Functional Composite Materials: Manufacturing Technology and Experimental Application

Optimization of Machining Process Parameters to Calculate Surface Finish, Metal Removal Rate, and Cutting Force on AA6061-T6 Alloy

Karthik Pandiyan G.¹, *, T. Prabaharan², Jafrey Daniel James D.³

¹ Department of Mechanical Engineering, Sri Vidya College of Engineering and Technology, Virudhunagar, Tamil Nadu, India

² Department of Mechanical Engineering, Mepco Schlenk Engineering College, Sivakasi, Tamil Nadu, India

³ Department of Mechanical Engineering, K.Ramakrishnan College of Engineering, Trichy, Tamil Nadu, India

Abstract

In this study, CNC turning process parameters are optimized while machining AA6061 – T6 alloy. The process variables, namely, cutting spindle speed or spindle velocity (Cs), rate of feed (Fr), rate of the depth of cut (Dc), and nose radius (Nr), were optimized with output responses such as surface roughness (Ra), feed power (Fx), thrust power (Fy), cutting power (Fz), and removal rate of material (MRR). The optimized results showed higher MRR, Fy, and Fz and lower Fx and Ra were obtained at a cutting spindle speed or spindle velocity (Cs) of 100 rev/min, rate of feed (Fr) of 0.2 mm/min, depth of cut (Dc) of 1.5 mm/min, and nose radius (Nr) of 0.8 mm. TOPSIS and ANOVA were used to assess the experimental results to determine the most favourable factors that were superior and suitable for selecting the optimal variable condition.

Keywords: AA6061-T6 alloy, ANOVA, CNC Turning, Cutting power, Depth of the cut, Feed power, Material removal rate, Multiresponse Optimization, Nose radius, Rate of feed, Spindle Speed, Surface roughness, Thrust power, TOPSIS.

* Corresponding author Karthik Pandiyan G.: Department of Mechanical Engineering, Sri Vidya College of Engineering and Technology, Virudhunagar, Tamil Nadu, India; E-mail: [email protected]

INTRODUCTION

Aluminium is ductile and has an FCC (Face Centred Cubic) structure. Al is a lighter metal with 2900 kg/m³ as density. It has higher electrical and thermal conductivity due to its lower density. Various wrought aluminium products include mechanical finishes, chemical finishes, and coatings.

Mechanical finishes include cold finished, buffed, and textured coatings, whereas chemical finishes include etched, bright dipped, and chemical conversion coatings. Anodizing, painting, plating, and chemical conversion coatings are included in the coating category. Due to low density to high strength ratio, toughness, and corrosion resistance, aluminium alloys are mainly used in automobiles and aerospace Industries. Zn, Si, Mg, Cu, are major alloying elements that are classified into cast-based alloys and wrought-based alloys with heat-treatable and non-heat-treatable properties.

Chemical composition of AA6061-T6 (T6- heat-treated, rapid cooling and aged) is Al - 97.25%; Cu - 0.32%; Mg - 1.08%; Mn - 0.52%; V - 0.01%; Fe - 0.17%;Sic - 0.63%; Ti - 0.02%, which is a work-hardened alloy with a low density to high strength ratio, increased hardness, improved toughness, capability of welding, and greater resistance to corrosion. The strain hardening method is highly effective in improving the strength of materials.

Due to tempering, materials lose their strength but show increased ductility, high tool wear and require elevated cutting power, and fatigue life of the tool. This results in difficulty of machining. An attempt was made to analyse multi-response characteristics based optimization of AA6061-T6 alloy, utilizing the TOPSIS method. In the previous examination, optimization was carried out based on the Taguchi S/N ratio-based technique. It is used as a tool to ensure better quality and efficient yield production at minimum expenditure and optimizes only a single response characteristic.

Normally, the Taguchi method is used for a single objective. Larger criteria are used for MRR and smaller criteria are used for surface roughness. Multi-responses are more difficult in the Taguchi method and are undertaken for converting the multi-response to a single response. MCDM uses Taguchi S/N ratio, Taguchi-based GRA, VIKOR, TOPSIS, and Analytic Hierarchy Process to determine the weights of each response in Decision-making methods.

An investigation was undertaken into the turning process where the input variable varied with different levels of input variable and output variables achieved as Ra, MRR, FX, FY, and FZ. A lot of research is available on the use of Grey, TOPSIS, and FUZZY TOPSIS to solve multi-objective problems, of which TOPSIS was found the most effective in the choice of variables and machining processes. Optimization combining TOPSIS and Taguchi methods achieved the most favourable conditions with satisfactory turning operations. Evolutionary computational methods have evolved, such as RSM, GA, PSO, and NN. But TOPSIS is still highly capable of solving MCDM problems to optimize the most favourable result.

In this study, Ra, MRR, FX, FY, and FZ were considered to have equal weights. Optimization study was undertaken to achieve surface roughness and wear rate of the tool by employing RSM, GRA, and TOPSIS for prediction [1]. TOPSIS was used at the end milling process to achieve Ra, lower energy, reduce the tool's wear rate, and improve yield in the manufacturing process [2]. Bing et al. experimented with 7050-T7451 aluminium alloy in the milling process to attain enhanced surface features by the ANOVA technique while evaluating specific cutting energy [3]. Gopal et al. used Taguchi-based S/N ratio, the TOPSIS and GRA ranking methods of milling on a carbide tool to predict the best optimum responses like cutting forces, cutting temperature, and Ra. Machining studies were conducted in the end milling process of Mg-based Hybrid metal matrix composite materials by GRA, TOPSIS, and Taguchi techniques [4]. Al alloys with superior surface finish were used to achieve lower surface irregularity, greater fatigue time, higher strength, high rate of wear, corrosion, and creep resistance to produce optimum surface quality [5].

End milling of Al6061 alloy was performed to achieve the best set of variables for the cutting force, which achieved tool deflection and cutting power [6]. Micro-milling of Al6061 alloy was performed to achieve Non-Uniformity of Surface Generation to ensure low feed rate and high depth of cut [7]. Szymon et al. investigated micro-milling of alloy steel to achieve good surface cease and minimum vibrations during milling [8]. Wojciechowski et al. investigated ball milling of hardened steel and achieved decreased deflection of tool diameter and tool extension [9]. Malghan et al. used the milling of AA6061 alloy with Taguchi-based optimization technique to evaluate process variables for greater surface irregularity [10].

Pereira et al. used the milling of Al 7075 alloy through TOPSIS to achieve an elevated removal rate of material, roundness, cutting power, and minimized energy [11]. Nipanikar et al. observed the tool wear of Ti-6Al-4V in a MQL environment using GRA and TOPSIS methods [12]. MQL based turning process was examined to determine the Multi-Objective characteristics of tooltip interaction using the Grey-Taguchi Method [13]. Design and Modelling of Inconel 718 alloy were studied to evaluate Tool Wear rate utilizing MQL [14]. Machining of Al-6082 T6 was performed to understand the characteristics of surface roughness [15].

Milling of Aluminum Alloy by using both Cryogenic enhanced CO2 and LN2 Coolants was studied to evaluate cutting temperature in varying environments to achieve surface roughness [16]. A review was conducted on Aluminium alloys for usage and properties [17]. Taguchi-Grey Relational Approach on End Milling was performed to determine Multiple Response Characteristics [18]. AWJ Cutting of D2 Steel was used to predict multi-response characteristics on process parameters [19]. End Milling Parameters were studied in Cryogenic Cooling Conditions [20].

Karthik et al. studied machining parameters on AA6351 alloy using RSM [21]. Karthik et al. predicted surface roughness using a soft computing method ANN [22]. The study represented an attempt to investigate multi-response characteristics of Turning process variables in AA6061-T6 alloy by using the TOPSIS method

MATERIALS AND METHODS

Experiments were carried out at various levels like cutting spindle speed or spindle velocity (Cs), rate of feed (Fr), depth of cut(Dc), and Nose radius in the turning process. Experiments s(Nr) were conducted in BATLIBOI model on CNC turning equipment. Table 1 demonstrates the machining variables of the AA6061-T6 alloy. Experiments were conducted for three levels of cutting spindle speed or spindle velocity (Cs) at 100, 200, and 300 rev/min, three rates of feed (Fr) at 0.10, 0.15, and 0.20 mm/min, depth of cut (Dc) at 0.5,1.0, and 1.5 mm/min and Nose radius (Nr) of 0.4, 0.6, and 0.8 mm.

Aluminium 6061-T6 alloy specimen with 150 mm length and 25 mm diameter was used as the workpiece. Table 2 shows the elements and composition of the AA 6061-T6 alloy, while Table 3 exhibits the properties of the AA 6061-T6 alloy considered for the study.

TPUN110408 116 type tool holder was used with CNMG12408-PM carbide inserts made by ISCAR. Table 4 displays the technical specifications of the CNMG12408-PM cutting insert.

Ra, Material Removal Rate (MRR), and cutting forces were measured in each experiment. Fig. (1) shows the decision model of the turning process variables.

Table 1 Machining variables of AA6061-T6 alloy.

Table 2 AA 6061-T6 alloy Elements and its Composition.

Table 3 AA 6061-T6 alloy Properties.

Table 4 Specifications of CNMG 120408-PM cutting insert.

Fig. (2) shows the pictures of the experimental setup, and Fig. (3) shows the Turning process of the workpiece.

Fig. (1))

Decision Model for Turning process variable.

Fig. (2))

Experimental setup.

Fig. (3))

Turning process of the workpiece.

Fig. (4) shows the uncoated CNMG 120408-PM grade-K20 carbide insert tool geometry, and Fig. (5) shows the force measure1ment setup, respectively.

Fig. (4))

Uncoated CNMG 120408-PM grade-K20 carbide insert geometry.

Fig. (5))

Force Measurement Setup.

The experimental setup consisted of a three component piezoelectric type dynamometer used to measure force. The machined workpiece’s surface roughness values were measured by a Mitutoyo surface tester (Make-Japan –Model SJ-301) with a cut-off length of 2.5 mm, speed of 0.5 mm/s, and a measuring range of 12.5 mm. Experiments were conducted with four variables at three levels.

Table 5 describes the Machining process variables at different levels. Table 6 shows the L27 orthogonal array with normalized values used to perform the experiments with each process variable on AA6061-T6 (150 mm (length) and 25 mm (diameter)). Ra, MRR, Fx, Fy, and Fz were considered output responses to locate the performance measures.

Table 5 Machining process variables with different levels.

Table 6 L27 orthogonal array with normalized values.

Steps Involved in TOPSIS Method

Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS) methodology was proposed by Hwang Yoon in 1981. This method uses positive ideal theoretical and negative ideal theoretical solution evaluation methods to calculate the coefficient value.

A Hypothetical solution for all experimental results consists of negative and positive ideal solutions. The obtained results should fall within the minimum and maximum values of the solution and the value of closeness coefficient between the minimum and maximum values.

TOPSIS is a ranking method to find the most favourable value from a set of results obtained from several experiments. The major principle of this method is that the final optimized result should be nearer to the positive values and distant from the negative values.

Fig. (6) shows the mean response value of the process variables. A specific set of output variables [Ra, MRR, Fx, Fy, and Fz] was given equal weight. TOPSIS method was mathematically used as follows [1, 4, 12, 19].

Fig. (6))

Mean Response value of process variable.

Step-I: First of all, the values must be normalized to the matrix by the equation (1) as follows.

Where i=1,2,…..27 j=1,2,…..5,

i denotes the number of observations

j denotes the number of output variables

aij denotes the normalized values of output variables

rij denotes the normalized matrix values of output variables

Step-II: The set of output variables [Ra, MRR, Fx, Fy, Fz] are given equal weight.

Step-III: The normalized weighted matrix with weights is determined using multiplication of the normalization matrix with its related weights in equation (2)

Where i=1, 2, …..27 j=1, 2, …..5,

i denotes the number of observations

j denotes the number of output variables

Vij denotes the normalized weighted matrix of output variables

Wi denotes the weight of the output variable

Step-IV: Determine the weight decision matrix where V+ is the maximum value of the particular output variable and V- is the minimum value of the particular output variable

V+ (V1+ V2+,........ Vn+ maximum values

V- (V1- V2-,........ Vn- maximum values

Step-V: The below equation is used to calculate the values of S+ and S-

Where S+ denotes the best performance, S- denotes the worst performance

where i = 1,2,….27.

Step-VI: The value of Closeness Co-efficient (CCs) is determined by equation (5).

Step-VII: Ranking the CCs result.

The highest CCs value is close to the optimized value.

Results and Discussion

Table 7 displays the L27 orthogonal array with the closeness coefficients standards and its position at every observation.

Table 7 Closeness coefficients standards and its position.