Measurement 44 (2011) 1697–1704

Contents lists available at ScienceDirect

Measurement
journal homepage: www.elsevier.com/locate/measurement

Determining the effect of cutting parameters on surface roughness

in hard turning using the Taguchi method
_
Ilhan Asiltürk ⇑, Harun Akkusß
Technical Education Faculty, Selcuk University, 42075 Konya, Turkey

a r t i c l e i n f o a b s t r a c t

Article history: This study focuses on optimizing turning parameters based on the Taguchi method to min-
Received 20 January 2011 imize surface roughness (Ra and Rz). Experiments have been conducted using the L9
Received in revised form 14 June 2011 orthogonal array in a CNC turning machine. Dry turning tests are carried out on hardened
Accepted 5 July 2011
AISI 4140 (51 HRC) with coated carbide cutting tools. Each experiment is repeated three
Available online 23 July 2011
times and each test uses a new cutting insert to ensure accurate readings of the surface
roughness. The statistical methods of signal to noise ratio (SNR) and the analysis of vari-
Keywords:
ance (ANOVA) are applied to investigate effects of cutting speed, feed rate and depth of
Hard turning
Taguchi method
cut on surface roughness. Results of this study indicate that the feed rate has the most sig-
Surface roughness nificant effect on Ra and Rz. In addition, the effects of two factor interactions of the feed
DOE rate-cutting speed and depth of cut-cutting speed appear to be important. The developed
model can be used in the metal machining industries in order to determine the optimum
cutting parameters for minimum surface roughness.
Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction Aslan et al. [4] conducted an optimization study by machin-

ing a hardened AISI 4140 grade (63 HRC) steel on a lathe by
Hard turning is a turning process in which work pieces using Al2O3 + TiCN coated ceramic inserts. They deter-
whose hardness ranges between 50 and 70 HRC are mined that Al2O3-base ceramics are required for cutting
machined by using single point cutting tools which have tools in machining hard steels during wear resistance and
high hardness and wear resistance. The hard turning high hardness. They ensured optimization in their experi-
exhibits a unique behavior, which is different than conven- mental studies by using the Taguchi method. The experi-
tional turning operations. The application of hard turning mental parameters chosen were: three different cutting
technology can be improved by utilizing advanced optimi- speeds, feed rates and depths of cut. Blank wear (VB) and
zation algorithms, which helps manufacturers to make surface roughness were chosen as criteria for performance.
educated decisions in the presence of multiple objectives The obtained results were analyzed by using variance anal-
that need to be satisfied [1,2]. ysis (ANOVA). As a result, it was seen that the VB value de-
Significant advances have been seen in cutting tools and creased as the cutting speed and the depth of cut increased;
machine tools in recent years. Cutting parameters may be however, it first decreased and then increased as the feed
specified according to hardness of materials and roughness rate increased. On the other hand, the surface roughness
of the surface of a work piece. The advantages in machining decreased as the cutting speed increased. In contrast sur-
materials with higher hardness are decreasing machining face roughness increased when the feed rate increased.
costs, saving time, improving surface quality, and eliminat- Ucun and Aslantasß [5] investigated the effects of the
ing of deformities in parts caused by temperature [3,4]. cutting parameters on surface roughness in hard turning.
They examined machining performance with respect to
⇑ Corresponding author. Tel.: +1 317 464 30 86; fax: +1 317 274 97 44. tool wear and surface roughness in machining hardened
_ Asiltürk).
E-mail address: [email protected] (I. AISI 52100 grade bearing steels with coated carbide cutting

0263-2241/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.measurement.2011.07.003
1698 _ Asiltürk, H. Akkusß / Measurement 44 (2011) 1697–1704
I.

tools. Four different cutting speeds, three different feed concluded that the effects of cutting speed, tool hardness
rates and two different depths of cut were used during and feed rate on tool life were 41.63%, 32.68% and
their experimental studies. As a result, they observed that 25.22%, respectively.
carbide cutting tools were not suitable for the hard turning Abu-Sinna et al. [13] investigated the effect of the load-
process, especially at high cutting parametric values. ing frame stiffness on the deadweight force machine–
Derakhshan and Akbari [6] used Cubic Boron Nitride loadcell interaction. In order to perform the finite element
(CBN) tools in their study on hard turning. They researched analysis efficiently, they utilized the Taguchi technique, as
the effects of hardness on the work piece and cutting speed this typically reduces the total required simulations. The
on surface roughness in machining with CBN tools. They validity of the finite element analysis was confirmed
machined the AISI 4140 grade work piece with two differ- through experiments.
ent tools at five different hardness values: HRC 45, 50, 55, The Taguchi method for finding out the optimal value of
60, and 65. The types of tools, the hardness value of the surface roughness under optimum cutting condition in
material and the cutting speed as input were used while turning SCM 440 alloy steel was applied by Thamizhmanii
Ra, Rz and VB were used as output. Chou et al. [7] obtained et al. [14]. They detected that the causes of poor surface
the effect of various parameters in machining hard steels finish were machine tool vibrations and tool chattering
with CBN inserts and blank wear depending on surface whose effects were ignored for analysis. The authors con-
roughness. They specified cutting speed, CBN content per- cluded that depth of cut was the only significant factor
centage and length of insert as factors. Yang and Tarng [8] which contributed to the surface roughness.
used the Taguchi method for determining optimum cutting Kilickap [15] investigated the use of the Taguchi method
parameters. They machined S45C steel work piece with and the Response Surface Methodologies (RSM) for mini-
carbide cutting tools. The cutting parameters chosen were mizing the burr height and the surface roughness in drilling
cutting speed, feed rate and depth of cut. Taguchi’s signal– Al-7075. The Taguchi method, a powerful tool to optimize
noise ratio and variance analysis were used to determine design quality, was used to find optimal cutting parame-
the effect of the cutting parameters on tool life and surface ters. The optimization results showed that the combination
roughness. They obtained optimum cutting parameters as of low cutting speed, low feed rate and high point angle
a result of their studies. Zhang and Chen [9] experimented were necessary to minimize burr height. The best results
to optimize the surface quality in a CNC drilling operation of the surface roughness were obtained at lower cutting
using Taguchi design. The control factors in their study speed and feed rates and at higher point angles.
were feed rate, spindle speed, peck rate and tool type, Wang and Lan [16] used orthogonal array of Taguchi
while the noise factors simulated were machine shop method coupled with the Grey Relational Analysis (GRA)
vibration and the presence or absence of magnetism in in considering the four parameters of cutting speed, depth
the work piece material. Through statistical analysis of of cut, feed rate and tool nose radius. The stated four
response variables and signal-to-noise ratios, the deter- parameters were used for optimizing the following three
mined significant factors were the spindle speed, tool type responses: surface roughness, tool wear and material re-
and peck rate, and the optimal combination of cutting moval rate in precision turning on an ECOCA-3807 CNC
parameters were selected. They verified that the selected lathe. They studied the proposal of optimization ap-
optimal combination through Taguchi design achieved proaches using Orthogonal Array and the GRA and contrib-
the desired surface roughness. Bhattacharya et al. [10] uted a satisfactory technique for improving the multiple
investigated the effects of cutting parameters on finish machining performances in precision CNC turning with
and power consumption by employing Taguchi techniques. profound insight.
Their results showed a significant effect of cutting speed on Nesßeli et al. [17] observed on the influence of tool
the surface roughness and power consumption, while the geometry on the surface finish obtained in turning of AISI
other parameters did not substantially affect the re- 1040 steel. Their study focused on the effect of tool geom-
sponses. Therefore, optimal cutting parameters were etry parameters on the surface roughness during turning.
obtained. The RSM and a prediction model were developed related
Kopac et al. [11] considered cutting speed, cutting tool to average surface roughness (Ra) using experimental data.
materials, feed rate and depth of cut as cutting parameters The results indicated that the tool nose radius was the
in machining C15 E4 steel on a lathe. They used the Tagu- dominant factor on the surface roughness.
chi orthogonal array of L16 (25), which has two levels and Shetty et al. [18] used Taguchi and Response Surface
a degree of freedom of 13 in the experimental design. The Methodologies for minimizing the surface roughness in
quality determinant of ‘‘the smaller the better’’ was used in turning of discontinuously reinforced aluminum compos-
calculating the signal–noise ratio. It was observed that the ites (DRACs) having aluminum alloy 6061 as the matrix
control parameter having the highest effect on surface and containing 15 vol.% of silicon carbide particles of mean
roughness is the cutting speed, and better surface rough- diameter 25 lm under pressured steam jet approach. The
ness values were obtained at higher cutting speeds. effect of cutting parameters on surface roughness was
ß ahin [12] compared the tool life of CBN and ceramic
S evaluated and the optimum cutting condition for minimiz-
inserts in turning hard steels using the Taguchi method. ing the surface roughness was also determined.
The effects of cutting parameters on tool life were deter- The literature survey reveals that traditional experi-
mined by using orthogonal array, signal–noise ratio and mental design procedures are too complicated and not
variance analysis. He chose cutting speed, feed rate and easy to use. A large number of experimental works have
tool hardness as cutting parameters. As a result, it was to be carried out when the number of the process
 _ Asiltürk, H. Akkusß / Measurement 44 (2011) 1697–1704
I. 1699

parameters increases. To solve this problem, the Taguchi Ø110  600 mm. As thermal treatment, the steel was
method uses a special design of orthogonal arrays to study tempered at 880 °C for an hour and quenched at 280 °C
the entire parameter space with only a small number of for 2 hour to eliminate stresses and to reduce hardness.
experiments [19]. As a result, hardness of the material was decreased from
The objective of this study is to obtain optimal turning 62 HRC to 56–57 HRC. The experimental studies were car-
parameters (cutting speed, feed rate and depth of cut) for ried out on a Mori Seiki NL 2500 CNC Lathe. The experi-
minimum on surface roughness, while turning hardened ments were conducted under dry cutting conditions. The
AISI 4140 steel with Al2O3 and TiC-coated carbide tools. tool holder used was model MWLNR 2525M-0.8W.
Taguchi’s parameter design approach has been used to Al2O3 and TiC-coated (WNMA 080408) inserts were used
accomplish this objective. Furthermore, a statistical analy- as the cutting tool material. The surface roughness was
sis (ANOVA) is performed to see which process parameters measured using a Mitutoyo SJ-301P portable device with-
are statistically significant. in the sampling length of 2.5 cm. Fig. 1 shows the exper-
imental arrangement.
2. Material and method The level of cutting parameter ranges and the initial
parameter values were chosen from the manufacturer’s
In this study, a work piece made of AISI 4140 grade handbook recommended for the tested material. These
(DIN 42CrMo4) steel was used. Its sizes were cutting parameters are shown in Table 1.

TURNING

SURFACE
ROUGHNESS

SOFTWARE

CUTTING
PARAMETERS

OPTIMIZED
PARAMETERS
USER

Fig. 1. Experimental scheme.

1700 _ Asiltürk, H. Akkusß / Measurement 44 (2011) 1697–1704
I.

Table 1 technical terms, signal-to-noise ratio compares the level

Cutting parameters. of a desired signal (such as music) to the level of back-
Symbol Cutting parameters Level 1 Level 2 Level 3 ground noise. The higher the ratio, the less obtrusive the
A Cutting speed (m/min) 90 120 150 background noise is. ‘‘Signal-to-noise ratio’’ is sometimes
B Feed rate (mm/rev) 0.18 0.27 0.36 used informally to refer to the ratio of useful information
C Depth of cut (mm) 0.2 0.4 0.6 to false or irrelevant data in a conversation or exchange.
In other words, the standard deviation cannot be mini-
The Taguchi method and L9 Orthogonal Array were mized first and the mean brought to the target [4,24,25].
used to reduce number of the experiments. The design of Taguchi has empirically found that the two stage optimiza-
experiments (DOE) and measured Ra/Rz values are shown tion procedure involving S/N ratios indeed gives the
in Table 2. The experiments were conducted with three parameter level combination, where the standard devia-
replicates. tion is minimum while keeping the mean on target [26].
This implies that engineering systems behave in such a
way that the manipulated production factors can be di-
3. The experimental design using the Taguchi method
vided into three categories:
The traditional experimental design methods are too
1. Control factors, which affect process variability as mea-
complex and difficult to use. Additionally, large numbers
sured by the S/N ratio.
of experiments have to be carried out when the number
2. Signal factors, which do not influence the S/N ratio or
of machining parameters increase. Therefore, the factors
process mean.
causing variations should be determined and checked un-
3. Factors, which do not affect the S/N ratio or process
der laboratory conditions. These studies are considered un-
mean.
der the scope of off-line quality improvement [20–22].
The Taguchi method is an experimental design tech-
In practice, the target mean value may change during
nique, which is useful in reducing the number of experi-
the process development. Two of the applications in which
ments dramatically by using orthogonal arrays and also
the concepts of S/N ratio are useful are the improvement of
tries to minimize effects of the factors out of control. The
quality through variability reduction and the improvement
basic philosophy of the Taguchi method is to ensure qual-
of measurement. The S/N ratio characteristics can be di-
ity in the design phase. The greatest advantages of the
vided into three categories given by Eqs. (1)–(3), when
Taguchi method are to decrease the experimental time,
the characteristic is continuous [27]:
to reduce the cost and to find out significant factors in a
shorter time period [23].
 Nominal is the best characteristic,
The most reliable of Taguchi’s techniques is the use of
parameter design, which is an engineering method for
S y

product or process design that focuses on determining ¼ 10 log 2 ð1Þ
N Sy
the parameter (factor) settings producing the best levels
of a quality characteristic (performance measure) with
 Smaller is the better characteristic,
minimum variation. The overall aim of quality engineering
is to make products that are robust with respect to all noise
factors. The most important stage in the design of an
experiment lies in the selection of control factors. As many S 1 X 2 
¼ 10 log y ð2Þ
factors as possible should be included in order to it would N n
be possible to identify non-significant variables at the ear-
 Larger is the better characteristic,
liest opportunity [4,25]. Taguchi creates a standard orthog-
onal array to accommodate this requirement. Taguchi used
the signal-to-noise (S/N) ratio as the quality characteristic
S 1 X1
 
of choice. S/N ratio is used as a measurable value instead ¼  log ð3Þ
of standard deviation because as the mean decreases, the N n y2
standard deviation also decreases and vice versa. In less

Table 2
The results of experiments and S/N ratios values.

Experimental number A B C Ra Rz Ra for S/N ratios Rz for S/N ratios

1 1 1 1 1.6 8.79 4.23 18.88
2 1 2 2 1.79 14.22 3.37 23.058
3 1 3 2 4.22 19.87 12.48 25.964
4 2 1 2 1.21 8.62 2.32 18.75
5 2 2 2 2.47 11.72 7.87 21.379
6 2 3 1 3.78 15.94 10.79 24.05
7 3 1 3 1.17 8.66 3.37 18.76
8 3 2 1 2.21 9.25 6.99 19.323
9 3 3 2 2.51 12.12 7.76 21.67
 _ Asiltürk, H. Akkusß / Measurement 44 (2011) 1697–1704
I. 1701

This implies that engineering systems behave in such a method. Thus, the optimum cutting condition was found as
way that the manipulated production factors can be di- 2.32 and 18.75 S/N ratios for Ra and Rz respectively in
vided into three categories: where y , is the average of ob- L9 orthogonal array in Table 2. The optimum cutting condi-
served data, S2y the variation of y, n the number of tions, which were the cutting speed of 120 m/min, the feed
observations, and y the observed data or each type of the rate of 0.18 mm/rev and the depth of cut of 0.4 mm (2 1 2
characteristics. With the above S/N ratio transformation, orthogonal array) were obtained for the best Ra and Rz val-
the smaller the S/N ratio, the better the result when we ues. Level values of the factors obtained for Ra, according
consider tool wear, surface roughness, cutting force and to the Taguchi design, are given in Table 3. Fig. 2 shows
trust force. the graphic of the level values given in Table 3. Therefore,
Deviation between experimental and desired values is interpretations may be made according to the level values
defined as loss function in the Taguchi method. This loss of A, B, and C factors given in Table 3 and Fig. 2 in deter-
function is further converted into S/N signal–noise ratio mining optimum cutting conditions of experiments to be
[22]. In the Taguchi method, the optimum cutting condi- conducted under the same conditions. The average S/N ra-
tions required for the best surface roughness were ob- tio for every level of experiment is calculated based on the
tained by using Eq. (1), ‘‘the smaller-the better’’ signal– recorded value as shown in Table 3.
noise ratio. S/N ratios and level values were calculated by The different values of S/N ratio between maximum and
using Eq. (2), ‘‘the smaller-the better’’ in the MINITAB 14 minimum are (main effect) also shown in Table 3. The feed
Program. S/N ratios obtained from this equation are given rate and the depth of cut are two factors that have the
in Table 2. highest difference between values, 6.32 and 3.6 respec-
tively. Based on the Taguchi prediction that the larger dif-
ferent between value of S/N ratio will have a more
4. Analyzing and evaluating results of the experiments
significant effect on surface roughness (Ra). Thus, it can
using the Taguchi method
be concluded that increasing the feed rate will increase
the Ra significantly and also the depth of cut. The results
The most essential criterion in the Taguchi method for
of data analysis of S/N ratio for Ra values, which are calcu-
analyzing experimental data is signal/noise ratio. In this
lated by Taguchi method, are shown in Table 4 (where df is
study, the S/N ratio should have a maximum value to ob-
degree of freedom, F variance ratio, and P significant fac-
tain optimum cutting conditions, according to the Taguchi
tor). Thus, it is seen in Fig. 2 and Table 3 that the third level
of A factor (cutting speed), the first level of B factor (feed
Table 3 rate) and the third level of C factor (depth of cut) are
S/N response table for Ra factor. higher.
Level A B C
Consequently, the optimum cutting conditions deter-
mined under the same conditions for the experiments to
1 6.343 4.024 7.338
2 7.709 5.723 4.846
be conducted will be 150 m/min for the cutting speed,
3 6.039 10.344 7.907 0.18 mm/rev for the feed rate and 0.6 mm for the depth
D 1.67 6.32 3.06 of cut.

Main Effects Plot (data means) for SN ratios

A B
-4

-6

-8
Mean of SN ratios

-10

1 2 3 1 2 3
C
-4

-6

-8

-10

1 2 3
Signal-to-noise: Smaller is better

Fig. 2. The graphic of mean of S/N ratios versus factor levels (Ra).
1702 _ Asiltürk, H. Akkusß / Measurement 44 (2011) 1697–1704
I.

Table 4
The interaction between S/N – cutting speed for Ra factor.

Sum of squares df Mean square F P

Ra – V Between groups (Combined) 4.747 2 2.374 0.159 0.856
Linearity 0.138 1 0.138 0.009 0.926
Deviation from linearity 4.609 1 4.609 0.309 0.598
Within groups 89.369 6 14.895
Total 94.116 8

Table 5
The interaction between S/N – feed rate for Ra factor.

Sum of squares df Mean square F P

Ra – f Between groups (Combined) 64.174 2 32.087 6.430 0.032
Linearity 59.908 1 59.908 12.005 0.013
Deviation from linearity 4.266 1 4.266 0.855 0.391
Within groups 29.942 6 4.990
Total 94.116 8

Table 6
The interaction between S/N – depth of cut for Ra factor.

Sum of squares df Mean square F P

Ra – a Between groups (Combined) 12.675 2 6.338 0.467 0.648
Linearity 7.741 1 7.741 0.570 0.479
Deviation from linearity 4.934 1 4.934 0.364 0.569
Within groups 81.441 6 13.573
Total 94.116 8

Table 7
S/N response table for Rz factor.
The Taguchi method was used in determining optimum
cutting conditions according to the S/N ratio, while the Level A B C
interaction between the cutting parameters was deter- 1 22.63 18.78 20.75
mined with the help of the variance analysis. Variance 2 21.39 21.25 21.16
analyses of S/N ratios were conducted through the SPSS 3 19.9 23.89 22.02
D 2.73 5.11 1.27
16.0 Program and the interactions between the S/N-cutting
speed, the S/N-feed rate and the S/N-depth of cut were
evaluated. The results of the conducted S/N-variance anal-
yses (ANOVA) are given in Tables 4–6.
Table 4 shows the interaction between the S/N and the cutting speed, 0.18 mm/rev for feed rate and 0.2 mm for
cutting speed considering that, according to P (significant) depth of cut.
values; it does not produce a significant level within the The interaction between cutting parameters was deter-
reliability interval of 95%. In other words, cutting speed mined through variance analysis, while optimum cutting
has no effect on S/N ratio. According to Table 5, The P value conditions were determined from S/N ratio obtained
is effective for every three feed rate levels at the reliability through the Taguchi method. The variance analysis of the
level of 95%, because the results are lower than 0.05. interactions of the S/N-cutting speed, the S/N-feed rate
According to Table 6, three depth levels have no effect and the S/N-depth of cut were evaluated. The results of
on surface roughness at the reliability level of 95%. As a re- the conducted S/N analysis (ANOVA) are given in Tables
sult, any interaction between the S/N-cutting speed and 8–10.
the S/N-depth of cut was not observed. However, a signifi- Table 8 shows the interactions between the S/N and cut-
cant correlation between the S/N and the feed rate was ting speed, do not produce a significant level within the
observed. reliability interval of 95%. According to Table 9, P value is
The Taguchi design was conducted to obtain Rz values effective for every three feed rate levels at the reliability le-
too. The level values obtained from MINITAB 14 Program vel of 95%, because the results are lower than 0.05. Effec-
according to the Taguchi design are given in Table 7. tiveness was observed at three depth levels at a
Accordingly, Fig. 3 and Table 7 show that the third level reliability level of 95% according to Table 10. Previous
of A factor (cutting speed), the first level of B factor (feed researchers suggested similar results. They claimed that
rate) and the first level of C factor (depth of cut) are higher. the surface roughness was affected significantly by the
Consequently, optimum cutting conditions for the experi- feed rate. The cutting speed had a lower effect on the sur-
ments to be conducted will be (3 1 1) 150 m/min for face roughness [1,3,4,6,20,27,28].
 _ Asiltürk, H. Akkusß / Measurement 44 (2011) 1697–1704
I. 1703

Main Effects Plot (data means) for SN ratios

A B

-20

-22
Mean of SN ratios

-24
1 2 3 1 2 3
C

-20

-22

-24
1 2 3
Signal-to-noise: Smaller is better

Fig. 3. The graphic of mean of S/N ratios versus factor levels (Rz).

Table 8
The interaction between S/N – cutting speed for Rz factor.

Sum of squares df Mean square F P

Rz – V Between groups (Combined) 11.235 2 5.617 0.761 0.508
Within groups 44.287 6 7.381
Total 55.522 8

Table 9
The interaction between S/N – feed rate for Rz factor.

Sum of squares df Mean square F P

Rz – f Between groups (Combined) 39.252 2 19.626 7.238 0.025
Within groups 16.270 6 2.712
Total 55.522 8

Table 10
The interaction between S/N – depth of cut for Rz factor.

Sum of squares df Mean square F P

Rz – a Between groups (Combined) 11.346 2 5.673 0.771 0.504
Within groups 44.176 6 7.363
Total 55.522 8

5. Conclusion a result, nine experiments were conducted instead of

the full factorial 27 experiments. Ra and Rz’ S/N ratios
The Taguchi experimental design was used to obtain were found as a result of experiments conducted
optimum cutting parameters on hard turning. Experimen- according to the L9 orthogonal array. The maximum
tal results were analyzed using ANOVA. The results ob- value was found by using the S/N ratio equation of
tained in this study are given below: ‘‘the smaller-the better,’’ the maximum S/N ratio
yielded optimum cutting parameters. Optimum cutting
 L9 orthogonal array was selected for three different lev- conditions-which correspond to maximum 2.32 S/N
els of cutting speed, feed rate and depth of cut, which value of the smaller Ra value for the smaller surface
were cutting factors, by using the Taguchi method. As roughness in hard turning operation (2 1 2)-were found
1704 _ Asiltürk, H. Akkusß / Measurement 44 (2011) 1697–1704
I.

to be 120 m/min for the cutting speed, 0.18 mm/rev for [9] Z. Zhang, J.C. Chen, Surface roughness optimization in a drilling
operation using the Taguchi design method, J. Mater. Manuf.
the feed rate and 0.4 mm for the depth of cut. Optimum
Processes 24 (4) (2009) 459–467.
cutting conditions-which correspond to maximum [10] A. Bhattacharya, S. Das, P. Majumder, A. Batish, Estimating the effect
18.75 S/N value of Rz value (3 1 1)-were found to be of cutting parameters on surface finish and power consumption
120 m/min for the cutting speed, 0.18 mm/rev for the during high speed machining of AISI 1045 steel using Taguchi design
and ANOVA, Prod. Eng. Res. Dev. 3 (2009) 31–40.
feed rate and 0.4 mm for the depth of cut. [11] J. Kopac, M. Bahor, M. Sokovic, Optimal machining parameters for
 Variance analysis was applied to S/N ratios to discover achieving the desired surface roughness in fine turning of cold pre-
interactions between cutting parameters relating to Ra formed steel work pieces, Int. J. Mach. Tools Manuf. 42 (2002) 707–
716.
and Rz. According to the ANOVA analysis, the feed rate [12] Y. Sß ahin, Comparison of tool life between ceramic and cubic boron
has an effect on Ra and Rz at a reliability level of 95%. nitride (CBN) cutting tools when machining hardened steels, J.
Any difference (variance) was not observed for the cut- Mater. Process. Technol. 209 (2009) 3478–3489.
[13] A. Abu-Sinna, Yon-Kyu Park, Dae-Im Kang, Min-Seok Kim, The
ting speed and the depth of cut at the reliability level of influence of loading frame stiffness on loadcell–deadweight force
95%. machine interaction, Measurement 42 (6) (2009) 830–835.
 The number of experiments in the same or similar area [14] S. Thamizhmanii, S. Saparudin, S. Hasan, Analysis of surface
roughness by using Taguchi method, Ach. Mater. Manuf. Eng. 20
in hard turning operations were reduced by using the (1–20) (2007) 503–505.
Taguchi experimental design to determine optimum [15] E. Kilickap, Modeling and optimization of burr height in drilling of
cutting conditions. Satisfying results were obtained so Al-7075 using Taguchi method and response surface methodology,
Int. J. Adv Manuf. Technol. 49 (9–12) (2010) 911–923.
that they may be used in future academic and industrial
[16] M.Y. Wang, T.S. Lan, Parametric optimization on multi-objective
studies. precision turning using grey relational analysis, Inf. Technol. J. 7
(2008) 1072–1076.
[17] S. Nesßeli, S. Yaldız, E. Türkesß, Optimization of tool geometry
parameters for turning operations based on the response surface
Acknowledgment methodology, Measurement 44 (3) (2011) 580–587.
[18] R. Shetty, R. Pai, V. Kamath, S.S. Rao, Study on surface roughness
minimization in turning of DRACs using surface roughness
This study is supported by Scientific Research Projects methodology and Taguchi under pressured steam jet approach,
Coordinators (BAP) of Selcuk University. This support is ARPN J. Eng. Appl. Sci. 3 (1) (2008) 59–67.
[19] E. Bagci, S. Aykut, A study of Taguchi optimization method for
greatly appreciated.
identifying optimum surface roughness in CNC face milling of
cobalt-based alloy (stellite 6), Int. J. Adv. Manuf. Technol. 29 (9–10)
(2006) 940–947.
References [20] A. Hasçalık, U. Çaydasß, Optimization of turning parameters for
surface roughness, and tool life based on the Taguchi method, Int. J.
[1] D. Singh, P.V. Rao, A surface roughness prediction model for hard Adv. Manuf. Technol. 38 (2008) 896–903.
turning process, Int. J. Adv. Manuf. 32 (11–12) (2007) 1115–1124. [21] T.R. Lin, Optimization technique for face milling stainless steel with
[2] Y. Karpat, T. Özel, Hard turning optimization using neural network multiple performance characteristics, Int. J. Adv. Manuf. Technol. 19
modeling and swarm intelligence, Trans. North Am. Manuf. Res. Ins. (2002) 330–335.
XXXIII (2005) 179–186. [22] O. Köksoy, Z.F. Muluk, Solution to the Taguchi’s problem with
[3] S.Z. Chavosh, M. Tajdari, Surface roughness modeling in hard turning correlated responses, Gazi Univ. J. Sci. 17 (1) (2004) 59–70.
operation of AISI 4140 using CBN cutting tool, Int. J. Mater. Form. 3 [23] M. Sßanyılmaz, Design Of Experiment And an Application for Taguchi
(2010) 233–239. Method in Quality Improvement Activity, M.S. Thesis, Dumlupınar
[4] E. Aslan, N. Camusßcu, B. Bingören, Design optimization of cutting University, Turkey, 2006.
parameters when turning hardened AISI 4140(63 HRC) with [24] J.A. Ghani, I.A. Choudhury, H.H. Hassan, Application of Taguchi
Al2O3 + TiCN mixed ceramic tool, Mater. Des. 28 (2007) 1618–1622. method in the optimization of end milling operations, J. Mater.
[5] I. Ucun, K. Aslantasß, Investigation of Performance of Carbide Cutting Process. Technol. 145 (2004) 84–92.
Tool in Turning Hardened 52100 Tool Steel, IATS’09, Karabük, [25] S.H. Park, Robust Design and Analysis for Quality Engineering,
Turkey, 2009, pp. 1–6. Chapman and Hall, London, U.K, 1996.
[6] E.D. Derakhshan, A.A. Akbari, Experimental investigation on the [26] H.S. Kim, A combined FEA and design of experiments approach for
work piece hardness and cutting speed on surface roughness in hard the design and analysis of warm forming of aluminum sheet alloys,
turning with CBN tools, in: Proceedings of the World Congress on Int. J. Adv. Manuf. Technol. 51 (1–4) (2010) 1–14.
Engineering WCE, London, UK, 2009, pp. 1–3. [27] R. Shetty, R.B. Pai, S.S. Rao, R. Nayak, Taguchi’s technique in
[7] Y.K. Chou, M.M. Barash, C.J. Evans, Experimental investigation on machining of metal matrix composites, J. Braz. Soc. Mech. Sci. Eng.
CBN turning of hardened AISI 52100 steel, J. Mater. Process. Technol. 31 (1) (2009) 12–20.
124 (3) (2002) 274–283. [28] I. Asilturk, M. Cunkas, Modelling and prediction of surface roughness
[8] W.H. Yang, Y.S. Tarng, Design optimization of cutting parameters for in turning operations using artificial neural network and multiple
turning operations based on the Taguchi method, J. Mater. Process. regression method, Expert Syst. Appl. 38 (5) (2011) 5826–5832.
Technol. 84 (1998) 122–129.