Dependence of pre-treatment structure on spheroidization and turning characteristics of AISI1040 steel

Abstract

During machinability, the combination of machining process parameters and the material properties of the component to be machined plays an important role. Material properties depend upon the type of phase form present and the grain size of the formed phases, which in turn depends upon the prior treatment given to alter the initial room temperature types and form. Accordingly, spheroidization treatment was carried out on medium carbon steel (AISI1040) by altering the initial room temperature structure through normalizing and hardening treatment. Machinability experiments were performed on CNC machine by varying machining process constraints. Tool wear and surface roughness of the machined component obtained by turning were analyzed and correlated. Using Minitab and full factorial design, the ANOVA study was carried out. With the help of regression analysis, residual and main effect plots combined optimization (tool wear and surface roughness) was targeted. ANOVA result shows excellent machinability for the as-bought-spheroidized condition where feed has a 67% contribution to tool wear (T_W) whereas the depth of cut has a 71.91% contribution to surface roughness (S_R). Also, the optimized regression values obtained for machining parameters are feed (0.39 mm/rev), depth of cut (0.6 mm), and spindle speed (780 rpm) with composite desirability of 0.8174. T_W and S_R experimental values for the optimized machining parameters are 0.039 mm and 2.89 μm, respectively, and the difference between the actual and optimized values is less than 5%.

Keywords:

• Heat treatment
• tool wear
• surface roughness
• machinability
• DoE

1. Introduction

Medium carbon steels are one of the important group members of structural steels, which are highly versatile in nature concerning the application point of view. These steels can be easily alloyed with Cu, Mn, Ni, Cr, and so on, to enhance the surface-related properties. In steel, the carbon content in the range of 0.25% to 0.65 wt.% is the best composition required for heat treatment. The considerable property alterations possible are ductility by annealing, grain refinement by normalizing, and hardness and tensile strength by hardening (S & Rajan, 2010). These steels are available as high-temperature deformed or warm/room temperature deformed forms, with cross sections ranging from circular, square, or rectangular type, or even as strips.

Plain carbon steels (AISI1040) are moderate strength, heat treatable type, whereas AISI4140 steel has ferrite stabilizers as alloying addition. Ferrite stabilizers reduce the lower critical temperature and the size of the gamma loop of the iron-carbon binary phase diagram (Avner, 2015). Hence, these steels increase the carbide content because ferrite stabilizers act as carbide former. Due to this content, the hardness, wear resistance and tensile strength of the steel increase. AISI4340 steel is still versatile in nature compared to the other two types mentioned. The high strength and hardness combination of AISI4340 steel is due to the possession of austenite (Ni) and ferrite (Cr and Mo) stabilizers as alloying elements. Also, the nickel content in the steel increases the high-temperature property.

Heat treatment is a solid state processing technique where the heating and cooling cycles show an essential role in property alteration. Among several heat treatments available, hardening is the primary treatment to improve hardness and tensile properties. In the present work, a comparison of turning characteristics is presented by subjecting AISI1040 to spheroidization treatment (Retd & Singh, 1926). Spheroidizing treatment is one of the techniques available in the heat treatment family to improve the machinability of medium/high carbon steels. For machinability, conventional normalizing treatment may not be suitable for all steels if steel contains alloying elements. Spheroidizing treatment changes the phase particles, especially carbides, into the globular form (Finkel’shtein et al., 1963). Harder carbides cannot be softened by the available techniques, but the sharp edges of the phase particles are blunted to form a spherical shape. The sharp cornered particles present in the workpiece increase the friction between the cutting tool and the workpiece, leading to higher wear and tear of the cutting tool. Also, the size and number of the phase particles as carbides or ferrite play an important role in machinability (Naylor et al., 1976).

To analyze the influence of phase morphology on machinability, an attempt is made to alter the initial room temperature structure of as-bought steel. Accordingly, steel is first normalized or hardened before being subjected to spheroidizing treatment. Normalizing is a grain refinement treatment where the phases present are similar to as-bought steel (ferrite and carbides) but are in finer form. Since they are in finer form, the number of particles in a given spot is greater than the as-bought (hot worked) condition. Once normalized condition steel is spheroidized, tiny globes of phase particles are seen in the microstructure. Finer particles easily get spheroidized with lesser soaking time. So, spheroidization process may be partially or fully completed as compared to as-brought (coarser) structure within the stipulated duration (Tian & Kraft, 1987). Hence, spheroid particle density is more in normalized compared to as-brought steel. On the other hand, when steel is hardened, a supersaturated single-phase martensite forms instead of a two-phase mixture of ferrite and carbides as a room temperature structure (Chattopadhyay & Sellars, 1977). This phase is highly stressed with an enormous number of nucleation sites and crystal defects (vacancy and dislocation). These enormous numbers of crystal defects contribute as nucleation sites during spheroidization to alter grain shape and size. Hence, particles are finer in normalized-spheroidized state and still finer in hardened-spheroidized conditions (Cabanas-Moreimo & Morales, 1992).

In the present work, as-bought steel was spheroidized following the normalizing and hardening paths, respectively, and compared with that of the as-bought spheroidized condition for turning machinability considering tool wear and machined surface roughness data (MADEN, 1984; Nalbant et al., 2007). Turning was performed on a CNC lathe with cemented carbide inserts. The machining parameters (speed, feed, and depth of cut) obtained from the literature were considered for machinability analysis. The objective of this research work is to reduce tool wear with an improved surface finish. To minimize the work content, obtain accurate results, and reduce the time required to do the analysis, DoE technique is applied (Dhar et al., 2002; Kamarudin et al., 2007). The novelty of the work is the alteration in the initial room temperature structure for the intentional phase transformation, which indirectly affects the spheroidization rate and spheroid morphology (size and shape).

2. Methodology

This section provides the details of the heat treatment of materials, heat treatment methods used, and different tests carried out to fulfill the set objectives for the present work. The initial part provides the details regarding conventional heat treatments, namely, normalizing, hardening, and spheroidizing procedures, for the materials and the various mechanical tests used for the heat-treated steel (Harisha et al., 2019; Çalik, 2009). The latter part provides the details regarding the DoE for the machinability of the heat-treated materials and optimization. Notations used for different heat treatment conditions are provided in Table 1.

Table 1. Spheroidization treatment cycle representations

Notation used	Heat treatment conditions
A	As-bought AISI1040 specimen
A3	As-bought 3 h spheroidized AISI1040 specimen
A6	As-bought 6 h spheroidized AISI1040 specimen
A9	As-bought 9 h spheroidized AISI1040 specimen
N	Normalized AISI1040 specimen
N3	Normalized 3 h spheroidized AISI1040 specimen
N6	Normalized 6 h spheroidized AISI1040 specimen
N9	Normalized 9 h spheroidized AISI1040 specimen
H	Hardened AISI1040 specimen
H3	Hardened 3 h spheroidized AISI1040 specimen
H6	Hardened 6 h spheroidized AISI1040 specimen
H9	Hardened 9 h spheroidized AISI1040 specimen

In order to do the heat treatment, a muffle furnace was used to heat the first set of as-bought workpieces at 850°C for 2 h. The workpieces were immediately transferred to another muffle furnace maintained at 725°C and soaked for 10 min and later transferred to a furnace maintained at 740°C isothermally for different time durations. The inhomogeneity in the chemical composition increases the spheroidization rate so that the cycle time is reduced (Arruabarrena et al., 2014). Out of the three sets of specimens, the first set of specimens was spheroidized for 3 h, the second set for 6 h, and the last set for 9 h, followed by air cooling to room temperature. For spheroidizing as-bought steel specimens, the heat treatment cycle is provided in Figure 1(a). For normalizing, the subsequent set of specimens were heated to 900ºC for 2 h. The specimens were then allowed to cool to room temperature in still air condition. The spheroidization process was carried out according to the heat treatment paths shown in Figure 1(b). Similarly, the third set of specimens were immediately quenched in water at room temperature to form a martensite structure followed by spheroidization. The spheroidization cycle with hardening is provided in Figure 1(c).

Figure 1. Heat treatment cycle for AISI1040 steel. (a) As-bought spheroidized specimen. (b) Normalized-spheroidized specimen. (c) Hardened-spheroidized specimen.

The heat-treated samples were subjected to a machinability study that was assessed in terms of T_W and S_R values. In the present study, machinability experiments were carried out on spheroidized medium-carbon alloy steel according to the ISO 3685:1993 tool life-testing method (E8, 2010). Machining and heat treatment parameters were varied at different levels to determine the effect of spheroidization on medium carbon alloy steels (Kumar et al., 2012). The test samples were obtained by machining a round bar of medium-carbon, low-alloy steel. The machinability test specification is as per the ISO 3685 standard given in Figure 2(a).

Figure 2. Machinability specifications. (a) Turning test sample. (b) Carbide insert.

The machinability experiments were performed in an abundant supply of coolant on the heat-treated specimens (ϕ 30 mm × 300 mm). A heavy-duty soluble oil having a consistency of 1 part oil to 10 parts water is recommended for turning. The machine offers a standard high-pressure coolant capability of 70/80 bar (1015/1160 psi) with a flow rate of 2.6 l/min. Initially, the specimen was subjected to rough turning to remove inhomogeneity present in the material due to scaling caused by the heat treatment. The turning practice was carried out for a total length of 300 mm and checked for tool flank wear (wear land) using the Tool maker’s microscope (Khan & Bhivsane, 2018). While conducting the machinability test by varying feed, the average value of spindle speed and depth of cut was taken to be constant. Similarly, on varying spindle speed and depth of cut, the average values of the other two parameters were retained constant (Sivaraman et al., 2012a). The carbide insert VNMG12t304 with carbide grade of WK20CT (HC K20) single point cutting tool was used for machining of heat-treated samples as shown in Figure 2(b).

The machining parameters used, namely, the speed of rotation of work (n, rpm), depth of cut (d, mm), and feed (f, mm/rev), are shown in Table 2. Machinability test results for the heat treatment cycles were recorded. T_W was quantified by the flank wear only. In the majority of the machining process of alloys, the crater wear was negligible. Hence, tool wear was quantified by the flank wear as wear land. Talysurf instrument was used for the measurement of R_a (average surface roughness) along the feed direction. T_w and S_R were used to assess the machinability of different heat-treated specimens (Baskar et al., 2018; Mongomery, 2017).

Table 2. Details of machining parameters for machinability tests on AISI1040 steel specimen

Machining parameters
Spindle speed, n (rpm)	720	750	780
Feed, f (mm/rev)	0.13	0.26	0.39
Depth of cut, d (mm)	0.2	0.4	0.6

3. Results and discussion

3.1. Influence of spheroidization on the microstructure of AISI1040 steel

Figures 3(a-c) represent the microstructure of as-bought, normalized, and hardened specimens, respectively. Figures 3(a-b) show the pearlite colony and ferrite grains where the as-bought ferrite grain is coarser than that normalized. Normalizing treatment gives more wt.% of pearlite compared to annealed or as-bought (Haddadi et al., 2014; Sk et al., 2018). Accordingly, the larger colony of pearlite in a normalized specimen (Figure 3(b)) is the indication of obtaining more quantity of pearlite compared to that of as-bought specimen (Figure 3(a)). The higher the pearlite content, the faster the rate of spheroidization (Arruabarrena et al., 2014).

Figure 3. Scanning electron microscope microstructure of AISI1040 workpiece. (a) As-bought. (b) Normalized. (c) Hardened. (d) As-bought 9 h spheroidized. (e) Normalized 9 h spheroidized. (f) Hardened 9 h spheroidized.

Figure 3. (Continued).

Figure 3. (Continued).

Ferrite and pearlite phases are generally seen as micro-constituents in medium carbon steels in as-bought and normalized conditions. Figure 3(c) represents the martensite plates without ferrite as well as pearlite phases. The martensite phase is observed in the hardened medium carbon steels. This is due to the slower critical cooling rate of the component to obtain pearlite free martensite (Ebrahimian & Ghasemi Banadkouki, 2017; Jahanara et al., 2019). This is justified by the microstructure obtained during hardening and hardened-9 h-spheroidization conditions (Figures 3(e-f) respectively). Figures 3(d-f) show spheroidized phases in the above mentioned conditions after 9 h spheroidization. In all the three cases, spheroidization is completed, showing nodular/globule particles. Depending on the fineness of grains in its pre-treatment process the spheroids formed are finer or coarser as shown in Figures 3(d-f).

3.2. Experimental turning characteristics on AISI1040 steel

Experimental results are investigated to determine the material condition (as-bought, normalized, hardened, and respective spheroidized) for good surface finish and minimum tool wear. From Figure 4 it can be seen that as the feed increases the surface roughness increases. At high feed, continuous chips formed at the workpiece tool interface increase friction that leads to high surface roughness due to excessive heat generation. At lower feeds, normalized spheroidized steel shows excellent surface roughness followed by the normalized condition. At higher feed (0.39 mm/rev), normalized steel shows poor surface finish that is 2.5 to 3.5 times increase in surface roughness compared to when the feed is moderate (0.26 mm/rev) as shown in Figure 4. As the feed increases, the deviation in surface roughness values between unspheroidized and spheroidized conditions decreases (Kamarudin et al., 2007). Also, an increasing trend in S_R in all the spheroidized specimens remains the same even though the least S_R is observed at the lowest feed and maximum at the highest feed. It indicates improvement in the machinability of the specimen during spheroidization (Saï, 2005). In an unspheroidized state, normalizing gives a better surface finish over hardened steel. The hardened steel shows a poor surface finish at all feed levels. This is due to the poor impact resistance values of the martensitic structure compared to the pearlitic or ferritic phase (Avner, 2015). As-bought spheroidized (A9) shows a better surface finish, almost the same value at all depths of the cut as shown in Figure 5. The as-bought steel shows poor surface finish due to the chemical inhomogeneity present in the specimen during casting (Harisha et al., 2019; YAMAMOTO & KUMAGAI, 1975). A moderate depth of cut (0.4 mm) is favorable for a better surface finish. A higher depth of cut (0.6 mm) increases the coefficient of friction, thus reducing the surface finish. A9 followed by H9 conditions show better surface finish at all levels of depth of cut compared to other heat-treated conditions (Figure 5).

Figure 4. Surface roughness (S_R) study with constant spindle speed (n, rpm) and depth of cut (d, mm) by varying feed (f, mm/rev) for AISI1040 steel.

Figure 5. Surface roughness (S_R) study with constant feed (f, mm/rev) and spindle speed (n, rpm) by varying depth of cut (d, mm) for AISI1040 steel.

A9 followed by H9 and N conditions show better surface roughness at all levels of spindle speeds as shown in Figure 6. A similar situation is observed for the depth of cut condition also (Figure 5). As the spindle speed increases, the contact time between the cutting tool and the work surface decreases. Lesser time is available for the heat transfer to the tool or chips, reducing the heat input on the tool and chips. However, the workpiece surface gets heated up, showing lesser resistance for the material removal process, thereby reducing surface roughness (MADEN, 1984; Pal et al., 2014). As the feed increases friction between the workpiece and tool increases, especially at the flank surface resulting in increased erosion of the flank surface of the tool by the machined surface (Das et al., 2015; Haq & Tamizharasan, 2006; Suresh et al., 2012). Hence, tool wear continuously increases as feed increases, as shown in Figure 7. A9 condition also shows higher wear due to the zonal variations of properties at the microscopic scale.

Figure 6. Surface roughness (S_R) study with constant feed (f, mm/rev) and depth of cut (d, mm) by varying spindle speed (n, rpm) for AISI1040 steel.

Figure 7. Tool wear (T_W) study with constant spindle speed (n, rpm) and depth of cut (d, mm) by varying feed (f, mm/rev) for AISI1040 steel.

A9 condition shows lesser tool wear when the depth of cut is varied by maintaining the other two machining parameters constant. The larger the depth of cut, the higher the friction, which accelerates tool wear (Suresh et al., 2012). Hence, a lower depth of cut is preferred. A9 condition shows excellent results with a mild increase in tool wear as the depth of cut increases in the chosen range of parameters (Figure 8). As shown in Figure 9, the A9 heat treatment condition shows reduced tool wear, especially at lower spindle speeds. The higher the spindle speed, the higher the friction at the interface, which yields more tool wear (Suresh et al., 2012). Severe wear is observed at higher cutting speeds (780 rpm) in all specimens. In the case of AISI1040 steel, A9 (as-bought-spheroidized) condition shows better surface finish (lower surface roughness) as well as reduced wear (minimum tool wear) as observed in Figures 4 to 9.

Figure 8. Tool wear (T_W) study with constant spindle speed (n, rpm) and feed (f, mm/rev) by varying depth of cut (d, mm) for AISI1040 steel.

Figure 9. Tool wear (T_W) study with constant depth of cut (d, mm) and feed (f, mm/rev) by varying spindle speed (n, rpm) for AISI1040 steel.

3.3. Machinability study using DoE

The full factorial design concept is used as part of DoE with a series of experimental research on tool wear and surface roughness by varying turning parameters (Puneet et al., 2016; Sivaraman et al., 2012a; Tamizharasan et al., 2006). The experiments carried out are summarized in Table 3.

Table 3. T_w and S_R values for AISI1040 as-bought-spheroidized steel

Sl. No.	Feed, f (mm/rev)	Depth of cut, d (mm)	Spindle speed, n (rpm)	Tool wear, Tw (mm)	Surface roughness, SR (μm)
1	0.13	.2	720	0.035	3.73
2	0.13	.2	750	0.046	4.56
3	0.13	.2	780	0.052	5.3
4	0.13	.4	720	0.034	2.73
5	0.13	.4	750	0.024	2.81
6	0.13	.4	780	0.04	3.12
7	0.13	.6	720	0.02	4.68
8	0.13	.6	750	0.037	4.73
9	0.13	.6	780	0.047	5.62
10	0.26	.2	720	0.068	2.38
11	0.26	.2	750	0.058	3.54
12	0.26	.2	780	0.072	3.53
13	0.26	.4	720	0.08	2.38
14	0.26	.4	750	0.067	2.42
15	0.26	.4	780	0.087	2.84
16	0.26	.6	720	0.072	4.63
17	0.26	.6	750	0.081	5.38
18	0.26	.6	780	0.09	5.27
19	0.39	.2	720	0.072	3.61
20	0.39	.2	750	0.064	3.02
21	0.39	.2	780	0.071	3.64
22	0.39	.4	720	0.065	3.61
23	0.39	.4	750	0.07	2.86
24	0.39	.4	780	0.112	3.8
25	0.39	.6	720	0.053	4.98
26	0.39	.6	750	0.07	5.61
27	0.39	.6	780	0.085	5.36

3.3.1. Turning characteristics analysis of as-bought-spheroidized AISI1040 steel using DoE

From the turning experiments, the measured T_w and S_R values for AISI1040 as-bought spheroidized steel are tabulated in Table 3. Using MINITAB software, based on a full factorial design analysis, ANOVA for a mean of T_w and S_R was carried out. Following that, regression was applied and a statistical model was obtained (Baskar et al., 2018; Ganesh et al., 2014). The ANOVA (Tables 3 and 4) was used to study the significance and effect of the cutting parameters on the response variables (T_w and S_R.).

Table 4. ANOVA for T_w for the as-bought-spheroidized specimen of AISI1040 steel

Parameter	DF	Seq SS	Contribution %	Adj SS	Adj MS	P-Value
Feed (mm/rev)	2	0.008248	67.53	0.008248	0.004124	<.001
Depth of cut (mm)	2	0.000094	0.77	0.000094	0.000047	.738
Spindle Speed (rpm)	2	0.000812	6.65	0.000812	0.000406	.095
Error	20	0.003059	25.05	0.003059	0.000153
Total	26	0.012214

From the ANOVA results for T_w, provided in Table 4, it is seen that feed is the major contributing factor, with a relative contribution of 67.53%. The speed and depth of cut do not contribute much to the variations in T_w of the material in the range of study.

The result of ANOVA for S_R is provided in Table 5. Depth of cut is the major contributing factor that contributes to 71.91% of the variations in the S_R of the material. The spindle speed and feed do not contribute much to the variations in S_R of the material in the selected range of study. In medium carbon steel, generally, T_W and S_R are the functions of spindle speed (Bhuyan et al., 2018; Davis et al., 2014). At moderate spindle speed, excellent T_W and S_R are obtained. It indicates speed selected (720 to 780 rpm) is not optimum because the contribution of spindle speed to T_W and S_R is a minimum in AISI1040 steel (Tables 4 and 5). The feed value selected is in the range of (0.13 to 0.39 mm/rev). It has a very good contribution to reducing the T_W. The depth of cut (0.2 to 0.6 mm) range selected is good for T_W but not for S_R. As the depth of cut increases, the MRR increases. Accordingly, interface temperature also increases which reduces surface finish. Hence, even though T_W is minimum, the turned surface deteriorates, thereby increasing S_R (Kumar et al., 2012; Mabrouki et al., 2017).

Table 5. ANOVA for S_R for the as-bought-spheroidized specimen of AISI1040 steel

Parameter	DF	Seq SS	Contribution %	Adj SS	Adj MS	P-Value
Feed (mm/rev)	2	1.5447	4.99	1.5447	0.7723	.059
Depth of cut (mm)	2	22.2528	71.91	22.2528	11.1264	<.001
Spindle speed (rpm)	2	2.4260	7.84	2.4260	1.2130	.016
Error	20	4.7214	15.26	4.7214	0.2361
Total	26	30.9449

The regression equations were fit in order to predict T_W and S_R. The regression equation for T_W is given in equation 1, where feed (f) is in mm/rev, depth of cut (d) is in mm, and spindle speed (n) is in rpm.

${\mathrm{T}}_{\mathrm{W}}\mathit{\hspace{0.17em}}\left(\mathrm{mm}\right)=0.06193-0.02470\mathrm{f}-000215\mathrm{d}+0.00730\mathrm{n}$

The R squared values for T_W are as follows:

R squared = 74.95%

R-Sq (Adj) = 67.44%

The R-Sq (Adj) values for T_W indicate that the regression equation possesses a good fit with the experimental results.

Figure 10 shows the residual plot for T_W. The model of turning characteristics for T_W is suitable as signified by the points dropping on a conventional line in the normal probability plot, which in turn shows that the errors are normally distributed. Similarly, the plot of the residuals against the expected response is structureless, i.e., having no obvious pattern (Mongomery, 2017). The histogram shows a closely bell-shaped normal distribution. Furthermore, the high R-Sq and R-Sq (Adj) values indicate that the regression equations possess a good fit for the actual experiment conducted (Agrawal et al., 2015; Baskar et al., 2018). This equation can be used to predict T_W involving factors with the range of values under study.

Figure 10. Residual plots of T_W for as-bought-spheroidized AISI1040 steel. (a) Normal probability plot (b) versus fits. (c) Histogram (d) versus order.

The regression equation for S_R is given in equation 2, where feed (f) is in mm/rev, depth of cut (d) is in mm, and spindle speed (n) is in rpm.

$S_R\mathit{\hspace{0.17em}}\left(\mathit{\mu }\mathrm{m}\right)=3.9311+0.211\mathrm{f}-0.230\mathrm{d}+0.418\mathrm{n}$

The R squared values for S_R are as follows:

R squared = 89.45%

R-Sq (Adj) = 87.90%

The R-Sq (Adj) value for S_R indicates that the regression equation possesses a good fit with the experimental results.

Figure 11 shows the residual plot for S_R. The normal probability plot revealed that the residuals were near the fit line (mean line) indicating that the residuals were normally dispersed, which is called linearity (Sivaraman et al., 2012a). From Figures 10 and 11, it is inferred that the experimental T_W and S_R data collected were genuine and reliable and the slight deviation observed about the mean line could be neglected (Hegde et al., 2022). The residual fall is in a straight line as displayed in Figures 10 and 11, which was seen in the way that errors were normally distributed. The random dispersion of the points around the horizontal axis justifies the selection of a regression model for the data. The regression equation for S_R (equation 2), has a higher R squared value compared to T_W. This could be due to the extraneous factors (tool geometry and cutting conditions) which were not considered in this study (Akasawa et al., 2004; Saï, 2005).

Figure 11. Residual plots of S_R for as-bought-spheroidized AISI1040 steel. (a) Normal probability plot (b) versus fits. (c) Histogram (d) versus order.

The main effects plot for T_W is shown in Figure 12. As the feed increases, T_W increases at a steeper rate, reaching a maximum and decreases thereafter. As the depth of cut increases, a similar kind of observation is recorded but at a lower rate. Lower feed (0.13 mm/rev), lower depth of cut (0.2 mm), and moderate spindle speed (750 rpm) are the favorable conditions to keep T_W to a minimum. As speed increases, T_W drastically reduces in the beginning, reaching a minimum and then increasing gradually. Minimum T_W is observed at lower feed and depth of cut (1^st level) but moderate spindle speed (2^nd level). Among the various machining constraints, the feed has an added effect on T_W because the increase in feed increases the impact load on the tool to increase the tool erosion by tool chatter (Varaprasad et al., 2014). From Table 5 and Figure 12, it is clear that feed has more contribution to T_W. T_W inclines to rise with increasing d. When the depth of cut is lower, there is less workpiece material stuck to the flank than at larger depths of cut (Sivaraman et al., 2012a, 2012b). Subsequently, the heat generated during the cutting is higher at the larger depth of cut due to higher cutting force. Accordingly, higher temperature and cutting force are the foremost details that cause the bond of workpiece metal onto the tool flank face (built-up edge) to create a false cutting edge to change the tool nomenclature periodically, thus speeding up T_W (Sivaraman et al., 2012b). As the optimal speed increases, this built-up edge phenomenon slows down due to the erosion of material from the built-up edge region, to minimize T_W. Hence increase in spindle speed reduces the tool wear at a drastic rate. It is clear from Figure 12 that low feed (1^st level) and moderate spindle speed (second level) are the favorable conditions for minimum tool wear.

Figure 12. Main effect plots of T_W for as-bought-spheroidized AISI1040 steel. (a) Feed. (b) Depth of cut. (c) Spindle speed.

Figure 13 shows that the S_R initially drops sharply with the rise in spindle speed. Afterward, it steadily decreases with an auxiliary growth in spindle speed. Generally, in ductile metals, a built-up edge is formed at lower spindle speeds. As the speed increases, the metal deposited on the cutting tool (built-up edge) periodically washes away to reflect the actual tool signature that reduces S_R. Similar phenomena happen in the case of feed from 1^st level to 2^nd level. Further increase in feed to the 3^rd level increases the S_R slightly at a lower rate (Bhuyan et al., 2018; Dhar et al., 2002). The increase in depth of cut decreases S_R till the 2^nd level and then further increases from the 2^nd level which enhances the heat generation by friction. This phenomenon increases the chatter and yields incomplete machining that yields higher S_R values (Passanha et al., 2022). It is clear from Figure 13 that moderate feed (0.26 mm/rev), depth of cut (0.4 mm), and high spindle speed (780 rpm) are the favorable conditions for minimum S_R. From Table 5 and Figure 13 it is clear that depth of cut has more contribution to S_R.

Figure 13. Main effect plots of S_R for as-bought-spheroidized AISI1040 steel. (a) Feed. (b) Depth of cut. (c) Spindle speed.

Figure 14 provides the systematic response optimization for T_W and S_R values. The result obtained shows that regression optimization can be used to obtain the optimum combination of T_W and S_R. Consequently, the values for dissimilar machining factors are found. The composite desirability, D value of 0.8174 indicates that the achieved optimized results have a good fit. Optimum T_W and S_R have been witnessed among the mentioned steel conditions within the chosen range of turning parameters. A validation test is carried out at the optimized constraint level to define the feasibility of the model (Saï, 2005; Sharma et al., 2008; Tamizharasan et al., 2006). The optimized values obtained from the regression equation are feed 0.39 mm/rev, depth of cut 0.6 mm, and spindle speed of 780 rpm. For the optimized parameter the experimental T_W is 0.039 mm and from the regression equation, it is 0.038 mm that provides the actual and optimized values that are nearer and within the range (Marimuthu & Edwin, 2014; Nikam & Kadam, 2016). Similarly, the experimental assessment of S_R is 2.89 µm and the optimized process result is 2.91 µm. The difference (error) between actual and optimized values in T_W and S_R is less than 5%. Table 6 provides the validation result for predicted and experimental values.

Figure 14. Combined optimized factors for T_W and S_R for as-bought-spheroidized AISI1040 steel. (a) Composite desirability. (b) Surface roughness. (c) Tool wear.

Table 6. Validation result for predicted and experimental values

Machining parameters			Predicted values		Experimental values		% Error of tool wear (TW)	% Error of surface roughness (SR)
Cutting speed (n) (rpm)	Feed (f) (mm/rev)	Depth of cut (d) (mm)	Tool wear (TW) (mm)	Surface roughness (SR) (μm)	Tool wear (TW) (mm)	Surface roughness (SR) (μm)
780	0.39	0.6	0.038	2.91	0.039	2.89	<5%	<5%

4. Conclusion

Spheroidization and related heat treatments are carried out for medium-carbon, low-alloy (AISI1040) steel, and a machinability study is performed by varying the initial room temperature structures. Experimental results of machinability are correlated with statistical analysis. Analysis of T_W and S_R by the statistical method is in good agreement with the experimental result. The machinability test is performed on a CNC turning center with a carbide-tipped tool at three levels of turning parameters by varying one parameter with respect to the other two parameters fixed. Results are analyzed, and the following conclusions are derived. As-bought spheroidized (A9) specimen shows better machinability (good surface finish and less tool wear) compared to other conditions. ANOVA result shows that feed has a 67% contribution to T_W whereas the depth of cut has a 71.91% contribution to S_R. The optimized regression values obtained for machining parameters are feed 0.39 mm/rev, depth of cut 0.6 mm, and spindle speed 780 rpm with composite desirability of 0.8174. The experimental values obtained for T_W and S_R values for the optimized machining parameters are 0.039 mm and 2.89 μm, respectively, and the difference between the actual and optimized values is less than 5%.

Nomenclature

AISI	=	American Iron and Steel Institute
wt.%	=	Weight percentage
CNC	=	Computer Numerical Control
TW	=	Tool wear
SR	=	Surface roughness
n	=	Spindle speed
f	=	Feed
d	=	Depth of cut
MRR	=	Material removal rate

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Notes on contributors

Harisha S R

Harisha S. R. is working as an Assistant Professor-Senior Scale in the Department of Mechanical & Industrial Engineering, Manipal Institute of Technology, Manipal Academy of Higher Education, Manipal. He holds B.E. (Mechanical Engineering), M.Tech. (Manufacturing Engineering) and Ph.D. (Heat treatment) degrees. He has 13 years of teaching experience. His areas of interest include Engineering materials, Heat treatment of metals and composites, and Machinability.

Sathyashankara Sharma

Sathyashankara Sharma is working as a Professor and Head in the Department of Mechanical & Industrial Engineering, MIT, MAHE, Manipal. He holds B.E. (Industrial and Production Engineering), M.Tech. (Materials Engineering) and Ph.D. (Materials Engineering) degrees.

Ramakrishna Vikas Sadanand

Ramakrishna Vikas Sadanand is working as an Assistant Professor-Senior Scale in the Department of Mechanical & Industrial Engineering, Manipal Institute of Technology, Manipal Academy of Higher Education, Manipal. He holds B.E. (Mechanical Engineering) and M.Tech. (Manufacturing Engineering).

Achutha Kini U

Achutha Kini U is a Professor of Mechanical & Industrial Engineering, at MIT, MAHE, Manipal. He has more than 30 years of experience in teaching.

Raviraj Shetty

Raviraj Shetty is a Professor of Mechanical & Industrial Engineering, at MIT, MAHE, Manipal. He has about 22 years of experience in teaching.

Sathish Rao U

Sathish Rao U is an Associate Professor of Mechanical & Industrial Engineering, at MIT, MAHE, Manipal. He has about 22 years of experience in teaching.