Hardener HY 956
is generated along the fibre in the hole wall surface as the Matrix modulus of elasticity 2.15 GPa
Matrix maximum strength stress 82.55 MPa
surface roughness increases. According to these authors, Reinforcing E-glass fibre (240 g/m2)
the thickness of the fibre bundle affects the internal damage
 ARTICLE IN PRESS
J. Campos Rubio et al. / International Journal of Machine Tools & Manufacture 48 (2008) 715–720 717

laminate 2.570.2 mm thick. Table 1 shows the identifica- 3. Results and discussion
tion and properties of the matrix and reinforcing materials.
Three cemented carbide drills (ISO grade K20) with Composite laminates usually present more severe dela-
5 mm diameter and 251 helical angle manufactured by mination at the drill exit. However, when drilling double
Guhring oHG were used as cutting tools: two twist drills board laminates and using higher speeds, the damage is
coded DIN 6539 RN and DIN 8038 RN with 1181 and 851 more critical at the drill entrance (peel-up delamination).
point angle, respectively, and one ‘‘Brad & Spur’’ special Table 3 shows the calculated Fd values for distinct test
geometry drill [11] code WN R FK. conditions. The effect of both feed and spindle speeds on
An appropriate clamping system was devised to fix the peel-up delamination can thus be assessed.
laminates in the machining centre. Two composite boards Figs. 1(a)–(c) show the effect of distinct drill geometries
were drilled simultaneously. Table 2 shows the cutting and machining parameters on the conventional delamina-
parameters (spindle speed and feed speed) used in the tion factor, measured at the drill entrance side of the
experimental work. Each test was replicated twice. The laminate. In general, it can be noticed that the delamina-
damage around the holes (conventional delamination tion factor increases as the feed speed is elevated but
factor—Fd) was measured using a Mitutoyo TM 500s decreases as the spindle speed is increased, irrespectively of
toolmaker’s microscope with 30  magnification and 1 mm
resolution. In order to identify the machining parameters
associated to minimal damage of the laminates, digital 115°
analysis was employed to measure the adjusted delamina- 2.2
4000 rpm

Delamination Factor (Fd)

tion factor (Fda) for the drill geometry responsible for the 2 8000 rpm
best results concerning the conventional delamination 40000 rpm
1.8
factor.
1.6

1.4
Table 2
Cutting parameters tested 1.2

1
Test run Spindle speed (rpm) Feed speed (mm/min) 1000 3000 6000 9000
1 4000 1000 Feed speed (mm/min)
2 3000
85°
3 6000
4 9000 2.2
5 8000 1000 4000 rpm
Delamination Factor (Fd)

2 8000 rpm
6 3000
40000 rpm
7 6000 1.8
8 9000
9 40 000 1000 1.6
10 3000
1.4
11 6000
12 9000 1.2

1
1000 3000 6000 9000
Table 3 Feed speed (mm/min)
Conventional delamination factor (Fd) measured at the drill entrance
2.2
Test run Delamination factor (Fd) 4000 rpm
Delamination Factor (Fd)

2 8000 rpm
DIN 6539 DIN 8038 WN R FK 40000 rpm
1.8
1 1.396 1.151 1.258
2 1.719 1.463 1.551 1.6
3 2.037 1.626 2.041
1.4
4 2.143 1.796 2.094
5 1.438 1.183 1.178 1.2
6 1.663 1.238 1.511
7 1.938 1.422 1.507 1
8 1.744 1.541 1.741 1000 3000 6000 9000
9 1.296 1.288 1.237 Feed speed (mm/min)
10 1.513 1.307 1.192
11 1.608 1.258 1.207 Fig. 1. The effect of feed speed on the delamination factor (Fd) for
12 1.521 1.322 1.225 different tool geometries: (a) 1151 point angle (DIN 6539 RN); (b) 851
point angle (DIN 8038 RN); and (c) ‘‘Brad & Spur’’ drill (WN R FK).
 ARTICLE IN PRESS
718 J. Campos Rubio et al. / International Journal of Machine Tools & Manufacture 48 (2008) 715–720

the tool geometry. In addition to that, under the highest delamination factor values, see Fig. 1(a). The drill coded
spindle speed (40 000 rpm) the delamination factor seems to DIN 8038 RN (851 point angle) gave best results at low and
be less sensitive the feed speed. This behaviour may be intermediate spindle speeds, as shown in Fig. 1(b). Finally,
explained by the cutting temperature, i.e., at lower spindle when high speed drilling at 40 000 rpm, drill WN R FK
speeds lower temperatures are observed and delamination accounted for the lowest delamination factors, as indicated
takes place as a result of the stresses imposed by shearing, in Fig. 1(c).
however, at a spindle speed of 40 000 rpm the heat The effect of the material removal rate on the adjusted
generated by friction between the major and minor cutting delamination factor for the three drills tested is given in
edges and the work material promotes the softening of the Fig. 2. In this case only the highest spindle speed
matrix. As a consequence, when drilling at the highest (40 000 rpm) was tested. The results suggest that the drill
spindle speed, delamination is not drastically affected by geometry WN R FK not only provides lowest delamina-
feed speed. Comparing the effect of the three drill tion factor, but also is also capable of maintaining the
geometries tested, one can conclude that the twist drill delamination factor at stable values as the material
with 1151 point angle (DIN 6539 RN) provides higher removal rate is elevated. The superior performance of the
Brad & Spur drill may be explained by the fact that, in
contrast to the other drills, the shear plane area is smaller,
DIN 6539 RN 115°
DIN 8038 RN
thus probably requiring lower thrust forces and resulting in
1.7
less delamination of the work materials, acting similarly to
Delamination Factor (Fd)

WN R FK
1.6
a trepanning tool.
1.5 85° Fig. 3 shows the captured and processed images of the
1.4 holes entrance generated in GFRP using the WN R FK
1.3 drill. The features observed can be used to evaluate the
1.2 characteristic damage after high speed drilling. It can be
1.1 seen that for spindle speeds of 4000 and 8000 rpm the
1 damage increases considerably with feed speed and
19.64 58.91 117.81 176.72 decreases as the spindle speed is elevated. In contrast, the
Material removal rate (cm3/min) holes produced at a spindle speed of 40 000 rpm possess a
Fig. 2. The effect of material removal rate on the delamination factor (Fd)
uniform damage pattern, irrespectively of the feed speed
measured at the drill entrance at high spindle speed (40 000 rpm) using used, thus indicating that the cutting temperature governs
different drill geometries. the extension of the damage. Furthermore, the tests

4000 rpm 8000 rpm 40000 rpm 4000 rpm 8000 rpm 40000 rpm
1m/min
3 m/min
6 m/min
9 m/min

Fig. 3. Captured (a) and processed (b) images at the hole entrance on GFRP using drill WN R FK.
 ARTICLE IN PRESS
J. Campos Rubio et al. / International Journal of Machine Tools & Manufacture 48 (2008) 715–720 719

employing 40 000 rpm were undertaken using the aerostatic Fda

headstock, while the tests at 4000 and 8000 rpm were Fd
Ad / (Amax-Ao)
carried out connecting the tool holder directly to the main 1.5
spindle of machining centre. It is believed that the

Damage area (Ad/Amax-Ao)

aerostatic headstock would lead to less delamination due 1.3

Delamination Factor
to its inferior run out deviation.
The value of the conventional (Fd) and adjusted (Fda) 1.1
delamination factors was obtained using the approach
proposed by Davim et al. [5]. Therefore, b is considered as 0.9
the ratio of the damage area (Ad) to the area corresponding
to Dmax minus the nominal hole area (A0Amax). The 0.7
parameter a is the complement of b (a ¼ 1b). Thus,
Eq. (2) can be rewritten as 0.5
Ad 1000 3000 6000 9000
F da ¼ Fd þ ðF 2  F d Þ. (3) Feed speed (mm/min)
ðAmax  A0 Þ d
Fig. 5. The effect of feed speed on the conventional (Fd) and adjusted
Table 4 presents the results obtained when Eq. (3) is
(Fda) delamination factor using ‘‘Brad & Spur’’ drill WN R FK when high
applied and Fig. 4 shows both delamination factors (Fd and speed drilling at 40 000 rpm.
Fda) obtained under distinct cutting conditions using the

Table 4
The effect of spindle speed and feed speed on both conventional (Fd) and WN R FK drill. The effect of the damage area on the
adjusted (Fda) delamination factors
adjusted delamination factor can be assessed for spindle
Test Delamination factors speeds of 4000 and 8000 rpm. It can be seen that for low
run spindle speeds the conventional delamination factor is
Fd Ad Ad =ðAmax  A0 Þ Fda ðF da  F d Þ=F da  100% closer to the adjusted factor, indicating that the damage
1 1.258 3.346 0.292 1.353 7.01 area (Ad) is small; however, the difference between Fd and
2 1.551 5.005 0.181 1.706 9.08 Fda increases as feed speed is elevated. Furthermore, for a
3 2.041 8.400 0.135 2.328 12.33 spindle speed of 4000 rpm, the difference between the
4 2.094 12.127 0.183 2.512 16.65 conventional and adjusted delamination factors remains
5 1.178 1.858 0.244 1.229 4.16
almost unaltered.
6 1.511 5.165 0.205 1.669 9.48
7 1.507 5.238 0.210 1.667 9.61 Furthermore, the results presented in Fig. 4 suggest that
8 1.741 8.935 0.224 2.030 14.25 the adjusted delamination factor is more sensitive to the
9 1.237 7.100 0.681 1.437 13.90 machining parameters (spindle speed and feed speed) than
10 1.192 5.670 0.685 1.349 11.63 the conventional delamination factor; therefore, the former
11 1.207 4.998 0.558 1.347 10.36
seems to be more suitable for mathematical modelling of
12 1.225 5.310 0.539 1.374 10.82
the damage, as reported by [12,13].
Finally, Fig. 5 allows a comparison between Fd and Fda
and the area relationship for drill WN R FK when high
2.6 speed drilling at 40 000 rpm. It can be seen that both
Fda - 4000 rpm
Fd -4000 rpm
delamination factors are not considerably affected by feed
2.4
Fda - 8000 rpm speed, whereas the area relationship is reduced as feed
2.2 Fd -8000 rpm speed is increased.
Delamination Factor

2
4. Conclusions
1.8
Based on the experimental results obtained concerning
1.6
the damage induced after high speed drilling glass fibre
1.4 reinforced laminates using three cemented carbide drills
with different geometries, the following conclusions can be
1.2
extracted:
1
1000 3000 6000 9000  Within the cutting range tested, delamination decreases
Feed speed (mm/min) as the spindle speed is elevated, probably owing to the
Fig. 4. The effect of feed speed on the conventional (Fd) and adjusted fact that cutting temperature is elevated with spindle
(Fda) delamination factors using ‘‘Brad & Spur’’ drill WN R FK and lower speed, thus promoting the softening of the matrix and
spindle speeds. inducing less delamination.
 ARTICLE IN PRESS
720 J. Campos Rubio et al. / International Journal of Machine Tools & Manufacture 48 (2008) 715–720

 For spindle speeds of 4000 and 8000 rpm delamination [2] A. Velayudham, R. Krishnamurthy, T. Soundarapandian, Evalua-
increases with feed speed. This can be explained by the tion of drilling characteristics of high volume fraction fibre glass
reinforced polymeric composite, International Journal of Machine
fact that using higher feed rates, higher thrust force
Tools and Manufacture 45 (2005) 399–406.
values are obtained, resulting in larger delamination. [3] W.D. Callister, Materials Science and Engineering: An Introduction,
However, for the highest spindle speed (40 000 rpm), an sixth ed., Wiley, Canada, 2002.
increase in feed speed did not result in appreciable [4] M.A. Boyle, C.J. Martin, J.D. Neuner, Epoxy resin, in: H. Hansmann
elevation of the damage. In this case, delamination is (Ed.), Composites Compendium, ASM Handbook FB MVU,
dictated by the cutting temperature, i.e., with the Werkstofftechnologien, Kunststofftechnik, 2003.
[5] J.P. Davim, J.C. Campos Rubio, A.M. Abrão, A novel approach
softening of the matrix, delamination will not be based on digital image analysis to evaluate the delamination factor
affected by feed speed. after drilling composite laminates, Composites Science and Technol-
 At high spindle speeds drill WN R FK produced less ogy 67 (2007) 1939–1945.
delamination compared to the twist drills with 1151 and [6] J.P. Davim, P. Reis, C.C. Antonio, Experimental study of drilling
851 point angle, owing to the fact that drill WN R FK glass fibre reinforced plastics (GFRP) manufactured by hand lay-up,
Composites Science and Technology 64 (2004) 289–297.
acts similarly to a trepanning tool, promoting lower [7] S. Arul, L. Vijayaraghavan, S.K. Malhotrab, R. Krishnamurthy, The
shear area and, consequently, lower thrust force values. effect of vibratory drilling on hole quality in polymeric composites,
 In order to obtain larger material removal rates International Journal of Machine Tools and Manufacture 46 (2006)
associated with minimal delamination, higher spindle 252–259.
[8] E. Aoyama, H. Nobe, T. Hirogaki, Drilled hole damage of small
speeds should be used when drilling GFRP. At a spindle
diameter in printed wiring board, Journal of Materials Processing
speed of 40 000 rpm, delamination is minimal, irrespec- Technology 118 (2001) 436–441.
tively of the feed speed employed. [9] H. Inoue, E. Aoyama, T. Hirogaki, K. Ogawa, H. Matushita, Y.
 The inclusion of the damage area in the adjusted Kitahara, T. Katayama, Influence of tool wear on internal damage in
delamination factor (Fda) obtained through digital small diameter drilling in GFRP, Composite Structures 39 (1–2)
image processing allows a more realistic assessment of (1997) 55–62.
[10] W.H. Kao, Tribological properties and high speed drilling applica-
drilling induced damage. Consequently, the effect of the tion of MoS2–Cr coatings, Wear 258 (2005) 812–825.
drilling parameters on the damage observed in GFRP [11] J.P. Davim, P. Reis, C.C. Antonio, Drilling fibre reinforced plastics
laminates can be more accurately determined. (FRPs) manufactured by hand lay-up: influence of matrix (Viapal
VUP 9731 and ATLAC 382-05), Journal of Materials Processing
Acknowledgements Technology 156 (2004) 1828–1833.
[12] R.Q. Sardiñas, P. Reis, J.P. Davim, Multi-objective optimization of
cutting parameters for drilling laminate composite materials by using
The authors would like to thank the Polytechnic genetic algorithms, Composites Science and Technology 66 (2006)
Institute of Viseu for supporting the experimental work. 3083–3088.
Part of the work reported here is sponsored by the CAPES/ [13] C.C. Tsao, H. Hocheng, Effects of special drill bits on drilling-
GRICES Research Program. induced delamination of composite materials, International Journal
of Machine Tools and Manufacture 46 (2006) 1403–1416.
References

[1] I. El-Sonbaty, U.A. Khashaba, T. Machaly, Factors affecting the

machinability of GFR/epoxy composites, Composite Structures 63
(2004) 329–338.