A. R. MOTORCU et al.: ANALYSIS OF THE CUTTING TEMPERATURE AND SURFACE ROUGHNESS ...
343–351
ANALYSIS OF THE CUTTING TEMPERATURE AND SURFACE
ROUGHNESS DURING THE ORTHOGONAL MACHINING OF AISI
4140 ALLOY STEEL VIA THE TAGUCHI METHOD
ANALIZA TEMPERATURE REZANJA IN HRAPAVOSTI POVR[INE
S TAGUCHI METODO PRI ORTOGONALNI STROJNI OBDELAVI
LEGIRANEGA JEKLA AISI 4140
Ali Riza Motorcu1, Yahya Isik2, Abdil Kus2, Mustafa Cemal Cakir3
1Çanakkale Onsekiz Mart University, Engineering Faculty, Department of Industrial Engineering, 17100 Çanakkale, Turkey
2Uludað University, Vocational School of Technical Science Machinery Program, 16059 Bursa, Turkey
3Uludað University, Engineering Faculty, Department of Mechanical Engineering, 16059 Bursa, Turkey
armotorcu@comu.edu.tr
Prejem rokopisa – received: 2015-01-22; sprejem za objavo – accepted for publication: 2015-06-08
doi:10.17222/mit.2015.021
In this research, the tool-chip interface temperature (TCTI), the tool temperature (TT) and the average surface roughness (Ra) were
measured experimentally during the turning of AISI 4140 alloy steel with TiAlN-TiN, PVD-coated, WNVG 080404-IC907
tungsten carbide inserts using an IR pyrometer technique, a K-type thermocouple and a portable surface-roughness
measurement device, respectively. The workpiece material was heat treated by an induction-hardening process and hardened up
to a value of 50 HRC. The Taguchi method L18 (21 × 37) was used for the determination of the optimum control factors. The
depth of cut, the cutting speed and the feed rate were taken as control factors. The analysis of variance was applied in order to
determine the effects of the control factors on the tool-chip interface temperature, the tool temperature and the surface
roughness. The optimum combinations of the control factors for TCTI, TT and Ra were determined as a2v1f3, a1v3f2 and a2v3f1,
respectively. Second-order predictive models were developed with a linear-regression analysis, and the coefficients of
correlation for TCTI, TT and Ra were calculated as R2 = 92.8, R2 = 68.1 and R2 = 82.6, respectively.
Keywords: tool temperature, thermocouple, pyrometer, machining, Taguchi method
V raziskavi so bile eksperimentalno izmerjenene tempetratura na stiku orodje-ostru`ek (TCTI), temperatura orodja (TT) in
povpre~na hrapavost povr{ine (Ra) pri stru`enju legiranega jekla AISI 4140, z volfram karbidnimi vlo`ki WNVG 080404-IC907
s PVD prevleko iz TiAlN-TiN, z uporabo IR pirometra, termoelementi vrste K in s prenosnim merilnikom hrapavosti.
Obdelovanec je bil toplotno obdelan z indukcijskim ogrevanjem in hlajenjem na trdoto 50 HRC. Za dolo~anje optimalnih
kontrolnih faktorjev je bila uporabljena Taguchi metoda L18 (21 × 37). Globina rezanja, hitrost rezanja in hitrost podajanja so
bile vzete kot kontrolni faktorji. Analiza variance je bila uporabljena za dolo~anje vpliva kontrolnih faktorjev na temperaturo
prehoda orodje-ostru`ek, temperaturo orodja in hrapavost povr{ine. Dolo~ene so bile optimalne kombinacije kontrolnih
faktorjev za TCTI, TT in Ra , kot a2v1f3, a1v3f2 and a2v3f1. Z linearno regresijsko analizo so bili razviti modeli drugega reda za
napovedovanje in izra~unani so bili koeficienti korelacije za TCTI, TT in Ra kot R2 = 92,8, R2 = 68,1 in R2 = 82,6.
Klju~ne besede: temperatura orodja, termo~len, pirometer, strojna obdelava, Taguchi metoda
1 INTRODUCTION
In order to overcome the difficulties in terms of
efficiency and the quality of production encountered in
the metal-cutting industries, all the stages of the
machining process need to be monitored. During the
metal-cutting processes, one of the key factors is the
cutting temperature, which directly affects the surface
quality, the tool wear, the tool life, and the cost of
production. The amount of heat generated varies with the
type of material being machined and the cutting
parameters (especially the cutting speed, which had the
biggest influence on the temperature).1
Temperature monitoring is one of the most difficult
and complicated procedures in metal-cutting operations.
It is extremely complex to develop a model for
measuring the temperature due to the complexity of the
different events at the point of contact between the tool
and the workpiece. Therefore, an accurate and repeatable
temperature prediction still remains as a challenge due to
this complexity of the contact phenomenon.2 It is quite
difficult to measure the temperature since the heat in the
region is very close to the cutting edge. Due to a lack of
sufficient experimental data, it is not possible to verify a
mathematical model. Numerous attempts have been
made to measure the temperature during machining ope-
rations.3
Amongst the many experimental methods to measure
the temperature directly, only a few systems have used
the temperature as an indicator of machine performance
and for industrial applications.4 Therefore, the tempe-
rature can be controlled using the appropriate cutting
parameters to design and develop the system and it will
be beneficial to increase the efficiency in production.
In recent years, experimental studies related to
metal-cutting processes have made use of the Taguchi
Materiali in tehnologije / Materials and technology 50 (2016) 3, 343–351 343
UDK 620.181.4:621.9.015:669.15 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 50(3)343(2016)
method. This method has been used successfully for a
determination of the appropriate cutting parameters and
in the optimization of parameters related to tool wear,
tool life, and the surface quality. The Taguchi method
and Analysis of Variance (ANOVA) can conveniently
optimize the cutting parameters with several experimen-
tal runs that are well designed. Taguchi parameter design
can optimize the performance characteristics through the
settings of the design parameters and reduce the sen-
sitivity of the system’s performance to the source of
variation.5 On the other hand, ANOVA is used to identify
the most significant variables and interaction effects.6,7
In the Taguchi method, quality is measured by the
deviation of a quality characteristic from its target value.
Therefore, the objective is to create a design that is
insensitive to all possible combinations of uncontrollable
factors and is at the same time effective and cost efficient
as a result of setting the key controllable factors at their
optimum levels.8 Taguchi’s parameter design offers a
simple and systematic approach that can reduce the num-
ber of experiments to optimize the design for perfor-
mance, quality and cost. The signal-to-noise (S/N) ratio
and the orthogonal array (OA) are two major tools used
in robust design.9
A lot of research has been conducted for determining
the optimal cutting parameters. W. H. Yang and Y. S.
Tarng10 employed the Taguchi method, and the optimal
cutting parameters for the turning of S45C steel bars
were successfully obtained. B. M. Gopalsamy et al.11
applied the Taguchi method to find the optimum machin-
ing parameters while machining hard steel and used the
L18 orthogonal array. The S/N ratio and ANOVA were
used to study the performance characteristics of the
machining parameters. F. Ficici et al.12 used the Taguchi
method to study the wear behaviour of boronized AISI
1040 steel. They used the S/N ratio to investigate the
optimum setting parameters.
M. Adinarayana et al.13 presented the multi-response
optimization of the turning parameters for the turning of
AISI 4340 alloy steel. The experiments were designed
and conducted based on Taguchi’s L27 orthogonal array
design. They discussed an investigation into the use of
Taguchi parameter design to predict and optimize the
surface roughness, the metal removal rate and the power
consumption during turning operations. E. D. Kirby14
discussed an investigation into the use of Taguchi para-
meter design for optimizing the surface roughness gene-
rated by a CNC turning operation. He used a standard
orthogonal array for determining the optimum turning
parameters with an applied noise factor. The controlled
factors include the spindle speed, the feed rate, and the
depth of cut.
In this paper, the measurement of temperature during
the turning of AISI 4140 alloy steel was performed using
various cutting parameters. The tool-chip interface
temperature TCTI was measured by infrared thermometer,
the tool temperature TT was measured with a K-type
thermocouple in the cutting zone, and the average
surface roughness Ra was measured using a portable
surface-roughness measurement device. The Taguchi
design was selected to find the relationships between the
control factors. The depth of cut (ap), the cutting speed
(vc), and the feed rate (f) were taken as the control
factors.
2 TEMPERATURES DURING METAL CUTTING
In the cutting process, nearly all of the energy
dissipated during plastic deformation is converted into
heat, which in turn raises the temperature in the cutting
zone. Since the heat generation is closely related to the
plastic deformation and friction, we can specify three
main sources of heat when cutting:
• plastic deformation by shearing in the primary shear
zone;
• friction on the cutting face and friction between the
chips;
• tool on the tool flank.
Temperature results in dimensional errors on the
machined surface. The cutting tool elongates as a result
of the increased temperature, and the position of the
cutting tool edge shifts towards the machined surface,
resulting in a dimensional error of about 0.01–0.02 mm.
Since the processes of thermal generation, dissipation,
and solid-body thermal deformation are all transient,
some time is required to achieve the steady-state con-
dition.
Heat is mostly dissipated by: the discarded chip that
carries away about 60–80 % of the total heat, the
workpiece acts as a heat sink drawing away 10–20 % of
heat, while the cutting tool draws away ~10 % of the
heat. The balance between heat generation and heat
dissipation during metal cutting is shown in Figure 1.
3 MATERIALS AND METHOD
3.1 Workpiece and cutting tool
The workpiece material is AISI 4140 alloy steel. The
chemical composition of the workpiece material (in
volume fractions) is shown in Table 1. The machining
process was performed using a NR 2020K-08 tool holder
A. R. MOTORCU et al.: ANALYSIS OF THE CUTTING TEMPERATURE AND SURFACE ROUGHNESS ...
344 Materiali in tehnologije / Materials and technology 50 (2016) 3, 343–351
Figure 1: The balance of heat generation and heat dissipation during
metal cutting
Slika 1: Izravnava med spro{~eno in odvedeno toploto pri rezanju
kovin
and a TiAlN-TiN, PVD-coated, WNVG 080404-IC907
solid carbide insert. Figure 2 and Table 2 show a
schematic of the tip geometry and the specifications of
the insert.
Table 1: Chemical composition of AISI 4140 alloy steel (in volume
fractions, x/%)
Tabela 1: Kemijska sestava legiranega jekla AISI 4140 (v volumen-
skih odstotkih, x/%)
C Cr Ni Mn P S Si Mo
0.38 0.80 9.58 0.75 0.035 0.04 0.15 0.15
Table 2: The specifications of the insert
Tabela 2: Specifikacije vlo`ka za rezanje
TiAlN-TiN PVD-coated WNVG 080404-IC907
d1 S I r HRA TRS d
12.70 4.83 8.70 0.40 92.80 560 4.70
Property Value
ISO Range – P/M/K (P10-P30)(M05-M20)
ISO Range – H/S/N (H05-H15)(S05-S20)
Grade or coating type PVD
Coating layers TiAlN-TiN
3.2 Experimental conditions, temperature and sur-
face-roughness measurements
In this study, two methods of tool-temperature eva-
luation are presented:
• the placement of the K-type thermocouple on the
tool,
• the infrared pyrometer.
A schematic view of the experimental setup is shown
in Figure 3. Cylindrical workpieces (Ø45 × 300 mm)
were fixed between the chuck and the tailstock and were
pre-machined using a separate insert. The workpiece
samples were heat treated by induction hardening and a
hardness of 50 HRC was maintained. The samples were
then solution heat treated and oil quenched in order to
achieve the proper hardness.
In this study, an Optris CF4 infrared thermometer
was used to measure TCTI. The maximum temperature
(which was about 525 °C) was recorded around the
cutting zone. A total of 18 trials were conducted
throughout these experiments and brand new inserts
were used for each temperature measurement. Hence, the
cutting temperature increased with the cutting speed, the
feed rate and the depth of cut. The experiments were
repeated three times for the same cutting conditions and
the measured values were averaged. TT was measured
using a K-type thermocouple. The thermocouple
measurements were recorded every five seconds.
The Ra surface roughness was measured to charac-
terize the surface quality. The Ra measurements were
carried out using a Time TR 200 device by obtaining
values from different points that were parallel to the
workpiece axis at a cut-off length of 5.6 mm. According
to the experimental design, three measurements were
made on the surfaces at the specified values of the con-
trol factors, and the Ra values were determined by taking
the average of the measurement results.
3.3 Experimental design using the Taguchi method
The Taguchi design was selected to find the relation-
ships between the control factors and the quality charac-
teristics. The cutting speed (vc), feed rate (f) and depth of
cut (ap), whose levels are given Table 3, were selected as
the control factors. The quality characteristics were the
tool-chip interface temperature (TCTI), the tool tempe-
rature (TT) and the average surface roughness (Ra). As
the total degree of freedom of the factor group was 5, a
standard Taguchi experimental plan with the notation
L18 (21 × 37) was chosen as the orthogonal array. The
rows in the L18 orthogonal array used in the experiment
corresponded to each trial and the columns contained the
factors to be studied. The first column consists of the
depth of cut; the second and the third columns contain
the cutting speed and the feed rate, respectively. In the
Taguchi method, the experimental results are trans-
formed into a S/N ratio. The S/N ratio is used while
approaching or moving away from the desired value and
measuring the quality characteristics.15-18 The smaller-is-
better (SB), the nominal-best (NB) and the larger-is-
better (LB) approaches are found according to the results
of the S/N ratio.15-18 As the tool-chip interface tempera-
ture (TCTI), the tool temperature (TT) and the surface
roughness (Ra) values were required to be the lowest, the
S/N ratios of these quality characteristics were calculated
in dB using Equation (1) according to the SB option in
the study.15–18
A. R. MOTORCU et al.: ANALYSIS OF THE CUTTING TEMPERATURE AND SURFACE ROUGHNESS ...
Materiali in tehnologije / Materials and technology 50 (2016) 3, 343–351 345
Figure 3: Thermocouple and IR pyrometer connections to the lathe
Slika 3: Povezava termo~lena in IR-pirometra na stru`nici
Figure 2: Schematic of tip geometry
Slika 2: Shema geometrije rezilne konice
S/N y i
i
n
SB lg= − ⋅
⎛
⎝
⎜ ⎞
⎠
⎟
=
∑10 1
2
2
1
(1)
In the Equation (1), n is the number of the experi-
ment and yi is the ith data point obtained.15–18 ANOVA
was applied in order to determine the percentage effects
of the control factors on TCTI, TT and Ra.
Table 3: Control factors and their levels
Tabela 3: Kontrolni faktorji in njihovi nivoji
Sym-
bol
Control
factors Unit
Level
1
Level
2
Level
3
Degree of
freedom
(DoF)
ap
Depth of
cut mm 0.40 0.60 – 1
vc
Cutting
speed m/min 76 114 170 2
f Feed rate mm/rev 0.05 0.08 0.12 2
3.4 Predictive models for temperature and surface
roughness with multiple regression analysis
Equations were developed for the prediction of TCTI,
TT and Ra using the experimental results in a multiple
regression analysis. The second-order linear models
containing the main effects of the control factors and
their interactions are signified with the Equation (2):
Y y b x b x b x
b x b x b x b x
1 0 0 1 1 2 2
3 3 4 12 5 13 6 23
= − = + + +
+ + + +

 (2)
where Y1 is the estimated answer of the second-order
equation and y is the tool-chip interface temperature
(TCTI), tool temperature (TT) or surface roughness (Ra)
measured on a the logarithmic scale, x0 = 1 is the fixed
variable, the x1, x2 and x3 control factors are the
logarithmic transformations of the depth of cut, the
cutting speed and the feed rate and, the x12, x13 and x23
interactions of the control factors are the logarithmic
transformations of the depth of cut–cutting speed, the
depth of cut–feed rate and cutting speed–feed rate. The
coefficient of the experimental error is 
, and the b
values (b0, b1, b2, b3, b4, b5 and b6) are the coefficients
of related factors.
4 ANALYSIS OF THE RESEARCH RESULTS
The present study was performed to understand and
evaluate the infrared- and thermocouple-based tempera-
ture measurements during metal cutting and to consider
the practical difficulties. TCTI, TT and Ra were used as the
quality characteristics. The experimental results are
shown in Table 4.
The TCTI, TT and Ra measurement results from the
turning of the quenched and tempered AISI 4140 steel
with coated carbide tools were resolved and analyzed by
means of the Minitab 16.0 package software. From
Table 4 it is clear that the overall means for TCTI, TT and
Ra were calculated as 446.11 °C, 70.78 °C and 0.578 μm,
respectively.
Table 4: The experimental results for the quality characteristics and
S/N ratios
Tabela 4: Rezultati preizkusov za opis kvalitete in S/N razmerja
Exp.
no
Control
factors Measured values S/N Ratios (dB)
ap vc f
Tool-
chip in-
terface
tempera-
ture, TCTI
(°C) (IR
Pyro-
meter)
Tool
tempera-
ture, TT
(°C)
(Thermo-
couple)
Surface
rough-
ness, Ra
(μm)
S/N
TCTI
S/N
TT
S/N
Ra
1 0.4 76 0.05 410 57 0.295 –52.26 –35.12 10.60
2 0.4 76 0.08 405 66 0.483 –52.15 –36.39 6.32
3 0.4 76 0.12 410 72 0.958 –52.26 –37.15 0.37
4 0.4 114 0.05 460 65 0.484 –53.26 –36.26 6.30
5 0.4 114 0.08 465 61 0.579 –53.35 –35.71 4.75
6 0.4 114 0.12 425 67 0.988 –52.57 –36.52 0.10
7 0.4 170 0.05 520 65 0.410 –54.32 –36.26 7.74
8 0.4 170 0.08 500 67 0.492 –53.98 –36.52 6.16
9 0.4 170 0.12 475 71 0.872 –53.53 –37.03 1.19
10 0.6 76 0.05 400 72 0.489 –52.04 –37.15 6.21
11 0.6 76 0.08 390 80 0.530 –51.82 –38.06 5.51
12 0.6 76 0.12 395 76 0.720 –51.93 –37.62 2.85
13 0.6 114 0.05 430 80 0.429 –52.67 –38.06 7.35
14 0.6 114 0.08 435 75 0.547 –52.77 –37.50 5.24
15 0.6 114 0.12 420 83 0.722 –52.46 –38.38 2.83
16 0.6 170 0.05 485 81 0.354 –53.71 –38.17 9.02
17 0.6 170 0.08 525 67 0.406 –54.40 –36.52 7.83
18 0.6 170 0.12 480 69 0.643 –53.62 –36.78 3.84
Overall mean of TCTI = 446.11 °C, S/N ratio of TCTI = –52.95 dB
Overall mean of TT = 70.78 °C, S/N ratio of TCTI = –36.95 dB
Overall mean of Ra = 0.578 ìm, S/N ratio of Ra = 5.24 dB
The variation of the tool temperature and the tool-
chip interface temperature with the cutting parameters
are shown in Figures 4a and 4b. Obviously, it is clear
that the tool-chip interface temperature and the tool tem-
perature increase with an increase in the cutting speed
(Figure 4a). The influence of the tool temperature and
the feed rate on the surface roughness is shown in Figure
5a. It was observed that the lowest feed rate produced a
better surface quality. The experiments showed that the
cutting speed and the feed rate are the main factors
affecting the surface roughness (Figure 5b).
4.1 Analysis of the control factors for the temperature
and surface roughness
The responses for the S/N ratios (smaller is better) of
TCTI, TT and Ra are presented in Table 5 and the
responses for the means in Table 6. While the signal
value represents the real desired value that the system
gives and which is to be measured, the noise factor
represents the portion of the undesired factors in the
measured value. The S/N ratio analysis provided signi-
ficant information about the nature of the process of
turning hardened AISI 4140 steel with coated carbide
cutting tools under selected conditions. The fact that the
differences between the highest and the lowest S/N ratio
values of each control factor calculated at different levels
A. R. MOTORCU et al.: ANALYSIS OF THE CUTTING TEMPERATURE AND SURFACE ROUGHNESS ...
346 Materiali in tehnologije / Materials and technology 50 (2016) 3, 343–351
A. R. MOTORCU et al.: ANALYSIS OF THE CUTTING TEMPERATURE AND SURFACE ROUGHNESS ...
Materiali in tehnologije / Materials and technology 50 (2016) 3, 343–351 347
Figure 4: The influence of cutting speed and feed rate on the tempera-
ture: a) cutting speed, b) feed rate
Slika 4: Vpliv hitrosti rezanja in hitrosti podajanja na temperaturah:
a) hitrost rezanja, b) hitrost podajanja
Figure 5: Influence of tool temperature and feed rate on surface
roughness: a) tool temperature, b) feed rate
Slika 5: Vpliv temperature orodja in hitrosti podajanja na hrapavost
povr{ine: a) temperatura orodja, b) hitrost podajanja
Table 5: Response table for S/N ratios (smaller is better) of TCTI, TT and Ra
Tabela 5: Razpredelnica odgovorov za S/N razmerja (manj{e je bolj{e) za TCTI, TT in Ra
Level Tool-chip interface temperature, TCTI(dB) Tool temperature, TT (dB) Surface roughness, Ra (dB)
ap vc f ap vc f ap vc f
1 –53.07 –52.08 –53.04 –36.33 –36.91 –36.84 4.838 5.313 7.873
2 –52.83 –52.85 –53.08 –37.58 –37.07 –36.78 5.632 4.429 5.969
3 – –53.93 –52.73 – –36.88 –37.24 – 5.963 1.864
 0.25 1.85 0.35 1.25 0.19 0.46 0.793 1.534 6.008
Rank 3 1 2 1 3 2 3 2 1
Table 6: Response table for means of TCTI, TT and Ra
Tabela 6: Tabela odgovorov za pomen TCTI, TT in Ra
Level Tool-chip interface temperature, TCTI(°C) Tool temperature, TT (°C) Surface roughness, Ra (μm)
ap vc f ap vc f ap vc f
1 452.2 401.7 450.8 65.67 70.50 70.00 0.6179 0.5792 0.4102
2 440.09 439.2 453.3 75.89 71.83 69.33 0.5378 0.6248 0.5062
3 – 497.5 434.2 – 70.00 73.00 – 0.5295 0.8172
 12.2 95.8 19.2 10.22 1.83 3.67 0.0801 0.0953 0.4070
Rank 3 1 2 1 3 2 3 2 1
are higher or lower was used in the determination of the
factors effective on TCTI, TT and Ra. The most effective
parameters on TCTI were the cutting speed, the feed rate
and the depth of cut because there were (1.85, 0.35 and
0.25) dB differences between their levels (Table 5). The
most effective parameters on TT were determined to be
the depth of cut, the feed rate and the cutting speed, with
differences of (1.25, 0.46 and 0.19) dB, respectively (Table
5). The most effective parameters on Ra were determined
to be the feed rate, the cutting speed and the depth of cut,
with differences of (6.008, 1.534 and 0.793) dB, res-
pectively (Table 5). The optimum values for the surface
roughness and the dimensional accuracy were reported to
be a2v1f3, a1v3f2 and a2v3f1, respectively (Table 6).
The main effects of the control factors on the perfor-
mance characteristics during the turning of the quenched
and tempered AISI 4140 steel with coated carbide
cutting tools were demonstrated using the "Graphical
Representation of Factor Effects" and evaluated.8–11 The
main effect graphs showing the effects of the control
factors on TCTI, TT and Ra are given in Figures 6 and 7,
respectively.
In Figure 6, the optimum levels of the control factors
for the tool-chip interface temperature are a2 (ap = 0.6
mm), v1 (vc = 76 m/min) and f3 (f = 0.12 mm/rev),
respectively. TCTI increases depending on the increase of
the cutting speed and the decrease of the depth of cut and
the feed rate. From the same graphic it is clear that the
most effective control factor on TCTI is the cutting speed.
In Figure 7, the optimum levels of the control factors for
the tool temperature are a1 (ap = 0.4 mm), v2 (vc = 114
m/min) and f2 (f = 0.08 mm/rev), respectively.
In Figure 7, when the effects of the control factors on
tool temperature were examined, a significant increase
was observed on TT, depending on the increase in the
depth of cut. With an increase of the cutting speed from
76 m/min to 114 m/min and an increase of the feed rate
from 0.08 mm/rev to 0.12 mm/rev the tool temperature
was increased (Figure 7). Similarly, the optimum levels
for the minimum Ra surface roughness were observed to
be a2 (ap = 0.6 mm), v3 (vc = 170 m/min) and f1 (f = 0.05
mm/rev), respectively (Figure 5). The most effective
parameter on Ra was the feed rate (Figure 7). With a
further increase in the feed rate value the Ra surface
roughness value increased.
ANOVA is a statistically based, objective, decision-
making tool used for determining any difference in the
average performance of a group of items being
tested.15–18 In the case when the F value of a process
parameter is greater than the tabulated F ratio, it shows
that the control factor has a significant effect on the per-
formance characteristic. An analysis of variance
(ANOVA) with a 95 % confidence interval was carried
out for each experiment using the L18 orthogonal array in
order to determine the effects of the control factors and
their interactions on selected performance/quality cha-
racteristics. The results of the ANOVA carried out for
TCTI, TT and Ra are presented in Tables 7, 8 and 9. The
cutting speed became the most effective factor for the
tool-chip interface temperature, with a contribution of
86.57 % followed by the feed rate with 4.03 % (Table 7).
The effects of other control factors and their interactions
on TCTI became insignificant with a smaller 5 % contri-
bution (Table 7). The results of the ANOVA for the tool
A. R. MOTORCU et al.: ANALYSIS OF THE CUTTING TEMPERATURE AND SURFACE ROUGHNESS ...
348 Materiali in tehnologije / Materials and technology 50 (2016) 3, 343–351
Figure 6: Mean effect plots for temperatures: a) tool-chip interface
temperature, b) tool temperature
Slika 6: Diagram srednjega vpliva na temperature: a) temperatura
stika orodje-ostru`ek, b) temperatura orodja
Figure 7: Mean effect plots for Ra surface roughness
Slika 7: Diagrami srednjega vpliva na hrapavost povr{ine Ra
temperature (TT) indicate that the depth of cut (ap) has
more influence on the tool temperature, with a contri-
bution of 52.65 % and a cutting speed–feed rate (vxf)
16.75 %, depth of cut–cutting speed (axv) 9.12 %, depth
of cut–feed rate (axf) 7.51 %, feed rate (f) 5.13 % and
cutting speed (vc) 1.21 % followed by a contribution %,
respectively (Table 8). Finally, from Table 9, it is
concluded that the feed rate with a contribution of
76.63 % has more influence on the surface roughness
(Ra) followed by the depth of cut–feed rate (axf), the
depth of cut (ap) and the cutting speed (vc) to obtain the
minimum surface roughness (Table 9).
Table 7: Results of ANOVA for tool-chip interface temperature (TCTI)
Tabela 7: Rezultati ANOVA za temperaturo stika orodje-ostru`ek
(TCTI)
Source DoF SS V F-Ratio Prob.>F Contr.(%)
ap 1 672.3 672.2 3.44 0.137 2.08
vc 2 27986.1 13993.1 71.71 0.001 86.57
f 2 1302.8 651.4 3.34 0.140 4.03
axv 2 302.8 151.4 0.78 0.519 0.94
axf 2 369.4 184.7 0.95 0.461 1.14
vxf 4 913.9 228.5 1.17 0.441 2.83
Res.Err. 4 780.6 195.1 2.41
Total 17 32327.8 100.00
R2 = 97.6, R2 (adj) = 89.7 (significant at 95 % confidence level)
Table 8: Results of ANOVA for tool temperature (Tt)
Tabela 8: Rezultati ANOVA za temperaturo orodja (Tt)
Source DoF SS V F-Ratio Prob.>F Contr.(%)
ap 1 470.22 470.22 27.57 0.006 52.65
vc 2 10.78 5.389 0.32 0.746 1.21
f 2 45.78 22.889 1.34 0.358 5.13
axv 2 81.44 40.722 2.39 0.208 9.12
axf 2 67.11 33.556 1.97 0.254 7.51
vxf 4 149.56 37.389 2.19 0.233 16.75
Res.Err. 4 68.22 17.056 7.64
Total 17 893.11 100.00
R2 = 92.4, R2 (adj) = 67.5 (significant at 95 % confidence level)
Table 9: Results of ANOVA for surface roughness (Ra)
Tabela 9: Rezultati ANOVA za hrapavost povr{ine (Ra)
Source DoF SS V F-Ratio Prob.>F Contr.(%)
ap 1 0.028880 0.028880 10.73 0.031 4.18
vc 2 0.027281 0.013641 5.07 0.080 3.95
f 2 0.543172 0.271586 100.89 0.000 78.63
axV 2 0.014830 0.007415 2.75 0.177 2.15
axf 2 0.062656 0.031328 11.64 0.022 9.07
Vxf 4 0.003192 0.000798 0.30 0.867 0.46
Res.Err. 4 0.010767 0.002692 – – 1.56
Total 17 0.690779 – – – 100.00
R2 = 98.4, R2 (adj) = 93.4 (significant at 95 % confidence level)
4.2 Developed second-order predictive equations for
the temperature and surface roughness
The equations that were developed with multiple
linear regression analysis to predict TCTI, TT and Ra in the
turning of quenched and tempered AISI 4140 steel with
coated carbide cutting tools and the equations that
contain the main effects of the control factors and their
interaction effects are presented in Equations (3) to (5),
respectively.
T a v f av vfCTI = − + + + −382 146 0 966 179 0 708 358. . . (4)
T a v f av vfT = − + + − −2 0 968 0328 255 0381 174. . . . . (5)
R a v f
av
a = − − + + −
− −
0114 034 0 00332 715
0 00617
. . . .
. 0 0102. vf
(6)
These equations were developed according to the
un-coded values of the control factors (i.e., 0.4, 0.6 mm,
etc. for ap; i.e., 76, 114, 170 m/min, etc. for vc; i.e., 0.05,
0.08, 0.012 mm/rev, etc. for f). af is highly correlated
with other variables, so af has been removed from all of
the equations. The correlation coefficients (R2) and the
adjusted correlation coefficients (R2(adj)) of the second-
order equations developed for the predictive tool-chip
interface temperature (TCTI) measured with an IR pyro-
meter, the tool temperature (TT) measured with a thermo-
couple and the surface roughness (Ra) were calculated as
R2 = 92.8 %, R2(adj) = 89.8 %, R2 = 68.1 %, R2(adj) =
54.8 % and R2 = 82.6 % R2(adj) = 75.3 %, respectively.
R2(adj) determines the amount of deviation about the
mean that is described by the model. The predicted R2
value and the R2(adj) value were found to be in good
agreement. These values show that the equations deve-
loped are sufficient to determine all the response values
at a confidence interval of 95 %. The regression models
can be successfully adopted for estimating TCTI, TT and
Ra. Moreover, as seen in these equations, vc and f have
additive effects, while ap has a negative effect on TCTI, TT
and Ra.
The comparisons of the results of TCTI, TT and Ra
measured experimentally (Table 4) with the fits for TCTI,
TT and Ra estimated via the Taguchi method and fits for
TCTI, TT and Ra estimated via the Regression model
(Equation (3) to (5)) are given in Table 8. As can be seen
from this table, the TCTI results obtained from the
Taguchi method and the linear regression analysis were
found to be very close. The mean of the % error ratios of
the estimated results obtained by the Taguchi method
and the predictive equations were less than 14 %. This
reflects the reliability of the statistical analyses (Tables
10 and 11).
5 CONCLUSIONS
In this study, the Taguchi design was selected to
determine the effects of the control factors. The effects
of the depth of cut, the cutting speed and the feed rate on
the tool-chip interface temperature (TCTI), the tool
A. R. MOTORCU et al.: ANALYSIS OF THE CUTTING TEMPERATURE AND SURFACE ROUGHNESS ...
Materiali in tehnologije / Materials and technology 50 (2016) 3, 343–351 349
A. R. MOTORCU et al.: ANALYSIS OF THE CUTTING TEMPERATURE AND SURFACE ROUGHNESS ...
350 Materiali in tehnologije / Materials and technology 50 (2016) 3, 343–351
Table 10: The comparisons of measured of TCTI, and Tt experimentally with fits estimated via the Taguchi method and regression models
Tabela 10: Primerjava izmerjenih TCTI in eksperimentalnih Tt, z ujemanji, dolo~enimi po Taguchi metodi in z regresijskimi modeli
Exp.
no
Tool-chip interface temperature, TCTI (°C) Tool temperature, Tt (°C)
Meas-
ured
TCTI
Fits for TCTI
estimated via
Taguchi method
Error
%
Fits for TCTI
estimated via
Regression model
Error
%
Meas-
ured
Tt
Fits for Tt
estimated via
Taguchi method
Error
%
Fits for Tt
estimated via
Regression model
Error
%
1 410 418 2.0 414 0.9 57 56 1.0 60 5.7
2 405 401 0.9 411 1.4 66 68 2.9 64 3.2
3 410 406 1.1 407 0.7 72 71 1.9 69 4.5
4 460 462 0.5 454 1.3 65 62 3.9 64 2.1
5 465 458 1.5 447 3.8 61 61 0.1 65 7.0
6 425 430 1.1 438 3.1 67 70 3.9 68 0.8
7 520 510 2.0 514 1.1 65 68 4.8 69 5.5
8 500 511 2.1 501 0.2 67 65 2.8 67 0.5
9 475 475 0.1 484 1.9 71 70 1.7 66 7.5
10 400 392 2.0 395 1.2 72 73 0.8 74 2.5
11 390 394 0.9 392 0.6 80 78 2.4 77 3.1
12 395 399 1.1 389 1.6 76 77 1.8 82 8.4
13 430 428 0.5 441 2.6 80 83 3.2 74 7.1
14 435 442 1.6 434 0.2 75 75 0.1 76 1.3
15 420 415 1.1 425 1.2 83 80 3.1 78 5.8
16 485 495 2.1 509 4.9 81 78 3.8 75 7.4
17 525 514 2.0 496 5.5 67 69 2.8 74 10.1
18 480 480 0.1 479 0.2 69 70 1.8 72 4.5
Min 390 392 0.1 389 0.2 57 56 0.1 60 0.5
Max 525 514 2.1 514 5.5 83 83 4.8 82 10.1
Mean 446 – 1.3 – 1.8 71 – 2.4 – 4.8
Table 11: The comparisons of the measured Ra with fits estimated via the Taguchi method and the regression models
Tabela 11: Primerjava izmerjene Ra z ujemanji, dolo~enimi po Taguchi metodi in z regresijskimi modeli
Exp.no Measured Ra Fits for Ra estimated viaTaguchi method Error %
Fits for Ra estimated via
Regression model Error %
1 0.295 0.338 14.4 0.406 37.5
2 0.483 0.478 1.1 0.597 23.5
3 0.958 0.921 3.9 0.851 11.1
4 0.484 0.461 4.7 0.418 13.5
5 0.579 0.594 2.5 0.598 3.3
6 0.988 0.996 0.8 0.837 15.3
7 0.410 0.390 4.9 0.438 6.7
8 0.492 0.483 1.9 0.600 21.9
9 0.872 0.901 3.4 0.816 6.4
10 0.489 0.446 8.7 0.380 22.3
11 0.530 0.535 1.0 0.571 7.7
12 0.720 0.757 5.2 0.826 14.7
13 0.429 0.452 5.3 0.346 19.4
14 0.547 0.532 2.7 0.525 4.0
15 0.722 0.714 1.1 0.765 5.9
16 0.354 0.374 5.7 0.296 16.5
17 0.406 0.415 2.3 0.458 12.8
18 0.643 0.614 4.6 0.674 4.9
Min 0.295 0.338 0.8 0.296 3.3
Max 0.988 0.996 14.4 0.851 37.5
Mean 0.578 – 4.1 – 13.7
temperature (TT) and the surface roughness (Ra) were
investigated in the turning of the quenched and nor-
malized AISI 4140 steel workpieces that were machined
using TiAlN-TiN, PVD-coated, carbide tools and the
obtained results are as follows:
The most effective parameter on the tool-chip inter-
face temperature was the cutting speed with a contribu-
tion ratio of 86.57 %. The effective parameters for tool
temperature were the depth of cut, the cutting speed-feed
rate, the depth of cut–cutting speed, the depth of cut-feed
rate, the feed rate and the cutting speed with contribu-
tions of 52.65 %, 16.75 %, 9.12 %, 7.51 %, 5.13 % and
1.21 %.
The feed rate with a contribution of 76.63 % has
more influence on the surface roughness (Ra) followed
by the depth of cut–feed rate (axf), the depth of cut (ap)
and the cutting speed (vc) to obtain the minimum surface
roughness.
The optimum levels of the control factors were ap =
0.6 mm, vc = 76 m/min and f = 0.12 mm/rev for the
minimum tool-chip interface temperature; ap = 0.4 mm,
vc = 114 m/min and f = 0.08 mm/rev for the minimum
tool temperature, and ap = 0.6 mm, vc = 170 m/min and
f = 0.05 mm/rev for the minimum Ra surface roughness.
The tool-chip interface temperature increased signi-
ficantly depending on the increase of the cutting speed.
The depths of cut and feed rate do not have a significant
effect on the tool-chip interface temperature.
The tool temperature increased significantly depend-
ing on the increase of the depth of cut.
The surface roughness increased depending on the in-
crease of the feed rate, while the same tendency was not
observed for the depth of cut and the cutting speed.
The correlation coefficients of the predictive equa-
tions developed for the estimation of the minimum
tool-chip interface temperature, the tool temperature and
the surface roughness by multiple linear regression anal-
ysis were calculated as 0.928, 0.681 and 0.826, respec-
tively. Higher correlation coefficients reflect the reliabi-
lity of the developed equations.
The mean of the % error ratios of the estimated re-
sults obtained by Taguchi method and the predictive
equations were less than 14 %. This reflects the reliabi-
lity of the statistical analyses.
6 REFERENCES
1 X. L. Liu, D. H. Wen, Z. J. Li, L. Xiao, F. G. Yan, Cutting tempe-
rature and tool wear of hard turning hardened bearing steel, Journal
of Materials Processing Technology, 129 (2002), 200–206,
doi:10.1016/S0924-0136(02)00651-9
2 N. A. Abukhshim, P. T. Mativenga, M. A. Sheikh, Investigation of
heat partition in high speed turning of high strength alloy steel, Inter-
national Journal of Machine Tools and Manufacture, 45 (2005),
1687–1695, doi:10.1016/j.ijmachtools.2005.03.008
3 N. A. Abukhshim, P. T. Mativenga, M. A. Sheikh, Heat generation
and temperature prediction in metal cutting: a review and impli-
cations for high speed machining, International Journal of Machine
Tools and Manufacture, 46 (2006) 7–8, 782-800, doi:10.1016/
j.ijmachtools.2005.07.024
4 A. H. Suhail, N. Ismail, S. V. Wong, N. A. Abdul Jalil, Optimization
of cutting parameters based on surface roughness and assistance of
workpiece surface temperature in turning process, American Journal
of Engineering and Applied Sciences, 3 (2010) 1, 102–108,
doi:10.3844/ajeassp.2010.102.108
5 P. D. Berger, R. E. Maurer, Experimental design with applications in
management, Engineering and The Sciences, 1th ed., Duxbury Press,
USA 2001
6 G. R. Henderson, Six sigma: Quality improvement with MINITAB,
2nd ed., John Wiley and Sons, England 2006, 452–460
7 T. P. Ryan, Statistical methods for quality improvement, 2nd ed.,
John Wiley and Sons, USA 2000
8 A. Mishra, A. Gangele, Application of Taguchi method in optimi-
zation of tool flank wear width in turning operation of AISI 1045
steel, Industrial Engineering Letters, 2 (2012) 8, 11–18
9 M. S. Phadke, Quality engineering using design of experiment, quali-
ty control, Rebust design and Taguchi method, 1st ed., Warsworth
and Books, California 1998
10 W. H. Yang, Y. S. Tarng, Design optimization of cutting parameters
for turning operations based on Taguchi method, Journal of Materials
Processing Technology, 84 (1998), 122–129, doi:10.1016/S0924-
0136(98)00079-X
11 B. M. Gopalsamy, B. Mondal, S. Ghosh, Taguchi method and
ANOVA: An approach for process parameters optimization of hard
machining while machining hardened steel, Journal of Scientific and
Industrial Research, 68 (2009) 8, 686-695
12 F. Ficici, M. Kapsiz, M. Durat, Applications of Taguchi design
method to study wear behaviour of boronized AISI 1040 steel,
International Journal of Physical Sciences, 6 (2011) 2, 237–243
13 M. Adinarayana, G. Prasanthi, G. Krishnaiah, Parametric analysis
and multi objective optimization of cutting parameters in turning
operation of AISI 4340 alloy steel with CVD cutting tool, Inter-
national Journal of Research in Engineering and Technology, 3
(2014) 2, 449–456
14 E. D. Kirby, A parameter design study in a turning operation using
the Taguchi method, The Technology Interface, (2006), 1–14
15 P. J. Ross, Taguchi techniques for quality engineering: Loss function,
orthogonal experiments, parameter and tolerance design, McGraw-
Hill, New York 1988
16 D. C. Montgomery, Taguchi’s contributions to experimental design
and quality engineering, design and analysis of experiment, Wiley,
Canada 1991
17 E. Canyýlmaz, F. Kutay, An alternative approach to analysis of vari-
ance in Taguchi method, Journal of the Faculty of Engineering &
Architecture of Gazi University, 18 (2003) 3, 51–63
18 R. K. Roy, A primer on the Taguchi method, Van Nostrand Reinhold,
New York 1990
A. R. MOTORCU et al.: ANALYSIS OF THE CUTTING TEMPERATURE AND SURFACE ROUGHNESS ...
Materiali in tehnologije / Materials and technology 50 (2016) 3, 343–351 351