Journal of Materials Processing Technology, 94/1(1999), pp. 30-35.
A machinability study of polymer matrix composites using abrasive waterjet cutting technology
School of Mechanical, Manufacturing and Medical engineering,
Queensland University of Technology,
GPO Box 2434, Brisbane, Qld. 4001, Australia.
Fax: (+61.7) 3864 1469,
Abrasive Waterjet (AWJ) cutting is an emerging technology for material processing with the
distinct advantages of no thermal distortion, high machining versatility, high flexibility and
small cutting forces. In this paper, an experimental investigation of the machinability and
kerf characteristics of polymer matrix composite sheets under abrasive waterjets is presented.
It shows that this unique ‘cold’ cutting technology is a viable and effective alternative for
polymer matrix composite processing with good productivity and kerf quality. Plausible
trends of kerf quality with respect to the input parameters are discussed, from which
recommendations are made for process control and optimization.
Keywords: Abrasive Waterjet Cutting, Composite processing, Machinability, Kerf
Polymer matrix composites are being increasingly used in various applications due to their
superior physical and mechanical properties. However, there are a number of problems
associated with the processing of these kinds of materials. Traditional processing methods
such as band saw cutting result in not only low cut quality but also low productivity, while
the non-traditional Laser cutting technology has been found to yield large burr formation,
dimensional inaccuracy due to thermal distortion, heat affected zone (or burnt), and even fire
hazard for these heat sensitive materials . The unique ‘cold’ Abrasive Waterjet (AWJ)
cutting technology, due to its distinct advantages of no thermal distortion, high machining
versatility, high flexibility and small cutting forces , offers great potential for the
processing of polymer matrix composites. While considerable amount of work has been
reported on the investigation of Abrasive Waterjet cutting of various materials such as metals
and ceramics [3-7], including the studies of the mechanism of Abrasive Waterjet cutting
process, modeling for process control and optimization as well as the techniques to enhance
the cutting performance of Abrasive Waterjets [3,5,8-12], the study on the processing of
polymer matrix composites using AWJ cutting technology has received little attention ,
despite the increasing use of these materials in practice.
In this paper, the machinability of polymer matrix composites under abrasive waterjets is
studied based on an experimental investigation. A visualization study is carried out to
evaluate the microscopic features of the cut surfaces and the machinability of the materials.
This is followed by a statistical analysis of the experimental data to examine the kerf
characteristics as assessed by kerf geometry (kerf width and taper) and surface quality, from
which recommendations are made on selecting the optimum combination of the process
parameters for practical applications.
2. Experimental set-up and procedure
The experiment was conducted on a Flow Systems International waterjet cutter equipped
with a model 20X dual intensifier high output pump (up to 55,000 psi or 380 MPa) and a five
axis robot positioning system to cut 300×300 mm test specimens. The specimens were 3 mm
polymer based matrix compound being reinforced with teflon fabric using phenolic resin.
Some of the major mechanical and thermal properties of the specimen are given in Table 1.
Although AWJ cutting involves a large number of variables, as noted by Hashish , and
virtually all these variables affect the cutting results, only the major and easy-to-adjust
variables were considered. These included the water pressure, the nozzle traverse speed and
the stand-off distance between the nozzle and the workpiece. As such, the water pressures
within the common ranges of application and the equipment limit were tested at five levels
from 30,000 psi to 50,000 psi at an increment of 5,000 psi. For each level of the water
pressure, four levels of nozzle traverse speeds (1,000, 1,200, 1,500 and 1,800 mm/min) and
four levels of stand-off distance (2, 3, 4 and 5 mm) were used at a single level of abrasive
flow rate of 0.36 kg/min and a single level of impact angle of 90°. The other parameters were
kept constant using the system standard configuration, that is, the orifice diameter was 0.41
mm, the mixing tube diameter was 1.27 mm, and the length of mixing tube was 88.9 mm.
The abrasives used was almandite garnet sand with a mesh number of 80. Thus, a total of 80
straight cuts (slits) of about 100 mm length were carried out.
The approach to selecting the appropriate levels of traverse speed was such that at the
predetermined maximum stand-off distance and minimum water pressure, the traverse speed
was adjusted in actual operations to allocate the maximum traverse speed allowed for a
through cut. Other traverse speeds were accordingly selected at an appropriate spacing. Since
it has been known that the penetration depth of abrasive waterjet increases with an increase in
the water pressure and a decease in the stand-off distance , this approach could ensure
that all the combinations of the parameters so selected would produce through cuts for
evaluation. It should be noted that higher traverse speed may be possible at higher water
pressures, but this will result in negative effect on the kerf quality, as noted in the
experiments and reported in the paper. In addition, it has been known that water pressure will
be less effective if beyond a certain critical value [6,7]. It is therefore believed that the
selected range of nozzle traverse speeds was reasonable both practically and fundamentally.
3. General assessment of machinability
An observation study on all the cuts has revealed that AWJ produced clean slits of much
higher quality than other processes such as band saw and laser cutting  and the
productivity was comparable to the laser cutting process. The kerfs can be represented by
Figure 1 where the top kerf is commonly wider than the bottom as a unique feature of jet
cutting technology. Burrs were noticed at the jet exit side (bottom kerf) on some of the cuts
due to the material deformation when low water pressure and high traverse speed were used.
However, the burrs formed in AWJ cutting were found to be much smaller than the burrs
formed in band saw and laser cutting and could be easily removed. Therefore, burr formation
is not a major concern in AWJ cutting of the test material.
Some representative cuts were examined under an ‘Olympus’ stereo microscope to evaluate
the microscopic features of the waterjet machined surfaces. At 20x magnification, no
significant irregularity was observed in any of the examined cut surfaces although striations
exist on some of the cuts, typically at the lower portion of the cut surfaces. As expected for
this ‘cold’ cutting process, no heat affected zone was noticed on any cut surfaces.
Microscopic observations were also made under the Scanning Electron Microscope (SEM) to
study the cutting mechanisms and the features of the machined surfaces at 200x to 300x
magnifications. This again showed that AWJ can produce clean and good quality cut
surfaces. Consequently, AWJ cutting is a viable and alternative technology for polymer
matrix composite processing with good cut quality and productivity.
4. Kerf characteristics
4.1 Effect of cutting parameters on surface quality
It has been reported  that the topography of an AWJ cut surface may be divided into a
microscopic (roughness) component at the upper so-called cutting wear zone and a
macroscopic (striation or waviness) component at the lower deformation wear zone. While
surface roughness is a common phenomenon in all machining, striation or waviness is a
special feature of cuts with beam cutting technology, such as AWJ cutting. It is formed when
the ratio between the available energy of the beam and the required energy of the destruction
becomes comparatively small . In AWJ cutting, the cutting power of the jet decreases
from the top to the bottom of the cut surface and striations are formed at the lower portion of
the cut surfaces. Thus, surface roughness and striation characteristics have formed the criteria
to evaluate the quality of AWJ cut surfaces.
These characteristics of the cut surfaces was noted in the current study and confirmed by a
SEM analysis of selected samples. It has been found that while some cuts show clear patterns
of roughness and striations, others appear to be dominated by either roughness or striation
due to the large ranges of water pressure and traverse speed used. This result has made
quantitative analysis difficult. In general, the cut surface may be characterised by roughness
when high water pressure and low traverse speed are used. Conversely, at low water pressure
and/or high traverse speed, striation becomes the main feature of the cut surface.
Figure 2. Illustrations of kerf quality under Abrasive Waterjets.
Figure 2 gives three microphotographs showing some typical trends of the peak-to-valley
height (or amplitude) for either surface roughness or striation in terms of the process
variables. It has been found that based on the results taken at close to the top edge of the
kerfs, the surface roughness or striation amplitude increases with an increase in the water
pressure, as evidenced in Figures 2(a) and (b). These microscopic photographs show that for
the cutting conditions used, the maximum peak-to-valley height increased from about 20μm
to about 70μm when the water pressure increased from 30,000 psi to 40,000 psi. This may be
due to the fact that, at the same traverse speed, increased jet pressure causes more energy
disbursement from the abrasives in the area bombarded by the atomized waterjets, resulting
in locally widened kerf. It is also believed that at higher water pressure, the dynamic
behaviour of the waterjet will contribute more to the surface striation as the energy
fluctuation increases with an increase in the water pressure. However, a further quantitative
study is required to examine the surface characteristics in terms of surface roughness and
striation, respectively, under varied water pressure.
It may be anticipated that an increase in the traverse speed is associated with an increase in
the surface roughness and the amplitude of striation, since higher traverse speed allows less
overlapping machining action on the cut surface  and fewer abrasive particles impinging
the surface. The experimental data in this study follow this trends, as indicated in Figures
2(b) and (c). By contrast, the effect of stand-off distance on the surface roughness and
striation is hardly discernible.
4.2 Effect of cutting parameters on kerf geometry
Kerf geometry is a characteristic of major interest in abrasive waterjet cutting. As shown in
Figure 1, abrasive waterjets will generally open a tapered slot with the top kerf Wt being
wider than the bottom kerf Wb, and kerf taper or kerf taper angle θ is normally used to
represent this characteristic.
In the present work, the top and bottom kerf widths for all the cuts were measured with a
“Carl Zeiss” universal measuring microscope. Three measurements were taken for each cut at
the segment away from the ends of the slots to eliminate the effects of waterjet entry and exit
of the cutting process, from which the average reading was taken as the geometrical values.
The kerf taper angle θ was then calculated using the measured values of the top and bottom
kerf widths for each cut based on the equation θ = tan-1[(Wt – Wb)/(2 t)], where t is the
Figures 3 to 5 show some typical and representative trends and relationships between the kerf
geometry (top kerf width Wt, bottom kerf width Wb and kerf taper angle) and the cutting
parameters. It can be noted from Figure 3 that both the top and bottom kerf widths increase
approximately linearly with the water pressure, as higher water pressure results in greater jet
kinetic energy impinging onto the material and opens a wider slot. The kerf taper angle also
increases with the water pressure. This is because the bottom kerf width is not increased in
the same order as the top kerf width, as indicated in the figure. It follows that as the jet loses
its kinetic energy, it cannot remove the material adequately at the lower section, resulting in a
narrow bottom kerf.
The effect of traverse speed on the top kerf width, bottom kerf width and kerf taper is shown
in Figure 4. It can be seen from the figure that the traverse speed has a negative effect on both
the top and bottom kerf widths, while the kerf taper angle appears to be independent of the
traverse speed although the overall statistical results show that the taper angle decreases
slightly with an increase in the traverse speed. The negative effect of the traverse speed on
both the top and bottom kerf widths is due to the fact that a faster passing of abrasive waterjet
allows fewer abrasives to strike on the jet target and hence generates a narrower slot. The
nearly stabilised kerf taper is the result of the comparable rate of decreasing for the top and
bottom kerf widths.
It is interesting to note that the characteristics of the taper angle in terms of water pressure
and traverse speed discussed above are opposite to those reported in ceramic cutting . This
may stem from the different types of materials processed, different pressure and speed ranges
selected as well as different ratios of jet energy used to the energy required to cut the
Figure 5 shows that the top and bottom kerf widths increase with an increase in the stand-off
distance although a smaller rate associated with the bottom kerf width is observed. This may
be explained as the result of jet divergence when high-velocity waterjets spread out (at
different angles) as they exit from the mixing tube. Since the jet is losing its kinetic energy as
it penetrates into the work material, the outer rim of the diverged jet will not take effect as it
approaches the lower part of the kerf. As such, the stand-off distance has a lesser effect on the
bottom kerf width than the top kerf width. As a consequence of this effect, the kerf taper
angle is increasing with the stand-off distance, as shown in Figure 5.
It should be noted that if the kerf width and kerf taper can be predicted, they may be
compensated in the design and process planning stages and by controlling the nozzle in the
machine. For this purpose, a multiple variable regression analysis has been carried out at a
confidence interval of 95% which resulted in the following empirical models for the kerf
geometry in terms of process variables:
Wt = 0.842 + 0.009P + 0.0593Sd − 0.00089V (1)
θ = 0.71+ 0.064P+ 0.51Sd − 0.000895V (2)
where Wt is the top kerf width in mm, θ is the kerf taper angle in degrees, P is the water
pressure in 1000psi, Sd is the stand-off distance in mm, and V is the traverse speed in
mm/min. The bottom kerf width may be obtained from equations (1) and (2). These equations
may be used in practice for kerf geometry estimation as well as for process optimization
when the parameters are selected within the domain of the current study.
4.3 Cutting parameter selection consideration
In order to evaluate the overall kerf characteristics and to recommend the optimum
combination of the process parameters for cutting the material under consideration, the above
analyses and trends are summarised and given in Table 2 where the upwards and downwards
arrowheads indicate the increasing and decreasing trends of the quantities, respectively, with
an increase in each of the three variables. It appears that an increase in the stand-off distance
will result in an increase in the kerf width and kerf taper though the effect on the surface
smoothness is not significant. As such, it may be deduced that for the tested material and
machine setting conditions, stand-off distance should be selected as small as possible.
Increasing the traverse speed reduces the kerf widths but increases the surface roughness and
striation. In addition, traverse speed is directly proportional to the productivity and should be
selected as high as possible unless surface roughness is a primary concern. By contrast, an
increase in the water pressure will yield increased kerf width, kerf taper and cut surface
roughness (and striation). However, higher water pressure will allow higher traverse speed to
be used for through cuts. From this study, it is recommended that a water pressure of 30,000
psi with a traverse speed of 1,800 mm/min be used along with a minimum possible stand-off
A study of Abrasive Waterjet cutting of polymer matrix composites has been presented based
on an experimental investigation. It has been shown that Abrasive Waterjet cutting is a viable
and effective alternative for polymer matrix composite processing with good productivity and
kerf quality. The analysis and empirical models of kerf characteristics in terms of process
parameters have provided a means of estimating kerf geometry and compensating the
inclination and width of the kerf in the design and processing stages. The combination of
process parameters recommended for the material under consideration may be used for
maximizing the productivity while maintaining good kerf quality in practice. The study on
the cutting mechanisms and predictive models for the depth of cut in processing thick
materials is being carried out and it is hoped to report on this work shortly.
The author wishes to thank Mr. J. Gatt at Queensland Manufacturing Institute and Mr. N.R.
Sivakumar, a former Masters student of Queensland University of Technology, for their
assistance in the experimental work.
 Krishnan, V.K., “CO2 laser cutting of polymer matrix composites”, M. Eng. Sc. Thesis,
Queensland University of Technology, 1996.
 van Luttervelt, C.A., “On the selection of manufacturing methods illustrated by an
overview of separation techniques for sheet materials”, Annals of CIRP, 38(2), 1989,
 Hashish, M., “A modeling study of metal cutting with abrasive waterjets”, Journal of
Engineering Materials and Technology, 106(1984), pp.221-228.
 Matsui, S,. Matsumura, H., Ikemoto, Y., Tsujita, K. and Shimizu, H., “High precision
cutting methods for metallic materials by abrasive waterjet”, Proc. 10th International
Symposium on Jet Cutting Technology, Cranfield, UK, 1990, pp. 263-278.
 Guo, N.S., Louis, G. and Meier, G., “Surface structure and kerf geometry in abrasive
waterjet cutting: formation and optimization”, 7th American Waterjet Conference,
Seattle, Washington, USA, 1993.
 Chen, L., Siores, E. and Wong, W.C.K., “Abrasive waterjet cutting of alumina
ceramics”, Proc. 6th International Conference on Manufacturing Engineering,
Melbourne, Australia, 1995, pp. 709-713.
 Wang, J. and Wong, W.C.K. “Kerf characteristics in abrasive waterjet cutting of
metallic coated sheet steels”, Proc. World Congress-Manufacturing Technology
Towards 2000, Cairns, Australia, 1997, pp.157-166.
 Hashish, M., “A model for abrasive waterjet performance optimization”, Journal of
Engineering Materials and Technology, 111(1989), pp.154-162.
 Hashish, M., “On the modeling surface waviness produced by abrasive waterjets”,
Proc. 11th Int. Symposium on Jet Cutting Technology, Dordrecht, NL, 1992, pp.17-34.
 Tan, D.K.M., “A model for the surface finish in abrasive-waterjet cutting”, Proc. 8th
Int. Symposium on Jet Cutting Technology, Crianfield, UK, 1986, pp.309-313.
 Chao, J., Geskin, E.S., and Chung, Y., “Investigation of the dynamics of surface
topography formation during abrasive waterjet cutting”, Proc. 11th International
Symposium on Jet Cutting Technology, Dordrecht, NL, 1992, pp.593-603.
 Siores, E. and Wong, W.C.K., Chen, L. and Wager, J.G., “Enhancing abrasive waterjet
cutting of ceramics by head oscillation techniques”, Annals of CIRP, 45(1), 1996,
 Koenig, W., “Machining of fibre reinforced plastics”, Annals of CIRP, 34(2), 1985,
 Chen, L., Siores, E. and Wong, W.C.K., “High-pressure abrasive waterjet erosion
process”, Proc. Pacific Conference on Manufacturing, Seoul, Korea, 1996, pp.642-647.