Effect of turning parameters on surface roughness of ... - Springer Link

Journal of Mechanical Science and Technology 26 (12) (2012) 3973~3979 www.springerlink.com/content/1738-494x

DOI 10.1007/s12206-012-0914-5

Effect of turning parameters on surface roughness of A356/5% SiC composite produced by electromagnetic stir casting† S. P. Dwivedi, Sudhir Kumar* and Ajay Kumar Noida Institute of Engineering Technology, Greater Noida, Gautam buddha Nagar, U.P. 201310, India (Manuscript Received May 10, 2012; Revised July 26, 2012; Accepted August 9, 2012) ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Abstract In the present investigation, A356 alloy 5 wt% SiC composite is fabricated by electromagnetic stir casting process. An attempt has been made to investigate the effect of CNC lathe process parameters like cutting speed, depth of cut, and feed rate on surface roughness during machining of A356 alloy 5 wt% SiC particulate metal-matrix composites in dry condition. Response surface methodology (Box Behnken Method) is chosen to design the experiments. The results reveal that cutting speed increases surface roughness decreases, whereas depth of cut and feed increase surface roughness increase. Optimum values of speed (190 m/min), feed (0.14 mm/rev) and depth of cut (0.20 mm) during turning of A356 alloy 5 wt% SiC composites to minimize the surface roughness (3.15µm) have been find out. The mechanical properties of A356 alloy 5 wt% SiC were also analyzed. Keywords: Electromagnetic stir casting; CNC Lathe; RSM; A356 alloy 5 wt% SiC composite; Tensile strength; Hardness; Surface roughness ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

1. Introduction Composite materials have now replaced traditional engineering materials. Metal matrix composite (MMCs) are especially preferred for high temperature applications in many areas such as the automotive, space, and aviation industries as well as sport applications due to their light weight, high mechanical properties, excellent wear, and corrosion resistance. The production of composite materials and the development of subsequent shaping processes have become very relevant for technological applications. In engineering designs, great interest is to search for new materials exhibiting good mechanical properties. For the development of such materials, metal matrix composites (MMCs) have been proved to be one of the best selections. Composite materials are engineered materials made from two or more constituent’s material with significantly different physical and chemical properties and which remain separate and distinct on a macroscopic level with the finished structure. MMCs cannot be easily fabricated using conventional methods. The fabrication techniques can vary considerably according to the choice of the matrix and reinforcement materials. In general, there are two types of fabrication methods, namely, the solid phase and liquid phase. The liquid metallurgy tech*

Corresponding author. Tel.: +919910300196, Fax.: +911202320062 E-mail address: [email protected] † Recommended by Editor Jai Hak Park © KSME & Springer 2012

nique is the most economical of all the available routes for metal matrix composite production, and allows very large sized components to be fabricated. On the basis of all the above facts, the most popular types of MMCs are aluminum alloys reinforcing with ceramic particles. These low cost composites provide higher strength, stiffness and fatigue resistance with a minimal increase in density over the base alloy. Owing to the addition of reinforcing materials which are normally harder and stiffer than the matrix, machining becomes significantly more difficult than in the case for conventional material. The SiC particles used in aluminum matrix composites are hardener than tungsten carbide. But, Poly Crystalline Diamond (PCD) is approximately 3-4 times hardener than SiC particles [1-3].

2. Literature review H. K. Moffatl [4] described electromagnetic stirring and stated that an alternating magnetic field apply to a conductor, whether solid or fluid will induces electric current in the conductor. S. Milind et al. [5] presented about design and analysis of a linear type electromagnetic stirrer, electromagnetic stirring (EMS) of metallic alloys is mainly used to refine the grain structure of casting. Han Jian-min et al. [6] investigated about the defects in SiC/A356 composites made by a stir casting method. Zhirui Wang et al. [7] studied mechanical behavior of cast particulate SiC/AI (A356) metal matrix composites, mechanical tests were carried out to study the deformation

3974

S. P. Dwivedi et al. / Journal of Mechanical Science and Technology 26 (12) (2012) 3973~3979

Table 1. Chemical composition of A356 alloy (wt%). Si

Fe

Cu

Mn

Mg

Zn

Ti

Al

6.5-7.5

0.2

0.2

0.1

0.25-0.45

0.1

0.1

Balance

behavior of particulate SiC-reinforced Al (A356) matrix composites produced through direct casting using the molten metal mixing method. Rajesh Kumar Bhushan et al. [8] made an attempt to study effect of machining parameters on surface roughness and tool wear for 7075 Al alloy SiC composite and investigated the influence of cutting speed, depth of cut, and feed rate on surface roughness during machining of 7075Al alloy and 10 wt.% SiC particulate metal-matrix composites. Dayananda Pai et al. [9] analyzed the application of response surface methodology on surface roughness in grinding of aerospace materials (6061Al-15Vol% SiC25P) and investigated the effects and the optimization of machining parameters on surface roughness in the grinding of 6061Al-SiC25P (MMCs) specimen. Electromagnetic stir casting process is a low cost effective process. It does not break the SiC particles because stirrer is not the part of the setup. In this case, molten metal rotates by electromagnetic field. Thus, this process gives better mechanical properties. It is well known that input process parametrs play a major role in determining the surface quality. As the process facts have not been disclosed so far, the selection of input parameters to machining A356 alloy 5%wt SiC composite is very difficult. While machining of aluminium alloy composite with conventional lathe is a problem. Hence, the problem of getting optimized surface parametrs to attain minimum surface roughness is attempted in this investigation.

3. Selection of the material 3.1 Matrix alloy In this study, A356 alloy is selected. It has very good mechanical strength, ductility, hardness, fatigue strength, pressure tightness, fluidity, and machinability. The chemical composition of A356 is shown in Table 1. 3.2 Reinforcement material To select a suitable reinforcement material for aluminum, important facts such as density, wettability and thermal stability were considered. Silicon carbide is a widely used reinforcement material because of its good wettability with the aluminum matrix [3].

4. Fabrication of metal matrix composite

Fig. 1. Electromagnetic stir casting set up.

Fig. 2. Produced metal matrix composite (A356 alloy 5wt% SiC).

graphite crucible and heated to above its liquidus temperature in an open hearth furnace. The temperature was recorded using chromel-alumel thermocouple, which was 700oC. Secondly, the liquid A356 aluminum alloy at a given temperature was poured into a stainless steel crucible which was packed very well with the help of glass wool (between crucible and winding) inside the motor as shown in Fig. 1. SiC particles with an average size of 50 µm were chosen as the reinforcement particles. The 5% SiC reinforcing particles were added on the surface of the mol ten liquid A356 at 700oC to the crucible. The SiC particles disperse into the melt material. By providing 240 V supply to three phase induction motor melt material (A356 + 5%Sic) stirred by an electromagnetic field for a 5 minutes and the stirring speed of melt material was 210 rpm. The prepared sample was removed from the stainless steel crucible which is shown in Fig. 2.

4.1 Electromagnetic stir casting setup

4.2 Selection of process parameters of CNC lathe and their levels

Fig. 1 shows the schematic of electromagnetic stirring set up which was fabricated for the composite processing. About 2 kilograms of the A356 alloy is cleaned and loaded in the

There are various process parameters of CNC lathe affecting the machining characteristics. On the basis of literature review and same pilot investigations, the following process

3975


Table 2. Process parameters with their ranges.

Table 3. Design matrix and Experimental results.

S. No.

Input parameters

Range

1

Speed (m/min)

100 – 190

Standard order

Speed (m/min)

Feed rate (mm/rev)

Depth of cut (mm)

2

Feed (mm/rev)

0.14 – 0.24

1

100

0.14

0.40

4.7

3

Depth of cut (mm)

0.20 – 0.60

2

190

0.14

0.40

3.798

3

100

0.24

0.40

5.7

4

190

0.24

0.40

4.8

5

100

0.19

0.20

4.0

6

190

0.19

0.20

3.1

5. Response surface methodology

7

100

0.19

0.60

5.2

The response surface methodology is used to design the experiment for the given problem and the problem formulated with the following steps. In this study, three parameters are used as levels that maximize the yield (y) of a process. The process yield is a function of the different constituents, say

8

190

0.19

0.60

4.3

9

145

0.14

0.20

3.2

10

145

0.24

0.20

3.5

11

145

0.14

0.60

4.1

12

145

0.24

0.60

5.2

13

145

0.19

0.40

4.2

14

145

0.19

0.40

4.15

15

145

0.19

0.40

4.29

16

145

0.19

0.40

4.27

17

145

0.19

0.40

4.25

parameters have been selected for study. Their ranges are given in Table 2.

y = f(x1, x2, x3) + Є

(1)

where Є represents the noise or error observed in the response y. If we denote the expected response (tensile strength) by E(y) = f(x1, x2, x3, x4, x5) = η, then the surface represented by η = f(x1, x2, x3)

Surface Roughness

(2)

is called a response surface [10]. Box-Behnken design is used to further study the quadratic effect of factors after identifying the significant factors using screening factorial experiments.

6. Planning of experiments Turning of A356 alloy 5wt% SiC composites was carried out by tungsten carbide inserts on CNC turning machine. As per the plan of experiments tabulated in Table 3 and average measured surface roughness are also given in Table 3.

(a)

(b)

Fig. 3. Microstructure of A356/ 5% SiC composite at stirring speed 195 rpm and 210 rpm.

7. Result and discussion 7.1 Microstructural analysis The microstructure of A356 alloy 5wt% SiC composite is composed of an aluminium matrix containing silicon carbide. In general, the silicon carbide is not uniformly distributed, but tends to be connected at inter-dendiritic boundaries. Fig. 3(a) shows the microstructure of A356 alloy 5wt % SiC composite in the as-cast condition formed at the stirring speed 195 rpm. It can be seen that the silicon carbide is not uniformly distributed and most of silicon accumulated at the grain boundaries due to the low stirring speed. It is known that density of SiC is much more than A356 alloy. During the experiment at low speed SiC particles whose density are high with respect to melt material (A356 alloy), settles down at the bottom. By increasing the stirring speed up to 210 rpm, it has been observed that the silicon carbide parti-

cles were not settles down and distribution is uniformed as shown in Fig. 3(b). 7.2 Mechanical properties Brinell hardness of A356/ 5% SiC composite is related with the distribution of SiC particles in the A356 alloy. If distributions of the SiC particles are good then hardness of A356/ 5% SiC composites is high. The toughness of A356/ 5% SiC composite is listed in Table 4. It was found that the value of toughness of A356 alloy is 5 Joule. While the average value of toughness of A356/ 5% SiC composite is 7.33 Joule. Theoretically, the toughness of SiC particles is very high as compared to A356 alloy. By mixing the SiC particles properly in A356, the toughness of A356/ 5% SiC composites is much better than alloy A356.

3976


Table 4. Observations of hardness, toughness and tensile strength of Composites. Ultimate tensile No. strength (N/mm2) 1 2 3 4 5 6

297 298.27 306.69 302.94 301.23 304.49

(a)

Load (KN)

8.397 8.433 8.671 8.565 8.517 8.609

Elongation (mm) 2.25 2.30 2.11 1.634 2.16 1.84

Dia of the Percentage indenta- BHN Elongation tion (mm) 6.25 6.38 5.86 4.53 6.00 5.11

(b)

1.23 1.24 1.18 1.22 1.24 1.20

77.3 76 85 80 79 82

Table 5. Analysis of variance (ANOVA) for surface roughness after pooling. Source

Sum of square

DF

Mean square

F value

Prob. > F

Model

7.63

7

1.09

155.22

< 0.0001

A

1.62

1

1.62

231.02

< 0.0001

7.0

B

1.45

1

1.45

206.08

< 0.0001

6.5

C

3.13

1

3.13

445.14

< 0.0001

8.5

A2

0.47

1

0.47

66.81

< 0.0001

7.5

B2

0.14

1

0.14

20.25

0.0015

2

7.0

C

0.73

1

0.73

103.67

< 0.0001

7.5

BC

0.16

1

0.16

22.79

0.0010

Residual

0.063

9

7.020E-003

Lack of fit

0.050

5

.010

3.12

0.1462

Pure error

0.013

4

3.220E-003

Cor total

7.69

16

Toughness (J)

(c)

Fig. 4. Specimen for the testing of hardness, toughness and tensile respectively.

Std. dev.

0.084

R-Square

Mean

4.28

Adj-R squared

C.V.

1.96

Pred R–squared

Press

0.52

Adeq precision

7.3 Analysis of surface roughness Surface roughness plays an important role in determining the machining accuracy. The study of surface roughness characteristics of A356 alloy 5 wt% SiC composites dependent on many factors, it is more influenced by the cutting parameters like cutting speed, feed, depth of cut, etc., for a given machine tool (PCD-10) and work piece set-up. The selected experimental design is box Behnken design and the design matrix is shown in Table 3. The analysis of response was done using Design Expert Software. Analysis of variance for surface roughness after pooling is shown in Table 5. Values of “Prob>F” less than 0.0500 indicate that model terms are significant. From the Table 5, linear terms speed, feed, depth of cut, square terms of speed, feed, depth of cut and interaction term of feed and depth of cut are significant model terms. Values are greater than 0.10 indicate that model terms are not significant. The final empirical relationship was constructed using only these coefficients, and the developed final empirical relationship is given below: Surface roughness = +8.7924 - 0.057802 * Speed 27.42500 * Feed + 7.64000 * Depth of Cut +1.64815E-004 * Speed2 +73.50* Feed 2 - 10.39 * Depth of Cut2 + 20.00 * Feed * Depth of Cut (3) Analysis of variance (ANOVA) technique was used to check the adequacy of the developed empirical relationship. In this investigation, the desired level of confidence was considered to be 95%. The model F value of 155.22 implies that the model is significant. There is only a 0.01% chance that a model F value this large could occur due to noise. The lack of

fit F value of 3.12 implies that the lack of fit is insignificant. There is only a 14.62% chance that a lack of fit F value this large could occur due to noise. The goodness of fit of the model was checked by the determination coefficient (R2). The coefficient of determination (R2) was calculated to be 0.9918 for response. This implies that 99.18% of experimental data confirms the compatibility with the data predicted by the model, and the model does not explain only 0.82% of the total variations. The R2 value is always between 0 and 1, and its value indicates correctness of the model. For a good statistical model, R2 value should be close to 1.0. The adjusted R2 value reconstructs the expression with the significant terms Each predicted value matches its experimental value well as shown in Table 6. The value of the adjusted determination coefficient (Adj R2 = 0.9854) is also high to advocate for a high significance of the model. The Pred R2 is 0.9320 that implies that the model could explain 95% of the variability in predicting new observations. This is in reasonable agreement with the Adj R2 of 0.9854. The value of coefficient of variation is also low as 1.96 indicates that the deviations between experimental and predicted values are low. Adeq precision measures the signal to noise ratio. A ratio greater than 4 is desirable. In this investigation, the ratio is 44.30, which indicates an adequate signal. Fig. 5 shows the correlation between the predicted and experimental values for surface roughness (Ra). The normal probabilities of residuals are shown in Fig. 6. After the regression model of surface roughness was developed, the model adequacy checking was performed in order to verify


3977

Table 6. Actual value, predicted value and residual for surface roughness. Standard order

Actual value

Predicted value

Residual

1

4.70

4.77

-0.075

2

3.80

3.83

-0.076

3

5.70

5.63

0.075

4

4.80

4.72

0.075

5

4.00

3.98

0.025

6

3.10

3.07

0.025

7

5.20

5.23

0.025

8

4.30

4.32

0.025

9

3.20

3.15

0.050

10

3.50

3.60

-0.10

11

4.10

4.00

0.10

12

5.20

5.25

-0.050

13

4.20

4.23

-0.032

14

4.15

4.23

-0.082

15

4.29

4.23

0.058

16

4.27

4.23

0.038

17

4.25

4.23

0.018

Fig. 6. The normal probability of residuals.

(a)

Fig. 5. Correlation between the predicted and experimental values.

that the underlying assumption of regression analysis is not violated. (b)

Fig. 6 illustrates the normal probability plot of the residual, which shows no sign of the violation since each point in the plot follows a straight line pattern. The normal probability plot is used to verify the normality assumption. The data are spread roughly along the straight line. Hence, it is concluded that the data are normally distributed.

face roughness of machined A356 alloy 5 wt% SiC composite. 3D graphs are shown in Fig. 7(a), (b).

7.3.1 Effect of process parameters on surface roughness In this section, the influence of cutting process parameters i.e. speed, feed and depth of cut, were evaluated against sur-

7.3.1.1 Influence of cutting speed on surface roughness It can be seen from Fig. 7(a) that surface roughness decreases with the increase in speed because at high speed, the

Fig. 7. (a) Effect of speed and feed on surface roughness; (b) Effect of speed and depth of cut on surface roughness.

3978


Fig. 8. Three dimensional plots of surface roughness at optimal process parameters levels.

surface temperature of work piece becomes high. High temperature of surface of work piece causes metallic bond of material becomes soft due to the resistance offered by a material against tool becomes low, hence at high speed, surface finish is good. 7.3.1.2 Influence of feed on surface roughness As seen in Fig. 7(a), feed rate increases from minimum to maximum limit, the surface roughness increases because at high feed rate, MRR of work piece becomes high. High MRR causes friction developed between tool and material. Friction is developed on tool surface due to pitting marks. This is due to high feed rate. Higher surface roughness is obtained. 7.3.1.3 Influence of depth of cut on surface roughness In experimental studies, surface roughness increased while depth of cut increased. This effect of depth of cut was observed due to the higher chip thickness. By analyzing the response surfaces, the minimium achievable surface roughness value is found to be 3.15 µm. The corresponding parameters that yields this minimium value are cutting speed of 190 m/min, feed of 0.14 mm/rev. and depth of cut 0.20 mm. Importance of process parameters can be ranked from their F ratio which is mentioned in Table 4. I can be concluded that depth of cut is contributing more and it is followed by cutting speed and feed. 7.4 Confirmation experiment Three confirmation experiments were conducted at the optimum setting of the cutting parameters. The cutting speed was set at 190 m/min., feed at 0.14 mm/rev, and depth of cut at 0.20 mm. The average surface roughness was found to be 3.12 µm which was near to the optimal value (3.15 µm). The two-dimensional plots of surface roughness taken by using optical profilometer at optimal value for confirmation experiments are shown in Fig. 8.

8. Conclusions The following conclusions can be drawn from the analysis:

(1) The electromagnetic stir casting method was found to be successful to fabricate A356 alloy 5 wt% SiC Composite. Electromagnetic stirring greatly refines the structure of A356/ 5% SiC composite. The microstructure of composite is more homogenous at 210 rpm than 195 rpm. (2) The hardness of A356 alloy increased when SiC were added. From the results, hardness of A356 alloy is 63 BHN, by addition of 5% SiC in A356 alloy, 27% hardness improved. (3) The value of toughness of A356 alloy was 6 Joule. While the average value of toughness of A356/ 5% SiC composites is 7.43 Joule. Improvement in toughness is 46%. (4) The tensile strength of A356 alloy was 235 N/mm2. While the tensile strength of A356/ 5% SiC composites is 301.77 N/mm.2 Tensile strength has improved by 28.4%. (5) An empirical relationship was developed to predict the surface roughness incorporating machining parameters at 95% confidence level. It was found that the parameters which affect the surface roughness in descending order are as follows: depth of cut, speed and feed. The best combination of parameters for surface roughness (3.15 µm) was found to be speed of 190 m/min, feed of 0.14 mm/rev. and depth of cut 0.20 mm.

References [1] A. Pakdel and H. Farhangi, Influence of extrusion ratio on the mechanical behavior of AA6061/Sic composites, Arabian Journal for Science and Engineering, 34 (1) (2009) 167-174. [2] R. K. Bhushan, S. Kumar and S. Das, Cutting forces and surface roughness analysis of aluminum silicon alloy sic composite, International Journal of Manufacturing Technology and Industrial Engineering, 1 (1) (2010) 57-66. [3] S. Kumar, K. L. Sivasubramanian and K. Kalaiselvan, Selection of optimum parameters in non conventional machining of metal matrix compsite, Portugaliae Electrochimica Acta, 27 (4) (2009) 477- 486. [4] H. K. Moffatt, Electromagnetic Stirring, Phys. Fluids A, 5 (1990) 1336-1343. [5] S. Milind and V. Ramanarayanan, Design and analysis of a linear type electromagnetic stirrer, Indian institute of science (2004) 188-194. [6] J. Han, Z. Wu, S. Cui, W. Li and Y. Du, Investigation of defects in Sicp/A356 composites made by a stir casting method, Journal of Ceramic Processing Research, 8 (1) (2006) 74-77. [7] Z. Wang and J. Zhang Ruby, Mechanical behavior of cast particulate Sic/AI (A356) metal matrix composites, Mettalurgical and Materials Transactions, A27 (12) (2004) 38933901. [8] R. K. Bhushan, S. Kumar and S. Das, Effect of machining parameters on surface roughness and tool wear for 7075 al alloy sic composite, Int J Adv Manuf Technol. (2010) 459469. [9] Dayananda, Pai, Rao, Shrikantha, Shetty, Raviraj and Rajesh.


3979

Nayak, Application of response surface methodology on surface roughness in grinding of aerospace materials (6061Al15Vol%Sic25p), ARPN Journal of Engineering and Applied Sciences, 5 (6) (2010) 23-28. [10] D. C. Montgomery, Design and analysis of experiments, 5th edition, Wiley-India (427) (2007).

Sudhir Kumar received his Ph.D from IIT Roorkee, India. He has published 75 research papers in international and national journals and conferences.

S. P. Dwivedi received his M.Tech. degree in Mechanical Engg. (CAD) in 2010. He has published 3 research papers in international and national journals and conferences.

Ajay Kumar received his M.Tech. degree from IIT Roorkee, India. He has published 5 research papers in international and national journals and conferences.