Introduction

Machinability is especially important when it comes to process design and machining processes. For production engineers, knowing the machinability of a work material ahead of time is critical so that manufacturing can be scheduled efficiently [1,2,3]. The machining process performance is influenced by input factors like usage of energy source, dimensional accuracy, tool life, cutting pressures, surface finish, temperature, chip characteristics, vibration, and noise [4,5,6,7]. The input variables used in machining processes are functions of process variables used economically and technically to the output of work materials, which is connected to the various machining conditions needed to achieve high machinability. In order to illustrate the machinability of a chosen workpiece material for a certain machining technique, these factors are employed. When evaluating machinability, some manufacturers place the greatest emphasis on tool life, while others place the greatest emphasis on processed surface polish. When assessing the machinability of various work materials, some researchers have simply taken one performance characteristic for the machining process into account [8,9,10,11,12]

Various researchers proposed the selection procedures took into account a several machining process performance variables, which were analysed in relation to the properties of work material and their characteristics. Each work material’s efficiency as an output variable of the machining process was examined independently. Then, in view of the overall results, a subjective decision is taken about the selection of work material [13,14,15,16,17,18].

The turning process was the traditional method of transforming raw materials into the desired result. The efficiency and output characteristics of CNC machining techniques are superior than manual methods like manual and semi-automatic lathes. The quality of the completed product is assessed by its surface roughness. For a decent surface, choosing the best parameters is essential, particularly in the case of surface roughness [19, 20]. The surface quality may be evaluated using Minitab software. Between the tool and work material, more amount of heat is produced in the machining operation. Hence, coolant is supplied to avoid failure of tool. Since, Coolant itself acts as a lubricant [21,22,23,24]. In most commonly used cutting technique i.e., turning, maintaining component quality in terms of surface finish is critical [25,26,27,28].

B. Kumaragurubaran et.al [8] have optimized the turning parameters of EN8 steel based on DOE concept. From the investigation it has been discovered that feed rate is an important factor affecting the MRR in turning operation. In case of surface roughness, spindle speed is the dominant factor. Ranganath.M S et.al [9] have reported that turning parameters affecting Ra during dry turning process. Ra was shown to be mostly affected by speed. The ideal Ra value may be attained by maintaining a low feed rate and shallow cut. Shukry H. Aghdeb et al. [10] researched to find the ideal value of surface roughness with best turning parameters to use while milling an aluminium shaft. Sanchit Kumar Khare et al.’s [11] evaluation of the surface texture analysis utilising the turning process and the taguchi design. They discovered through the experiment that the rake angle and cutting speed both had an impact on RA. Turning machinability research was done on PH SS with coated inserts by D. Palanisamy et al. [12]. They recommended setting the nose radius to the ideal value and maintaining the parameter cutting speed at a high level with a reduced feed rate in order to obtain improved machinability. The impact of these factors on MRR and Ra is the primary focus of many studies on machinability parameter optimisation.

In their research, Vijay Kumar M et al. [13] examined a turning sample using a CNC machine and determined the impact of the cutting process factors on both MRR and Ra. Spindle speed and feed rate are two factors that affect MRR and Ra, it was found via analysis. It was shown that lubrication and cut depth had no impact on MRR and Ra. Irfan Sayed and others [7] The evaluation of process parameter optimisation for turning EN45 with a simple carbide tool was done using Taguchi’s experiment. It was shown that MRR increases with increasing depth of cut, speed, and feed rate V.V.D. Sahithi et al. [14] examined the impact of machining factors on roughness during turning operations for aluminium 6061 alloy. Since the results of this machined sample of the aluminium 6063 alloy were encouraging, it was compared to the Al6061 alloy. Al6082 was able to provide a great surface polish, nonetheless, by using optimal conditions. Evaluating of the turning process’s long-term viability in the manufacturing industry by Karmjit Singh et al. Experiments on the spherical 6061-T6511 aluminium alloy was turned. Air emissions and energy use are the two categories of environmental concerns that are important for sustainability. The criteria impacting the RA for various types of materials were discussed by M H M Ghazali et al. [15]. Ra reduces steadily as cutting speed rises. The feed rate increases as Ra value increases. Md. Zishanur Rahman et al. [16] investigated the Taguchi turning of AISI-1055 and AISI 4340 steels using regression modelling and comparative analysis. When compared to AISI-1055 steels, AISI-4340 steels exhibit acceptable machinability, which was anticipated based on Ra and MRR values. In this study, depth of cut, feed rate and cutting speed are among the turning factors whose effects on MRR and Ra generation are examined. To increase MRR with the least amount of surface roughness, the machinability parameters must be tuned using the Taguchi optimisation technique.

Experimental Details

Composition of Cu-Al-Mn SMAs

The compositions of Cu-Al-Mn SMAs are shown in Table 1. The compositions for the prepared samples were determined by using a spark emission spectrophotometer.

Table 1 Shows the composition of Cu Al-Mn SMAs
Full size table

Selection of Material

Comparing the current Cu-Al-Mn SMAs to other copper-based SMAs, they show good machinability characteristics. The ductility of the alloy can be improved by increasing the amount of manganese or aluminium present in the copper in a certain range [17, 29,30,31]. SMAs’ machinability can change depending on how they are made, which can impact the components’ precision and quality. Before it is made to the desired shape and size, it must first be properly studied to see if the SMAs can be used. The present analysis chooses copper-based SMAs after taking these aspects into account.

CNC Lathe Machine

Cu-Al-Mn SMAs were turned on a Jobber XL CNC Lathe Machine in this experiment. The CNC lathe machine for the turning process is presented in Fig. 1, and the experimental setup picture of Cu-Al-Mn SMAs mounted on the CNC lathe machine is shown in Fig. 2.

Fig. 1
figure 1

CNC Lathe Machine

Full size image
Fig. 2
figure 2

Experimental setup of Cu-Al-Mn SMA

Full size image

Cutting Tool Selection

According to the literature that is currently accessible, no study has been done on CNC machining of Cu-Al-Mn SMAs up until this point, and no reports are published. In light of the literature previously utilised for other materials, the tool material was chosen. Tungsten carbide tip tools were selected in the current experiment as the cutting tools to perform the CNC turning procedure for Cu-Al-Mn SMAs.

Taguchi Method Using MINITAB

Design of Experiments (DOE)

The Design of Experiments (DOE) technique, which is based on the Taguchi methodology, is used to produce the ideal number of experiments. With an orthogonal array and a variety of potential parameter and range combinations for every trial, Taguchi’s experimental approach is described. This article’s method makes it possible to explore the parameter space in the most effective way possible. For this inquiry, the L9 OA was chosen as a result of the DOE-based DOE. Based on the literature, the current study took into account the ranges of the CNC turning parameters. In order to reduce tool breakage, tool life constraints, and machine capabilities while taking into account all of these factors, the research’s three primary cutting parameters depth of cut, spindle speed, and feed rate were chosen. Cu-Al-Mn SMAs are manufactured as cylindrical tubes with a 100 mm length and a 12 mm diameter. Table 2 lists the cutting process variables and their values, which were taken from the literature to serve as a guide for the current study. The MRR and Ra response factors for the current investigation were selected based on the literature that is currently available [33,34,35,36,37,38,39,40].

Table 2 Cutting parameters at experimentation levels
Full size table

Turning Process

Based on DOE, tests were carried out, and using a Mitutoyo surface roughness tester, the relevant output values of MRR and Ra were recorded. For CAM1, CAM2, and CAM3 SMAs, machining has been done. The data for CNC machining the CAM3 SMA is provided in Table 3.

Table 3 L9 Orthogonal array chosen for Turning of CAM3 SMA
Full size table

Material Removal Rate (MRR)

MRR is calculated by change in workpiece volume to the time taken during turning. Figure 3 indicates the turned samples of Cu-Al-Mn SMAs. MRR is calculated using the formulas represented in Eq. (1),

$$ {\text{MRR}} = {\text{ Initial Volume }} - {\text{Final volume}}/{\text{Time taken during turning}} $$
(1)
Fig. 3
figure 3

Turned samples of Cu-Al-Mn SMAs

Full size image

Surface roughness (Ra)

A Mitutoyo surface roughness tester was used to record the surface roughness results after each machining test. Mean values were recorded after measuring the Ra on three distinct machined surfaces. A machined surface that is uneven will have a rougher surface. The lack of roughness is shown by the surface’s lustre. Using Mitutoyo roughness equipment Model No. SJ-410, which has an accuracy of 0–1 microns, it was possible to measure the Ra values of the machined surface of Cu-Al-Mn SMAs [41,42,43,44,45], (Fig. 4).

Fig. 4
figure 4

Mitutoyo surface roughness tester

Full size image

Results and Analysis

Cutting Parameters Impact on MRR

The impact of various factors on MRR is depicted in Fig. 5. The feed rate, spindle speed, and depth of cut all directly affect MRR. Figure 6 depicts the S/N plot for MRR. Depth of cut and cutting speed are shown in Table 4 to have the most effects on the performance metric, or MRR. With a feed rate of 0.20 mm/rev, a depth of cut of 0.8 mm, and a cutting speed of 1500 m/min, the MRR achieves its maximum value. This factor has a greater impact on the performance measure. The statistics in Table 4 show this. The delta value shows the response value’s middle value, which is calculated by averaging its greatest and lowest values. The MRR’s S/N ratio is shown in Table 5. The “larger is better” approach has been modified to determine MRR. The percentage contributions for cutting speed, depth of cut, and feed rate were observed in ANOVA Table 6 to be 51.92%, 27.81%, and 16.83% respectively. For both cutting speed and cut depth, the probability (P) value was determined to be low (zero), however the p-value for feed rate is 0.004. As a result, the P-values for the chosen input components are less than α = 0.005 with a 95% confidence level.

Fig. 5
figure 5

MRR’s main impact graph

Full size image
Fig. 6
figure 6

MRR’s S/N ratio plot

Full size image
Table 4 Response table for MRR
Full size table
Table 5 Response for MRR’s S/N ratio
Full size table
Table 6 ANOVA for MRR ( (larger is preferable)
Full size table

Cutting Parameters’ Impact on Surface Roughness

According to Table 7, the depth of incision is the primary cause of Ra. Surface finish is inversed by the Ra value. Therefore, the “smaller is better” approach is employed in the current research. Figure 7 shows that the roughness of the surface is minimal at the maximum cutting speed of 1500 m/min, but its value only rises as the cutting speed lowers. Ra was measured at a lower feed rate of 0.10 mm/rev, as it is only at a feed rate increase from 0.15 to 0.20 mm/rev that Ra reaches its maximum value. Roughness of the surface is minimal at 0.3 mm and maximal at 0.8 mm of cut depth. Roughness of the surface S/N ratios is represented in Fig. 8. The roughness of the surface value was shown to decrease as cutting speed increased. A3B1C1 is the ideal option to get the least amount of roughness of the surface.

Table 7 Surface roughness response chart
Full size table
Fig. 7
figure 7

Roughness of the surface main effect graph

Full size image
Fig. 8
figure 8

S/N ratio for roughness of the surface main effect plot

Full size image

The depth of cut and feed rate were kept constant at 0.10 mm/min and 0.30 mm, while the cutting speed varies between 1000 and 1500 m/min. The slope of the curve increases and varies within the range. While the cutting speed and depth of cut were maintained constant, the feed rate varies between 0.10 and 0.20 mm/rev. Here, the slope is typically not as steep. Consequently, it is believed that a lower feed rate will result in a superior surface quality. As the feed rate is lowered, Ra value decreases, as seen in Fig. 8 [18]. Similar to this, feed rate and cutting speed were kept constant as fixed components, however depth of cut was a variable factor. The Fig. 8 showed that the steep slope significantly varies as a linear shape. Because a shallower cut is the essential component that results in the highest surface quality. Similar to this, it was noticed that the Ra value decreased as cutting speed rose. However, as depth of cut and feed rate decline, its roughness value does have a tendency to increase [46,47,48,49,50].

The Ra’s S/N ratio is shown in Table 8. The “Smaller is better” approach has been modified to determine Ra. The percentage contributions for depth of cut, cutting speed and feed rate were observed in ANOVA Table 9 to be 67.93%, 19.51%, and 10.82%, respectively. For cut depth, the probability (P) value was determined to be low (zero), however the p-value for cutting speed and feed rate are 0.001 and 0.003. As a result, the P-values for the chosen input components are less than α = 0.005 with a 95% confidence level.

Table 8 Response chart for roughness of the surface signal-to-noise ratios
Full size table
Table 9 Analysis of variance for RA (Smaller is better)
Full size table

Regression Analysis

With the use of experiments and the MINITAB 17 programme, this analysis was done to forecast the theoretical MRR and Ra values. The residual plots were plotted in accordance with the results. Figure 9 illustrated the residual plot for Ra. The lines that divide each point into two equal halves (i.e., the maximum residual values will fall on or near a straight line) show that the experimental data follow a normal distribution, as shown by the normal probability plot, which shows that the experimental data. Consequently, the values obtained were quite near to the plot’s average line. In versus fit plot also the obtained values are near to the average line. Histogram also shows the considerable results [51,52,53,54,55]. The versus order plot also shown the results such that nearly 6 values have crossed the zero line.

Fig. 9
figure 9

Residual plots for MRR

Full size image

Figures 9 and 10 respectively provide the residual plots for MRR and Ra. The experimentation was carried out correctly since the mistake spots were discovered to be very near to the normal probability plot. Both the MRR and Ra regression equations are shown in Eqs. 2 and 3, respectively.

Fig. 10
figure 10

Residual plots for Ra

Full size image

Regression formula for the rate of material removal

$$\begin{aligned} {\text{MRR }} &= \, - 0.0{362 }\; + \; 0.0000{46} \;{\text{Cutting}}\; {\text{speed }}\\&\quad + \;0.{26}00\; {\text{Feed}}\; {\text{rate }}\; + \;0.0{664} \;{\text{Depth}}\; {\text{of}}\; {\text{cut}} \end{aligned} $$
(2)

Regression formula for the surface roughness

$$\begin{aligned} {\text{Ra }}& = \, - 0.{257}\; \, - \; 0.000{493}\; {\text{Cutting}} \;{\text{speed}}\; \\&\quad + \;{3}.{673} \;{\text{Feed}}\; {\text{rate}}\; \, + \;{1}.{829} \;{\text{Depth}} \;{\text{of}} \;{\text{cut}} \end{aligned} $$
(3)

Figure 10 combines the plots for Ra. The results of the normal probability plot were quite close to the average line of the plot. The results are also quite close to the average line versus plot. The histogram’s findings were also noteworthy. Additionally, the acceptable outcome was that about 5 numbers in the reverse order crossed the zero line [56,57,58,59,60].

Confirmation test

In order to improve MRR and lower Ra, the test was run utilising the optimal set of parameters. The ideal values are A3B3C3 and A3B1C1, respectively, for MRR and Ra. The test’s MRR and Ra calculations’ outcomes are contrasted with theoretical and experimental values in Table 10. When roughness of the surface was less than 10%, it was discovered that the inaccuracy produced for MRR had been increased to 0.355% and lowered to − 0.3165%.

Table 10 Confirmation test for MRR and Ra
Full size table

SEM Analysis of Cu-Al-Mn SMAs

SEM [TESCAN VEGA3] was used to analyse the surface texture of a Cu-Al-Mn SMAs turned sample. The SEM picture of the sample’s rotated top surface from CAM3 SMA is shown in Fig. 11. Surface fissures and tiny pit holes were visible on the microscopy. Cu-Al-Mn SMAs were turned at a higher cutting speed of 1500 m/min and a feed rate of 0.2 mm/rev, and a longitudinal structure was seen. The tool used was a tungsten carbide tip. Because of the vibration of the tool that occurs when it comes into contact with the workpiece at a greater cutting speed, wavy tracks have been seen on the machined surface of Cu-Al-Mn SMAs [61,62,63,64,65]. A straight strip type was seen on the microscopy because to a quick tool moment. Due to increased heat production at the contact between the tool and the workpiece, charred chips were discovered at a high spindle speed of 1500 m/min. The primary goal of adding manganese as filler to the binary Cu-Al matrix is to provide the alloy the ductility qualities that are essential for optimal machinability in all machining operations. The machining procedure can only be used on ductile SMA alloys; brittle SMA alloys cannot be machined. First and foremost, employing this alloy has minimal manufacturing costs and is simple to fabricate. Second, because of the alloy’s SME and PE characteristics. It is also appropriate for usage in low- and high-temperature environments. Thirdly, the alloy adapts to environmental changes on its own and conforms to the necessary shape and size, which will undoubtedly be advantageous for industrial production sectors in terms of cost and time savings without the need for post processing. There are certain limits to this work that have been noted [66,67,68,69,70]. First, the alloy only exhibits ductility for a narrow range of compositions. Beyond that, the alloy turns very brittle and hard. Before employing SMA in applications, more thorough research is needed to evaluate its machinability using different machining techniques and procedures [71,72,73,74].

Fig. 11
figure 11

SEM Images of the machined surface of SMA

Full size image

Conclusions

Based on the findings, the machinability of Cu-Al-Mn SMA has been examined, and the impacts of cutting parameters on MRR and Ra have also been examined. The conclusions made are listed below. The experiments were planned and carried out using the Taguchi approach, and their outcomes were assessed using ANOVA and regression analysis, respectively.

  • ANOVA revealed that the factors with the greatest effect on performance metrics were feed rate, cutting speed, and depth of cut. Cutting speed and cut depth have the most effects on MRR’s turning process. The two main factors that affect the quality of the surface are feed rate and cutting depth. Ra value was discovered to rise with both a higher feed rate and a deeper cut.

  • According to the Taguchi approach, a material may be machined to produce a good surface quality when the right parameters are met, such as a cutting speed of 1500 m/min, a feed rate of 0.10 mm/rev, and a depth of cut of 0.3 mm. For MRR to produce the greatest results, the recommended set of parameters is A3B3C3. Therefore, every type of machining is possible in these optimum working circumstances.

  • The texture that developed throughout the turning process was examined using SEM micrographs at varying cutting speeds, feed rates, and cut depths. The surface that has been machined has less surface flaws such surface cracks, pinholes, wavy tracks, and burned chips.

  • The machinability behaviour of Cu-Al-Mn SMAs is entirely different compared to other materials because of these two properties (i.e., SME and PE). During the process, both the wear and tear of the alloy were slightly observed. The cutting performance shown by Cu-Al-Mn SMAs using CNC machining (turning) exhibited better and favourable results. Hence these Cu-Al-Mn SMAs are easily machinable to required shape and size.