UDK 621.941:519.233.4	ISSN 1580-2949
Original scientific article/Izvirni znanstveni članek	MTAEC9, 49(5)765(2015)
APPLICATION OF THE TAGUCHI METHOD TO OPTIMIZE THE CUTTING CONDITIONS IN HARD TURNING OF A RING BORE
UPORABA TAGUCHI-JEVE METODE ZA OPTIMIZACIJO TRDEGA STRUŽENJA ROBA IZVRTINE
Mehmet Boy1, Ibrahim Ciftci2, Mustafa Gunay3, Feridun Ozhan4
1Karabuk Vocational College, Karabuk University, Karabuk, Turkey 2Department of Manufacturing Engineering, Karabuk University, Karabuk, Turkey 3Department of Mechanical Engineering, Karabuk University, Karabuk, Turkey 4ORS Bearing, Ankara, Turkey mboy@karabuk.edu.tr
Prejem rokopisa - received: 2014-09-29; sprejem za objavo - accepted for publication: 2014-11-28
doi:10.17222/mit.2014.246
This paper is focused on optimizing the cutting conditions for the surface roughness, inner-diameter error and roundness obtained in hard turning of an inner ring bore. The hard-turning experiments were conducted on hardened and tempered AISI 52100 bearing rings using the L9 orthogonal array on a CNC lathe. The cutting speed, feed rate and number of the machined part were selected as control factors. The optimum cutting conditions were determined using the signal-to-noise (S/N) ratio. S/N ratios were calculated using the lower-the-better approach. An analysis of variance (ANOVA) was also employed to determine the level of the effect of the control factors for the surface roughness, inner-diameter error and roundness. The statistical analysis showed that the feed rate was the most significant factor for the surface roughness while the cutting speed was the most significant factor for the roundness and inner-diameter error. Finally, the optimum cutting conditions were further confirmed with confirmation tests.
Keywords: diameter error, hard turning, roundness, surface roughness, Taguchi method
Članek je osredinjen na optimiranje razmer pri rezanju glede na hrapavost površine, napake notranjega premera in okroglosti pri trdem struženju notranjega roba izvrtine. Preizkusi trdega struženja so bili izvršeni pri kaljenem in popuščenem AISI 52100 obroču ležaja z uporabo L9 ortogonalne matrike na CNC-stružnici. Hitrost rezanja, hitrost podajanja in število obdelanih kosov so bili izbrani kot kontrolni faktorji. Optimalne razmere rezanja so bile izbrane z uporabo razmerja signal-hrup (S/N). S/N-razmerja so bila izračunana s približkom čim manj tem boljše. Analiza variance (ANOVA) je bila tudi uporabljena za določitev vpliva kontrolnih faktorjev na hrapavost površine, napak notranjega premera in okroglosti. Končno so bili optimalni parametri potrjeni s potrditvenimi preizkusi.
Ključne besede: napaka premera, trdo struženje, okroglost, hrapavost površine, Taguchi-jeva metoda
1 INTRODUCTION	turning, due to the demands for a geometric accuracy of
a few micrometers, its application is limited by the un-
Bearings are the most important components of	certainties with respect to the part quality and process
rotating mechanical systems. The highly precise surface	reliability 1-4 finish, material properties, cleanliness, dimensions and
For a turned part, the diameter error, the surface
tolerances of today s bearings contribute significantly to	, j j
roughness and the roundness are the three most import-
the product performance. Traditionally, bearing rings are
r , . r ATCT coiAA , 1 .u u	aut quality characteristics.There are many causes for
manufactured from the soft AISI 52100 steel through	j. . , j . • , j
forging, cold ring rolling and machining processes to	dimensional and geometric accuracy errors in hard
their near-net shapes. Then, they are hardened and tem-	turning. The roundness error is considered to be one of
pered to obtain the required hardness. The final geometry	the important geometrical errors of cylindrical compo-
and surface quality are achieved through grinding and	nents because it has a negative effect on the accuracy and
super-finishing processes. The hard-turning process has	other important factors such as the mating mechanical
been increasingly used as an alternative to grinding	components and wear of rotating elements.5,6 operations. The hard-turning process has the capacity of	The machine-tool rigidity and clamping system machining the hardened components such as roller	influence the form accuracy and dimensional tolerance in bearings, gears, shafts, cams, axles requiring a low	precision hard turning. Possible machining errors may be surface roughness and close dimensional and form tole-	caused due to different types of clamping, the number of rances. Hard turning, compared to grinding operations,	clamping jaws, clamping forces, the hydraulic pressure has many advantages such as a high material-removal	of a clamping system, the clamping rigidity and accurate, lower production costs, a shorter cycle time, an	racy. The effects of clamping techniques on the defor-improved overall product quality, and a lower environ-	mations of thin-walled rings have been investigated by mental impact. However, in the area of precision hard	many researchers.7-13
Geometric errors mostly result from the inaccuracy of tooling parts, such as inserts, tool holders and clamping devices.12 An investigation suggested that after replacing an insert, the repeatability errors at the tip of the insert can reach up to several microns, and the displacement of a tool tip under a cutting load can also reach several microns. Precision hard turning requires highly precise machine tools to eliminate the form and dimensional errors induced by the moving components of a machine tool such as the spindle, the slide bed and the tail stock.713
The surface roughness is one of the most important requirements in a machining process, as it is considered to be an indicator of the product quality. It indicates the irregularities of a surface texture. Achieving a desired surface quality is critical for the functional behavior of a part. The surface roughness influences the performance of mechanical parts and their production costs because it affects the factors such as friction, ease of holding the lubricant, electrical and thermal conductivities. A relatively better surface finish may involve a higher cost of manufacturing. The surface roughness and the roundness error are affected by several factors including the cutting-tool geometry, the cutting speed, the feed rate, the microstructure of a workpiece and the rigidity of a machine tool.1415
In recent years, much work has been performed using various statistical and experimental techniques. These studies were mostly based on the design and analysis of experimental methods to determine the effects of the cutting parameters on the diameter error, the roundness and the surface roughness of cylindrical parts. Islam5 investigated the effects of the cooling method, the blank size and the work material on the dimensional accuracy and surface finish of various turning parts using the traditional analysis, Pareto ANOVA and the Taguchi method. The results showed that the work material had the greatest effect on the diameter error and surface roughness, while the major contributor to the circularity was the blank size. Rafai and Islam6 experimentally and analytically investigated the effects of the cutting parameters on the dimensional accuracy and surface finish in dry turning. The Taguchi method and the Pareto ANOVA analysis were used to determine the effects of the major controllable machining parameters and, subsequently, to find their optimum combination. They reported that while the surface roughness could be optimized through a proper selection of the feed rate, optimization of the diameter error and circularity was difficult due to the complex interactions between the input parameters. Brinksmeier et al.8 analyzed the basic mechanisms causing a ring distortion in soft machining in order to derive the strategies for its minimization. They concluded that distortion was significantly influenced by the turning before the heat treatment in the manufacture of bearing rings and that the elastic ring deformation under the clamping force led to variations in the depth of cut and polygonal form. Brinksmeier and Sölter10 developed a new method to predict the shape deviation of machined
workpieces with a complex geometry. This new method combined the experimental results for the machining workpieces with simple geometries using finite-element simulations. This was achieved by making use of the known source stresses in simple parts for which the approach was validated. Finally, the method was applied to predict the shape deviation of a ground linear rail guide. Zhou et al.12 analyzed the possible error-driver factors and error sources in the precision hard turning and a strategy was proposed for an on-line compensation of dimensional errors, based on the tool-wear monitoring and a thermal-expansion prediction. They found that within a certain range of flank wear, the developed method could significantly improve the geometric accuracy. Sölter et al.13 analyzed the strategies for reducing the roundness deviations of turned rings using a three-jaw chuck. The simulations and measurements agreed reasonably well and showed that the minimum out-of-round-ness strongly depends on the interaction of the angular shift between the external and internal clamping with the deviation between the segment-jaw diameter and the inner-ring diameter.
The purpose of this study was to obtain the optimum cutting conditions (the cutting speed, the feed rate, the number of the machined part) for minimizing the surface roughness, diameter error and roundness when hard turning an inner-bearing ring bore. An L9 orthogonal array was used in the design of the experiment. The Taguchi method and an analysis of variance (ANOVA) were also used to achieve this purpose. Furthermore, ANOVA was used to determine the statistical significance of the cutting conditions. In addition, the optimum cutting conditions were aimed to be further confirmed by confirmation tests.
2 MATERIAL AND METHOD 2.1 Material
The hard turning tests were performed on the inner bearing ring bores of AISI 52100 steel, the composition of which is given in Table 1. The geometry of the inner-bearing ring is given in Figure 1. The rings were manufactured from soft AISI 52100 through forging, spheroidising, cold ring rolling, soft machining, hardening and, finally, tempering processes to their near-net shapes. The rings were hardened and tempered to 58-62 HRc.
Table 1: Chemical composition of AISI 52100 steel (w/%) Tabela 1: Kemijska sestava jekla AISI 52100 (w/%)
C	Si	Mn	P	Ni	Cr	Mo	Cu	Al	Fe
0.99	0.24	0.36	0.016	0.06	1.43	0.02	0.10	0.017	Balance
2.2 Cutting inserts and the tool holder
The cutting tools used were commercial-grade TiN-coated low-content CBN inserts with the geometry of DCGW 11T304 produced by Sandvik Coromant.
Figure 1: Geometry of the inner ring used for hard-turning tests Slika 1: Geometrija notranjega obro~a, uporabljenega pri preizkusih trdega struženja
These inserts were recommended for machining hardened steel by Sandvik and had the 7015 Sandvik designation. The inserts were rigidly mounted on a right-hand-style tool holder designated by ISO as A25T SDUCR 11 for hard turning inner ring bores.
2.3	Cutting conditions and measurements
The hard-turning tests were carried out on a twin-spindle CNC chucker equipped with a pair of highly precise collet chucks. Three cutting speeds (120, 140 and 160) m/min and three feed rates (0.04, 0.06 and 0.08) mm/r with a fixed depth of cut of 50 pm were tested. At each condition, a total of 12 rings were hard turned. Surface-roughness and roundness measurements were carried out on the hard-turned bore surfaces using a Taylor Hobson precision instrument. The inner diameter of an inner ring bore was measured with Diatest UD-1. The surface-roughness, roundness and diameter-error measurements were carried out on the 4"^, 8'h and 12"^ hard-turned rings and the readings were averaged.
2.4	Design and analysis of the experiments
Optimization of machining processes is essential for achieving a high responsiveness to production, which provides a preliminary basis for the survival in today's dynamic market conditions. The Taguchi method is widely used in engineering analyses and it is a powerful
design tool. This method dramatically reduces the number of tests using orthogonal arrays and minimizes the effects of the factors that cannot be controlled. Furthermore, it provides a simple, efficient and systematic approach to specifying the optimum cutting parameters for a manufacturing process.16,17
The Taguchi method converts objective function values to the signal-to-noise (S/N) ratio to measure the performance characteristics at the levels of the control factors against these factors. The S/N ratio is defined as the desired signal ratio for an undesired random-noise value and shows the quality characteristics of the experimental data.1617 Usually, there are three categories of the performance characteristic in an analysis of the S/N ratio, i.e., the lower-the-better, the higher-the-better and the nominal-the-better characteristics. The goal of this study was to minimize the surface roughness, the innerdiameter error and the roundness. Therefore, the lower-the-better quality characteristic was used as shown in Equation (1):
S
- - - (1)
r 1 ^ 21
where Yi is the observed data during the experiment and n is the number of experiments.16,17
In this study, the cutting speed (V), the feed rate (f) and the number of the machined part (MP) were selected as the control factors and their levels were determined as shown in Table 2. The L9 orthogonal array of the Ta-guchi method allowed us to determine the optimum cutting conditions and analyze the effects of the machining parameters.
Table 2: Control factors and their levels
Tabela 2: Kontrolni faktorji in njihovi nivoji
Control factors	Level 1	Level 2	Level 3
Cutting speed, V/(m/min)	100	120	140
Feed rate, f/(mm/r)	0.04	0.06	0.08
Machined part number, MP	4	8	12
ANOVA was used to find the relative contribution of the cutting conditions in controlling the response of a turning operation. The optimum combination of the
Table 3: Experimental results and corresponding S/N ratios Tabela 3: Rezultati preizkusov in odgovarjajo~a razmerja S/N
Run	Control factors			Experimental results			Signal-to-noise ratio (S/N)		
	V	f	MP	IDe	Ra	R	R	Ra	IDe
1	120	0.04	4	0.016	0.164	3.33	-10.448	15.703	35.917
2	120	0.06	8	0.012	0.293	3.04	-9.657	10.662	38.416
3	120	0.08	12	0.011	0.459	3.21	-10.130	6.763	39.172
4	140	0.04	12	0.010	0.176	2.63	-8.389	15.089	40.000
5	140	0.06	4	0.008	0.178	3.02	-9.600	14.991	41.938
6	140	0.08	8	0.005	0.363	2.35	-7.421	8.801	46.020
7	160	0.04	8	0.010	0.271	4.23	-12.526	11.340	40.000
8	160	0.06	12	0.013	0.342	4.6	-13.255	9.319	37.721
9	160	0.08	4	0.007	0.363	4.51	-13.083	8.801	43.098
control factors for the surface roughness, diameter error and roundness was determined with the ANOVA table and S/N ratios. Lastly, confirmation tests were done using the optimum cutting conditions found with the Taguchi optimization method and thereby the validity of the optimization was tested.
3 RESULTS AND DISCUSSION
The surface roughness, the inner-diameter error and the roundness achieved during the hard turning of inner-bearing rings were measured after the experiments performed according to the L9 orthogonal array. The lowest values of the surface roughness, the inner-diameter error and the roundness significantly improve the quality of the bearing assembly. For this reason, the lower-the-better quality characteristic was used for the calculation of the S/N ratio. The experimental results and S/N ratios are given in Table 3.
3.1 Evaluation of the surface roughness for an inner
ring bore
The interaction plots for the surface-roughness parameters are given in Figure 2. The minimum value of the surface roughness (Äa) was obtained for V =120 m/min, f = 0.04 mm/r and the 4th machined part. It can be seen from Figure 2 that the Ra value increases as the feed rate is increased from 0.04 mm/r to 0.08 mm/r when hard turning the inner-bearing ring bores. With the raise in the feed-rate value from 0.04 mm/r to 0.08 mm/r, a significant increase is observed in the Ra value. No significant relation between the surface roughness and the number of the machined ring bore is seen.
The S/N ratios of the Ra data obtained from the experimental results, used to determine the optimum level of each variable, were calculated according to Equation (1). Figure 3 illustrates the plots of the S/N ratios that were calculated for Ra in the hard turning of the inner ring bores. The S/N ratios of the factors for each level are shown in Table 4. Different values (A) of the S/N ratio between the maximum and the minimum
/	/ >♦' s // \
f	♦-----
OJ
MP
1413-
s
'S 12-
Z te
'S IH i
V	f	MP
A	\	\
' -	\	V----
120 140 160 0,m 0,06 0,08 4	8	12
Signal-to-noise: Smaiter is better
Figure 3: Main-effect plot for surface roughness Slika 3: Diagram glavnega u~inka na hrapavost povr{ine
are also shown in Table 4. Therefore, by considering the S/N ratios in Table 4 and Figure 3, the optimum cutting conditions for the surface roughness were V2 (V = 140 m/min), f1 (f = 0.04 mm/r) and MP1 (the 4th machined part). The smallest surface roughness and its S/N ratio that can be obtained with these levels were calculated using Equations (2) and (3). The Ra value and its S/N ratio were determined as 0.131 ^m and 17.621 dB, respectively:
" " (2)
v g = vg + (v0 - vg) + (f0 - vg) + (mp0 - ^g)
R = 10
"20
(3)
In these equations, Vg is the S/N ratio calculated at the optimum levels (dB)^ v^is the average S/N ratio of all the variables (dB), V), f0, MP0 are the mean S/N ratios when factors V, f and MP are at the optimum levels (dB), and Ra cal is the calculated surface-roughness (Ra) value.
Table 4: Response table of S/N ratios for surface roughness Tabela 4: Tabela odzivov razmerja S/N glede na hrapavost povr{ine
Factor	S/N ratio				Rank
	Level 1	Level 2	Level 3	A	
V	11.043	12.961	9.821	3.14	2
f	14.044	11.658	8.122	5.922	1
MP	13.166	10.268	10.391	2.897	3
Figure 2: Interaction plots for surface roughness Slika 2: Interakcijski diagrami za hrapavost povr{ine
The analysis of variance (ANOVA) was used to find which design parameters significantly affect the surface roughness. This analysis was carried out at the 95 % confidence level. The ANOVA results for the surface roughness (Ra) are shown in Table 5. This table also shows the degree of freedom (DF), the sum of squares (SS), the mean square (MS), the F-values (F), the probability (P) and the percentage-contribution ratio (PCR) of each factor. A low P-value of 0.05 shows a statistically significant level of the source for the corresponding response. The F-ratios and their PCR were taken into consideration to identify the significance levels of the variables. Table 5 indicates that the most effective variable for the Ra
value is the feed rate with the PCR of 62.84 %. It is well known that the theoretical geometrical-surface roughness
is primarily a function of the feed for a given nose radius
and that it changes with the square of the feed-rate value. It is once again shown that the feed rate has an important effect on the surface roughness in hard turning.The other variables having an effect on Ra are the number of the machined part with a PCR of 19.00 % and the cutting speed with a PCR of 17.74%.
Table 5: ANOVA for S/N ratios for surface roughness Tabela 5: ANOVA za S/N-razmerje glede na hrapavost površine
Source
MP
Error
Total
DF
SS
15.0351
53.265
16.1066
0.3544
84.7612
MS
7.5176
26.6325
8.0533
0.1772
F
42.42
150.28
45.44
0.023
0.007
0.022
PCR
17.74
62.84
19.00
0.42
100
3.2 Evaluation of the roundness for an inner ring bore
The interaction plots for the roundness and the cutting conditions are given in Figure 4. The minimum value of the roundness was obtained at V = 140 m/min and f = 0.08 mm/r for the 8th machined part. The roundness is generally affected by the cutting parameters such as the cutting speed, the feed rate and the depth of cut. In Figure 4 the roundness decreases for all the feed rates with the cutting speed increasing from 120 m/min to 140 m/min. This can be attributed to the cutting force decrease with the increasing cutting speed. This decrease is explained with a thermal softening of the machined material in the flow zone due to the increasing cutting temperature. The highest roundness values for all the feed rates were obtained at a cutting speed of 160 m/min. Although a moderate increase in the cutting speed decreases the cutting force, a further increase in the cutting speed was thought to increase the tool wear. This, in turn, increases the roundness. In addition to the cutting parameters, the roundness is also affected by other factors such as the clamping system and the machinetool rigidity.
The experimental results and the calculated S/N ratios for the roundness are given Table 3. The S/N ratios of the factors for each level are shown in Table 6. The values in Table 6 are given as the plots from Figure 5. It is seen from Figure 5 and Table 6 that the control factors can be used to reach the smallest roundness in the machining of inner ring bores. The smallest roundness is obtained in the following cutting conditions: V2, f3 and MP2. The optimum cutting conditions for the roundness were V2 (V = 140 m/min), f3 (f = 0.08 mm/r) and MP2 (the 8th machined part). The S/N ratios for the roundness of Rcal were calculated with Equation (3). Consequently, ^G and Rcal calculated for the optimum cutting conditions were found to be -7.5487 dB and 2.38 ^m, respectively.
Table 6: Response table of S/N ratios for roundness Tabela 6: Tabela odgovorov S/N-razmerje glede na okroglost
Symbol
MP
S/N ratio
Level 1
-10.079
-10.458
-11.044
Level 2
-8.474
-10.838
-9.869
Level 3
-12.955
-10.212
-10.595
4.482
0.626
1.176
Rank
Table 7: ANOVA for S/N ratios for roundness Tabela 7: ANOVA za S/N-razmerje glede na okroglost
Source
V
MP
Error
Total
DF
8
SS
30.9353
0.5965
2.1115
0.2809
33.9242
MS
15.4676
0.2982
1.0558
0.1405
F
110.12
2.12
7.52
P
0.009
0.320
0.117
PCR
91.18
1.76
6.23
0.83
100
The ANOVA results for the inner-ring-bore roundness are given in Table 7. The F-ratios and their PCRs were taken into consideration to identify the significance level of the variables. Table 7 shows that the most effective variable for the roundness is the cutting speed with a PCR of 91.18 %. The contributions of the number of the machined part and the feed rate were found to be 6.23 % and 1.76 %, respectively. Consequently, the feed rate and the number of the machined part do not have significant
Figure 4: Interaction plots for roundness
Slika 4: Diagrami medsebojnih vplivov na okroglost
Figure 5: Main-effect plots for roundness Slika 5: Diagrami glavnih vplivov na okroglost
P
2
2
2
2
8
V


2
2
2
2
	
	
Figure 6: Interaction plots for inner-diameter error
Slika 6: Interakcijski diagrami za napako notranjega premera
effects on the roundness (P < 0.05). The error ratio was calculated as 0.83 % and it is the smallest ratio.
3.3 Evaluation of the inner-ring-bore-diameter error
The interaction plots for the inner-diameter error and the cutting conditions are given in Figure 6. The minimum value of the inner-diameter error was obtained at V = 140 m/min and f = 0.08 mm/r for the 8th machined part. The cutting speed might have influenced the resulting cutting forces and this interaction might have consequently influenced the diameter error in a number of ways, i.e., by changing the elastic deformation of the workpiece, by altering the tool wear, by increasing the thermal distortion, by forming a built-up edge (BUE) and by increasing the radial-spindle error. In this case, the most likely cause for the change in the diameter error is considered to be the change in the elastic deformation of the workpiece.
The experimental results and the calculated S/N ratios for the roundness are given Table 3. Figure 7 shows the plots of the S/N ratios that were calculated for the inner-diameter error. The S/N ratios of the factors for each level are shown in Table 8. Hence, by considering the S/N ratios from Table 8 and Figure 7, the optimum cutting conditions for the inner-diameter error were V2 (V = 140 m/min), f3 (f = 0.08 mm/r) and MP2 (the 8th
Figure 7: Main-effect plots for inner-diameter error
Slika 7: Diagrami glavnih učinkov na napako notranjega premera
machined part). The S/N ratios for the roundness of IDe (the inner diameter error) were calculated with Equation (3). Consequently, ^g and IDccal calculated for the optimum cutting conditions were found to be 46.388 dB and 0.0048 ^m, respectively.
Table 8: Response table of S/N ratios for inner-diameter error Tabela 8: Tabela odgovorov razmerja S/N glede na napako notranjega premera
Symbol
MP
Mean S/N ratio
Level 1
37.84
38.64
40.32
Level 2
42.65
39.36
41.48
Level 3
40.27
42.76
38.96
4.82
4.12
2.51
Rank
3
Table 9: ANOVA for S/N ratios for inner-diameter error
Tabela 9: ANOVA za S/N-razmerje glede na napako notranjega pre-
Source
V
MP
Error
Total
DF
8
SS
34.815
29.1223
9.5031
0.6344
74.0748
MS
17.4075
14.5612
4.7515
0.3172
F
54.87
45.9
14.98
0.018
0.021
0.063
PCR
46.99
39.32
12.83
0.86
100
The ANOVA results for the inner-diameter error are shown in Table 9. The F-ratios from Table 9 and their PCR were taken into consideration to identify the significance levels of the variables. Table 9 shows that the most effective variable for the inner-diameter error is the cutting speed with a PCR of 46.99 %. The feed rate and the number of the machined part also affect the innerdiameter error with their PCR being 39.32 and 12.83 %, respectively. Therefore, the number of the machined part does not have a significant effect on the inner-diameter error (P < 0.05). As the P-value of the number of the machined part is less than 0.05, it is not significant.
3.4 Confirmation tests
Confirmation tests of the control factors were made for the Taguchi method. The optimization was verified with the confirmation tests after the determination of the control factors that give the optimum results. The confirmation tests were conducted at the optimum factor levels for each surface roughness, roundness and diameter error. The confirmation tests performed at the optimum variable levels determined for the surface roughness, the roundness and the inner-diameter error were evaluated by taking into consideration the confidence interval (CI) calculated with Equations (4) and (5)16,17:
CI ^ K^. (1, V. )y
^ 1 1
-+-
n eff r
N
n .. =
1 + V
(4)
(5)
V

r
mera
P
2
2
2
2
where Ve is the error degree of freedom, Ve is the error variance, neff is the repeating number of the experiments, N is the total number of the experiments, vT is the variable's degree of freedom and r is the number of the confirmation tests.
Table 10 gives a comparison of the results of the confirmation tests conducted according to the optimum levels of the variables and the values calculated using Equations (2) and (3). Besides, confidence-interval values for Ra, R and IDe were calculated using Equations (4) and (5).
Table 10: Comparison lated values Tabela 10: Primerjava ~unanimi vrednostmi
between confirmatory test results and calcu-med rezultati potrditvenih preizkusov in izra-
Confirmatory experiment results		Calculated values		Differences	
Ra mea/pm	mea/dB	Ra cal/pm	^cal/dB	Ra mea -Ra cal	^mea — ^cal
0.151	16.42	0.131	17.621	0.02	1.201
Rmea/pm	^mea/dB	Rcal/pm	^ca/dB	Rmea -Rcal	^mea — ^cal
2.35	-7.4214	2.38	-7.5487	0.03	0.1273
IDemea/pm	^mea/dB	IDecal/pm	^ca/dB	IDemea -IDecal	^mea — ^cal
0.005	46.020	0.0048	46.388	0.0002	0.368
The optimum conditions for the surface roughness were observed at V2-f1-MP1 (i.e., cutting speed V = 140 m/min, feed rate f = 0.04 mm/r and the 4'h machined part). The optimum conditions for the roundness and the inner-diameter error were observed at the same levels, V2-/3-MP2 (i.e., cutting speed V = 140 m/min, feed rate f = 0.08 mm/r and the 8'h machined part).
Confirmation tests were carried out at the optimum conditions. According to the confirmation-test results, the measured values were found to be within the 95 % confidence interval.
The outcomes for the roundness and the inner-diameter errors are very close to the predicted values. Thus, there is no need to carry out confirmation tests if the cutting conditions found with the optimization procedure are included in the cutting conditions within the Taguchi experimental design.
Acknowledgement
The authors would like to thank the Ministry of Science, Industry and Technology, Turkey (00980.STZ. 2011-2) and ORS Bearings, Turkey, for the financial support of this study.
Using Equation (5), confidence values of (2.56, 2.28 and 3.42) dB were obtained for the surface roughness (Ra), the roundness (R) and the inner-diameter error (IDe), respectively. Table 10 shows differences between the values obtained with the confirmatory tests and the values of the S/N ratios calculated with Equations (2) and (3). It is seen that a difference of 1.201 dB is under the 5 % confidence interval for the surface roughness of 2.56 dB, a difference of 0.1273 dB is under the 5 % confidence interval of for the roundness of 2.28 dB and, similarly, a difference of 0.368 dB is under the 5 % confidence interval for the inner-diameter error of 3.42 dB. Therefore, the optimum-control-factor settings for all the cutting conditions were confirmed as confident.
4 CONCLUSIONS
In this study, the cutting conditions for the surface roughness, the inner-diameter error and the roundness during the hard turning of an inner ring bore were optimized with the Taguchi method. The results obtained from this study are presented below:
According to the results of the statistical analysis, it was found that the feed rate was the most significant factor for the surface roughness with a PCR of 62.84 %. The cutting speed was the most significant parameter for the roundness and the inner-diameter error with PRC of 91.18 and 46.99 %, respectively.
The optimum levels of the control factors for minimizing the surface roughness, the roundness and the inner-diameter error using S/N ratios were determined.
5 REFERENCES
1 H. K. Tönshoff, C. Arendt, R. B. Amor, Cutting of hardened steel, Annals of the CIRP, 49 (2000) 2, 547-566, doi:10.1016/S0007-8506(07)63455-6
2G. Byrne, D. Dornfeld, B. Denkena, Advancing cutting technology, Annals of the CIRP, 52 (2003) 2, 483-507, doi:10.1016/S0007-c8506(07)60200-5
3	W. König, A. Berktold, K. F. Koch, Turning versus grinding - a comparison of surface integrity aspects and attainable accuracies, Annals of the CIRP, 42 (1993) 1, 39-43, doi:10.1016/S0007-8506(07) 62387-7
4	F. Klocke, E. Brinksmeier, K. Weinert, Capability profile of hard cutting and grinding processes, Annals of the CIRP, 54 (2005) 2, 22-45, doi:10.1016/S0007-8506(07)60018-3
5	M. N. Islam, Effect of additional factors on dimensional accuracy and surface finish of turned parts, Machining Science and Technology: An International Journal, 17 (2013) 1, 145-162, doi:10.1080/ 10910344.2012.747936
6	N. H. Rafai, M. N. Islam, An investigation into dimensional accuracy and surface finish achievable in dry turning, Machining Science and Technology: An International Journal, 13 (2009) 4, 571-589, doi:10.1080/10910340903451456
7	J. A. Malluck, S. N. Melkote, Modeling of deformation of ring shaped workpieces due to chucking and cutting forces, Journal of Manufacturing Science and Engineering, 126 (2004), 141-147, doi:10.1115/1.1643079
8	E. Brinksmeier, J. Sölter, C. Grote, Distortion engineering - identification of causes for dimensional and form deviations of bearing rings, CIRP Annals - Manufacturing Technology, 56 (2007) 1, 109-112, doi:10.1016/j.cirp.2007.05.028
9	L. Björn, T. Beekhuis, E. Brinksmeier, M. Garbrecht, J. Sölter, Improving the shape quality of bearing rings in soft turning by using a Fast Tool Servo, Prod. Eng. Res. Devel., 3 (2009), 469-474, doi:10.1007/s11740-009-0175-z
10	E. Brinksmeier, J. Sölter, Prediction of shape deviations in machining, CIRP Annals - Manufacturing Technology, 58 (2009) 1, 507-510, doi:10.1016/j.cirp.2009.03.123
11	D. Stöbener, B. Beekhuis, Application of an in situ measuring system for the compensation of wall thickness variations during turning of thin-walled rings, CIRP Annals - Manufacturing Technology, 62 (2013) 1, 511-514, doi:10.1016/j.cirp.2013.03.129
12	J. M. Zhou, M. Andersson, J. E. Stahl, Identification of cutting errors in precision hard turning process, Journal of Materials Processing Technology, 153-154 (2004), 746-750, doi:10.1016/j.jmatprotec. 2004.04.331
13	J. Sölter, C. Grote, E. Brinksmeier, Influence of clamping strategies on roundness deviations of turned rings, Machining Science and Technology, 15 (2011), 338-355, doi:10.1080/10910344.2011. 601207
14	K. Bouacha, M. A. Yallese, T. Mabrouki, J. F. Rigal, Statistical analysis of surface roughness and cutting forces using response surface methodology in hard turning of AISI 52100 bearing steel with CBN tool, Int. J. Refract. Met. Hard., 28 (2010) 3, 349-361, doi:10.1016/ j.ijrmhm.2009.11.011
15	P. G. Benardos, G. C. Vosniakos, Predicting surface roughness in machining: A review, International Journal of Machine Tools & Manufacture, 43 (2003) 8, 833-844, doi:10.1016/S0890-6955(03) 00059-2
16	M. Gunay, E. Yucel, Application of Taguchi method for determining optimum surface roughness in turning of high-alloy white cast iron, Measurement, 46 (2013), 913-919, doi:10.1016/j.measurement.2012. 10.013
17	E. Yucel, M. Gunay, Modelling and optimization of the cutting conditions in hard turning of high-alloy white cast iron (Ni-Hard), Proc. IMechE, Part C: J. Mechanical Engineering Science, 227 (2013) 10, 2280-2290, doi:10.1177/0954406212471755