Simulation of the influence of cutting speed and feed rate on tool life in hard turning of AISI 4140 steel

Tool life is the time during which a tool has cut satisfactorily. Tool life is highly important in cutting processes, which include sawing, turning, milling, drilling, grinding, and broaching, because considerable time and cost have been lost from replacing and resetting cutting tools. Generally, the tool life model is developed by applying Taylor’s tool life equation on the basis of the tool wear criterion because it is straightforward to measure quantitively. Taylor’s tool life is commonly employed to predict the tool life of drilling tools. For example, John et al. [1] proposed a modified Taylor’s tool life equation for high-speed steel tools when drilling polypropylene-based, natural fiber-reinforced composites. Moreover, Farid et al. [2] reported Taylor’s tool life equation on the basis of the criterion of 100-μm flank wear at an outer corner while highspeed drilling of A383 Al-Si alloy using uncoated, TiAlN- and AlTiN-coated tools with various cutting parameters based on a central composite design of experiments. In milling operations, Milan et al. [3] proposed an extended Taylor’s tool life equation based on the tool life criterion of average flank wear of 0.4 mm when milling the calcium-treated and nontreated mold steels (AISI P20 and P 20 UF steels) using the three coatings TiN-, Al₂O₃-, and TiCN-coated carbide cutting tools. Kuo et al. [4] formulated Taylor’s tool life equation as a function of cutting time based on the maximum flank wear criterion of 0.3 mm during milling of superalloy Inconel 718 using uncoated, TiN-, and TiCN-coated tools. In addition, the modified Taylor’s equations were used to estimate the tool life as a function of feed per tooth and cutting speed based on the proposed procedure [5]. In turning operation, Taylor’s tool life equation was also used to estimate the tool life of cutting tools when machining various workpiece materials, including AISI 52100 steel using polycrystalline cubic boron nitrides [6], AISI 302 stainless steel using carbide inserts of P10 grade with a triangular shape of 16 mm in side length and 3.18 mm in thickness [7], martensitic stainless steel using TiAlN [8], AISI H13 steel with WC-TiC-Co ultrafine cemented carbide [9], Ti6Al4V (UNS R56400) alloy utilizing with uncoated WC-Co inserts [10], martensitic S41000, and supermartensitic S41426 steels using carbide tools coated with TiC/TiCN/TiN by the chemical vapor deposition coating method [11].

Not only is Taylor’s tool life equation developed on the basis of the criterion of tool wear, but it also can be evaluated on the basis of other criteria. Many factors influence tool life, including tool materials, tool shapes, machining operating conditions, coolant types, and workpiece surface quality requirements. Taylor’s tool life equation is utilized to describe the relationship between tool life and machining parameters. For example, according to Taylor’s tool life model, Chen and Luo [12] successfully demonstrated the relationship between tool wear and chip morphology with an average error of experimental repeatability of 4.5%. This indicated that chip surface chromaticity was used to predict tool wear based on the development of Taylor’s tool life equation during the cutting of stainless steel (2316ISO-B MOD) with a TiAlN coated insert. In a drilling operation, the tool life equation was utilized to create a relationship between tool life and cutting speed and feed rate during machining of forging brass with uncoated tungsten carbide (WC) and AlCrN-coated WC tools using the criterion of exit burr height of 0.16 mm [13]. Kovac et al. [14] also proposed an extended Taylor’s functional relationship between tool life and cutting speed, feed per tooth, and depth of cut through measurements of tool-workpiece thermocouple temperature during milling of AISI 1060 steel with a carbide insert of 125 mm in diameter. They stated that the proposed method for Taylor’s tool life equation was faster work pace and lower cost than the traditional method. Kundrák and Pálmai [15] proposed tool life models, which were modified from Taylor’s tool life equation based on the number of turned boreholes in turning internal cylindrical surfaces of 100Cr6 hardened steel using cubic boron nitride K10. Meanwhile, the lifetime of a machine component (i.e., the fishing net weaving machine component of a fishing net weaving machine) was evaluated by the weight of material removed from the component during wear tests [16].

Surface quality and dimensional accuracy of machined parts are crucial for many applications. Manufacturers need to evaluate the machinability performances during computer numerical control (CNC) turning of various materials using a variety of tool inserts when a machined workpiece’s surface roughness (R_a) begins to degrade and fails to match the customer’s requirements for workpiece surface quality, leading to high manufacturing costs. Normally, tool wear increases, resulting in increasing Ra due to vibration of the tool tip at high acceleration [17]. Consequently, R_a can be indirectly used as a criterion for assessing cutting tool deterioration.

This research aims to develop and evaluate the tool life models of Al₂O₃+TiC and TiN+AlCrN during turning of AISI 4140 steel with different cutting speeds and feed rates using Taylor’s tool life equation with a maximum R_a of 0.8 μm for the tool life criterion. Monte Carlo simulation (MCS) was also used to evaluate the sensitivity of cutting speed and feed rate to tool life. It has been pointed out that AISI 4140 steel has been extensively used in a wide range of industrial applications including component parts for vehicles such as shafts, crankshafts, gears, connecting rods, and collars as well as machinery parts in the oil and gas industry and other industries like motor shafts, pump shafts, honed cylinder tubes for hydraulic, pneumatic cylinder barrels, bolts, forging die, jigs, and frames due to its toughness, abrasion and impact resistance, and high fatigue strength [18].

Figure 1 shows an overview of the experimental work and the machined samples for the hard-turning study. The workpiece material was AISI 4140 chromium molybdenum steel bars. The chemical compositions were as follows: 0.47% C, 1.05% Cr, 0.01% P, 0.01% S, 0.26% Si, 0.67% Mn, 0.2% Mo, 0.0042% V, 0.012% Al, and 0.19% Cu. Its hardness value was 58 HRC.

Fig. 1.

The turning experiments were carried out in a Fanuc CNC lathe machine (Takisawa model: NEX-106). The cutting tools were mixed ceramic insert (Al₂O₃+TiC) (Tungaloy, Japan) and TiN+AlCrN-coated insert. The ISO designation of the inserts and tool holder was SNGA120408TH10 and ASBNR2020K12-A coded external tool holder.

The experiments were carried out on AISI 4140 steel bar samples with two types of tool inserts at three cutting speeds (200, 220, 240 m/min) and three feed rates (0.04, 0.06, 0.08 mm/rev) while keeping the depth of cut constant at 0.1 mm.

A commercial surface roughness measuring tester (Mitutoyo model: Surf test SJ-210) with a cut-off length of 0.8 mm and a sampling length of 5 mm was used to measure R_a. R_a measurements were taken at three different locations on each of the machined workpieces. The maximum R_a of 0.8 μm was set with the customers’ requirements for quality acceptance of automotive parts [19]. The wear of the tool insert was investigated using a scanning electron microscope (Hitachi model S-4700) with a 20-kV accelerating voltage.

The turning experiments were carried out to obtain data for tool life models of the two tool inserts. Taylor’s equation was used to obtain the relationships between tool life and cutting speed and feed rate of the two tool inserts. The equation is defined as: 1 T(v)1/n(f)1/m=c \[T{{(v)}^{1/n}}{{(f)}^{1/m}}=c\] where v represents cutting speed, T denotes tool life, and f indicates feed rate [20, 21]. The exponents n, m, and the constant c are obtained from Taylor’s equation fitting to the observations of turning of AISI 4140 samples using both tool inserts.

Monte Carlo simulation (MCS) is a sampling experiment used for approximating the distribution of an outcome factor that is dependent on various probabilistic input factors [22]. It is regularly used to evaluate the sensitivity of processes in various areas. This method has been used in numerous studies and many application areas since World War II [23]. Marci et al. [24] compared the thin films sputtering yields for low-energy noble gas ions bombardment on MgO and Mg(OH)₂ targets in the binary collision approximation. This method demonstrated that the sputtering depends on both the incident particle and the targets. In the cold spray process of mixtures of alumina content and aluminum particles, the effects of particle size, particle size distribution, and behavior of deposition efficiency were evaluated and compared between the numerical simulation and the experimental data [25]. In the grinding process of machined parts, MCS was adopted to identify the sensitivities in surface roughness of machined parts after obtaining the appropriate operating conditions of wheel speed and feed rate of the grinding process using a central composite design of experiment and response surface methodology [26]. In addition, Kahraman and Öztürk [27] used MCS in sensitivity analyses to evaluate the effects of grinding process parameters on the average surface roughness of the hard-brittle materials [27] and flat glass [28]. To improve the operational reliability of a thermal power plant, Kostić et al. [29] simulated the failure interaction of the system components based on the Weibull probability distribution. In milling operation of aluminum 7075, Kahraman et al. [30] investigated the effect of cutting speed, feed rate, and depth of cut on average surface roughness using Taguchi design of experiment and examined the sensitivity analysis of R_a of the machined parts using MCS. In a turning operation of AISI D2 steel, Sarnobat and Raval [31] evaluated the influence of tool edge geometry and work piece hardness on averages of the tri-axial surface residual stresses, surface roughness, and work hardening in the affected machining region. They studied the sensitivity of the tri-axial stresses and surface roughness of the machine parts to the process parameters by applying MCS. Thomas et al. [32] presented three cases regarding design, development, and implementation of MCS to define a bearing replacement strategy due to bearing failures within a steel-processing industry for a total productive maintenance program. They stated that the bearing replacement strategy contributed a simple approach to significantly improving system reliability and achievement through less downtime and higher cost savings. Timata and Saikaew [13] studied the sensitivity of cutting speed and feed rate to tool life prediction in tool life evaluation by adopting an MCS based on Taylor’s tool life equation.

Therefore, MCS was utilized in this study to evaluate tool life predictions. Sensitivity of cutting speed and feed rate to tool life was studied while turning AISI 4140 samples using the criterion of average R_a of 0.8 μm. The simulation was performed by varying cutting speeds below and above the cutting speed of 220 m/min while keeping the feed rate constant at 0.06 mm/rev and by varying feed rates below and above the feed rate of 0.06 mm/rev while keeping the cutting speed constant at 220 m/min. Minitab software package version 16 was used to run MCS.

The tool life performance of Al₂O₃+TiC and TiN+AlCrN tool inserts were studied for dry turning of AISI 4140 steel. Based on the comparison of test results of Taylor’s tool life equations and MCS, the following conclusions could be drawn:

The machined part’s surface quality was effectively employed as a criterion for monitoring cutting tool deterioration.