Characterization of material strain and thermal softening effects in the cutting process

Highlights

• The high-speed filming and induction heating are introduced to the cutting process.

• At higher heating temperatures, the thickness of the PSZ becomes greater.

• The strain softening and reduced thermal softening are experimentally observed.

Abstract

Accurate descriptions of workpiece behaviors are indispensable for achieving reliable simulations of the cutting process. However, the material plastic constitutive models obtained through conventional material tests do not apply in the ranges of strain and strain rate encountered in the realistic machining processes. To address this issue, in this study, we attempt to develop a methodology to understand and identify the plastic deformation behaviors based on high-speed filming and induction preheating during the cutting tests. Different levels of strain, strain rate, and temperature are realized by varying the rake angle, cutting velocity, and initial workpiece temperature, respectively. The plastic deformation and temperature rise in the primary shear zone are characterized by the fine-scale digital image correlation technique and heat convection–conduction equation, respectively, thus rendering the machining test into a high-dynamic-material testing method. The material exhibits strain softening in the primary shear zone and a reduced thermal softening effect under rapid heating conditions. These initial findings can deepen the understanding of material behaviors during the cutting process and can be further developed for implementation in numerical machining models.

Introduction

Numerical simulations of the metal cutting process are necessary to increase productivity and optimize tool wear [1], [2] and surface integrity [3], [4]. To apply such approaches, the workpiece behaviors at high plastic strain, strain rate, and temperature as encountered during the cutting process must be determined. However, characterization of the material behaviors under realistic machining conditions is still a challenge because the extremely intense plastic deformation and rapid heating may lead to dynamic recrystallization [5], [6] and hinder the microstructure transformation [7], respectively.

The split Hopkinson bar (SHB) tests depicted in Fig. 1(a) have been conventionally adopted to identify the material’s constitutive parameters for machining simulation. In the SHB tests, by varying the incident velocity of the incident bar and the heating temperature, compressive tests can be conducted with strain rate and temperature reaching up to several $10^3$/s and melting point, respectively.

However, the large strains, high strain rates, and high temperatures concentrated locally in the deformation zones near the cutting tool edge are significantly greater than those achieved by the SHB test. Consequently, extrapolations of the flow stress outside the SHB testing ranges need to be performed. To address the issue of insufficient testing ranges of the SHB, the orthogonal cutting process has been employed as a material test method [8], [9] for intense shear deformation in the primary shear zone (PSZ) as illustrated in Fig. 1(b). With the many developed analytical machining predictive models, the strain, strain rate, temperature, and flow stress along the shear plane (denoted as SP in Fig. 1(b)) can be derived using trial material parameters [9], [10], [11]. Selecting the shear plane for analysis rather than the rear part of the PSZ, which has larger strains, is to simplify the calculation of the shear stress through direct decomposition of the measured cutting forces without having to consider the stresses in the secondary or tertiary shear zone. Subsequently, the material parameters are tuned until a close match is achieved between the calculated and experimentally observed quantities, such as cutting forces, shear angles, and cutting temperatures [12], [13]. The comparison of the SHB with the metal cutting process as a material test method is presented in Table 1.

The samples commonly used for the SHB tests are cylinders with a size of a few millimeters, yielding a quasi-homogeneous plastic deformation, thus reducing the calculation to a one-dimensional (1D) problem. Although the plastic deformation in the through-thickness direction of the workpiece is almost uniform in the case of orthogonal cutting, the thickness of the PSZ is commonly around tens to a few hundred micrometers, exaggerating the heterogeneous behaviors of the material microstructure. Consequently, the shear angle oscillates during the cutting process, which has been observed using high-speed filming [14], [15]. In addition, the material deformations along the streamlines in the PSZ tend to be non-uniform, violating the general assumption of a uniformly distributed strain rate made by analytical models [16]. Therefore, accurate determinations of the strain rate, strain, and temperature are required when the metal cutting process serves as a material test method.

The most common sensors to determine the effective stress of the material are strain gauges attached to the incident and transmitted bars in the SHB tests and a dynamometer attached to the cutting tool or workpiece in the cutting process, respectively. Owing to the high dynamic response of strain gauges, rapid changes in the effective stress can be captured. In contrast, the dynamometer is nearly impossible to capture the dynamic changes in the flow stress, particularly for serrated chip formation owing to its low natural frequency. Without high-bandwidth cutting force measurements, i.e. through the measurements of the dynamic elastic deformation field of the workpiece [17], a steady-state cutting process is preferred when a dynamometer is adopted to derive the flow stress of the material in the PSZ.

In the SHB tests, the achievable maximum strain rate is approximately a few thousand per second, limited by the speed of the incident bar. Meanwhile, strains from 0 to 0.2 can generally be obtained. With regard to the temperature dependence of the yield stress, a heating device using induction or resistive heating is commonly used to achieve thousands of degrees Celsius. The plastic deformation in the cutting process can possess a strain rate of a few thousand to hundreds of thousands per second by varying the cutting velocity. However, the material temperature in the PSZ is typically low because the inherent plastic deformation serves as the sole heat source, thus making the identified material’s model incapable of describing the thermal softening in the secondary or tertiary shear zone. Therefore, an external heat source is required to further enhance the temperature rise, which has rarely been addressed in the inverse identification methods based on the cutting process [18], [19].

In addition, it has been demonstrated that the commonly used parameters, such as cutting forces and chip shapes, are not generally sensitive to changes in the material parameters. For example, an increase in strain hardening or strain rate sensitivity causes an increase in temperature [20], [21]. However, the subsequent thermal softening causes a reduction in the flow stress, indicating that multiple combinations of hardening and softening may lead to similar cutting forces and chip shapes. Therefore, different levels of strain, strain rate, and temperature are required to avoid non-uniqueness during the inverse identification of material parameters based on the orthogonal cutting process.

In this study, a material testing and identification method based on experimental measurements of the orthogonal cutting process is proposed using the flowchart depicted in Fig. 2. The measurements of the strain and strain rate are realized by high-speed filming combined with the digital image correlation (DIC) technique, which has been widely used to study subsurface deformation [17], [22] and chip formation processes [23], [24], [25], [26].

Furthermore, the initial workpiece temperature is elevated by high-frequency induction heating, and the temperature rise in the PSZ is calculated based on a 1D steady-state heat transfer equation. Considerably wider ranges of temperature, strain, and strain rate are achieved by varying the heating temperature, rake angle, and cutting velocity in comparison with previous studies [18], [19]. Furthermore, the plastic parameters of the material are genetically optimized to minimize the discrepancy between the calculated and measured shear stresses along the shear plane.