Wear of rotary carbide tools in machining of AlrSiCp composites
S.S. Joshi a , N. Ramakrishnan
a,)
, H.E. Nagarwalla b, P. Ramakrishnan
c
a
c
Department of Mechanical Engineering, Indian Institute of Technology, Powai, Mumbai, 400 076, India
b
Nagarwalla Consultants, Mumbai, 400 021, India
Department of Metallurgical Engineering and Material Science, Indian Institute of Technology, Powai, Mumbai, 400 076, India
Abstract
With the projected widespread application of Metal Matrix Composites, it is necessary to develop an appropriate technology for their
efficient and cost-effective machining. This paper deals with the study of feasibility of rotary carbide tools in the intermittent machining
of AlrSiCp composites. A rotary tool holder was designed and fabricated for this work. Experiments were designed using Taguchi
Methods to analyse the influence of various factors and their interactions on the flank wear of rotary carbide tools during machining. A
tool-life model describing the effect of process, tool and material dependent parameter on the magnitude of flank wear of a rotary carbide
tool is proposed.
Keywords: Metal Matrix Composites; Machining; Tool wear; Taguchi Methods; Statistical design of experiments; Rotary tools
1. Introduction
Metallic composites with aluminium alloy reinforced
with discontinuous ceramic reinforcements are rapidly replacing the conventional materials in various industrial and
engineering applications w1,2x. A key to obtain maximum
benefits from this potential technology lies in its early
commercialisation. One of the ways for this could be to
direct efforts to blend the entire technology for these
materials into a common program consisting of processing,
characterisation and machining w3,4x. A global view of the
technology for these materials envisages a need for a
sound processing technology with predictability in the
properties of resulting composites, adequate methods and
standards for property evaluation and a deeper understanding of their machining aspects. A review of literature on
the machining aspects of these composites reveals that: Ža.
Majority of the literature deals with the study of wear of
various types cutting tool materials during the continuous
machining of the composites by the processes such as
turning, drilling, etc. w5–11x. Žb. It is now established that
the cost-effective continuous machining of composites can
be accomplished by using PCDrCBN tools only. Žc. The
intermittent machining of composites has not received
enough attention. PCD tools have exhibited tendencies to
chip-off easily under the impact encountered during shaping. Žd. Taylor’s tool-life equation is not sufficient to
represent the tool wear pattern during machining of composite w6x. Že. The mechanism of tool wear during machining of these composites was found to be predominantly
abrasive and no evidence of chemical wear was observed
w5,7x.
In view of the above, it was perceived that the concept
of rotary carbide tools could be a good alternative to the
stationary carbide tools in the continuous machining, and
to PCDrCBN tools in the intermittent machining. The
carbide tool material being more impact resistant than
PCDrCBN tools can overcome the problem of chipping
associated with PCDrCBN tools. Besides, the rotation of
cutting edge would give increased tool-life as compared to
stationary carbide tools. The pioneering work in the area of
rotary tools was done in as early as 1968 w12,13x. Recently,
the continuous machining of the composites by rotary tools
has been discussed in Ref. w14x.
It was also evident from the literature that the use of
conventional Taylor’s tool-life equation for these composites yields very high value for n, the Taylor’s tool-life
exponent, indicating less dependency of tool wear on the
cutting speed w6x. However, in reality, the predominant
abrasive action of the reinforcement particles in composites is a strong function of cutting speed Žor abrasion
125
Table 1
Independent variables and their levels
Fig. 1. Concept of inclination angle.
velocity.. Hence, a need for a tool-life model specific to
machining of these materials was envisaged.
Since the past experience regarding the application of
rotary carbide tools for composite materials is not comprehensive, a statistically designed experimentation would
highlight the influence of various factors on the magnitude
of tool wear. Thus, the primary aim of this work is to
conduct an investigation to understand the influence of
various factors and their interactions on the magnitude of
wear of rotary carbide tools during intermittent machining
of composites. Further, modeling of tool wear process has
been carried out so as to include volume of reinforcement,
a material dependent parameter in the tool life model
besides the process and tool setting dependent parameters.
The experiments were designed using Taguchi Methods.
2. Design of the experiment
2.1. Selection of response Õariables
The selection of response variables was done on the
basis of pilot experiments performed using four criteria:
Ž1. Magnitude of flank wear, Ž2. Magnitude of face wear,
Variables)
Cutting
speed
Žmrmin.
Feed
rate
Žmmrrev.
Inclination
angle
Ždeg.
Volume
reinforcement in
the composite
Žpercent of SiCp.
Level 1
Level 2
22
88
0.084
0.17
15
45
10
30
Ž3. Surface finish, and Ž4. Magnitude of forces. The
discussion in this paper pertains to the analysis of the flank
wear. While the detailed analysis of cutting forces during
the machining of composites was presented in Ref. w15x,
the other two factors were not found significantly influenced during machining.
2.2. Selection of independent Õariables
The selection of independent variables for machining of
the composites can be attempted based on a basic understanding of the process in view of the non-availability of
sufficient proven data. Again, from the preliminary experimentation it was thought that four independent variables
Ž1. cutting speed, Ž2. feed rate, Ž3. inclination angle Žrefer
to Fig. 1 for more details., and Ž4. volume of reinforcement in the composite material, could influence the magnitude of flank wear. Of these, the first two are process
parameters, the third one is a tool setting dependent parameter and the last one is a material dependent parameter.
The volume of reinforcement is chosen as material dependent parameter because it is the cause of improvement in
their properties. Also, wear of the tools during machining
of composites is more concerned with the hardness of
reinforcements than the hardness of bulk material w5x. The
levels of these factors chosen for the experimentation are
given in Table 1. The depth of cut was kept constant at 0.2
mm for all the experimental runs.
Fig. 2. Logic for the selection of interactions.
126
Table 2
Degrees of freedom
Factor and A
B
C
D
AB BC AC Total
interactions Žspeed. Žfeed. Žangle. Žvolume.
Levels
DOF
2
1
2
1
2
1
2
1
2
1
2
1
2
1
7
Table 3
Assignment of factors and interactions to an L8 orthogonal array
Column
1
2
3
4
5
6
7
Experimental A
B
A=B C
A=C B=C D
Žfeed. Žspeed.
Žangle.
Žvolume.
runs
1
2
3
4
5
6
7
8
1
1
1
1
2
2
2
2
1
1
2
2
1
1
2
2
1
2
1
2
1
2
1
2
1
2
2
1
2
1
1
2
Fig. 4. Rotary tool holder.
mentation because such interactions are found to be negligible in most engineering phenomena w16,17x.
2.4. Selection of an orthogonal array and linear graph
2.3. Selection of interactions
There can be six 2-factor interactions for the present
experiment. An engineering understanding of the experiment shows that of the six 2-factor interactions, three
would probably influence both of the response variables.
Logic behind the selection of these interactions taking
flank wear as a response variable is depicted in the block
diagram, Fig. 2. It was thought that, of the four factors,
cutting speed, feed rate and inclination angle would interact with each other by influencing the speed of rotation of
the rotary insert. For example, an increase in the cutting
speed causes an increase in the speed of rotation of the
round insert, which subsequently can result in a decrease
in the wear of carbide inserts. Similarly, other two factors
also can interact and influence the response variables. It
may be noted that the interactions of the volume of
reinforcement with the other three factors are not considered because they cannot influence any common factor
such as the speed of round insert. Furthermore, three or
more factor interactions are also ignored during the experi-
An orthogonal array and a linear graph can be selected
based on the total degrees of freedom required for an
experiment w18x. The total number of degrees of freedom
for this experiment is 7, Table 2. Accordingly, an L8
orthogonal array was selected. Assignment of various factors and interactions to the columns of this array and
corresponding linear graph are shown in the Table 3 and
Fig. 3, respectively.
3. Experimental procedure
The AlrSiCp composites with 10 and 30% of the
volume of reinforcement required for this experimentation
were fabricated using a liquid metallurgy route consisting
of rheocasting followed by squeeze casting and hot extrusion; the details are described elsewhere w19x.
A rotary tool holder as shown in Fig. 4 was designed
and fabricated for these experiments. It has a round insert
mounted on a cylindrical shaft supported in a needle roller
bearing. An indexing bracket helps changing inclination
angle from 0 to 758 in a step of 158. After setting a specific
Fig. 3. Linear graph.
127
Fig. 5. Experimental specimen.
inclination angle, the indexing bracket can be located by
the locating pin and clamped using the clamping screw
during experiments.
In order to simulate intermittent machining action, the
experiments were carried out on a lathe using a cylindrical
specimen provided with a keyway slot Žrefer to Fig. 5..
The keyway slots had to be machined by electric discharge
machining as the use of HSS or HSS coated with TiN
slitting saws to produce the slot was ruled out due to
excessive wear.
During experiments, the flank wear width was measured after specific time intervals of about 10 to 30 s using
a Nikon 2D Tool Makers’ Microscope at eight to 10
locations along the periphery of the round insert. Each
experimental run was replicated three times.
vs. time plot, refer to Fig. 6. These points correspond to
the absolute magnitude of flank wear after 10 and 50 s of
machining, respectively. Statistical analysis of results was
carried out by preparing response tables for Analysis of
Means ŽAOM., normal probability plots, means plots and
Analysis of Variance ŽANOVA. as suggested in Refs.
w16–18x.
The absolute magnitude of the flank wear is important
in certain cases during manufacturing where the tool is
considered to be unsatisfactory even before the wear
reaches the standard tool-life criterion. Such situations
occur especially during finishing operations or machining
slender or delicate workpieces. Hence, apart form the time
dependent tool wear limit, the absolute magnitude of the
tool wear also becomes important.
4.1. Statistical results
4. Results and analysis
Analysis of the effect of various factors and their
interactions was carried out using two data points per
experimental run as shown in a typical flank wear width
The normal probability plots, i.e., the plot of normal
probability vs. the contribution of various factors and their
interactions for two stages of wear are shown in Figs. 7
and 8. These plots show that during both the stages of
Fig. 6. Typical plot of flank wear width vs. time plot Žfor Run 2..
Fig. 7. Normal probability plot for flank wear Žafter 10 s of machining..
128
are structureless; hence, the ANOVA models are adequate
w16x.
The means plots for the two stages of wear show that
the cutting speed has the largest influence during both
stages of wear, Figs. 9 and 10. In the initial stage, other
three factors have almost the same effect on the tool wear
ŽFig. 9.. Whereas in the second stage of wear, there is an
increase in the contribution of feed rate and the volume of
reinforcement but only a slight decrease in the contribution
of the inclination angle ŽFig. 10.. Also, an increase in all
these factors causes an increase in the flank wear of the
tool in both stages of wear.
4.2. Analysis of statistical results
Fig. 8. Normal probability plot for flank wear Žafter 50 s of machining..
wear, only one point corresponding to cutting speed Žpoint
‘B’. appears to lie distinctly away from rest of the points.
Similarly, the AOM for flank wear width after 10 and 50 s
of machining carried out preparing the response tables Žnot
presented in the paper. has given the same results. This is
an indication of the predominant influence of cutting speed
on the response variable. The contribution of rest of the
factors and interactions is not very clear from respective
AOM tables and normal plots. In such situations, ANOVA
analysis needs to be looked into w17x. ANOVA tables for
the primary and secondary wear stages are shown in
Tables 4 and 5, respectively. These tables show that during
both stages of wear, all the four factors are significant at
the 95% confidence level. It is also evident that in the
primary wear zone, only feed rate–inclination angle ŽA =
C. interaction is significant, Table 4. In the later stage,
speed-angle ŽB = C. and feed-speed ŽA = B. interactions
become significant but not ŽA = C. ŽTable 5.. Here, it is
important to note that the residual plots Žnot shown here.
4.2.1. Effect of cutting speed and Õolume of reinforcement
The predominant effect of cutting speed on the tool
wear during both stages of wear can be due to an increase
in the abrasion velocity with the increase in the cutting
speed. In fact, while selecting the interactions, it was
thought that an increase in cutting speed could increase the
speed of rotation of the round insert; thereby reducing the
tool wear. But in fact, the net effect of the two actions was
to increase the flank wear. Here, it is felt that the significant influence of the presence of harder reinforcements
could be the reason for the reduction in the effect of the
rotary motion of the tool. The reinforcement in composites
may cause rapid abrasion of the tool material, thereby
sacrificing the advantage of insert rotation in substantially
reducing the wear. It may be noted from ANOVA Tables 4
and 5 that the volume of reinforcement significantly influences flank wear during both stages of wear.
An interesting example describing the superiority life of
rotary carbides over stationary carbides and their importance in intermittent machining is given below w15x. During
the earlier experiments on measurement of cutting forces
while machining composites, it was evident that the life of
stationary carbide tools is extremely small of the order of a
few seconds even at very low cutting speeds of about 7–10
Table 4
ANOVA for magnitude of flank wear Žafter 10 s of machining.
Source of variation
Sum of squares
DOF
Mean square
F-ratio
Significance level
Model
A: Feed rate
B: Cutting speed
C: Inclination angle
D: SiCp Volume
A=B
A=C
B=C
Residual
Total Žcorrected.
0.04552917
0.00198017
0.03300417
0.00170017
0.00370017
0.00028017
0.00476017
0.00010417
0.00065667
0.04618583
7
1
1
1
1
1
1
1
16
23
0.00650417
0.00198017
0.03300417
0.00170017
0.00370017
0.00028017
0.00476017
0.00010417
0.00004104
158.48
48.25
804.16
41.43
90.16
6.83
115.98
2.54
0.0001
0.0001U
0.0001U
0.0001U
0.0001U
0.0188
0.0001U
0.1307
U
Indicates that these factors are statistically significant.
129
Table 5
ANOVA for magnitude of flank wear Žafter 50 s of machining.
Source of variation
Sum of squares
DOF
Mean square
F-ratio
Significance level
Model
A: Feed rate
B: Cutting speed
C: Inclination angle
D: SiCp volume
A=B
A=C
B=C
Residual
Total Žcorrected.
0.0659678
0.0087402
0.0466402
0.0006615
0.0014415
0.0001602
0.0013202
0.0070042
9.24 = 10y4
0.0668918
7
1
1
1
1
1
1
1
16
23
0.0094240
0.0087402
0.0466402
0.0006615
0.0014415
0.0001602
0.0013202
0.0070042
5.77 = 10y5
163.186
151.345
807.622
11.455
24.961
2.773
22.860
121.284
0.0000
0.0000U
0.0000U
0.0038U
0.0001U
0.1153
0.0002U
0.0000U
U
These factors are statistically significant.
mrmin. On the other hand, the rotary carbide tools can
machine up to 60–80 s even at very high cutting speed
such as 88 mrmin. The mechanism of tool wear is abrasion and the rate of abrasion increases with the increase in
the cutting speed, hence the magnitude of cutting forces.
Thus, the dependency of cutting forces on tool wear is
somewhat reduced in case of rotary carbide tools so that
they can machine up to larger duration than stationary
tools. Going further, while machining these composites
with PCD tools, the dependency of forces is almost negligible and these tools have almost infinite life compared to
the stationary carbide tools. However, they are sensitive to
impacts; in such situations, the rotary carbide tools could
be of some use. Thus, rotary tools could possibly overcome deficiencies of both stationary carbide and
PCDrCBN tools especially in the intermittent machining
of composites. In addition, further efforts are necessary to
improve the life rotary carbide tools to a practical level.
angle increases the flank wear as can be seen from the
means plots Figs. 9 and 10. This could be due to an
increase in area of chip cross-section with the increase in
both these factors. The effect of inclination angle on the
area of chip cross-section is shown in Fig. 11Ža–b.. Note
the increase in the length ‘OA’ with the increase in
inclination angle. Similarly, it is well-known that the increase in the feed rate increases the area of chip cross-section.
4.2.2. Effect of feed rate and inclination angle
The other two factors significantly influencing the tool
wear are feed rate and inclination angle, refer to ANOVA
Tables 4 and 5. An increase in feed rate and inclination
4.2.3. Effect of interactions
The significant effect of interaction A = C in the primary wear zone ŽTable 4. could be due to the sharp cutting
edge of the tool resulting in almost no slipping. Interaction
A = C could also be due to the prominent influence of
these factors on the area of chip cross-section. In the latter
stages, as the tool wear becomes more rapid, there could
be some slip, thereby reducing the A = C interaction effect
ŽTable 5.. Once the wear process stabilises, i.e., in the
secondary wear zone, cutting speed becomes the most
important factor influencing the wear process. Hence, its
interaction with other factors such as A = B and B = C
becomes significant. Thus, with the progress of wear from
Fig. 9. Means plot for flank wear Žafter 10 s of machining..
Fig. 10. Means plot for flank wear Žafter 50 s of machining..
130
Fig. 11. Ža–b. Effect of inclination angle on the area of chip cross-section.
the primary to secondary wear zone, the following phenomena occur.
Ø A reduction in the contribution of inclination angle.
Ø A reduction in the feed rate–inclination angle ŽA = C.
interaction.
Ø An increase in the contribution of the feed rate.
Ø The interaction between cutting speed and feed rate
and, cutting speed and inclination angle becomes significant.
4.3. Model for tool life
It was evident form the literature that the use of conventional Taylor’s tool-life equation for the composite mate-
Fig. 12. Ža. Flank wear width: Experimental vs. Predicted ŽRun 2.; Žb. Flank wear width: Experimental vs. Predicted ŽRun 4.; Žc. Flank wear width:
Experimental vs. Predicted ŽRun 6.; Žd. Flank wear width: Experimental vs. Predicted ŽRun 8..
131
rial is not sufficient w6x. Besides the cutting speed and
time, the volume fraction of reinforcement in composites
and feed rate influences the magnitude of flank wear. In
the case of the rotary tools, the inclination angle of the tool
also plays an important role. Considering all these factors,
the objective function for the time dependent flank wear
ŽWft . of a rotary carbide tool during machining of composites is defined as:
Wft s f Ž f , Vc , l , Vp , t . .
Ž 1.
In the analysis, the time dependent plots Žtool-life plots.
similar to the one shown in Fig. 6 were obtained for all the
eight experimental runs and the entire data is used for the
analysis. It was felt that the following model on the similar
lines as that of the Taylor’s tool-life equation but inclusive
of other factors could be helpful in describing the experimental data.
Wft s K w f a Vcb lc Vpd t c .
Ž 2.
Taking logarithm of both sides of above equation gives:
ln Ž Wft . s ln Ž k w . q a ln Ž f . q b ln Ž Vc . q c ln Ž l .
q d ln Ž Vp . q e ln Ž t . .
Ž 3.
Let, Y s lnŽWft ., K s lnŽ k w ., X s lnŽ f ., Z s lnŽ Vc ., W s
lnŽ l., V s lnŽ Vp ., U s lnŽ t ., A s a, B s b, C s c, D s d,
E s e. After substitution, Eq. Ž3. becomes:
Y s K q AX q BZ q CW q DV q EU
Wft s 0.008730 f 0.304 Vc0.398 l0.082 Vp0.2215 t 0.303 .
Ø Finally, having established the utility of the rotary
tool concept and the parameters which influence the magnitude of tool wear, further study would be necessary to
establish an appropriate tool material for commercial application of rotary tools.
Ž 4.
If y is the estimated value of the response variable, then,
by minimising the sum of square of errors, by partial
differentiation with respect to constants K, A, B, C, D, E.
Solving resulting equations by the Cramer’s rule, the toollife model for a rotary tool is obtained as:
Wft s 0.008730 f 0.304 Vc0 .398 l0.082 Vp0.2215 t 0.303
Ø All the four factors selected for the analysis were
found to be statistically significant in influencing the absolute magnitude of the flank wear. The cutting speed was
found to have the most predominant influence. As wear
progresses from the primary to secondary zone, the following phenomena occur:
–Reduction in the contribution of the inclination angle
–Reduction in the feed rate–inclination angle ŽA = C.
interaction
–Increase in the contribution of the feed rate
–Interaction of cutting speed with feed rate and inclination angle becomes significant.
Ø A tool-life model has been proposed. It is emphasised here that not only the process and tool parameters but
volume of reinforcement in the composite material has
been incorporated in the model. It is felt that this will be a
very useful information for a process-planning engineer.
The proposed model as given below was found to agree
satisfactorily with the experimental data.
Ž 5.
The above model shows that cutting speed has maximum influence on the process of tool wear, followed by
feed rate, duration, volume of reinforcement and inclination angle. Typical comparisons of the experimental and
predicted values among the runs from ‘1’ to ‘8’ are shown
in Fig. 12Ža–d.. It can be evident that in most of the cases
the experimental data agreed well with the predicted data.
The average difference between experimental and predicted data considering all the runs is 9.1%. Hence, it
could be concluded here that the proposed model agrees
with the experimental data satisfactorily.
6. Nomenclature
Vt
Vf
l
Wft
kw
Vc
Vp
f
t
y
xi
bi
Velocity of rotation of round insert Žmrmin.
Feed speed Žmrmin.
Inclination angle Ždeg.
Time dependent flank width wear Žmm.
Constant in time dependent flank wear equation
Cutting speed Žmrmin.
Volume of reinforcement in composite Žpercent of
SiCp.
Feed rate Žmmrrev.
Duration of cut Žs.
Dependent variable in regression equations
Independent variables in regression equations Ž i s 1
to 4.
Regression coefficients in regression equations
Ž i s 0 to 4.
Acknowledgements
5. Concluding remarks
Ø A concept of rotary carbide tools in the intermittent
machining of AlrSiCp composites has been tried out and
is found to be an attractive proposition.
Ø The statistical analysis for the rotary carbide tools
has brought out the influence of process, tool and material
dependent parameters on the magnitude of tool wear.
The authors wish to express their gratitude to Mrs.
WIDIA ŽIndia. Bangalore for the tools and financial assistance provided during this project work. The authors also
wish to acknowledge the technical support and encouragement provided by Mr. Ajeet Khare ŽDy. Managing Director. and Mr. D. Sarathy ŽDivisional Manager, Research
D & D., WIDIA ŽIndia. for carrying out this work.
132
References
w1x J.E. Allison, G.S. Cole, Metal Matrix Composites in the Automotive
Industry: Opportunities and Challenges JOM, Jan. 1993, pp. 19–24.
w2x P. Rohatgi, Cast Aluminium Matrix Composites for Automotive
Applications, JOM, April 1991, pp. 10–15.
w3x S.S. Joshi, N. Ramakrishnan, D. Sarathy, P. Ramakrishnan, Development of Technology for Discontinuously Reinforced Aluminium
ŽDRA. Composites, Integrated Design and Process Technology, The
First World Conference on Integrated Design and Process Technology, Austin, TX, USA, IDPT-v1, 1995, pp. 492–497.
w4x N. Ramakrishnan, Some Experiences in Machining of Composite
Materials, Processing and Fabrication of Advanced Materials—V,
in: T.S. Srivatsan, J.J. Moore ŽEds.., TMS Annual Meeting, Oct.
6–10, 1996, Cincinnati, OH, USA, 1996, pp. 81–89.
w5x M.K. Brun, M. Lee, Wear characteristics of various hard materials
for machining of SiC-reinforced aluminium alloy, Wear 104 Ž1985.
21–29.
w6x N. Tomac, K. Tonnessen, Machinability of particulate Aluminium
Matrix Composites, Annals of CIRP 41 Ž1. Ž1992. 55–58.
w7x L. Cronjager, D. Meister, Machining of fibre and particle-reinforced
aluminium, Annals of CIRP 41 Ž1. Ž1992. 63–66.
w8x K. Weinert, W. Konig, A consideration of tool wear mechanisms
when machining Metal Matrix Composites ŽMMC., Annals of CIRP
42 Ž1. Ž1993. 95–98.
w9x L.A. Looney, J.M. Monaghan, P. O’Reilly, D.M.R. Taplin, The
turning of an AlrSiC Metal Matrix Composite, J. Mater. Proc.
Technol. 33 Ž1992. 453–468.
w10x J.M. Monaghan, P. O’Reilly, Machinability of an aluminium
alloyrsilicon carbide Metal Matrix Composites, Processing of Advanced Materials 2 Ž1992. 37–46.
w11x J. Monaghan, T. O’Reilly, The drilling of an AlrSiC Metal Matrix
Composite, J. Mater. Proc. Technol. 33 Ž1992. 469–480.
w12x P.K. Venu Vinod, An Investigation to Explain and Evaluate the
Performance of a Rotary Turning Tool, M. Tech Thesis, Indian
Institute of Technology, Bombay, 1965.
w13x N. Ramaswamy, F. Koenigsberger, Experiments with Self-propelled
Rotary Cutting Tools, Proc. of 9th IMTDR Conference, Part 2,
1968, pp. 945–959.
w14x P. Chen, T. Hoshi, High performance machining of SiC whisker-reinforced aluminium composite by self-propelled rotary tools, Annals
of CIRP 41 Ž1. Ž1992. 59–63.
w15x S.S. Joshi, N. Ramakrishnan, D. Sarathy, P. Ramakrishnan, Study of
cutting forces while turning AlrSiCp composites using rotary carbide tools, Materials and Manufacturing Processes 13 Ž1. Ž1998.
65–84.
w16x D.C. Montgomerry, Design and Analysis of Experiments, 4th edn.,
Wiley, 1997.
w17x R.H. Lonchner, J.E. Martar, Design for Quality—An Introduction to
the Best of Taguchi and Western Methods of Statistical Experimental Designs, Chapman & Hall, New York, 1990.
w18x M.S. Phadke, Quality Engineering Using Robust Design, Prentice
Hall, NJ, 1989, p. 151.
w19x S.S. Joshi, Some Studies on Machining of Squeeze Cast and Extruded AlrSiCp Composites, PhD Thesis, Indian Institute of Technology, Bombay, India, 1997.