Optimization of surface roughness and tool wear in hard turning of austempered ductile iron (grade 3) using Taguchi method

Accepted Manuscript Optimization of surface roughness and tool wear in hard turning of austempered ductile iron (grade 3) using Taguchi method D. Manivel, R. Gandhinathan PII: DOI: Reference:

S0263-2241(16)30347-5 http://dx.doi.org/10.1016/j.measurement.2016.06.055 MEASUR 4175

To appear in:

Measurement

Received Date: Revised Date: Accepted Date:

19 February 2016 9 June 2016 23 June 2016

Please cite this article as: D. Manivel, R. Gandhinathan, Optimization of surface roughness and tool wear in hard turning of austempered ductile iron (grade 3) using Taguchi method, Measurement (2016), doi: http://dx.doi.org/ 10.1016/j.measurement.2016.06.055

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Optimization of surface roughness and tool wear in hard turning of austempered ductile iron (grade 3) using Taguchi method D.Manivel1 *, R.Gandhinathan2 1

Research scholar, Department of Production Engineering, PSG College of Technology, Coimbatore-641004, India 2 Professor, Department of Production Engineering, PSG College of Technology, Coimbatore-641004, India (* Corresponding author. Tel: +91 9842365519, E-mail address: [email protected])

Abstract In this work, the cutting parameters are optimized in hard turning of ADI using carbide inserts based on Taguchi Method. The cutting insert CVD coated with AL2O3/ MT TICN. Experiments have been carried out in dry condition using L18 orthogonal array. The cutting parameters selected for machining are cutting speed, feed rate and depth of cut with each three levels, nose radius in two levels maintaining other cutting parameters constant. The ANOVA and signal to noise ratio are used to optimize the cutting parameters. The cutting speed is the most dominant factor affecting the surface roughness and tool wear. In optimum cutting condition, the confirmation tests are carried out. The optimum cutting condition results are predicted using signal to noise ratio and regression analysis. The predicted and experimental values for surface roughness and tool wear adhere closer to 9.27 % and 1.05 % of deviations respectively. Keywords: Austempered ductile iron; Surface roughness; Taguchi; Tool wear. 1. Introduction and literature review Hard Turing is a turning of hard material with a hardness range from 45 to 68 HRC. Generally, coated carbide, cubic boron nitride (CBN), Ceramic and polycrystalline cubic boron nitride (PCBN) inserts are used to turn the hard materials in CNC lathe. The machining of hardened materials using CBN, PCBN and ceramic tools are generally used and it is a good alternate to expensive grinding operations [1]. Hard turning has many advantages other than the cost advantage such as faster metal removal rate, reduced cycle time, good surface finish and environmental free [2]. ADI materials have been used in many engineering applications because of their high strength, high hardness, ductility and toughness. These materials have been widely used for many applications such as automotive, agricultural, railroad, construction and mining industries due to their excellent mechanical properties, such as high strength to weight ratio, high wear resistance and inexpensive material compared to other materials [3]. Austempered ductile iron (ADI) is a comparatively latest material for industrial applications. ADI is a heat treated form of as cast Ductile Iron. The Austempering process consists of three stages they are austenitizing, isothermal quenching and cooling to room temperature. Austempering results, the microstructure consists of ferrite in high carbon austenite matrix [4]. ADI is difficult to machine compared to ductile cast irons in the austempered condition, because of relatively high hardness and high strength. In machining, the material is strained hardening takes place due to the presence of retained austenite. The strain hardening of retained austenite increases mechanical loads and reduces the contact length on the cutting insert tool’s edge [5]. The higher cutting tool wear was observed in machining of ADI in austempered condition, when compared to other hardened material. The higher tool wear occurs at the cutting tool’s edge due to high temperature, adhesion resting on the cutting tool and higher ductility of ADI [6]. In turning of ADI, the cutting tool’s edge subjected thermal softening due to higher cutting temperatures and low thermal diffusivity of ADI [7].The aim of the new machining industries is to produce components at low product cost with good quality in minimum time. To achieve a good cutting performance in turning, selection of optimum cutting parameters is important. Machinability of hardened materials was evaluated by cutting force surface roughness and tool wear. Turning the hard material to get a minimum surface roughness with minimum tool wear is difficult. The following literature survey indicates that most of the hard material is turned by CBN, PCBN and ceramic inserts. Katuku et al. [8] conducted experimental work in dry cutting condition on austempered ductile iron (ASTM Grade 2). The cutting forces, chip characteristics and tool wear were analyzed with PcBN cutting tools. The result revealed that the optimum cutting speed for better tool life and flank tool wear is 150 to 500 m/min. In another work Marcelo Vasconcelos de Carvalho et al. [4] investigated machinability of ADI (ASTM grades 2 and 3). It has been reported that, minimum surface roughness and higher tool wear observed when turning ADI grade3 with higher tool nose radius. In another work, Tug˘rul O¨ zel and, Yig˘it Karpat [9] developed the prediction model using regression and neural networks in hard turning for surface roughness and tool wear by

CBN inserts. Minimum surface roughness was obtained at high work piece hardness with high cutting speed. Higher tool wear were obtained with higher cutting speed at lower feed rate. Lower in feed rate gives good surface finish. Zahia Hessainia et al. conducted experimental work on hard turning. The surface roughness was predicted with the use of cutting parameters and tool vibrations. The mixed ceramic cutting tool Al2O3/TiC was used. They found that feed rate was the most dominating factor than the tool vibration in affecting the surface roughness. [10]. Mustafa Gunay and Emre Yuce applied Taguchi for cutting conditions optimizing for surface roughness in turning of white cast iron (high alloy). The hardness of the material was 50 HRC and 62 HRC. The turning was carried out by ceramic and cubic boron nitride. The Taguchi orthogonal array L18 was used. Optimum cutting conditions were calculated using the signal-to-noise (S/N) ratio based on ‘the-smaller-the-better’ approach. It was noted that cutting speed and feed rate was the dominant factor disturbing the surface roughness [11]. Sanjeev Saini et al. [12] studied the Influence of cutting parameters in hard turning. The AISI H11 tool steel material and the ceramic insert were selected for hard turning. The effect of cutting parameters was analyzed by response surface methodology on tool wear and surface roughness. They found that the depth of cut has no effect on tool wear and surface roughness. Insert nose radius and cutting speed have the maximum effect on surface roughness. In higher cutting speed the tool nose radius subjected to maximum pressure and temperature results in increased tool wear. Abhijeet S et al. [13] have been conducted the experiments and compared the performance of CBN–TiN coated and PCBN inserts on hardened AISI 4340 steel about HRC = 53. The performance of cutting tools was compared in terms of tool wear, surface roughness, and cutting forces. The result indicated that PCBN insert has better tool life, lower cutting force, and minimum surface roughness than the cBN–TiN coated carbide inserts at given cutting conditions. But the cBN–TiN coated carbide tool has less cost than the PCBN; therefore, it reduces the machining cost based on a single cutting edge. The surface quality of the machined part is significant. The surface quality depends on the surface finish. The good surface increases fatigue strength, wear resistance and minimizes the corrosion attack of the finished parts. The surface finish depends on various parameters like material hardness, type of inserts, type of coating, heat transfer and cutting parameters. Therefore the optimization of surface roughness and tool wear is important. The optimized parameter gives a good surface finish in the first machining itself and also increases the production rate and decreases the production cost. Some of the following literature survey about optimization. Mandal et al. [14] investigated that optimization of cutting parameter for tool flank wear using newly developed cutting tool Zirconia Toughened Alumina (ZTA). Taguchi method and regression analysis were used to optimize the cutting parameters. It has been observed that the tool wear was highly affected by the depth of cut. Ilhan Asiltürk and Süleyman Neseli [15] conducted the experiment on austenitic stainless steel. The cutting parameters were multi optimized by Taguchi method and response surface analysis. It found that the feed rate has the most effect on surface roughness. It was most leading parameters affecting the surface roughness comparability to other parameters. In another work D. Philip Selvaraj et al. [16] applied the Taguchi method for dry turning of nitrogen alloyed duplex stainless steel by coated carbide cutting inserts. The cutting parameters were optimized for surface roughness, tool wear and cutting force. The results revealed that the surface roughness and cutting force was mostly dominated by feed rate where as tool wear was mostly dominated by the cutting speed. In another work, Palanikumar [17] optimized the drilling parameters for glass fiber-reinforced plastics, composites by Taguchi method with grey relational study. The result indicated that feed rate was the most dominant parameter than the spindle speed. Muhammad munawar et al. [18] studied the effect of cutting parameters in internal turning. Taguchi method was applied. Turning was carried out using L18 orthogonal arry. It was noted that low cutting speed high feed rate developed minimum surface roughness. The cost of the CBN, PCBN and ceramic inserts are very high compared to coated carbide inserts. Turning of hard material by coated carbide insert reduces the production cost. Some of the researcher used carbide inserts for turning of hard material as follows, Ucun and Aslantas [19] studied the effects cutting speed, depth of cut and feed rate on surface roughness in hard turning. The hard material used for experimental was bearing steel with the grade of AISI 52100. The hard material was turned by coated carbide inserts. They observed that, in the hard turning process, carbide cutting tools were not fit especially at high cutting speed. Ilhan Asilturk and Harun Akkus applied the Taguchi method in the hard turning process using coated carbide tools; the experimental result indicated that the feed rate was the dominant factor effect on surface roughness [20]. In another work, AISI D2 steel a hardness of 66 HRC was machined with coated carbide insert decreases in surface roughness with the increase in cutting speed and surface roughness increased with increased feed rate and depth cut [2]. Y. Kevin Chou and Hui Song have investigated the effect of tool nose radius on hardened steel (AISI 52100) in hard turning; the better surface finish was obtained in larger tool nose radius [21]. The literature survey shows that CBN, PCBN, and ceramic tool were widely used for turning of hardened materials, but the hard turning using coated carbide insert is very limited. In this work, the coated carbide tool is used for turning the hard material. It is a cost effective method. In this experimental work, the surface roughness and tool wear only were considered.

In tthis sstud dy, th he e effecct of machin ning parrame eterss such a as nose rad dius, cuttting spe eed, feed ra ate a and d deptth cu ut on the su urfacce rou ughn nesss and d to ool w wearr in the turn ning of A ADI (gra ade 3) w with h CV VD ccoated ccarb bide inse erts wass invvesttigatted. Exp perim men nts were ccond ducte ed u using g Ta agucchi’ss L188 orth hogo onal arra ay. Ana alysiss of varriancce (A ANO OVA A) an nd S S/N ratio o we ere u used d to ana alysiis, nditio ons (No ose rradiu us, ccuttiing ttool,, cuttting g spe eed and d fee ed rrate)) for surrface e rou ughn nesss an nd to ool w wear. An nalyysis of cuttting con varriancce (A ANO OVA A) wa as u used d to ffind outt the e sig gnificcant facttors. S/N N ra atio w wass use ed to o ca alcullate the opttimum ccuttin ng fa acto ors and d the eir le evelss. In a addittion, the me easured valu ues w were e prrediccted by u usin ng re egressio on an nalyysis. At tthe e end, the e con nfirm matio on exp perim ments w were use ed to o che eck the relia abilitty off the e devvelo oped d mo odel’s re eliability. Expe erim menttal D Deta ails 2. E T The exp perim menttal w workk dettailss, me easu urem ments, ccalcu ulatiionss and d prroce edure e for thiis exxperrime entall stu udy are disccusssed in follo owin ng chap pterss. ateria al 2.1 Ma The e wo ork m mate erial use ed fo or exxperrime ental work w was a austtemp pere ed ductille iro on A ASTM grade e 3. The e ma aterial iss aussten nitize ed o 880 C for 2 hrrs. A Afterr austen nitizing tthe sspeccime ens werre quenched d in saltt ba ath o of 50 0 % of ssodium nitra ate ((NaN NO3) an nd at 8 o 50% % po otasssium m niitrate e (K KNO3) att 230 0 C forr 2 h hrs. The e Te ensile e Sttreng gth 1241 M MPa, Yie eld S Stren ngth 893 3 MP Pa, Harrdne ess 4 45 HR RC a and E Elon ngatiion 4 4.88 8 % werre ob btain ned afte er the e au ustem mpe ered d hea at tre eatm mentt. Th he ssize o of th he w work piecce m mate erialss wa as 35 mm m dia ametter a and 9 90 m mm long g. Th he chem micall com mpo ositio on off AD DI grrade e 3 iss givven iin Ta able e 1. 2 Cu uttin ng C Cond ditio ons 2.2 Th he e expe erime entss we ere ccond ducte ed in ga alaxxy Midas M s6C CNC C latthe und der d dry ccutting ccond ditions, whicch h havin ng the m maxximum spin ndle e pow wer of 7 7.5 K Kw a and the spin ndle spe eed from m 40 0 to 4 4000 rp pm. T The cuttting toolls ussed werre ca arbid de in nserrts C CVD coa ated of T Taeg guTe ec C Company (T TT73 310) with h a specificcation off CN NMG G 120 0408 RT T, a and C CNM MG 120404 4 MT T (wiith a an IS SO d desigual to o K1 10 – K20 0), h havin ng n nose e rad dius 0.8 8 mm m an nd 0.4 m mm rresp pectivelyy we ere u used d. Th he fig.1 shows tthe tturning of nattion equ AD DI exxperime ental se et-up p. E Everry e expe eriment a fressh ccuttin ng edg ge w was utilized d. T The right h hand d sstyle too ol h holder PC CLNR R202 20K K12 R RH d desiigna ated by IISO is u used d to hold d the e inssertss. Th he m mate erial 1mm m de epth h wa as re emovved to a avoid d oxxidization n, de ecarrburiization a and also o to rem move e irre egullaritiies b befo ore testin ng. The e cuttting g parrameterr ran ngess selecte ed a as per ccuttin ng too ol ma anuffactu urer’’s re ecom mme enda ation n, ma achiine ccapa abilitty an nd liitera ature e revview w [2],, [10 0], [2 20]. T The e dessign invo olve es va ariattion of four cutting g pa aram meterrs att diffferen nt le evelss as sho own in T Table e 2. g. 1 T Turn ning of A ADI experim menttal se etup p Fig

Tab ble 1 Chem mical compo osition o of ADI A AST TM g grade e3 E Elem mentt

%C C

% %Si

%M Mn

%C Cu

%C Cr

%P P

%S S

% %Mg g

% %Sn n

W Weig ght %

3.42 2

2.4 47

0.41 15

0 0.45 58

0 0.02 27

0 0.02 29

0 0.00 05

0 0.02 20

0 0.079

2.3 3 Tool W Wea ar meas sure emen nt Th he M Mituttoyo too ol ma akerr’s m micro osco ope ((TM M 505 5 se eriess) wa as u used d to m mea asure e the e tool w wear. It iss ussed tto m meassure e the e wo ork piecce cconto ourss and d exxamine tthe ssurfa ace feattures. T The m mag gnificcatio on off the e insstrum mentt is 3 30 X and d lea ast ccoun nt is 0.0 001m mm.

2.4 Surface roughness measurement The surface roughness was measured by using stylus type surface roughness tester (surfcorder SE 1200 series). According to JIS -1994 standard [18], cut off length and sampling length are 0.8 mm and 4 mm respectively were selected to evaluate the surface texture. The surface roughness were measured at three locations and then, the average surface roughness (Ra) calculated. Table.2 Cutting parameters for Turning and their levels Cutting Symbol parameter Unit Nose r Radius mm Cutting m/ V speed min Feed f rate mm/rev Depth of d cut mm

Level 1

Level 2

Level 3

0.8

0.4

50

100

150

0.04

0.08

0.12

0.2

0.3

0.4

3. Taguchi method and design of experiment The Taguchi design is widely accepted and used in engineering analysis and optimization. This design is developed by Dr.Genichi [20]. It is a power full design. InTaguchi methods, the number of experiments were reduced by orthogonal array and also reduce the effects of uncontrollable factors [20]. The quality of the Taguchi design is ensured in the design phase itself. The Taguchi method is used to reduce no. of trails, decreased experimental time, decreased the production cost, simple and precision are the most advantages in this technique. This also used to determine the significant factors in a minimum time [22]. The Taguchi method used to calculate loss function. Loss function is the difference among the experimental and desired values. Then the loss function is converted in the form of signal–noise (S/N) ratio. Generally, three types of quality S/N ratio characteristics written in equation (1-3), when the characteristic is continuous [20] • “Nominal is the best” approach, S/N = 10 log y̅ / S2y •

(1)

“Smaller is the better” approach, S/N = -10 log 1/n (Σ y2)

•

(2)

“Larger is the better” approach, S/N = -10 log1/n (Σ 1/y2)

(3)

This indicates that engineering systems perform in such method. The manipulated production factors are divided into three categories: where, y̅ is the observed data average, S2y the variance of y, n is the observation number and y the observed data of each characteristics. The S/N ratio calculated for each level of parameters by S/N ratio analysis. The objective of this study is to achieve minimum surface roughness and minimum tool wear. Therefore the “smaller is the better” characteristic was chosen and shown in Eq. (2). The orthogonal array L18 ((2**1 3**7) was selected to analysis and optimize cutting parameters [11 &23]. The experiments were conducted using L18 mixed orthogonal array and it is shown in Table 3.

Table.3 L18 orthogonal array S.N

1 2 3 4 5 6 7 8 9

Nose Radius (mm) 1 1 1 1 1 1 1 1 1

Cutting speed (m/min) 1 1 1 2 2 2 3 3 3

Feed Rate (mm/rev) 1 2 3 1 2 3 1 2 3

Depth of cut (mm) 1 2 3 1 2 3 2 3 1

10 11 12 13 14 15 16 17 18

2 2 2 2 2 2 2 2 2

1 1 1 2 2 2 3 3 3

1 2 3 1 2 3 1 2 3

3 1 2 2 3 1 3 1 2

4. Results and discussion In this experimental work, the Taguchi method is used to determine the optimum cutting condition. The orthogonal array (OA) is used to check the quality characteristics with minimum trial [15]. The experiments conducted based on orthogonal array and then the results are transformed to signal to noise ratio (S/N ratio) to perform characteristic analysis. The parameter design is performed to get optimum conditions. The parameter design is also called as robust design. The figure 2 shows the Taguchi method flow chart.

Start Objective

Definition of Problem

Factors & Levels OA

Experimental Design Conduct Experiment ANOVA & S/N ratio

Results analysis

No

Contour Plot

Results Validated? Yes

End 4.1 Analysis of Variance (ANOVA) for surface roughness (Ra) and Tool Wear (Vb) Analysis of variance (ANOVA) can be useful in calculating the influence of given input parameters and also used to interpret experimental data. In this work, ANOVA was used analyze the effect of cutting tool nose radius, cutting speed, feed rate and depth cut on surface roughness and tool wear. This analysis was done with 95 % confidence level and 5 % significance level. The ANOVA table consists of degrees of freedom, mean square, sum of square, F ratio and % of contribution. In ANOVA, the F values of each control factor were compared to determine the significance each control factors. The higher F contribution, the higher the influence a factor has on the result. F ratio is the ratio between mean square and the mean square of the experimental error. Table 4 shows the ANOVA results for surface roughness and tool wear. The percentage contributions of the nose radius 14.6 %, cutting speed 49.1 %, feed rate 21.6 % and depth of cut 9.7 % on the surface roughness is found. The cutting speed (49.1 %) is the most predominant factor affecting the surface roughness. From the ANOVA results, the percentage contribution of the nose radius 5 %, cutting speed 50.2 %, feed rate 30.2 % and depth cut 11.3 % on the tool wear is found. The cutting speed (50.2 %) is the most predominant factor affecting the tool wear. The percent of error for surface roughness 5 % and tool wear 3.3 % is considerably low. 4.2 Analysis of the S/N ratio for surface roughness (Ra) and Tool Wear (Vb)

As per Taguchi technique, the experiments were conducted for every combination of all the control factors. The surface roughness and tool wear measured in off line. Signal to noise ratios was calculated using the condition of “smaller is the better”. Table 5 shows the values of observed S/N ratios for every combination of surface roughness and tool wear. The average surface roughness values were calculated as 0.550889 µm and the average values of tool wear were calculated as 0.120722 mm during the end of the turning tests. Similarly, the S/N ratio average values are calculated for surface roughness is 5.788311 dB and the S/N ratio average values are calculated for tool wear is 18.40208 dB Table.4 Analysis of Variance results for surface roughness and tool wear, Cutting parameter

Degrees of freedom

Sum of square

Nose Radius Cutting speed Feed Depth of cut Error Total

1 2 2 2 10 17

0.11940 0.40080 0.17605 0.07891 0.04053 0.81570

Nose Radius Cutting speed Feed Depth of cut Error Total

1 2 2 2 10 17

0.0001125 0.0011254 0.0006774 0.0002528 0.0000734 0.0022416

Mean F ratio square Surface roughness 0.11940 29.46 0.20040 49.44 0.08803 21.72 0.03946 9.73 0.00405 Tool Wear 0.0001125 15.32 0.0005627 76.62 0.0003387 46.12 0.0001264 17.21 0.0000073 -

P

Contribution (%)

0.000 0.000 0.000 0.005

14.6 49.1 21.6 9.7 5 100

0.003 0.000 0.000 0.001

5 50.2 30.2 11.3 3.3 100

Table 5 .The experimental layout for the L18 orthogonal array and cutting condition

S.N

Nose Radius (mm)

Cutting speed (m/min)

Feed Rate (mm/rev)

Depth of cut (mm)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4

50 50 50 100 100 100 150 150 150 50 50 50 100 100 100 150 150 150

0.04 0.08 0.12 0.04 0.08 0.12 0.04 0.08 0.12 0.04 0.08 0.12 0.04 0.08 0.12 0.04 0.08 0.12

0.2 0.3 0.4 0.2 0.3 0.4 0.3 0.4 0.2 0.4 0.2 0.3 0.3 0.4 0.2 0.4 0.2 0.3 Mean

Surface Roughness (Ra) (µm) 0.274 0.324 0.438 0.364 0.340 0.508 0.803 0.470 0.704 0.338 0.421 0.709 0.585 0.446 0.614 0.836 0.636 1.106 0.550889

S/N for Ra (dB) 11.2450 9.7891 7.1705 8.7780 9.3704 5.8827 1.9057 6.5580 3.0485 9.4217 7.5144 2.9871 4.6569 7.0133 4.2366 1.5559 3.9309 -0.8751 5.788311

Tool Wear(Vb) (mm) 0.137 0.124 0.118 0.126 0.110 0.104 0.134 0.132 0.124 0.128 0.125 0.108 0.104 0.112 0.104 0.132 0.135 0.116 0.120722

S/N for (Vb) (dB) 17.2656 18.1316 18.5624 17.9926 19.1721 19.6593 17.4579 17.5885 18.1316 17.8558 18.0618 19.3315 19.6593 19.0156 19.6593 17.5885 17.3933 18.7108 18.40208

The surface roughness and tool wear response table for S/N are shown in Table 6. According to the Taguchi, to get optimum cutting condition the S/N ratio should have a maximum value. In Fig.3, Fig.4 and Table 6 indicates the level of S/N response and cutting parameters effect on surface roughness and tool wear. The optimal level for each control factor was calculated based on highest S/N and shown in Table.6 as bolded values for surface roughness and tool wear. According to this, the factors giving the minimum surface roughness were specified as nose radius (r 1, S/N = 7.083), cutting speed (v1, S/N = 8.021) feed rate (f 2, S/N = 7.363) and depth of cut (d1, S/N 6.459). In another way, a good surface finish value was obtained with a nose radius 0.8mm, at cutting speed 50 m/min at a feed rate 0.04 mm/rev and at a depth of cut 0.2 mm.

Similarly, factors giving the minimum tool wear were specified as nose radius (r2, S/N = 18.59), cutting speed (v2, S/N = 19.19) feed rate (f 3, S/N = 19.01) and depth of cut (d2, S/N=18.74) based on factor level and S/N ratio. In another way, the minimum tool wear was obtained with a nose radius of 0.4mm, at a cutting speed of 100 m/min at a feed rate 0.12 mm/rev and at a depth of cut 0.3mm. Table.6 S/N response for surface roughness and tool wear Control factors Nose Cutting Feed Radius speed rate Surface Roughness 7.083 8.021 6.261 4.494 6.656 7.363 2.687 3.742 2.590 5.334 3.621 Tool wear 18.22 18.20 17.97 18.59 19.19 18.23 17.81 19.01 0.37 1.38 1.04

Levels

1 2 3 Delta 1 2 3 Delta

Depth of cut 6.459 4.639 6.267 1.820 18.08 18.74 18.38 0.66

Fig.3. The effect of cutting parameters on average S/N ratios for surface roughness. Main Effects Plot for SN ratios Data Means Nose Radius

Cutting Speed

8

Mean of SN ratios

6 4 2 1

2

1

2 Depth of Cut

3

1

2

3

Feed Rate 8 6 4 2 1

2

3

Signal-to-noise: Smaller is better

Fig.4.The effect of cutting parameters on average S/N ratio for tool wears Main Effects Plot for SN ratios Data Means

Mean of SN ratios

Nose Radius

Cutting Speed

19.2 18.9 18.6 18.3 18.0 1

2

1

2 Depth of Cut

3

1

2

3

Feed Rate 19.2 18.9 18.6 18.3 18.0 1

2

Signal-to-noise: Smaller is better

3

4.3 Evaluation of experimental results The variation in the surface roughness and flank wear were obtained from experimental work as shown in Figs. 5 and 6 respectively. The contour plot is a graphical representation of the relationships between three numeric variables in two dimensions. From the contour plot, larger tool nose radius had the better finish than the smaller nose radius in hard turning. However, in flank wear, smaller nose radius showed better flank wear resistant than larger tool nose radius. In both the nose radius, surface roughness and flank wear increased with increase in cutting speed because of the cutting tool was subjected to highest temperature and pressure at the nose in hard turning. This leads to thermal softening of cutting tool [6]. At lower cutting speed the wear rate is high when cutting speed increases with the decrease in tool wear; this is due to increased cutting speed decreases the friction between tool and chip. It was noted that the second most effective parameter affecting surface roughness is feed rate. An increased feed rate caused the increase in the roughness values. This is due to material hardness and tendency of strain hardening [5]. However, the increasing feed rate with the decrease in tool wear due to increased feed rate decreases the contact time of tool and material. Higher feed rate with cutting speed increases the surface roughness and tool wear. Surface roughness is also affected by another parameter is the depth of cut, increased depth of increases the surface roughness. The contour plot shows (Fig. 5 and 6) the effects of control factors obtained from Taguchi Method on the changes of roughness and tool wear to verify the outcome obtained based on the experimental work. 4.4 Surface roughness and tool wear prediction The relationship between a one dependent and several independent variables were calculated and analyzed by Regression analysis [24]. Surface roughness and tool wear is dependant variables and the independent variables are tool nose radius, cutting speed, feed rate and depth cut. Regression analysis was used to calculate predictive equations for surface roughness and tool wear. The predictive equations were prepared with full quadratic regression models. The predictive equations for the full quadratic regression of surface roughness are given in equation (4) and tool wear given in equation (5): Quadratic regression equation for surface roughness= 0.198452 -0.0890224 r - 0.00646256V -15.1912 f + 7.23537d + 0.0000482308V2 +102.775 f2 -13.3059d2 – 0.000558974 r V + 0.0981571 r f - 1.08926 r d - 0.0155000V f +0.00660000 V d + 8.49359 f d. (4) R-Sq = 97.86% R-Sq (adj) = 90.90% Quadratic regression equations for Tool wear: 0.201192 + 0.0536327 r - 0.00107598V + 0.409075 f – 0.470971 d + 0.00000644872V2 -2.52445 f2 + 0.843816 d2 -0.0000878205r V – 0.335391 r f - 0.0192526r d – 0.0000265152V f 0.000343939V d + 0.0384615 f d. (5) R-Sq = 99.59% R-Sq (adj) = 98.25% In Fig. 7 and Fig 8 shows the relationship between the predicted values by regression and experimental values. There is good adherence between regression predicted values and experimental values. The R2 value was obtained by full quadratic regression model for surface roughness was found 90.90 % and tool wear was found 98.25 %. Fig.5. Effect of cutting parameters on surface roughness Contour Plot of Surface Roughness vs Nose Radius, Cutting Speed 0.8

Surface Roughness < 0.4 0.4 – 0.6 0.6 – 0.8 0.8 – 1.0 > 1.0

Nose Radius

0.7

0.6

0.5

0.4 50

75

100 Cutting Speed

125

150

Contour Plot of Surface Roughness vs Feed Rate, Cutting Speed 0.12

Surface Roughness < 0.4 0.4 – 0.6 0.6 – 0.8 0.8 – 1.0 > 1.0

0.11

Feed Rate

0.10 0.09 0.08 0.07 0.06 0.05 0.04 50

75

100 Cutting Speed

125

150

Contour Plot of Surface Roughness vs DoC, Cutting Speed 0.40

Surface Roughness < 0.4 0.4 – 0.6 0.6 – 0.8 0.8 – 1.0 > 1.0

DoC

0.35

0.30

0.25

0.20 50

75

100 Cutting Speed

125

150

Contour Plot of Surface Roughness vs Nose Radius, Feed Rate 0.8

Surface Roughness < 0.4 0.4 – 0.6 0.6 – 0.8 0.8 – 1.0 > 1.0

Nose Radius

0.7

0.6

0.5

0.4

0.04 0.05 0.06 0.07 0.08 0.09 0.10 0.11 0.12

Feed Rate

Fig 6.Effect of cutting parameters on Tool Wear Contour Plot of Tool Wear vs Nose Radius, Cutting Speed 0.8

Tool Wear < 0.105 0.105 – 0.115 0.115 – 0.125 0.125 – 0.135 > 0.135

Nose Radius

0.7

0.6

0.5

0.4 50

75

100 Cutting Speed

125

150

Contour Plot of Tool Wear vs Feed Rate, Cutting Speed 0.12

Tool Wear < 0.105 0.105 – 0.115 0.115 – 0.125 0.125 – 0.135 > 0.135

0.11

Feed Rate

0.10 0.09 0.08 0.07 0.06 0.05 0.04 50

75

100 Cutting Speed

125

150

Contour Plot of Tool Wear vs Nose Radius, Feed Rate 0.8

Tool Wear < 0.105 0.105 – 0.115 0.115 – 0.125 0.125 – 0.135 > 0.135

Nose Radius

0.7

0.6

0.5

0.4 0.04 0.05 0.06 0.07 0.08 0.09 0.10 0.11 0.12 Feed Rate

Contour Plot of Tool Wear vs DoC, Cutting Speed 0.40

Tool Wear < 0.105 0.105 – 0.115 0.115 – 0.125 0.125 – 0.135 > 0.135

DoC

0.35

0.30

0.25

0.20 50

75

100 Cutting Speed

125

150

Fig.7. Comparison of the regression model and experimental results for surface roughness Variable Predicted Roughness Actual Roughness

1.1 1.0

Roughness values

0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 2

4

6

8 10 12 Samples

14

16

18

Fig.8. Comparison of the regression model and experimental results for Tool Wear 0.14

Variable Predicted Tool Wear Actual Tool Wear

Tool Wear

0.13

0.12

0.11

0.10 2

4

6

8 10 12 Samples

14

16

18

Table.7. The effect of cutting parameters on means response table for surface roughness and tool wear Levels

Control factors Nose Cutting Feed Radius speed rate Surface Roughness 0.4694 0.4173 0.5333 0.6323 0.4762 0.4395 0.7592 0.6798 0.1629 0.3418 0.2403 Tool wear 0.1232 0.1233 0.1268 0.1182 0.1100 0.1230 0.1288 0.1123 0.0050 0.0188 0.0145

1 2 3 Delta 1 2 3 Delta

Depth of cut 0.5022 0.6445 0.5060 0.1423 0.1252 0.1160 0.1210 0.0092

4.5. Optimum surface roughness and tool wear Prediction In the Fig.9, Fig.10 and Table 7 shows the surface roughness and tool wear values of the S/N means response. The optimized values for the surface roughness and tool wear were calculated as r1v1f2d1 and r2v2f3d2 respectively. The equation (6) was used for calculation of optimized surface roughness and equation (7) used for calculation of optimized tool wear. Raopt = (r1-AvRa) + (v1- AvRa) + (f2- AvRa) + (d1- AvRa) + AvRa

(6)

Vbopt = (r2-AvVb) + (v2- AvVb) + (f3 - AvVb) + (d2 - AvVb) + AvVb

(7)

Raopt denotes the predicted mean of surface roughness and Vbopt denotes the tool wear at optimum condition. The AvRa states the average value of surface roughness and AvVb state the average value of tool wear calculated from the experimental study (Table 5). As per the equations (6) & (7), it was calculated that Raopt =0.176 µm and Vbopt =0.094 mm. In Taguchi optimization technique, the optimized conditions need to be evaluated. The confirmation experiment was used to evaluate. Fig.9.The effect of cutting parameters on means response characteristics for surface roughness. Main Effects Plot for Means Data Means

Nose Radius

0.8

Cutting Speed

0.7

Mean of Means

0.6 0.5 0.4

1

2

1

Feed Rate

0.8

2

3

Depth of Cut

0.7 0.6 0.5 0.4

1

2

3

1

2

3

Fig.10.The effect of cutting parameters on means response characteristics for Tool Wear. Main Effects Plot for Means Data Means

Nose Radius

0.130

Cutting Speed

0.125

Mean of Means

0.120 0.115 0.110 1

2

1

Feed Rate

0.130

2

3

Depth of Cut

0.125 0.120 0.115 0.110 1

2

3

1

2

3

The following equation (8) & (9) were used to calculate confidence interval [14]. The assumed condition for the reliability is 95% CI = (F (α, 1, fe) Vc {(1/ Neff) + (1/R)}) 1/2

(8) and

Neff = {N/ (1+Tdof)}

(9)

Where, the F-ratio required for 95%CI, F (α, 1, fe), α is the significance level=0.05, fe is the error of degrees-of-freedom = 10, Vc is error variance = 0.00405, R is the confirmation experiments number of replications =3, Neff is effective no. of replication, N total number of trails in the experiment = 18 and Tdof is total main degrees of freedom = 7. F (0.05, 1, 10) = 4.9646 (from standard F table). Substituting the above values in equation (8) and (9), we calculated that Neff =2.25 and CIRa = ± 0.0177 and CIVb = ± 0.0032.The predicted optimal surface roughness is to be 0.176 ± 0.125 µm and tool wear is to be 0.094 ± 0.0053 mm at 95% confidence level. 4.6 Confirmation Test The table 8 shows the comparison of confirmation test results and predicted results calculated by Taguchi and full quadratic regression equations. The experimental values and predicted values are adhering close to each other. The error value should be lower than 20 % for reliable statistical analysis [24]. The maximum percentage of error calculated for surface roughness and tool wear 19.4 % and 4.326 % respectively, there are within the acceptable range. The test result obtained from confirmation test indicates the successful optimization. 5. Conclusions In this paper, Taguchi orthogonal array L18 was used. The optimal cutting parameters were determined in the turning of austempered ductile iron (grade 3) using CVD-coated carbide inserts under dry turning conditions the Taguchi method was used. The confirmation experiments were used to validate the optimum cutting parameters. Based on the ANOVA and S/N ratio analysis, the following conclusions can be made •

•

•

According to results of ANOVA analysis, the main contributing factors affecting the surface roughness and tool wear were cutting speed with a contribution of 49.1 % and 50.2 % respectively. The significant contributions of the nose radius, feed rate and depth of cut on the surface roughness were found to be 14.6 %, 21.6 % and 9.7 % respectively and on tool wear found to be 5 %, 30.2 % and 11.3 % respectively. Based on the S/N ratio using “smaller is the better” approach, the optimum level of cutting parameters for surface roughness r1v1f2d1 (i.e., Nose radius= 0.8 mm, cutting speed=50 m/min, Feed rate = 0.08 mm/rev, and depth of cut 0.2mm) and tool wear were r2v2f3d2 (i.e., Nose radius= 0.4 mm, cutting speed=100 m/min, Feed rate = 0.12 mm/rev, and depth of cut 0.3 mm). Good surface finish was achieved in 0.8 mm nose radius and minimum tool wear achieved in 0.4 mm nose radius. Surface roughness and tool wear values predicted using quadratic regression analysis was in close adhere to the experimental values with high correlation coefficients.

•

The confirmation test result shows that the measured values were 95% within the confidence level.

In this work, it was found that the Taguchi method was successfully used in reduction of production cost and production time in the turning of ADI. The results obtained in this work can be used as standards both academic research and industrial applications. In future, the other factors could be considered like different coating materials, lubricants, and cutting tool geometry, all these factors would affect on the surface roughness and tool wear. Table .8 Comparison of predicted and confirmation test result. Test

r1v1f2d1 (optimum) r2v1f2d1 (Random)

Quadratic regression Prediction ExperiPreError ExPreError mental dicte (%) perim dicted (%) d ental Surface Roughness 0.194 0.176 9.3 0.194 0.231 19.1

r2v2f3d2 (optimum) r1v2f3d3 (Random)

Taguchi Prediction

0.421

0.339

19.4

0.421

0.362

14.0

0.095

Tool Wear 0.094 1.1

0.095

0.096

1.1

0.104

0.102

1.3

0.104

0.108 5

4.3

Acknowledgement The authors acknowledge with thanks to M/s Indoshell Cast (Pvt) Limited, Coimbatore, India for providing castings, technical ideas and help in this experimental work. Abbreviations CVD - Chemical Vapor Deposition ADI - Austempered ductile iron HRC- Rockwell hardness measured on “C” scale

References [1] Satish Chinchanikar, and S.K. Choudhury, Effect of work material hardness and cutting parameters on performance of coated carbide tool when turning hardened steel: An optimization approach, Measurement 46 (2013) 1572–1584. [2] A.Srithar, K.Palanikumar and B.Durgaprasad, Experimental Investigation and Surface roughness Analysis on Hard turning of AISI D2 Steel using Coated Carbide Insert, Procedia Engineering 97 ( 2014 ) 72 – 77. [3] Marcelo Vasconcelos de Carvalho, Davi Melo Montenegro and Jefferson de Oliveira Gomes, An analysis of the machinability of ASTM grades 2 and 3 austempered ductile iron, Journal of Materials Processing Technology 213 (2013) 560– 573. [4] Yamamoto, S., Nakajima, H and Miyaji, H., Mechanisms of poor machinability of austempered ductile iron. The Iron and Steel Institute of Japan 81 (7), 2005, 721–726. [5] Klocke, F and Klöpper, C., Machinability characteristics of austempered ductile iron (ADI). Proceedings of the World Conference on ADI for Casting Producers, Suppliers and Design Engineers, Ductile Iron Society and American Foundry Society, Galt House Hotel, Louisville, Kentucky, USA In: (2002). [6] Gekonde, H.O and Subramanian, S.V., High speed machining of ductile iron. AFS Transactions 120(1995), 309–317. [7] Ahmet Akdemir, Sakir Yazman, Hacı Saglam and Mesut Uyaner, Effects of Cutting Speed and Depth of Cut on Machinability Characteristics of Austempered Ductile Iron, Journal of Manufacturing Science and Engineering,134 (2012), 021013-1 to 021013-9. [8] K. Katuku, A. Koursaris and I. Sigalas, Wear, cutting forces and chip characteristics when dry turning ASTM Grade 2 austempered ductile iron with PcBN cutting tools under finishing conditions, journal of materials processing technology 2 0 9 (2009) 2412–2420.

[9] Tug˘rul O¨ zel and Yig˘it Karpat, Predictive modeling of surface roughness and tool wear in hard turning using regression and neural networks International Journal of Machine Tools & Manufacture 45 (2005) 467–479. [10]Zahia Hessainia, Ahmed Belbah, Mohamed Athmane Yallese , Tarek Mabrouki and Jean-François Rigal , On the prediction of surface roughness in the hard turning based on cutting parameters and tool vibrations., Measurement 46 (2013) 1671–1681. [11]Mustafa Gunay and Emre Yuce., Application of Taguchi method for determining optimum surface roughness in turning of high-alloy white cast iron Measurement 46 (2013) 913–919. [12] Sanjeev Saini, Inderpreet Singh Ahuja and Vishal S. Sharma., Influence of cutting parameters on tool wear and surface roughness in hard turning of AISI H11 tool steel using ceramic tools, International Journal of Precision Engineering and Manufacturing 13(2012)1295-1302 . [13]Abhijeet S. More, Wenping Jiang, W.D. Brown and Ajay P. Malshe., Tool wear and machining performance of cBN– TiN coated carbide inserts and PCBN compact inserts in turning AISI 4340 hardened steel. Journal of Materials Processing Technology 180 (2006) 253–262. [14]N.Mandal, B.Doloi, B.Mondal, R.Das, Optimization of flank wear using Zirconia Toughened Alumina (ZTA) cutting tool: taguchi method and regression analysis, Measurement 44 (2011) 2149–2155. [15]Ilhan Asiltürk and Süleyman Neseli., Multi response optimisation of CNC turning parameters via Taguchi methodbased response surface analysis. Measurement 45 (2012) 785–794. [16]D. Philip Selvaraj , P. Chandramohan and M. Mohanraj., Optimization of surface roughness, cutting force and tool wear of nitrogen alloyed duplex stainless steel in a dry turning process using Taguchi method Measurement 49 (2014) 205–215. [17]K. Palanikumar., Experimental investigation and optimisation in drilling of GFRP composites Measurement 44, 10, (2011), 2138–2148. [18]Muhammed munawar, Joseph ching-Shihn chen and Nadeem Ahmad Mufti, Investigation of cutting parameters effect for minimization of surface roughness in internal turning International journal of precision engineering and manufacturing, Vol. 12, No. (2011), pp. 121-127. [19]I. Ucun and K. Aslantas, Investigation of Performance of Carbide Cutting Tool in Turning Hardened 52100 Tool Steel, IATS’09, Karabuk,Turkey, 2009, pp. 1–6. [20]Ilhan Asilturk, Harun Akkus, Determining the effect of cutting parameters on surface roughness in hard turning using the Taguchi method, Measurement 44 (2011) 1697–1704. [21]Y. Kevin Chou, Hui Song., Tool nose radius effects on finish hard turning, Journal of Materials Processing Technology 148 (2004) 259–268. [22]M. Sanyılmaz, Design Of Experiment And an Application for Taguchi Method in Quality Improvement Activity, M.S. Thesis, Dumlupınar University, Turkey, 2006. [23]Turgay Kıvak, Optimization of surface roughness and flank wear using the Taguchi method in milling of Hadfield steel with PVD and CVD coated inserts, Measurement 50 (2014) 19–28. [24]M.H. Cetin, B. Ozcelik, E. Kuram, E. Demirbas, Evaluation of vegetable based cutting fluids with extreme pressure and cutting parameters in turning of AISI 304L by Taguchi method, J. Cleaner Prod. 19 (2011)2049–2056.