Introduction

CAM SMAs have superior strength, extremely high damping, and shape memory effects when compared to conventional metals or alloys. Since these CAM SMAs can be easily produced compared to Ni–Ti SMAs, they do not require performing a vacuum casting process. Due to their lower melting temperatures, the Mn-rich CAM SMAs also have greater castability [1].

CNC machining has become increasingly indispensable in the last few decades by providing improved reliability, accuracy, and productivity. Furthermore, when compared to a traditional milling process, CNC milling offers greater ease in selecting the levels of the cutting parameters [2, 3]. Various milling methods, including side milling, end milling, plain milling, and gang milling, are used in industries. Due to its high precision, accuracy, and durability, the CNC end milling technology has gained the lead among these in the automotive, aerospace, and metal processing sectors [4]. One of the adaptable procedures that has already achieved success in the industrial sector by fulfilling needs is end milling. Any machining operation involves a number of variables that control the process. These variables are separated into ones that can be controlled and those that cannot. Rake angle, spindle rotation, feed rate, cutting speed, and depth of cut are all variables that may be changed to suit the needs. Non-controllable factors are ones that can be governed indirectly by controllable parameters rather than being directly controlled. Chip development during vibrations can happen for a variety of reasons, such as tool wear and surface roughness [5,6,7]. Modern CNC vertical milling technology has advanced greatly to satisfy the norms of numerous manufacturing sectors, particularly the precision metal cutting industry [8]. A multi-point cutting instrument called a cutter is one of the processes used in milling to create smooth, curved, and helical surfaces. In the milling process, a work item with several cutting edges is fed through a rotating cylindrical tool. The tool’s rotational axis is perpendicular to the feed vein. Surface finishing is a crucial component of any machining process in the creation of a high quality product. [9,10,11]. The milling technique is frequently employed in modern manufacturing due to its adaptability and precision [12, 13]. Taguchi design, which can quickly and precisely determine the appropriate surface roughness for controlling particular end milling processes [14,15,16].

For cutting samples to required sizes and shapes along with better surface quality and increased productivity in not much time and money. The most popular application for CNC milling is in manufacturing industries and machine shops, etc. In many machining procedures, high quality and productivity are two crucial yet crucial factors. End milling, which is conducted with the aid of a computer numerically controlled machine, is a popular technique for removing material that is used in the process of producing components with complex geometries and profiles. One component that contributes to chip formation during vibrations is tool wear, and two other factors include surface texture [17,18,19,20]. In the process of removing metal, CNC milling may create components with curved or flat shapes. The milling process may be greatly improved by improving the surface quality [21, 22]. Malay et al. [3] examined on optimization of input factors under CNC milling process on aluminium 6351. According to the results of the ANOVA, the spindle has a considerable impact on Ra, followed by the feed rate and the depth of cut. The surface finish can be improved by optimising these parameters [3]. Sahare et al. [6] conducted an evaluation of the Al2024 end milling operation. End milling’s primary output machining characteristics were thought to be Ra, cutting forces, and MRR. The surface roughness, cutting pressures, and metal removal rate are all influenced by the depth of cut, cutting speed, and feed rate, according to the S/N ratio response chart. From ANOVA, cutting depth has major influence on surface roughness than cutting speed. The predicted model’s predictions and the data from the experiment agreed well [6]. End milling of Inconel 7018 super alloys has been studied by Reddy Sreenivasulu et al. [7]. Taguchi’s DOE provided confirmation of the Inconel 718 super alloy’s machining capacity. S/N ratios and ANOVA results revealed that the feed rate and combination rotational speed had the greatest impact on burr height [7]. For aluminium alloy AA6063, Maiyar et al. [10] reported on an examination of the CNC milling parameters. The analysis revealed that Ra is mostly impacted by feed rate. With an increase in feed rate, Ra’s value rose. An increase in cut depth allowed for the achievement of the maximum MRR. The process parameters used determine the type of chip that is produced. On the machining time in this case, the spindle speed was determined to be inconsequential [10].

The length, width, and thickness of chips can change when feed rates and materials are changed during milling, according to a 2017 study by L Prasetyo et al. Measurements of the chip’s length, breadth, and thickness were used to ascertain its properties. The AISI 1020’s hardness is what causes a chip to develop in a discontinuous pattern. The length, breadth, and thickness of the chip will all grow when the feed rate is increased, regardless of the type of material being cut. The key conclusion of this study was that it is possible to determine the degree of hardness that various kinds of materials possess by looking at the amount of chips that are produced [11]. The effect of CNC milling process parameters on Ra using aluminium alloy 7024 was reported by Routara et al. in [12]. According to the experimental findings, Ra was most significantly influenced by the tool’s diameter, followed by cutting speed, depth of cut, and feed rate. Less surface roughness will be produced with a bigger tool diameter. Raising spindle speed will result in a reduction in surface roughness [12]. Raja and Baskar [13] Investigation of effect of input factors such as feed rate, speed, and depth of cut on Ra and MRR of the machined parts during milling of aluminium alloy 6065-T6 under a CNC vertical milling machine. A carbide tool cutter is used as a tool during the process. Based on Taguchi’s L16 OA, experiments were performed. The main factor affecting the output is the feed rate from the ANOVA method [13].

Syeddu and colleagues [14] When CNC end milling the alloy Al6061-T6, Taguchi’s method was utilised to forecast how different machining parameters will affect surface roughness. To mill the work material, uncoated and TiAlN coated carbide tools were employed. Tests comparing the two types of tools were carried out to establish which type gave a higher surface quality. According to an ANOVA, spindle speed is more important than cut depth for uncoated carbide tools. Variations in cut depth and spindle speed affect the quality of the surface roughness. The main factor affecting coated tools is also spindle speed rather than feed rate. Cut depth has little to no effect on surface finish. 2500 rpm, 150 mm/min, and 0.6 mm were found to be the ideal parameter values for the sample that was machined using a carbide tool coated in TiAlN. For the coated carbide tool, a surface roughness of 0.1406 mm is ideal [14]. George et al. [15] investigated on AISI 410 and AISI 420 Martensitic Stainless Steel (MSS) dry milling characteristics and were optimised. Optimal machining parameters improve component quality and productivity. Spindle speed affects surface roughness and cutting force for both MSS grades. Depth of cut affects both MSS classes’ cutting temperature. Depth of cut affects both MSS classes’ cutting temperature. Surface roughness and cutting force are higher in MSS AISI 420 because to its higher chromium and carbon composition. ANOVA shows that spindle speed, followed by feed and cutting depth, most affects Ra of both grades. According to ANOVA results for both grades, spindle speed, feed, and depth of cut affected Fc the most. ANOVA findings for both grades reveal that feed, spindle speed, and depth of cut all affect temperature [15]. The mechanism that governs the milling process is depicted in Fig. 1.

Fig. 1
figure 1

CNC milling operation [16]

Full size image

From the literature, it was found that many researchers have already worked on CNC milling machining using various materials such as steels and composites. But for Cu-based SMAs till today, no work was carried out and reports are not available on milling studies of CAM SMAs [16]. Hence, new trail is carried out to evaluate the factors and their effect on the output responses, such as MRR and surface roughness, using the Taguchi method. The most popular optimisation approaches are Taguchi method, full factorial designs, and others. The Taguchi strategy, however, is one of the best techniques for optimisation as it needs the fewest tests while taking into account all relevant sets of input variables that can possibly have a big impact on the result. The goal of the current effort is to assess the output parameters with different input elements, such as depth of cut, feed, and spindle speed in order to maximise MRR and minimise Ra and provide an alloy surface finish of a higher grade.

Experimental Details

Materials and Method

From the literature, the alloy chosen is ductile in nature because of the presence of manganese in the Cu-Al binary alloy. In this experiment study, pure Cu, Al, and Mn of weight 200 g by varying the percentage of Al (10–13wt.%) and Mn (0–7.5wt.%) and rest being copper have been added and prepared alloy using induction melting process. Figure 2 shows the prepared seven samples of CAM SMAs for CNC Milling process. In this investigation, the as-cast samples of dimension 100X50X4mm (LXWXT) are homogenised at 900 °C for 45mints to 1 h to remove the agglomeration in the prepared alloy and to enhance uniformity in the composition. Later on the samples are betatized for 900 °C for 1 h and step quenched in hot water and cold water to relive the internal stress present in the alloy and also the alloy is converted from austenite to martensite phase to obtain martensite structure.

Fig. 2
figure 2

Step quenched CAM SMA samples for CNC end milling machining

Full size image

Finally, the shape memory alloy is subjected to CNC end milling operation. Following the stage quenching procedure, Fig. 3 indicates that CAM SMA sample-7 has quenched cracks formation throughout its surface. As a result, alloy 7 exhibited poor machinability due to brittleness of the alloy; hence, this alloy is not suitable for CNC end milling operation and machining has been carried out for remaining six samples.

Fig. 3
figure 3

Brittle CAM-7 SMA sample

Full size image

Chemical Compositions of CAM SMAs

The compositions chosen based on wt.%. The chemical compositions obtained after the sample preparation using optical spectrophotometer. Table 1 shows CAM SMA compositions.

Table 1 Compositions of CAM SMAs based on wt.%
Full size table

Experimental Details

The current work was mostly completed using a CNC three-axis end milling machine. Based on a review of the literature, carbide was chosen as the material for cutting tools. A flat-end mill cutter with a 6 mm diameter has been chosen as the cutting tool. A CNC end milling machine was used in this particular experiment (Fig. 4). Figure 5 depicts the usage of a flat-type milling cutter as a cutting tool during a milling process. End milling was done on the workpiece material while it was dry utilising nine trails (i.e. a L9 orthogonal array) and adjusting the various parameters, as given in Table 2.

Fig. 4
figure 4

Three axis vertical end milling machine

Full size image
Fig. 5
figure 5

Carbide flat-end mill cutter

Full size image
Table 2 Process parameters and their ranges
Full size table

Machining Conditions

Cutting fluids harm the environment, economy, and health. Dry cutting, which fully forgoes the use of cutting fluids, is the greatest way to eradicate their effects. In addition to dry machining, there are also various machining conditions that can be utilised, such as wet machining, minimum quantity of lubricant (M.Q.L) machining, flooded machining, and cryogenic cooling. This type of machining uses no coolant, oil, or metal cutting fluid. It is mostly used in milling operations, not it is suggested for drilling operations. The use of cutting fluids in cutting operations causes significant economic, environmental, and health issues. Dry cutting, which fully forgoes the use of cutting fluids, is the greatest way to eradicate their effects which also lowers machining costs, environmental risks, and increase productivity [17,18,19,20,21,22]. However, this is not suitable for all machining operations and materials, and in some instances, dry cutting exhibits benefits including reduced thermal shock and extended tool life [23,24,25,26,27]. Dry machining does not need coolants. Consequently, the working atmosphere is more comfortable and clean, which eliminates the possibility of slippage when cutting [28,29,30,31,32,33,34,35]. Hence for the present investigation, a dry machining cutting condition is employed. Without the use of any cutting fluid, the machining process was carried out at 28 °C room temperature. Figure 6 shows the benefit in choosing dry condition for the present investigation as per the previous literature.

Fig. 6
figure 6

Advantage of dry machining [24]

Full size image

Results and Discussions

CAM SMA Samples after Milling Process

Figure 7 shows the CAM SMA samples obtained milling profile on the surface after undergoing CNC end milling operation. The samples are subjected to study the surface harshness using Mitutoyo surface roughness tester. Table 3 shows the milling data obtained based on Taguchi design and conducted the experiments using L9 orthogonal array. However, CNC milling was also carried out for all the seven samples of CAM SMAs. This combination of SMA was chosen because the addition of manganese concentration provides required ductility for the SMA [36,37,38,39,40,41,42]. For CAM-7 SMA, machining was unable to carry out due to the brittleness of the alloy taken place after step quenching process, because their parent phase has L21 structure due to the decrease in the degree of order [43,44,45,46,47,48]. A surface tester (Mitutoyo) was used to measure the texture that resulted from the machining process. For the analysis, the average roughness value (Ra) is taken into account. Figure 9 indicates the samples of CAM SMAs after milling operation.

Fig. 7
figure 7

Cu-Al-Mn SMA alloy samples after milling operation

Full size image
Table 3 CNC end milling data of CAM SMAs for MRR
Full size table

The formulae used for evaluating MRR are as follows,

$${\text{MRR }}\left( {{\text{gms}}/{\text{min}}} \right) = w_{x} d_{x} f_{x} \rho$$
(1)

where w—width of cut in mm, d—depth of cut, mm, f—feed rate in mm/min, and ρ—density of material in gm/cc.

S/N Plots for MRR and Ra

Figures 8 and 9 indicate the S/N curve for MRR and Ra. To determine MRR, “larger is better” method was chosen and “smaller is better”, for Ra. Figure 10 shows that the MRR was observed to be decreasing as spindle speed rose. Higher cut depths allow for the achievement of the maximum MRR [10]. According to the graph, MRR rises with rising feed rate and cutting depth, respectively. From Fig. 10, it was observed that an increase in feed rate and spindle speed tends to increase Ra value but decreases with an increase in feed rate. The MRR response data are shown in Table 4. Consequently, it was discovered that the feed rate is the primary characteristic that displayed rank-1. Because from the investigation, the feed rate is affecting more on MRR followed by depth of cut (rank-2) and spindle speed (rank-3). To evaluated MRR, larger the method was followed and for Ra the method followed is smaller the better. Table 5 shows that the spindle speed is ranked first, with depth of cut and feed rate coming in at positions two and three, respectively, as the key factors influencing Ra.

Fig. 8
figure 8

S.N curve for MRR

Full size image
Fig. 9
figure 9

S/N curve for Ra

Full size image
Fig. 10
figure 10

a–e CNC milling profile of CAM-4 SMA under dry condition

Full size image
Table 4 Response table for MRR
Full size table
Table 5 Response data for Ra
Full size table

ANOVA for MRR and Ra

From Tables 6 and 7, it was observed that the %contribution is more for feed rate (77.42%) followed by cutting depth (19.35%) and spindle speed (1.21%) for MRR but for Ra, the % contribution is more for spindle speed (69.78%) followed by feed rate (16.50%) and depth of cut (0.79%), respectively. The probability value for feed rate is almost zero, followed by depth of cut (0.001) and spindle speed (P > 0.144) for MRR. The function is deemed to be insignificant if the values surpass 0.10. As a result, there is a 0.01 per cent chance that noise will have little effect. Similarly, it was noticed that the P value is less for spindle speed (P < 0.003) followed by feed rate (p < 0.053) and depth of cut (P < 0.605), respectively, for Ra. In case of MRR, feed rate and depth of cut are the two main important factors responsible for affecting the output response. Here, spindle speed was found as insignificant parameters because it is not affecting on output characteristic (i.e. Ra). In case of Ra, both feed rate and spindle speed are the main significant parameters affecting the Ra of Cu-Al-Mn SMAs. But this parameter (depth of cut) is showing least effect on Ra. However, it was considered as the insignificant parameter for Ra.

Table 6 ANOVA for MRR
Full size table
Table 7 ANOVA for Ra
Full size table

Surface Morphology of Milled Profile of CAM SMAs

Both MRR and Ra are highly sensitive to the machining conditions, which can be altered by adjusting factors including feed rate, depth of cut, and spindle speed. SEM analysis was investigated for the samples machined under CNC Milling operation. Figure 10 represents the difference between up milling and down milling under CNC end milling process. The burr size can vary as a result of this type of milling procedure. It was observed that compared to down milling, the height of the burr obtained during the up-milling operation is short. The up-milling procedure is followed to produce burrs with decreased height and thickness. The height of the burr during the down-milling process is higher than up-milling process, the burr thickness must be greater. In such constraint, the burr thickness in down milling is less. The volume of the material should remain the same during the deformation process, as recommended by the principles of plastic deformation. Therefore, in order to satisfy the criterion of a constant burr volume, the burr thickness must be increased if, during the up-milling process, the burr height is lower than it was during the down-milling process. Under these circumstances, the burr thickness achieved through down milling is lower [49,50,51,52,53,54,55].

Figure 10a shows the milling profile of CAM-4 SMAs. Here the tool rotation direction up-milling side and down-milling side, feed applied direction can be observed and ring type of profile formed at the milled slot of the alloy. From Fig. 10b, milling burrs, individual milling grooves, flank profile of milling groove, bottom of milling grooves after each trail of milling operation during dry machining condition were observed. The slot side is the side portion which is not machined and slot base is the base of the milled profile [56,57,58,59,60].

SEM pictures of the machined surfaces taken in dry conditions, as shown in Fig. 10c, can be used to identify various surface flaws. It was clear that the ring-shaped grooves and smears spread over the machined surface throughout the dry machining process. It consists of smooth and rough surfaces during the operation, formation of voids and chips can also be observed at spindle speed—1200RPM, feed—100 mm/rev, and depth of cut—0.5 mm, respectively.

First, because of the slower feed rate of 100 mm/rev, no feed marks are visible. The second observation is that there are no waves or ridges of any type on the surface. This is because there is less of a cutting force, which results in less plastic deformation and, as a result, a smoother surface. It is possible that the vibrations produced are responsible for the creation of waves on the surface, but it might also be the result of tool wear, which led the material to shift from a surface that was machined by one flute of the tool to a surface machined by the second flute of the tool, respectively [61,62,63,64,65]. Regardless of the reason, the waves can be seen as an undesirable feature. In a similar fashion, there were found to be some micro-cracks observed over the machined surface. Due to insufficient material removal from the surface, micro-cracks developed on the surface [66, 67].

Figure 10d shows the uneven sized burrs with varying height and thickness at the milled groove end. At the sidewalls of the channel, recast layer development was also seen. Because of maximum amount of heat generated between the rotating tool and workpiece interface melts the material rapidly responsible for the structure of cast layer. Figure 10e indicates initially when machining starts with rotating tool movement leads to the structure of tiny burrs, further movement of tool results in wider burrs of varied thickness and height, and finally, the forms short burrs with small thickness and height in each trail can be observed, respectively [68,69,70,71].

Conclusions

Below is a summary of the findings from the experimental research on Cu-Al-Mn SMAs, along with any inferences that may be made.

  • Based on experimental findings of MRR and Ra, which are mostly connected to speed and feed rate followed by depth of cut, the approach employed in this work focuses on analysis of CNC milling parameters to achieve optimal values of the output parameters. The machining parameters have been optimised using Taguchi design.

  • By exhibiting relative contributions of 77.42%, 19.35%, and 1.21% in the ANOVA findings, spindle speed, depth of cut, and feed rate were identified as the parameters influencing MRR. When compared to the other two factors (such as depth of cut spindle speed), it was discovered that the feed rate’s percentage contribution has a greater impact on MRR. The rank values demonstrate that the influence of feed rate on MRR is greater. Therefore, extending the depth of cut will result in the greatest amount of material (MRR).

  • According to the analysis, the two primary factors most significantly determine surface roughness (Ra) relative to spindle speed, depth of cut, and feed rate. The ideal MRR and Ra values are achieved at the first level spindle speed (400 rpm), third level depth of cut (0.5 mm), and third level feed rate (300 mm/min). Ra at spindle’s third level (1200 RPM), first level feed rate (100 mm/Rev), and third level cut depth (0.5 mm). The smallest amount of roughness needed to provide a decent surface finish can be obtained by lowering spindle speed.

  • The profile was created via a CNC end milling operation, as seen by SEM pictures. The surface morphology of the machined produced changes according on the input parameter values used, as seen by images taken using a SEM. It has been discovered that the rotation and movement of the tool during the end milling process causes the production of wavy burrs, short burrs, long burrs, and rings by rings kinds of burrs.

  • From the entire analysis, it can be inferred that the brittle samples of CAM SMAs are not appropriate for the CNC end milling operation, which may be used to provide a higher surface finish for ductile SMAs. Brittle alloys are created as a result of the Cu-matrix’s increased Al content and decreased Mn concentration. As a result, the CAM-7 alloy (Cu—85.5%, Al—13%, and Mn—1.5%) displayed hard and brittle behaviour, and following milling, powdered sort rather than chips were produced. This specific composition has been shown to be poorly machinable and unsuitable for any machining operation.

  • This end milling process is suitable where the smart materials component manufacturing takes place. For industrial sectors where batch production of component manufacturing takes place, it is primarily low cost and simple to manufacture with higher precision, which will be unquestionably advantageous both economically and commercially.