- Email: [email protected]
6063 aluminium MMC using CNC Wire Electrical Discharge Machining
Composites Communications 6 (2017) 6–10
Contents lists available at ScienceDirect
Composites Communications journal homepage: www.elsevier.com/locate/coco
Examination of accuracy aspect in machining of ZrSiO4p/6063 aluminium MMC using CNC Wire Electrical Discharge Machining Mohinder Pal Garga, Anand Sharmab, a b
MARK
⁎
DAV University, Jalandhar, India Ansal University, Gurgaon, India
A R T I C L E I N F O
A B S T R A C T
Keywords: WEDM dimensional deviation RSM ANOVA
Wire Electrical Discharge Machining (WEDM) is a deep-rooted non-contact modern machining technique used to manufacture geometrically intricate shapes in hard materials which are difficult to machine using conventional methods. To achieve close dimensional accuracy in WEDM, the output parameter dimensional deviation is an essential response to be controlled during machining. Determination of dimensional deviation for a particular process parameter combination assists the manufacturing planners in setting a wire offset so that the actual (cut) work piece dimensions match the part geometry.This study investigates the dimensional deviation induced by WEDM of ZrSiO4p/6063 Aluminium metal matrix composite (MMC) by using response surface methodology (RSM). The key WEDM input process parameters namely, pulse-on time, pulse-off time, servo voltage and peak current were varied in order to examine their influence on dimensional deviation. Significant process parameters affecting the process are identified by carrying out analysis of variance (ANOVA) technique. To aid in selecting the best combination of input parameters during WEDM of ZrSiO4p/6063 Aluminium MMC the concept of desirability is utilized. Confirmation experiments have been conducted to verify the optimal parameter combinations.
1. Introduction There is a great need for materials with special properties with emergence of new stringent requirements posed by manufacturing industries. Traditional engineering materials are unable to meet these requirements of special properties like high strength, low weight, hot hardness, temperature resistance for various industries like aerospace, automotive etc. Composite materials emerged as a new class of engineering materials to cater the needs of these aforesaid industries. Composites are a unique class of materials formed by combining two or more distinct materials to form stronger, tougher and more durable material than each would have been separately. The extreme environment in space a typical spacecraft encounter with high temperature while entering the earth’s atmosphere, naturally occurring phenomena such as vacuum, thermal and ionizing radiation, along with factors such as micro or macro meteoroids and debris therefore, demand lightweight space structures with high strength and dimensional accuracy. Nevertheless, such materials are difficult to be machined by traditional machining methods. The presence of high strength reinforcements in the MMC leads to the rapid wear and tear of cutting tools during machining by conventional or traditional methods [1]. This leads to low
⁎
Correspondıng author. E-mail address: [email protected] (A. Sharma).
http://dx.doi.org/10.1016/j.coco.2017.07.002 Received 4 November 2016; Received in revised form 3 May 2017; Accepted 11 July 2017 2452-2139/ © 2017 Elsevier Ltd. All rights reserved.
cutting rate and subsequent increase in production cost. Therefore, there is a need to enhance the machining performance for these materials. Hence, non-traditional machining methods including abrasive jet machining (AJM), laser beam machining (LBM), water jet machining (WJM), electrical discharging machine (EDM) etc. are applied to machine such difficult-to-cut materials. Techniques like EDM and WEDM are quite successful for machining of high strength MMC’s. EDM can be used only for drilling purpose, while WEDM conforms to easy control and can machine intricate and complex shapes as a result of which WEDM technique is highly recommended for machining of MMC’s [2]. WEDM is a specialized machining technique which uses electro-thermal mechanism to produce complicated cut-outs through electrically conductive and difficult-to-machine materials. Generally, WEDM is perceived to be an accurate non-traditional process. Aluminium based MMC’s also known as aluminium matrix composites (AMC’s) have demonstrated improved mechanical properties compared to properties of pure or non-reinforced Al alloys. Because of their attractive properties, relative ease in fabrication technology and their potential to be available at low cost, Aluminium-matrix composites (AMCs) are being widely used by various industries [3]. AMCs reinforced with hard ceramic particles have emerged as a potential
Composites Communications 6 (2017) 6–10
M.P. Garg, A. Sharma
One of the main purposes of the present study is to investigate the effects of key input process parameters such as, pulse on time, pulse off time, servo voltage and peak current on the output response namely dimensional deviation (DD) and to obtain optimal parameter combinations using desirability approach. This study also aims to establish an empirical relationship between various input process parameters and output response using ANOVA for machining of MMC using Box Behnken Designs (BBD). BBD are one class of the experimental designs for response surface methodology. They are rotatable or nearly rotatable based on three-level incomplete factorial designs. Rotatable means that the model would possess a reasonably stable distribution of scaled prediction variance throughout the experimental design region. BBD allows calculations of the response function at intermediate levels and enables estimation of the system performance at any experimental point within the range studied through careful design and analysis of experiments [12]. For three factors BBD, its graphical representation can be seen as a cube that consists of the central point and the middle points of the edges as shown in Fig. 2 [13].
material especially for wear-resistant and weight critical applications such as brake drums, cylinder liners, pistons, cylinder blocks, connecting rods, and so on [13]. Past studies have indicated that efficient and economical WEDM processing of these materials has opened new areas of applications for MMC’s, but there is a lack of efforts made to investigate machining of MMC’s through WEDM process and particularly of Al-based MMC’s. Literature surveys indicate that most of the reported research is concentrated on machining of conventional materials like steel, brass, titanium etc. than on aluminium based materials [4–11]. 1.1. Fabrication of ZrSiO4p/6063 al metal matrix composite The fabrication methodology of a MMC can be clustered under solid state, liquid state and powder metallurgy, based on the physical state of the matrix i.e., solid phase and liquid phase. ZrSiO4p/6063 Al MMC’s can be widely used materials in many engineering applications due to their improved properties but none of the research effort is made to investigate the potential of WEDM in machining of ZrSiO4p/6063 Al MMC. ZrSiO4p/6063 Al MMC offers a wide variety of applications including wear-resistant components, punches, engine parts, high temperature applications, medical and biomedical purposes etc. Zirconium silicate or zircon (ZrSiO4) reinforced particles of average particle size 5μm, is used for casting of Al 6063 MMC by stircasting technique. Table 1 represents the chemical composition of Al 6063 alloy used as a base metal for fabrication of MMC. In this study, commercially available Al 6063 in the form of cylindrical rods is used as matrix reinforced with ZrSiO4 particles. The melting of measured quantity of (3 kg) pickled Al 6063 is carried out in a clay-graphite crucible (Fig. 1a) placed inside the muffle furnace. Fig. 1b shows the muffle furnace and the developed stirring setup for melting and mixing of Al 6063 and ZrSiO4 particles. Prior heating of required quantity of zirconium silicate (ZrSiO4p) particulates 150gm (5 wt %) to around 500 °C for 1 hour was carried out in a separate furnace. Preheating of ZrSiO4 particles before mixing it with molten metal was purposely done to eliminate the water vapours present within the particles and also to improve the wettability by removing the absorbed hydroxide and other gases. The furnace temperature is first raised above 1000 °C (50° above the melting point of Al 6063) for proper liquidification of Al 6063. At this stage, the pre-heated ZrSiO4 particles are added to the molten metal and mixed mechanically using stirrer. While adding the zircon particles to the molten aluminium alloy, to avoid atmospheric contamination nitrogen gas (Fig. 1c) is used as the shield of the mixer. The composite slurry is then reheated to a fully liquid state and mechanical mixing was carried out for 5 min at 100 rpm average stirring speed. Stirring is carried out to ensure even temperature distribution as well as particulate distribution. The mixer of reinforced particulate and the molten metal is then poured to a prepared sand mould. After pouring is over, the melt is allowed to cool and solidify in the mould under the shield of N2. After solidification, the casting is taken from the mould and is cut to the required shape and size (110mm x 80mm x 11mm) for subsequent WEDM. The improved properties of ZrSiO4p/6063 Al MMC and the conventional properties of base metal Al 6063 are shown in Table 1. It reveals that the important mechanical properties are improved to a good extent by addition of 5% ZrSiO4p to base metal Al 6063 and it can effectively replace conventional Al 6063 material in automobile and aerospace applications owing to its improved properties.
2. Design of experiments and experimentation Various input process parameters viz. Pulse on time (TON), Pulse off time (TOFF), Peak Current (IP) and Servo Voltage (SV) are varied to investigate their effects on DD during machining of ZrSiO4p/6063 Al MMC. The ranges of input parameters were selected on the basis of literature survey, machining capability of the machine and pilot study conducted by using one variable at a time approach [14]. The range of selected process parameters for further analysis is presented in Table 2. Table 2 also lists the parameters and there levels which were kept fixed in the study. The experiments were performed on a four axis Electronica Sprintcut 734 CNC Wire Cut Machine. Diffused brass wire of 0.25mm diameter was used as tool material and deionized water is used as dielectric. A plate of rectangular shape (110 mm × 80 mm × 11 mm) having 5% ZrSiO4 particles (by weight) as reinforcement is fixed on the machine table. An 8 mm × 8 mm rectangular cut is taken on the work piece as shown in Fig. 3a. Fig. 3(b,c) represents the selected machine tool and the rectangular pieces cut from the plate respectively in each run. Experiments are performed according to the run order as shown in Table 3. The experimental results are collected for DD on 4-axis Sprintcut 734 CNC wire cut machine. Table 3 summarizes the data related to dimensional deviation obtained for 29 experiments by following the design plan of BBD. The output response in the present study is DD in μm. ‘Dimensional Deviation’ is defined as the difference in the actual profile traced by the wire with the job profile required. DD of a square punch is equal to the half the width of cut. DD is measured using the digital micrometer, with a least count of 0.01mm. It is measured at two random places on sides AB, BC and CD and the average of these six values represents the dimensional deviation used in the present article. DD is calculated as [15]:
DD = 0.5 × (Wd − Wa )
(1)
where (Wd-Wa) = width of the cut. Wd = desired size of work piece = 8 mm. Wa = actual size of workpiece obtained after machining. 3. Results and discussion
Table 1 Mechanical properties of ZrSiO4p/6063 Al MMC and base metal Al 6063. Specimen
Tensile strength (KN/mm2)
% elongation
Hardness
Aluminium 6063 ZrSiO4(p)/6063 Al MMC
0.117 0.152
14 to 19 15
68 74.8
Quadratic model is recommended by design expert software (6.0.8) for dimensional deviation after performing three tests namely sum of square test, lack of fit test and model summary statics on the sets of observation data. ANOVA is applied to identify the significance of process parameters towards dimensional deviation. 7
Composites Communications 6 (2017) 6–10
M.P. Garg, A. Sharma
Fig. 1. (a) Crucible (b) Muffle furnace (c) Nitrogen oxide gas cylinder.
Table 3 Design and results for WEDM Output Response.
Fig. 2. Box–Behnken Design as Derived from a Cube [13].
Table 2 Process parameters with their ranges. Variable parameters
Pulse On Time (TON) Pulse Off Time (TOFF) Peak Current (IP) Servo Voltage (SV)
Coded factor
Levels
Units
Fixed parameters
Description
I
II
III
A
112
116
120
mu
Wire Feed
8 m/min
B
50
55
60
mu
8 Kg/cm2
C
120
150
180
amp
Water Pressure Wire Tension
D
50
65
80
volts
Servo Feed
2200
8g
S.No.
Run order
TON
TOFF
IP
SV
DD
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
6 15 16 19 9 25 10 5 1 3 11 2 12 7 14 28 27 17 24 13 18 20 22 8 29 26 21 4 23
112 120 112 120 116 116 116 116 112 120 112 120 116 116 116 116 112 120 112 120 116 116 116 116 116 116 116 116 116
50 50 60 60 55 55 55 55 55 55 55 55 50 60 50 60 55 55 55 55 50 60 50 60 55 55 55 55 55
150 150 150 150 120 180 120 180 150 150 150 150 120 120 180 180 120 120 180 180 150 150 150 150 150 150 150 150 150
65 65 65 65 50 50 80 80 50 50 80 80 65 65 65 65 65 65 65 65 50 50 80 80 65 65 65 65 65
115 460 110 440 255 425 225 415 135 480 85 470 190 270 440 410 77 390 170 490 430 380 415 350 360 340 315 310 297
Legend TON = Pulse on time (mu), TOFF = Pulse off time (mu), IP = Peak Current (A) and SV = Servo Voltage (V), DD = Dimensional deviation (μm).
Fig. 3. (a) Rectangular profile cut on work piece (8 mm × 8 mm) (b) Four axis Electronica Sprintcut 734 CNC wire cut machine (c) Square punches of 8 mm × 8 mm size.
8
Composites Communications 6 (2017) 6–10
M.P. Garg, A. Sharma
Table 4 ANOVA for response surface of reduced quadratic model of dimensional deviation. Source
Sum of squares
Degree of freedom
Mean square
F -value
Prob (p) > F
% contribution
Model A (TON) C (IP) A2 Residual Lack of Fit Pure Error Cor Total
443,561.6 346,120.3 74,104.08 23,337.14 32,908.31 30,351.11 2557.2 476469.9
3 1 1 1 25 21 4 28
147,853.9 346,120.3 74,104.08 23,337.14 1316.33 1445.29 639.3
112.32 262.94 56.29 17.73
< 0.0001 < 0.0001 < 0.0001 0.0003
72.64 15.55 4.89
2.26
0.2234
Significant Significant Significant Significant Non Significant
Standard Deviation = 36.28, R-Squared = 0.9309, Mean = 318.93, Adj R-Squared = 0.9226, Coefficient of variance = 11.38, Pred R-Squared = 0.9109, PRESS = 42,450.45, Adeq Precision = 36.87. Fig. 4. Main effects of input process parameters on dimensional deviation.
the non-significant terms and Eq. (2) represents the relation between dimensional deviation and process parameters.
Table 5 Most favorable solutions for dimensional deviation.
Dimensional Deviation (μm) = −53415.26961
S. no.
TON
TOFF
IP
SV
DD
Desirability
1 2 3 4 5 6 7 8 9 10
112.02 112.05 112.11 112.03 112.01 112.13 112.09 112.38 112.14 112.00
59.13 59.62 55.60 51.48 55.52 51.65 58.41 58.88 50.78 59.67
120.53 133.20 124.69 124.35 130.79 124.68 124.28 123.06 120.42 125.62
50.35 50.01 62.17 57.72 51.62 64.99 73.83 66.79 62.19 75.51
39.77 74.77 56.98 50.52 65.89 58.47 54.25 71.60 48.06 51.72
1 1 1 1 1 1 1 1 1 1
Selected
+ (877.62990 × TON) 2 + (2.61944 × IP)–(3.59988 × TON )
(2) Eq. 2 shows that the main effects of pulse on time(TON), peak current (IP) and quadratic effect of pulse on time(TON2) have significant effect on dimensional deviation where as pulse off time and servo voltage emerged as insignificant factors. Pulse on time indicates the duration during which current is on [14]. A large value of pulse on time produces discharge energy over a longer duration of time. Material is removed by regularly timed intense sparks and pressure forces resulting from larger discharge energy wave deflect the wire from its pre-programmed path and hence, increases the dimensional deviation. A high value of peak current further increases the energy content of discharge channel leading to larger thrust forces and pressure waves on the wire which deflects the wire electrode from the required path. This increases the dimensional deviation. Fig. 4 shows the effect of various input parameters viz. pulse on time, peak current and pulse off time, servo voltage on dimensional deviation. It indicates that when pulse on time was increased from 112 to 120 μs, the dimensional deviation was increased drastically from 115.33 to 455 μm and similar trend was observed while increasing the peak current from 120 to 180 V, dimensional deviation value increased from 264.181 to 421.348 μm. Dimensional deviation value increases if the energy contained in a pulse is increased to a high value as in case of pulse on time and peak current. A large amount of discharge energy on the work piece surface, results in the melting of large amount of material as well as in the formation of large craters as well as more deflection of wire due to thrust of powerful explosions that increases dimensional deviation. The effect of pulse off time and servo voltage on dimensional deviation was appreciably low. On increasing pulse off time from 50 to 60 μs, the dimensional deviation was decreased minutely from 363.16 to 334 μm and similar trend was observed while
Table 6 Confirmatory experiments for Dimensional Deviation (DD). No.
TON
TOFF
IP
SV
Predicted DD
Actual DD
Error %
1 2
112 112
59 60
120 133
50 50
39.77 74.77
40.212 73.225
-1.11 2.066
Legend: TON = Pulse on time (mu), TOFF = Pulse off time (mu), IP = Peak current (A), SV = Servo voltage (V) and DD = Dimensional Deviation (μm).
3.1. Effect of process parameters on dimensional deviation ANOVA analysis is carried out to determine the significant process parameters and is represented in Table 4. Table 4 shows that the pvalue for lack of fit is 0.223 which implies that it is insignificant. Moreover, F-value of the model is 112.32, suggesting that quadratic model adequately fits the data. An R2 value of 0.9309 indicates that model explain 93.09% of the total variability inherent in the system. Value of predicted R2 of 0.9109 is in reasonable agreement with the adjusted R2 of 0.9226, which indicates high correlation between the obtained and predicted values. The values of R2, adjusted R2 obtained in the study are in close agreement with the studies of Garg et. Al. [2] and Kumar et. Al. [16]. Backward elimination is employed to eliminate 9
Composites Communications 6 (2017) 6–10
M.P. Garg, A. Sharma
• Optimum value of dimensional deviation 40.212 μm is obtained
increasing the servo voltage from 50 to 80 V, dimensional deviation value decreased from 374.617 to 340.45 μm.
•
4. Optimization of process parameters In the present study an effort is made to find the machine settings to obtain minimum DD. In this method, the measured property of the output response is transformed to a dimensionless value (d) where d varies between 0 and 1. The desirability function value 0 suggests that the response is completely unacceptable and d = 1 indicates that the response is exactly at the target value. The value of d increases as the desirability of the corresponding response increases [17]. The factor setting with maximum total desirability is considered to be optimal parameter combination. Optimization is carried out using design expert 6.0.8 software. The 10 solutions obtained using desirability approach are shown in Table 5. The highest desirability value of 1 is obtained at TON = 112.02, TOFF = 59.13, IP = 120.53 and SV = 50.35. At these values minimum dimensional deviation is obtained while machining of ZrSiO4(p)/6063 Al MMC. Confirmatory experiments carried out for solution no. 1 and solution no. 2. The process parameter values are rounded off to nearest integer value and Table 6 indicates that error obtained is less than 5%. It confirms reproducibility of results.
References [1] B.H. Yan, C.C. Wang, Machinability of sic particle reinforced aluminum alloy composite material, J. Jpn. Inst. Light Met. 43 (4) (1993) 187–192. [2] M.P. Garg, A. Jain, G. Bhushan, Multi- objective optimization of process parameters in wire electric discharge machining of Ti 6-2-4-2 alloy, Arab. J. Sci. Eng. 39 (2) (2012) 1465–1476. [3] N.P. Hung, L.J. Yang, K.W. Leong, Electro discharge machining of cast metal matrix composites, J. Mat. Process. Technol. 44 (1994) 229–236. [4] A.E. Nitsham, New application for Al based MMC, Light Met. Age, USA 54 (1997). [5] P. Saha, D. Tarafdar, S.K. Pal, A.K. Srivastava, K. Das, Modelling of wire electrodischarge machining of Tic/Fe in situ metal matrix composite using normalized RGFN with enhanced K-means clustering technique, Int. J. Adv. Manuf. Technol. 43 (2009) 107–116. [6] N.G. Patil, P.K. Brahmankar, Some studies into wire electro-discharge machining of alumina particulate reinforced aluminium matrix composites, Int. J. Adv. Manuf. Technol. 48 (2010) 537–555. [7] A. Shah, N.A. Mufti, D. Rakwal, E. Bamberg, Material removal rate, kerf, and surface roughness of tungsten carbide machined with wire electrical discharge machining, J. Mater. Eng. Perform. 20 (1) (2010) 71–76. [8] S.K. Garg, A. Manna, A. Jain, Experimental investigation of spark gap and material removal rate of Al/ZrO2 (P)-MMC machined with wire EDM, J. Braz. Soc. Mech. Sci. Eng. 38 (2) (2016) 481–491. [9] R. Khanna, H. Singh, Performance analysis for D-3 material using response surface methodology on WEDM, Int. J. Mach. Mach. Mater. 14 (1) (2013) 45–65. [10] R.T. Yang, C.J. Tzeng, Y.K. Yang, M.H. Hsieh, Optimization of wire electric discharge machining parameters for cutting tungsten, Int. J. Adv. Manuf. Technol. 60 (2011) 135–147. [11] P. Shandilyan, P.K. Jain, N.K. Jain, Modeling and analysis of surface roughness in WEDC of SiCp/6061 Al MMC through response surface methodology, Int. J. Eng. Sci. Technol. 3 (1) (2010) 531–535. [12] E. Hamed, A. Sarkar, Application of multiple response optimization technique to extended release formulations design, J. Control. 73 (2001) 329–338. [13] S.L. Ferreira, R.E. Bruns, H.S. Ferreira, G.D. Matos, J.M. David, G.C. Brandao, E.G. Da Silva, L.A. Portugal, P.S. Dos Reis, A.S. Souza, W.N. Dos Santos, Box–Behnken design: an alternative for the optimization of analytical methods, Anal. Chim. Acta 597 (2013) 179–186. [14] A. Sharma, M.P. Garg, K.K. Goyal, Prediction of optimal conditions for WEDM of Al 6063/ZrSiO4(p) MMC, Proc. Mater. Sci. 6 (2014) 1024–1033. [15] S. Sarkar, S. Mitra, B. Bhattacharyya, Parametric analysis and optimization of wire electrical discharge machining of γ-titanium aluminide alloy, J. Mat. Process. Technol. 159 (3) (2005) 286–294. [16] A. Kumar, J. Garg, V. Kumar, Multi-response optimization of process parameters based on response surface methodology for pure titanium using WEDM process, Int. J. Adv. Manuf. Technol. 68 (9) (2013) 2645–2668. [17] G. Derringer, R. Suich, Simultaneous optimization of several response variables, J. Qual. Technol. 12 (9) (1980) 214–219.
5. Conclusion In the present research work, an experimental plan of the BoxBehnken design based on the RSM has been applied to perform the experimentation work. The quadratic model for dimensional deviation is developed to investigate the effect of machining parameters namely pulse on time, pulse off time, peak current and servo voltage in the WEDM of newly developed ZrSiO4(p)/6063 Al MMC. From this study, following conclusions are drawn:
• The properties of newly developed ZrSiO •
when the machine is set at pulse on time at 112 mu, pulse off time at 59 mu, peak current at 120 A and servo voltage at 50 V. The results of ANOVA and comparison of experimental results proved that mathematical models of DD are fairly well fitted with experimental values within 95% confidence level.
4p Al 6063 metal matrix composite are much improved as compared to conventional Al 6063 material. It was found experimentally that increasing the pulse on time and peak current, the Dimensional Deviation (DD) increases, whereas no significant effect of pulse off time and servo voltage is observed on dimensional deviation. Due to large values of pulse on time and peak current, intense discharge energy impinges over the workpiece for longer duration of time. This leads to larger pressure as well as thrust forces on wire that increases the deviation of wire from its programmed path and hence, increased dimensional deviation.
10
Contents lists available at ScienceDirect
Composites Communications journal homepage: www.elsevier.com/locate/coco
Examination of accuracy aspect in machining of ZrSiO4p/6063 aluminium MMC using CNC Wire Electrical Discharge Machining Mohinder Pal Garga, Anand Sharmab, a b
MARK
⁎
DAV University, Jalandhar, India Ansal University, Gurgaon, India
A R T I C L E I N F O
A B S T R A C T
Keywords: WEDM dimensional deviation RSM ANOVA
Wire Electrical Discharge Machining (WEDM) is a deep-rooted non-contact modern machining technique used to manufacture geometrically intricate shapes in hard materials which are difficult to machine using conventional methods. To achieve close dimensional accuracy in WEDM, the output parameter dimensional deviation is an essential response to be controlled during machining. Determination of dimensional deviation for a particular process parameter combination assists the manufacturing planners in setting a wire offset so that the actual (cut) work piece dimensions match the part geometry.This study investigates the dimensional deviation induced by WEDM of ZrSiO4p/6063 Aluminium metal matrix composite (MMC) by using response surface methodology (RSM). The key WEDM input process parameters namely, pulse-on time, pulse-off time, servo voltage and peak current were varied in order to examine their influence on dimensional deviation. Significant process parameters affecting the process are identified by carrying out analysis of variance (ANOVA) technique. To aid in selecting the best combination of input parameters during WEDM of ZrSiO4p/6063 Aluminium MMC the concept of desirability is utilized. Confirmation experiments have been conducted to verify the optimal parameter combinations.
1. Introduction There is a great need for materials with special properties with emergence of new stringent requirements posed by manufacturing industries. Traditional engineering materials are unable to meet these requirements of special properties like high strength, low weight, hot hardness, temperature resistance for various industries like aerospace, automotive etc. Composite materials emerged as a new class of engineering materials to cater the needs of these aforesaid industries. Composites are a unique class of materials formed by combining two or more distinct materials to form stronger, tougher and more durable material than each would have been separately. The extreme environment in space a typical spacecraft encounter with high temperature while entering the earth’s atmosphere, naturally occurring phenomena such as vacuum, thermal and ionizing radiation, along with factors such as micro or macro meteoroids and debris therefore, demand lightweight space structures with high strength and dimensional accuracy. Nevertheless, such materials are difficult to be machined by traditional machining methods. The presence of high strength reinforcements in the MMC leads to the rapid wear and tear of cutting tools during machining by conventional or traditional methods [1]. This leads to low
⁎
Correspondıng author. E-mail address: [email protected] (A. Sharma).
http://dx.doi.org/10.1016/j.coco.2017.07.002 Received 4 November 2016; Received in revised form 3 May 2017; Accepted 11 July 2017 2452-2139/ © 2017 Elsevier Ltd. All rights reserved.
cutting rate and subsequent increase in production cost. Therefore, there is a need to enhance the machining performance for these materials. Hence, non-traditional machining methods including abrasive jet machining (AJM), laser beam machining (LBM), water jet machining (WJM), electrical discharging machine (EDM) etc. are applied to machine such difficult-to-cut materials. Techniques like EDM and WEDM are quite successful for machining of high strength MMC’s. EDM can be used only for drilling purpose, while WEDM conforms to easy control and can machine intricate and complex shapes as a result of which WEDM technique is highly recommended for machining of MMC’s [2]. WEDM is a specialized machining technique which uses electro-thermal mechanism to produce complicated cut-outs through electrically conductive and difficult-to-machine materials. Generally, WEDM is perceived to be an accurate non-traditional process. Aluminium based MMC’s also known as aluminium matrix composites (AMC’s) have demonstrated improved mechanical properties compared to properties of pure or non-reinforced Al alloys. Because of their attractive properties, relative ease in fabrication technology and their potential to be available at low cost, Aluminium-matrix composites (AMCs) are being widely used by various industries [3]. AMCs reinforced with hard ceramic particles have emerged as a potential
Composites Communications 6 (2017) 6–10
M.P. Garg, A. Sharma
One of the main purposes of the present study is to investigate the effects of key input process parameters such as, pulse on time, pulse off time, servo voltage and peak current on the output response namely dimensional deviation (DD) and to obtain optimal parameter combinations using desirability approach. This study also aims to establish an empirical relationship between various input process parameters and output response using ANOVA for machining of MMC using Box Behnken Designs (BBD). BBD are one class of the experimental designs for response surface methodology. They are rotatable or nearly rotatable based on three-level incomplete factorial designs. Rotatable means that the model would possess a reasonably stable distribution of scaled prediction variance throughout the experimental design region. BBD allows calculations of the response function at intermediate levels and enables estimation of the system performance at any experimental point within the range studied through careful design and analysis of experiments [12]. For three factors BBD, its graphical representation can be seen as a cube that consists of the central point and the middle points of the edges as shown in Fig. 2 [13].
material especially for wear-resistant and weight critical applications such as brake drums, cylinder liners, pistons, cylinder blocks, connecting rods, and so on [13]. Past studies have indicated that efficient and economical WEDM processing of these materials has opened new areas of applications for MMC’s, but there is a lack of efforts made to investigate machining of MMC’s through WEDM process and particularly of Al-based MMC’s. Literature surveys indicate that most of the reported research is concentrated on machining of conventional materials like steel, brass, titanium etc. than on aluminium based materials [4–11]. 1.1. Fabrication of ZrSiO4p/6063 al metal matrix composite The fabrication methodology of a MMC can be clustered under solid state, liquid state and powder metallurgy, based on the physical state of the matrix i.e., solid phase and liquid phase. ZrSiO4p/6063 Al MMC’s can be widely used materials in many engineering applications due to their improved properties but none of the research effort is made to investigate the potential of WEDM in machining of ZrSiO4p/6063 Al MMC. ZrSiO4p/6063 Al MMC offers a wide variety of applications including wear-resistant components, punches, engine parts, high temperature applications, medical and biomedical purposes etc. Zirconium silicate or zircon (ZrSiO4) reinforced particles of average particle size 5μm, is used for casting of Al 6063 MMC by stircasting technique. Table 1 represents the chemical composition of Al 6063 alloy used as a base metal for fabrication of MMC. In this study, commercially available Al 6063 in the form of cylindrical rods is used as matrix reinforced with ZrSiO4 particles. The melting of measured quantity of (3 kg) pickled Al 6063 is carried out in a clay-graphite crucible (Fig. 1a) placed inside the muffle furnace. Fig. 1b shows the muffle furnace and the developed stirring setup for melting and mixing of Al 6063 and ZrSiO4 particles. Prior heating of required quantity of zirconium silicate (ZrSiO4p) particulates 150gm (5 wt %) to around 500 °C for 1 hour was carried out in a separate furnace. Preheating of ZrSiO4 particles before mixing it with molten metal was purposely done to eliminate the water vapours present within the particles and also to improve the wettability by removing the absorbed hydroxide and other gases. The furnace temperature is first raised above 1000 °C (50° above the melting point of Al 6063) for proper liquidification of Al 6063. At this stage, the pre-heated ZrSiO4 particles are added to the molten metal and mixed mechanically using stirrer. While adding the zircon particles to the molten aluminium alloy, to avoid atmospheric contamination nitrogen gas (Fig. 1c) is used as the shield of the mixer. The composite slurry is then reheated to a fully liquid state and mechanical mixing was carried out for 5 min at 100 rpm average stirring speed. Stirring is carried out to ensure even temperature distribution as well as particulate distribution. The mixer of reinforced particulate and the molten metal is then poured to a prepared sand mould. After pouring is over, the melt is allowed to cool and solidify in the mould under the shield of N2. After solidification, the casting is taken from the mould and is cut to the required shape and size (110mm x 80mm x 11mm) for subsequent WEDM. The improved properties of ZrSiO4p/6063 Al MMC and the conventional properties of base metal Al 6063 are shown in Table 1. It reveals that the important mechanical properties are improved to a good extent by addition of 5% ZrSiO4p to base metal Al 6063 and it can effectively replace conventional Al 6063 material in automobile and aerospace applications owing to its improved properties.
2. Design of experiments and experimentation Various input process parameters viz. Pulse on time (TON), Pulse off time (TOFF), Peak Current (IP) and Servo Voltage (SV) are varied to investigate their effects on DD during machining of ZrSiO4p/6063 Al MMC. The ranges of input parameters were selected on the basis of literature survey, machining capability of the machine and pilot study conducted by using one variable at a time approach [14]. The range of selected process parameters for further analysis is presented in Table 2. Table 2 also lists the parameters and there levels which were kept fixed in the study. The experiments were performed on a four axis Electronica Sprintcut 734 CNC Wire Cut Machine. Diffused brass wire of 0.25mm diameter was used as tool material and deionized water is used as dielectric. A plate of rectangular shape (110 mm × 80 mm × 11 mm) having 5% ZrSiO4 particles (by weight) as reinforcement is fixed on the machine table. An 8 mm × 8 mm rectangular cut is taken on the work piece as shown in Fig. 3a. Fig. 3(b,c) represents the selected machine tool and the rectangular pieces cut from the plate respectively in each run. Experiments are performed according to the run order as shown in Table 3. The experimental results are collected for DD on 4-axis Sprintcut 734 CNC wire cut machine. Table 3 summarizes the data related to dimensional deviation obtained for 29 experiments by following the design plan of BBD. The output response in the present study is DD in μm. ‘Dimensional Deviation’ is defined as the difference in the actual profile traced by the wire with the job profile required. DD of a square punch is equal to the half the width of cut. DD is measured using the digital micrometer, with a least count of 0.01mm. It is measured at two random places on sides AB, BC and CD and the average of these six values represents the dimensional deviation used in the present article. DD is calculated as [15]:
DD = 0.5 × (Wd − Wa )
(1)
where (Wd-Wa) = width of the cut. Wd = desired size of work piece = 8 mm. Wa = actual size of workpiece obtained after machining. 3. Results and discussion
Table 1 Mechanical properties of ZrSiO4p/6063 Al MMC and base metal Al 6063. Specimen
Tensile strength (KN/mm2)
% elongation
Hardness
Aluminium 6063 ZrSiO4(p)/6063 Al MMC
0.117 0.152
14 to 19 15
68 74.8
Quadratic model is recommended by design expert software (6.0.8) for dimensional deviation after performing three tests namely sum of square test, lack of fit test and model summary statics on the sets of observation data. ANOVA is applied to identify the significance of process parameters towards dimensional deviation. 7
Composites Communications 6 (2017) 6–10
M.P. Garg, A. Sharma
Fig. 1. (a) Crucible (b) Muffle furnace (c) Nitrogen oxide gas cylinder.
Table 3 Design and results for WEDM Output Response.
Fig. 2. Box–Behnken Design as Derived from a Cube [13].
Table 2 Process parameters with their ranges. Variable parameters
Pulse On Time (TON) Pulse Off Time (TOFF) Peak Current (IP) Servo Voltage (SV)
Coded factor
Levels
Units
Fixed parameters
Description
I
II
III
A
112
116
120
mu
Wire Feed
8 m/min
B
50
55
60
mu
8 Kg/cm2
C
120
150
180
amp
Water Pressure Wire Tension
D
50
65
80
volts
Servo Feed
2200
8g
S.No.
Run order
TON
TOFF
IP
SV
DD
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
6 15 16 19 9 25 10 5 1 3 11 2 12 7 14 28 27 17 24 13 18 20 22 8 29 26 21 4 23
112 120 112 120 116 116 116 116 112 120 112 120 116 116 116 116 112 120 112 120 116 116 116 116 116 116 116 116 116
50 50 60 60 55 55 55 55 55 55 55 55 50 60 50 60 55 55 55 55 50 60 50 60 55 55 55 55 55
150 150 150 150 120 180 120 180 150 150 150 150 120 120 180 180 120 120 180 180 150 150 150 150 150 150 150 150 150
65 65 65 65 50 50 80 80 50 50 80 80 65 65 65 65 65 65 65 65 50 50 80 80 65 65 65 65 65
115 460 110 440 255 425 225 415 135 480 85 470 190 270 440 410 77 390 170 490 430 380 415 350 360 340 315 310 297
Legend TON = Pulse on time (mu), TOFF = Pulse off time (mu), IP = Peak Current (A) and SV = Servo Voltage (V), DD = Dimensional deviation (μm).
Fig. 3. (a) Rectangular profile cut on work piece (8 mm × 8 mm) (b) Four axis Electronica Sprintcut 734 CNC wire cut machine (c) Square punches of 8 mm × 8 mm size.
8
Composites Communications 6 (2017) 6–10
M.P. Garg, A. Sharma
Table 4 ANOVA for response surface of reduced quadratic model of dimensional deviation. Source
Sum of squares
Degree of freedom
Mean square
F -value
Prob (p) > F
% contribution
Model A (TON) C (IP) A2 Residual Lack of Fit Pure Error Cor Total
443,561.6 346,120.3 74,104.08 23,337.14 32,908.31 30,351.11 2557.2 476469.9
3 1 1 1 25 21 4 28
147,853.9 346,120.3 74,104.08 23,337.14 1316.33 1445.29 639.3
112.32 262.94 56.29 17.73
< 0.0001 < 0.0001 < 0.0001 0.0003
72.64 15.55 4.89
2.26
0.2234
Significant Significant Significant Significant Non Significant
Standard Deviation = 36.28, R-Squared = 0.9309, Mean = 318.93, Adj R-Squared = 0.9226, Coefficient of variance = 11.38, Pred R-Squared = 0.9109, PRESS = 42,450.45, Adeq Precision = 36.87. Fig. 4. Main effects of input process parameters on dimensional deviation.
the non-significant terms and Eq. (2) represents the relation between dimensional deviation and process parameters.
Table 5 Most favorable solutions for dimensional deviation.
Dimensional Deviation (μm) = −53415.26961
S. no.
TON
TOFF
IP
SV
DD
Desirability
1 2 3 4 5 6 7 8 9 10
112.02 112.05 112.11 112.03 112.01 112.13 112.09 112.38 112.14 112.00
59.13 59.62 55.60 51.48 55.52 51.65 58.41 58.88 50.78 59.67
120.53 133.20 124.69 124.35 130.79 124.68 124.28 123.06 120.42 125.62
50.35 50.01 62.17 57.72 51.62 64.99 73.83 66.79 62.19 75.51
39.77 74.77 56.98 50.52 65.89 58.47 54.25 71.60 48.06 51.72
1 1 1 1 1 1 1 1 1 1
Selected
+ (877.62990 × TON) 2 + (2.61944 × IP)–(3.59988 × TON )
(2) Eq. 2 shows that the main effects of pulse on time(TON), peak current (IP) and quadratic effect of pulse on time(TON2) have significant effect on dimensional deviation where as pulse off time and servo voltage emerged as insignificant factors. Pulse on time indicates the duration during which current is on [14]. A large value of pulse on time produces discharge energy over a longer duration of time. Material is removed by regularly timed intense sparks and pressure forces resulting from larger discharge energy wave deflect the wire from its pre-programmed path and hence, increases the dimensional deviation. A high value of peak current further increases the energy content of discharge channel leading to larger thrust forces and pressure waves on the wire which deflects the wire electrode from the required path. This increases the dimensional deviation. Fig. 4 shows the effect of various input parameters viz. pulse on time, peak current and pulse off time, servo voltage on dimensional deviation. It indicates that when pulse on time was increased from 112 to 120 μs, the dimensional deviation was increased drastically from 115.33 to 455 μm and similar trend was observed while increasing the peak current from 120 to 180 V, dimensional deviation value increased from 264.181 to 421.348 μm. Dimensional deviation value increases if the energy contained in a pulse is increased to a high value as in case of pulse on time and peak current. A large amount of discharge energy on the work piece surface, results in the melting of large amount of material as well as in the formation of large craters as well as more deflection of wire due to thrust of powerful explosions that increases dimensional deviation. The effect of pulse off time and servo voltage on dimensional deviation was appreciably low. On increasing pulse off time from 50 to 60 μs, the dimensional deviation was decreased minutely from 363.16 to 334 μm and similar trend was observed while
Table 6 Confirmatory experiments for Dimensional Deviation (DD). No.
TON
TOFF
IP
SV
Predicted DD
Actual DD
Error %
1 2
112 112
59 60
120 133
50 50
39.77 74.77
40.212 73.225
-1.11 2.066
Legend: TON = Pulse on time (mu), TOFF = Pulse off time (mu), IP = Peak current (A), SV = Servo voltage (V) and DD = Dimensional Deviation (μm).
3.1. Effect of process parameters on dimensional deviation ANOVA analysis is carried out to determine the significant process parameters and is represented in Table 4. Table 4 shows that the pvalue for lack of fit is 0.223 which implies that it is insignificant. Moreover, F-value of the model is 112.32, suggesting that quadratic model adequately fits the data. An R2 value of 0.9309 indicates that model explain 93.09% of the total variability inherent in the system. Value of predicted R2 of 0.9109 is in reasonable agreement with the adjusted R2 of 0.9226, which indicates high correlation between the obtained and predicted values. The values of R2, adjusted R2 obtained in the study are in close agreement with the studies of Garg et. Al. [2] and Kumar et. Al. [16]. Backward elimination is employed to eliminate 9
Composites Communications 6 (2017) 6–10
M.P. Garg, A. Sharma
• Optimum value of dimensional deviation 40.212 μm is obtained
increasing the servo voltage from 50 to 80 V, dimensional deviation value decreased from 374.617 to 340.45 μm.
•
4. Optimization of process parameters In the present study an effort is made to find the machine settings to obtain minimum DD. In this method, the measured property of the output response is transformed to a dimensionless value (d) where d varies between 0 and 1. The desirability function value 0 suggests that the response is completely unacceptable and d = 1 indicates that the response is exactly at the target value. The value of d increases as the desirability of the corresponding response increases [17]. The factor setting with maximum total desirability is considered to be optimal parameter combination. Optimization is carried out using design expert 6.0.8 software. The 10 solutions obtained using desirability approach are shown in Table 5. The highest desirability value of 1 is obtained at TON = 112.02, TOFF = 59.13, IP = 120.53 and SV = 50.35. At these values minimum dimensional deviation is obtained while machining of ZrSiO4(p)/6063 Al MMC. Confirmatory experiments carried out for solution no. 1 and solution no. 2. The process parameter values are rounded off to nearest integer value and Table 6 indicates that error obtained is less than 5%. It confirms reproducibility of results.
References [1] B.H. Yan, C.C. Wang, Machinability of sic particle reinforced aluminum alloy composite material, J. Jpn. Inst. Light Met. 43 (4) (1993) 187–192. [2] M.P. Garg, A. Jain, G. Bhushan, Multi- objective optimization of process parameters in wire electric discharge machining of Ti 6-2-4-2 alloy, Arab. J. Sci. Eng. 39 (2) (2012) 1465–1476. [3] N.P. Hung, L.J. Yang, K.W. Leong, Electro discharge machining of cast metal matrix composites, J. Mat. Process. Technol. 44 (1994) 229–236. [4] A.E. Nitsham, New application for Al based MMC, Light Met. Age, USA 54 (1997). [5] P. Saha, D. Tarafdar, S.K. Pal, A.K. Srivastava, K. Das, Modelling of wire electrodischarge machining of Tic/Fe in situ metal matrix composite using normalized RGFN with enhanced K-means clustering technique, Int. J. Adv. Manuf. Technol. 43 (2009) 107–116. [6] N.G. Patil, P.K. Brahmankar, Some studies into wire electro-discharge machining of alumina particulate reinforced aluminium matrix composites, Int. J. Adv. Manuf. Technol. 48 (2010) 537–555. [7] A. Shah, N.A. Mufti, D. Rakwal, E. Bamberg, Material removal rate, kerf, and surface roughness of tungsten carbide machined with wire electrical discharge machining, J. Mater. Eng. Perform. 20 (1) (2010) 71–76. [8] S.K. Garg, A. Manna, A. Jain, Experimental investigation of spark gap and material removal rate of Al/ZrO2 (P)-MMC machined with wire EDM, J. Braz. Soc. Mech. Sci. Eng. 38 (2) (2016) 481–491. [9] R. Khanna, H. Singh, Performance analysis for D-3 material using response surface methodology on WEDM, Int. J. Mach. Mach. Mater. 14 (1) (2013) 45–65. [10] R.T. Yang, C.J. Tzeng, Y.K. Yang, M.H. Hsieh, Optimization of wire electric discharge machining parameters for cutting tungsten, Int. J. Adv. Manuf. Technol. 60 (2011) 135–147. [11] P. Shandilyan, P.K. Jain, N.K. Jain, Modeling and analysis of surface roughness in WEDC of SiCp/6061 Al MMC through response surface methodology, Int. J. Eng. Sci. Technol. 3 (1) (2010) 531–535. [12] E. Hamed, A. Sarkar, Application of multiple response optimization technique to extended release formulations design, J. Control. 73 (2001) 329–338. [13] S.L. Ferreira, R.E. Bruns, H.S. Ferreira, G.D. Matos, J.M. David, G.C. Brandao, E.G. Da Silva, L.A. Portugal, P.S. Dos Reis, A.S. Souza, W.N. Dos Santos, Box–Behnken design: an alternative for the optimization of analytical methods, Anal. Chim. Acta 597 (2013) 179–186. [14] A. Sharma, M.P. Garg, K.K. Goyal, Prediction of optimal conditions for WEDM of Al 6063/ZrSiO4(p) MMC, Proc. Mater. Sci. 6 (2014) 1024–1033. [15] S. Sarkar, S. Mitra, B. Bhattacharyya, Parametric analysis and optimization of wire electrical discharge machining of γ-titanium aluminide alloy, J. Mat. Process. Technol. 159 (3) (2005) 286–294. [16] A. Kumar, J. Garg, V. Kumar, Multi-response optimization of process parameters based on response surface methodology for pure titanium using WEDM process, Int. J. Adv. Manuf. Technol. 68 (9) (2013) 2645–2668. [17] G. Derringer, R. Suich, Simultaneous optimization of several response variables, J. Qual. Technol. 12 (9) (1980) 214–219.
5. Conclusion In the present research work, an experimental plan of the BoxBehnken design based on the RSM has been applied to perform the experimentation work. The quadratic model for dimensional deviation is developed to investigate the effect of machining parameters namely pulse on time, pulse off time, peak current and servo voltage in the WEDM of newly developed ZrSiO4(p)/6063 Al MMC. From this study, following conclusions are drawn:
• The properties of newly developed ZrSiO •
when the machine is set at pulse on time at 112 mu, pulse off time at 59 mu, peak current at 120 A and servo voltage at 50 V. The results of ANOVA and comparison of experimental results proved that mathematical models of DD are fairly well fitted with experimental values within 95% confidence level.
4p Al 6063 metal matrix composite are much improved as compared to conventional Al 6063 material. It was found experimentally that increasing the pulse on time and peak current, the Dimensional Deviation (DD) increases, whereas no significant effect of pulse off time and servo voltage is observed on dimensional deviation. Due to large values of pulse on time and peak current, intense discharge energy impinges over the workpiece for longer duration of time. This leads to larger pressure as well as thrust forces on wire that increases the deviation of wire from its programmed path and hence, increased dimensional deviation.
10