Influences of tool rake angle and cutting speed on ceramic tool failure in continuous and intermittent turning of hardened steel

Influences of tool rake angle and cutting speed on ceramic tool failure in continuous and intermittent turning of hardened steel

Ceramics International ∎ (∎∎∎∎) ∎∎∎–∎∎∎ Contents lists available at ScienceDirect Ceramics International journal homepage: www.elsevier.com/locate/c...

4MB Sizes 1 Downloads 39 Views

Recommend Documents

No documents
Ceramics International ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Contents lists available at ScienceDirect

Ceramics International journal homepage: www.elsevier.com/locate/ceramint

Influences of tool rake angle and cutting speed on ceramic tool failure in continuous and intermittent turning of hardened steel Xiaobin Cui a,n, Dong Wang b, Jingxia Guo c a

School of Mechanical and Power Engineering, Henan Polytechnic University, Jiaozuo 454003, PR China School of Mechanical Engineering, Xi'an Technological University, Xi'an 710021, PR China c School of Energy Science and Engineering, Henan Polytechnic University, Jiaozuo 454003, PR China b

art ic l e i nf o

a b s t r a c t

Article history: Received 28 March 2016 Received in revised form 29 April 2016 Accepted 4 May 2016

Theoretical and experimental analysis were conducted to reveal the influences of tool rake angle and cutting speed on ceramic tool failure in continuous and intermittent turning of hardened steel. Based on the tool material microstructure model, the initial state of the tool material microstructure was evaluated using micromechanics and damage mechanics. The initial damage of the tool material and tool stress were integrated using the concept of damage equivalent stress. Damage equivalent stress on the tool body was analyzed for continuous and intermittent turning. The size of the representative element was obtained as 21  21 grains or 47 mm  47 mm. The initial damage of the tool material was 0.0066. Peak values of damage equivalent stress arose in intermittent turning, which was not found in continuous turning. Compared to continuous turning, the position where the maximum damage equivalent stress on the tool body arose varied substantially in intermittent turning. The highest damage equivalent stress in the whole cutting process decreased first and then increased as the tool rake angle evolved from  10° to  2°. Relatively long tool life can be obtained at the tool rake angle of 6° and  8° for continuous and intermittent turning, respectively. When the cutting speed increased from 60 m/min to 140 m/min, the highest damage equivalent stress in the whole cutting process became larger, indicating that tool life decreased as the cutting speed increased. As the tool rake angle or cutting speed increased, the position where the highest damage equivalent stress appeared became closer to the cutting edge. Compared to continuous turning, tool fracture in intermittent turning was more likely to begin at a position further away from the cutting edge. The optimum combination of tool rake angle γo and cutting speed v was γo ¼  6° and v ¼ 60 m/min for continuous turning. The combination changed to be γo ¼  8° and v ¼60 m/min in intermittent turning. & 2016 Elsevier Ltd and Techna Group S.r.l. All rights reserved.

Keywords: Tool rake angle Cutting speed Ceramic tool Tool failure

1. Introduction Al2O3-based ceramics possess a series of advantages such as high hardness, high heat resistance, high wear resistance and good chemical stability [1]. They have been considered as one of the most suitable tool materials for the machining of hardened steel. However, drawbacks such as lower fracture toughness and lower strength make the ceramic tools more inclined to fracture in the machining of hardened steel. The manufacturing cost of Al2O3-based ceramic tools is also relatively high [2]. Tool geometry [3] and cutting parameters [4] have substantial influences on the failure of ceramic tools. It is of great importance to reveal the influences of tool geometry and cutting parameters on the ceramic tool failure. Thus, valuable information can be provided for prolonging the ceramic tool life and reducing machining cost. n

Corresponding author. E-mail address: [email protected] (X. Cui).

There have been great amounts of valuable studies relating to the influences of tool geometry parameters [5–8] and cutting parameters [9–11] on tool failure in continuous turning. 3-D finite element simulation and turning tests were performed by Karpat and Özel [5] to investigate the effects of cutting edge geometry on tool wear in hard turning. It was found that the use of polycrystalline cubic boron nitrite (PCBN) tools with honed microgeometry resulted in lower tool flank wear. Hard turning was conducted by Özel et al. [6] to investigate variable edge design PCBN inserts. The advantages of variable edge micro-geometry design were revealed by means of analyzing the temperature distributions and tool wear contours. Özel et al. [7] studied the wear of ceramic wiper design inserts in finish turning of AISI D2 hardened steel. The results showed that the established neural network models were suitable to predict tool wear for a series of cutting conditions. Hard turning of hardened steel was conducted by Gaitonde [8] with ceramic tools used. It was revealed that wiper insert performed better in terms of tool wear and the conventional

http://dx.doi.org/10.1016/j.ceramint.2016.05.015 0272-8842/& 2016 Elsevier Ltd and Techna Group S.r.l. All rights reserved.

Please cite this article as: X. Cui, et al., Influences of tool rake angle and cutting speed on ceramic tool failure in continuous and intermittent turning of hardened steel, Ceramics International (2016), http://dx.doi.org/10.1016/j.ceramint.2016.05.015i

X. Cui et al. / Ceramics International ∎ (∎∎∎∎) ∎∎∎–∎∎∎

2

insert was useful in reducing the machining force, power and specific cutting force. Yin et al. [9] investigated cutting performance and failure modes of an Al2O3/TiC micro-nano-composite ceramic tool in turning of austenitic stainless steel. The optimum cutting parameter combination was identified by means of orthogonal test and range analysis. Aslan et al. [10] studied the cutting parameter optimization in turning AISI 4140 steel with Al2O3 þTiCN mixed ceramic tool. Combined effects of cutting parameters on tool wear were investigated based on an orthogonal array and the analysis of variance (ANOVA). Chinchanikar and Choudhury [11] investigated the effect of cutting parameters on the performance of coated carbide tool in turning of hardened steel. The relationship between the cutting parameters and tool life was established using multiple linear regression models. Many researches [12–14] were conducted to investigate the influences of cutting parameters on tool failure in the field of intermittent turning. However, there existed scant study on the effects induced by tool geometry. Zhao and Ai [12] investigated the cutting performance of an Al2O3-(W, Ti)C functionally gradient ceramic tool in intermittent turning. The analysis results showed that the Functionally Gradient Material (FGM) tool exhibited similar wear mechanisms to those of homogeneous ceramic tools when the cutting speed was relatively low. However, the failure modes of FGM tool were different from those of homogeneous ceramic tools at higher cutting speed. Intermittent turning of AISI 1045 hardened steel was conducted by Cui et al. [13] to analyze the effects of cutting speed on the failure mechanisms of Al2O3-(W, Ti)C ceramic tools. It was found that there existed critical cutting speed above which adhesive wear of the cutting tool became less serious. Zhao et al. [14] analyzed the cutting performance and failure mechanisms of an Al2O3/WC/TiC micro-nanocomposite ceramic tool in intermittent turning. The analysis results indicated that the tool life increased when the cutting speed became higher. This can be attributed to the synergistic strengthening/ toughening mechanisms induced by the WC microparticles and TiC nanoparticles in the tool material. In the cutting process, the cutting tools with certain initial microstructure withstand the mechanical and thermal loads resulted from interaction between the cutting tool and the workpiece. Both the initial state of the tool material microstructure and the external loads influence the tool failure substantially. However, most of these previous researches relating to the influences of tool geometry parameters and cutting parameters on tool failure were performed without evaluation of the initial state of the tool material. Furthermore, the integrated effects of the tool's initial state and the external loads were scarcely discussed in the previous studies. Damage mechanics is often used to investigate the response of materials weakened by randomly distributed microcracks which are of irregular shapes and random in size and orientation [15]. The evolution of micro-defects density in the materials is observed as the gradual reduction of the stiffness on macro scale. Damage mechanics has been widely applied as an engineering tool in industry [16]. It provides the opportunity for quantitative evaluation of the initial state of tool material microstructure. Moreover, the concept of damage equivalent stress [16] in damage mechanics is closely related to material damage and stress distribution. It can be used to reflect the integrated effects of the initial state of the tool material and external loads. Taking the initial state of the tool material microstructure and the external loads into consideration, the influences of tool rake angle and

cutting speed on ceramic tool failure in continuous and intermittent turning of hardened steel were investigated and compared in the present work. Microstructure of the ceramic tool material was modeled using Voronoi Tessellation hybrid random algorithms. The initial state of the tool material microstructure was analyzed based on micromechanics and damage mechanics. Tool material microstructure was reflected on the cutting zone of the ceramic tool model. Finite element simulation was conducted for different combinations of tool rake angle and cutting speed to acquire the tool stress. On the basis of the initial state of the tool material and tool stress, the evolutions of damage equivalent stress in the cutting process were analyzed. Based on the analysis results, the influences of tool rake angle and cutting speed on tool failure were revealed. The optimum combination of tool rake angle and cutting speed was identified in terms of tool life.

2. Finite element simulation and experimental procedures 2.1. Finite element simulation of continuous and intermittent turning of hardened steel For the purpose of acquiring the tool stress in the cutting process, finite element simulation of continuous and intermittent turning were performed in the present work. Abaqus/Explicit software was employed in the simulation of cutting process. A micro-composite Al2O3/(W, Ti)C ceramic tool was studied in the simulation. AerMet100 hardened steel (46–48 HRC) was used as the workpiece. Material properties of the ceramic cutting tool and the workpiece are shown in Tables 1 and 2, respectively. The use of a proper material-constitutive model for the workpiece is essential for simulating the cutting process successfully. The Johnson-Cook model has been extensively applied to analyze the deformation of the workpiece in the machining process. The Johnson-Cook constitutive equation can be defined as:

⎡ ⎛ ε¯⋅ ⎞⎤⎡ ⎛ T − T ⎞m⎤ r σ¯ = [A + B(ε¯)n]⎢ 1 + Cln⎜ ⋅ ⎟⎥⎢ 1−⎜ a ⎟ ⎥ ⎝ ε¯0 ⎠⎦⎢⎣ ⎝ Tm − Tr ⎠ ⎥⎦ ⎣

(1)

where ε , ε⋅ , Ta and σ are shear strain, shear strain rate, absolute temperature and shear stress, respectively. It can be found that the material properties are dominated by the yield strength A, the hardening modulus B, the reference plastic strain ε⋅0 , the strain rate sensitivity C, the strain hardening exponent n, the thermal softening coefficient m, the reference temperature Tr and the melting temperature Tm. On the basis of the work by Su [17], these Johnson-Cook parameters were set as A¼ 1450 MPa, B¼ 455 MPa, C¼ 0.01, n ¼ 0.5 and m ¼1.8. Fig. 1(a) and (b) show the schematics of finite element simulation for continuous and intermittent turning, respectively. In intermittent turning, the ratio of cutting length to air-cutting length was fixed as 4.89 as shown in Fig. 1(b). Microstructure of the tool material was reflected on the quadrilateral region Rz close to the cutting edge as shown in Fig. 1. Five different tool rake angles γo (  10°,  8°,  6°,  4° and  2°) were applied in the simulation. The clearance angle α0 was set as 8°. Cutting speeds v ranging from 60 to 140 m/min at an interval of 20 m/min were used. Feed rate f was fixed to be 0.2 mm/r. Since both the tool rake angle γo and cutting speed v have 5 different values, there existed 25 combinations of γo and v. Finite element

Table 1 Material properties of the ceramic cutting tool. Density (ρ) 3

4100 kg/m

Young's modulus (E) 426 GPa

Poisson's ratio (ν) 0.26

Specific heat (Cp) 860 J Kg

1

K

1

Thermal conductivity (λ) 8.25 W m

1

K

1

Vicker's hardness (HV) 21.6 GPa

Please cite this article as: X. Cui, et al., Influences of tool rake angle and cutting speed on ceramic tool failure in continuous and intermittent turning of hardened steel, Ceramics International (2016), http://dx.doi.org/10.1016/j.ceramint.2016.05.015i

X. Cui et al. / Ceramics International ∎ (∎∎∎∎) ∎∎∎–∎∎∎

simulation was performed for each combination used in continuous and intermittent turning. 2.2. Experimental procedures Continuous and intermittent turning tests were conducted for each combination of tool rake angle and cutting speed used in the finite element simulation. The turning test was replicated three times for each combination. The cutting conditions were set to be the same to those in the simulations. The cutting forces were measured using a Kistler piezoelectric dynamometer (Kistler 9441) during the turning process as shown in Fig. 2(a). The charges generated at the dynamometer were amplified by means of a multi-channel charge amplifier (Kistler 5807A) as shown in Fig. 2 (b). The sampling frequency was 4000 Hz. In the cutting process, the tool flank wear was observed with a microscope (Fig. 2(c)) periodically. Tool life was recorded as the tool flank wear VB reached or increased over 0.3 mm [18]. The failed cutting tool was analyzed using scanning electron microscopy (SEM) (JSM-6510LV, Japan).

3. Results and discussion 3.1. Analysis of the initial state of the tool material In the present work, Voronoi diagram [19–21] and random method were used to build the two-dimensional model for tool material microstructure. Microstructure of the tool material matrix Table 2 Material properties of the workpiece. Density (ρw) Young's modulus (Ew)

Poisson's ratio (νw)

Specific heat (Cpw) Thermal conductivity (λw)

7840 kg/m3

0.29

464 J Kg  1 K  1

195.3 MPa

45 W m  1 K  1

3

was constructed based on Voronoi diagram. The incremental method was utilized to form the Voronoi diagram on the basis of hybrid programming of MATLAB and VCþ þ. The random method was applied to build the reinforced phase of the tool material. First, a set of cell points with uniform distribution were generated using MATLAB. The cell points were located in tool material matrix one after another. The increase of the cell points in the matrix led to the update of the Voronoi diagram. This process ceased when the average grain diameter became equal to the specified value. After that, convex quadrilaterals were inserted into the tool material matrix randomly in order to form the reinforced phase. The positions of the convex quadrilaterals complied with the uniform distribution function. During this process, Boolean operations were used to eliminate the overlapping of cells. The total area of the newly added convex quadrilaterals was evaluated and the whole process stopped as the total area increased to be the predetermined value. SEM analysis of the tool material microstructure showed that the average grain size was 2.24 mm. The volume fraction of the reinforced phase was 18%. On the basis of the work by Wang et al. [21], microstructure of the ceramic tool material can be modeled as shown in Fig. 3. The light-colored and deep colored cells in Fig. 3 correspond to the Al2O3 matrix and the (W, Ti)C reinforced phase of the tool material, respectively. For the purpose of identifying the size of the representative element, statistical properties and average values of the grain diameters were studied. Probability density distributions of grain diameters were analyzed for regions with different sizes. Square regions containing 7  7, 14  14, 21  21 and 28  28 grains were investigated and Fig. 4 shows the probability density distributions and average grain diameters for these regions. It can be found from Fig. 4 that the grain diameter followed a normal distribution. It has been mentioned that SEM images of the tool material microstructure were analyzed and the average grain diameter was obtained as 2.24 mm. It can be seen from Fig. 4 that when the size of the square region increased, the average grain diameter obtained from the microstructure model became closer to that acquired based on analysis of the material microstructure image. For example, when the square region contained 21  21 grains, the

Fig. 1. The schematic of finite element simulation for continuous and intermittent turning.

Please cite this article as: X. Cui, et al., Influences of tool rake angle and cutting speed on ceramic tool failure in continuous and intermittent turning of hardened steel, Ceramics International (2016), http://dx.doi.org/10.1016/j.ceramint.2016.05.015i

4

X. Cui et al. / Ceramics International ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Fig. 2. Experimental apparatus used in the turning tests.

Fig. 4. Statistical properties and average grain diameters for different square regions.

Fig. 3. Microstructure model of the ceramic tool material under consideration.

average grain diameter obtained in the model and the SEM image became very close to each other. Taking the analysis of statistical properties and average values of the grain diameters into account, the size of the representative element can be determined as 21  21 grains or 47 mm  47 mm. Damage mechanics analysis of the representative element was conducted to determine the initial damage of the ceramic tool material. In the present work, the initial damage was considered as the intrinsic property of the ceramic tool material. In the cutting process, the ceramic cutting tools withstood the tri-axial stress s induced by mechanical and thermal loads. Damage equivalent stress [16] s* of the tool material can be calculated as:

⎧ ⎫1/2 1−D σ* = ⎨ (1 + ν )⟨σ ⟩+ : ⟨σ ⟩+ −ν⟨trσ ⟩2 + (1 + ν )⟨σ ⟩− : ⟨σ ⟩− −ν⟨−trσ ⟩2 ⎤⎦⎬ [ ⎩ ⎭ 1−hD

(2)

where ν is the Poisson's ratio of the tool material, D is the damage of the tool material and h is on the order of 0.2. It can be found that when the tool material is under pure compressive stress sc, s* can be expressed as:

⎛ 1−D ⎞1/2 ⎟ σc σ* = ⎜ ⎝ 1−hD ⎠

(3)

On the basis of the hypothesis that the amount of elastic strain energy is the same for the tri-axial stress state and the uni-axial compressive stress state, the tri-axial stress s can be transformed to be the uni-axial compressive stress sc. Based on Eqs. (2) and (3),

Please cite this article as: X. Cui, et al., Influences of tool rake angle and cutting speed on ceramic tool failure in continuous and intermittent turning of hardened steel, Ceramics International (2016), http://dx.doi.org/10.1016/j.ceramint.2016.05.015i

X. Cui et al. / Ceramics International ∎ (∎∎∎∎) ∎∎∎–∎∎∎

the magnitude of sc can be calculated as:

constants and S12 ¼S21. With 1/2

⎧ ⎫ 1−D σc = ⎨ (1 + ν )⟨σ ⟩+ : ⟨σ ⟩+ −ν⟨trσ ⟩2 + [ (1 + ν)⟨σ⟩− : ⟨σ⟩−−ν⟨−trσ⟩2⎤⎦⎬⎭ ⎩ 1−hD ⎛ 1−D ⎞−1/2 ⎜ ⎟ ⎝ 1−hD ⎠

σc =

(4)

In the present work, analysis of the initial damage of the tool material focused on the representative element under pure compressive stress sc. According to the previous work [22–24], a sliding crack model with an array of sliding microcracks can be established for the representative material element as shown in Fig. 5(a). According to the work by Kemeny [24], the distance 2w between the initial microcracks is considered to be four times the microcrack length. On the basis of the study by Ashby and Hallam [25], the individual microcrack can be analyzed using the sliding crack model as shown in Fig. 5(b). In Figs. 5(b), 2c is the length of the initial crack. l is the length of the tensile crack. θ is used to represent the angle between the initial microcrack and the tensile crack. Based on the study by Fredrich et al. [26], the initial crack length 2c can be considered as the grain size of the ceramic tool material. Angle θ denoted in Fig. 5(b) is assumed to be π/4. Total strains of the representative element can be calculated as: e ⎛ ϵ1 ⎞ ⎛⎜ ϵ1 + Δϵ1 ⎞⎟ ⎜ ⎟ = ⎝ ϵ 2 ⎠ ⎜⎝ ϵ e + Δϵ 2 ⎟⎠ 2 ⎡ ( κ + 1)( ν + 1) ⎢⎢ 1 = ⎢ κ −3 4E ⎢ ⎣ κ+1

κ −3 ⎤ ⎥ ⎡S ⎤⎛ σc ⎞ κ + 1 ⎥⎜⎛ σc ⎟⎞ 11 S12 ⎜ ⎟ ⎥ + N⎢ ⎥⎝ 0 ⎠ ⎣⎢ S 21 S 22 ⎥⎦⎝ 0 ⎠ 1 ⎥ ⎦

where ε1e and ε2e are the elastic strains of the representative element without damage. Δε1 and Δε2 are the damage strains resulted from the growth of the sliding cracks and the sliding of initial microcracks. κ ¼[(3  ν)/(1 þ ν)] and κ ¼ 3–4ν are for plane stress and plane strain conditions, respectively. E, ν and N are Young's modulus, Poisson's ratio and the number of pre-existing microcracks, respectively. Based on the work by Ravichandran and Subhash [27], the area density f of the initial microcracks can be calculated as f ¼Ninc2, where Nin is the number of microcracks per unit area of the material. Taking this into consideration, the number N of the pre-existing microcracks in the tool material can be obtained using equation N ¼4abPt/c2, of which Pt is the porosity. The porosity of the tool material was obtained on the basis of Archimedes drainage method. The porosity of the tool material under consideration was found to be 0.017. Sαβ (α, β ¼1, 2) are

ε1 ¼ ε, sc can be expressed as:

E ε 1 + ENS11

(6)

The energy equilibrium equation for a body under compression was applied in the present work to acquire S11 in Eq. (6). For a body with one single sliding crack, the equation can be expressed as [27]:

W = Wf + 2Ue

(7)

where W is the work induced by the external uni-axial compressive stress, Wf is the frictional energy resulted from the sliding of the initial microcracks and Ue is the elastic strain energy caused by the growth of the tensile cracks. W can be calculated as:

W = 4ab(σ1Δε1 + σ2Δε2) = 4abS11σc2

(8)

where 2a and 2b are the size of the representative element. It has been obtained that 2a ¼2b¼47 mm. Wf can be defined as:

Wf = 2cτf δ

(9)

where τf is the shear traction on the crack surface caused by the normal traction resulted from the uni-axial compressive stress, δ is the sliding displacement along the pre-existing microcrack faces. The following equation can be used to determine τf:

τf = (5)

5

1 ⎡ 1 μ⎣ (σ1 + σ2)−(σ1−σ2)cos2θ ⎤⎦ = μ(σc−σccos2θ ) 2 2

(10)

where μ is the frictional coefficient. Friction tests were performed using pin-on-disk tribometer under dry conditions. The friction coefficient μ of the contacting tool material was determined as 0.51 based on the experimental data. δ can be obtained using the following equation [27,28]:

δ=

−1/2 (κ + 1)(1 + ν ) *⎡ π (l + l*) ⎤ cτ ⎢ wsin [2π (l + l**)] ⎥ ⎣ ⎦ E w

(11)

where l* is equal to 0.27c [22], l** is 0.083c [28] and w is half the distance between the cracks. τ* can be expressed by means of the following equation:

1 (σ1−σ2)sin2θ−μ[σ1 + σ2−(σ1−σ2)cos2θ ] 2 1 1 = σcsin2θ− μ[σc−σccos2θ ] 2 2

τ* =

(12)

Fig. 5. Sliding crack model with an array of sliding microcracks and individual microcrack model.

Please cite this article as: X. Cui, et al., Influences of tool rake angle and cutting speed on ceramic tool failure in continuous and intermittent turning of hardened steel, Ceramics International (2016), http://dx.doi.org/10.1016/j.ceramint.2016.05.015i

X. Cui et al. / Ceramics International ∎ (∎∎∎∎) ∎∎∎–∎∎∎

6

Ue can be obtained using Eq. (13):

∫0

Ue = 2

l

(κ + 1)(ν + 1) 2 KΙ dl 4E

(13)

where KI is the stress intensity factor of the crack array shown in Fig. 5(a). KI can be can be calculated as [23,29]: −1/2 1/2 ⎡ ⎛ π (l + l*) ⎤ πl ⎞ KΙ = F sinθ ⎢ wsin ⎟ = F sinθ −σ2⎜ 2w tan ⎥ ⎝ ⎣ ⎦ 2w ⎠ w −1/2 ⎡ π (l + l*) ⎤ ⎢ wsin ⎥ ⎣ ⎦ w

(14)

where

(15)

F = 2cτ*

On the basis of Eqs. (7)–(15), S11 in Eq. (6) can be obtained as:

S11 =

1 1 ( κ + 1)( ν + 1) ⎡ ⎣ csin2θ−cμ( 1−cos2θ )⎤⎦sinθ E 4ab π

{

2

}

ln

tan

(

)

π l + l*

tan

2w πl* 2w

1 ( κ + 1)( ν + 1) + cμ( 1−cos2θ )⎡⎣ csin2θ−cμ( 1−cos2θ )⎤⎦ 4ab 2E −1/2 ⎡ π l + l* ⎤ 1/2 ⎡ ⎤ ⎥ ⎢ ** ⋅⎣ 2π l + l ⎦ wsin ⎥⎦ ⎢⎣ w

(

)

(

)

(16)

The degradation of the Young's modulus [30] can be employed to evaluate the damage of the ceramic tool material:

σc = EDε = E(1−D)ε

(17)

where ED is the Young's modulus of the damaged tool material and D is the damage of the tool material. On the basis of Eq. (17), D can be expressed as:

D = 1−

1 1 + ENS11

Fig. 6. Comparisons of the average values of simulated and experimental resultant cutting forces.

(18)

Based on Eqs. (16) and (18), damage D of the tool material can be obtained. For the purpose of acquiring the initial damage Din of tool material, the tensile crack length l is set to be zero and Din can be acquired as 0.0066. 3.2. Evolution of damage equivalent stress in continuous and intermittent turning Based on the microstructure model shown in Fig. 3, characteristics of the tool material microstructure were reflected on the quadrilateral region Rz near the cutting tool edge as shown in Fig. 1. It has been discussed that the size of the representative element was 47 mm  47 mm. Taking the size of the representative element and the tool failure criterion into consideration, the length of lt denoted in region Rz was set to be 0.3 mm. It can be found that the size of region Rz was much larger than that of the representative element. The average values of simulated and experimental resultant cutting forces are compared in Fig. 6. It should be noted that the intermittent cutting process contains cutting and non-cutting periods. In the present work, cutting forces in the cutting period is focused on. The deviations Dev between the simulated cutting force Fr and the experimental cutting force Fre were calculated using the following equation:

Dev =

Fr − Fre ⋅100% Fre

(19)

It can be seen from Fig. 6 that cutting forces obtained in finite element simulation and cutting tests were close to each other. This validated the correctness of the simulation to some extent.

Fig. 7. The evolution of the maximum damage equivalent stress Sm on the cutting tool in the continuous cutting process (γo ¼  6°, v ¼100 m/min).

Damage equivalent stress s* defined in Eq. (2) can be used as a tool to analyze the rupture of brittle or quasi-brittle materials [16]. It can be deduced from Eq. (2) that the value of s* on the cutting tool is closely related to the magnitude of tool stress s and the damage D of the tool material. Therefore, tool stress and the damage of the tool material can be integrated using the concept of damage equivalent stress. In the present work, damage equivalent stress on the tool body was calculated using the tool stress obtained in the simulation and the initial damage of the tool material. Fig. 7 shows the evolution of the maximum damage equivalent stress Sm on the cutting tool in the continuous cutting process. It can be seen from Fig. 7 that Sm changed cyclically as the cutting

Please cite this article as: X. Cui, et al., Influences of tool rake angle and cutting speed on ceramic tool failure in continuous and intermittent turning of hardened steel, Ceramics International (2016), http://dx.doi.org/10.1016/j.ceramint.2016.05.015i

X. Cui et al. / Ceramics International ∎ (∎∎∎∎) ∎∎∎–∎∎∎

7

Fig. 8. The development of the maximum damage equivalent stress Sm on the cutting tool in a cutting period of the intermittent cutting process (γo ¼  6°, v¼ 100 m/min).

Fig. 9. The variation of shear angle when the cutting tool was about to leave the workpiece (γo ¼  6°, v¼100 m/min).

time increased. This can be attributed to the formation of serrated chip. The formation of serrated chip led to periodic change of the interaction between the cutting tool and the workpiece, resulting in the cyclical changes of the cutting forces and tool temperature. The tool stress was substantially influenced by the cutting forces and tool temperature. Fig. 8 shows the development of the maximum damage equivalent stress Sm on the cutting tool in a cutting period of the intermittent cutting process. Comparisons between Figs. 7 and 8 were conducted and it can be found that they share similar features. However, it should be noted that in intermittent turning, there existed peak values of damage equivalent stress which did not appear in continuous turning. When the cutting tool cut into the workpiece, the damage equivalent stress reached a relatively

high value abruptly. This can be attributed to the impact caused by sudden contact between the cutting tool and the workpiece. The damage equivalent stress suddenly increased as the cutting tool was about to leave the workpiece. The variation of shear angle was investigated in order to explain this phenomenon as shown in Fig. 9. It was found that the shear angle reversed from a positive value to a negative one. Because of the reverse of the shear angle, the value and distribution of the tool stress transformed substantially, leading to the sudden increase of damage equivalent stress. It can be deduced that the cutting tool was more prone to fracture when the ceramic tool cut into or was about to leave the workpiece. Fig. 10 shows the evolving trend of the position Pm where the maximum damage equivalent stress on the tool body arose in

Please cite this article as: X. Cui, et al., Influences of tool rake angle and cutting speed on ceramic tool failure in continuous and intermittent turning of hardened steel, Ceramics International (2016), http://dx.doi.org/10.1016/j.ceramint.2016.05.015i

8

X. Cui et al. / Ceramics International ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Fig. 10. The evolving trend of the position Pm in continuous and intermittent turning process (γo ¼  6°, v¼ 100 m/min).

continuous and intermittent turning process. The positive and negative values correspond to the tool rake and flank faces, respectively. The absolute value of Pm indicated the distance between the cutting tool edge and the position where the maximum damage equivalent stress appeared. It can be seen from Fig. 10 (a) that in continuous turning, the maximum value of damage equivalent stress appeared on the tool rake face and there was little variation in the position Pm. The position Pm was close to the cutting tool edge. It seems that tool fracture is more likely to begin on the tool rake face near the tool cutting edge. Fig. 10(b) shows that the position Pm changed substantially as the intermittent turning process proceeded. The position Pm can be on the tool rake face or tool flank face. Taking Figs. 8 and 10(b) into consideration, it can be deduced that when the ceramic tool cut into the workpiece and was about to leave the workpiece, there was greater possibility of tool fracture on the tool rake face. 3.3. Influences of tool rake angle and cutting speed on tool failure The highest value of damage equivalent stress in the whole cutting process was used as an indicator to reveal the influences of tool rake angle and cutting speed on tool failure. Fig. 11 shows the development of the value Sh and position pH of the highest damage equivalent stress with tool rake angle in continuous turning. It can be seen from Fig. 11(a) that as the tool rake angle increased from  10° to  2°, the highest damage equivalent stress decreased firstly and then increased. The lowest value of Sh can be obtained when the tool rake angle was  6°. It can be deduced that longer

Fig. 11. The development of the value Sh and position pH of the highest damage equivalent stress with tool rake angle in continuous turning (v ¼100 m/min).

tool life was expected to arise at the tool rake angle of  6°. Fig. 11 (b) indicates that the position pH stayed at the tool rake face despite the changes of tool rake angle. However, it can be found that the position pH was getting closer to the cutting edge when the tool rake angle increased. It seems that tool fracture will emerge at positions closer to the cutting edge when relatively large tool rake angle is adopted. Fig. 12 shows the evolution of the value Sh and position pH of the highest damage equivalent stress with tool rake angle in intermittent turning. It can be observed from Fig. 12(a) that the developing trend of Sh with tool rake angle in intermittent cutting was similar to that in continuous turning. Relatively low value of Sh can be acquired when the tool rake angle was  8°. Similarly, it can be inferred that relatively long tool life can be obtained at the tool rake angle of  8°. However, it can be found that the values of Sh in intermittent turning were higher than those in continuous turning. It can be deduced that when the cutting condition was the same, tool lives were relatively short in intermittent turning compared to those in continuous turning. Fig. 12(b) indicates that the position pH in intermittent turning changed in a similar way to that in continuous turning as the tool rake angle increased. The difference was that when the same cutting condition was adopted, the distance between position pH and cutting edge was relatively long when intermittent turning was applied. It can be deduced

Please cite this article as: X. Cui, et al., Influences of tool rake angle and cutting speed on ceramic tool failure in continuous and intermittent turning of hardened steel, Ceramics International (2016), http://dx.doi.org/10.1016/j.ceramint.2016.05.015i

X. Cui et al. / Ceramics International ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Fig. 12. The evolution of the value Sh and position pH of the highest damage equivalent stress with tool rake angle in intermittent turning (v ¼100 m/min).

from this phenomenon that tool fracture in intermittent turning was likely to start at a position further away from the cutting edge compared to continuous turning. Fig. 13 shows the development of the value Sh and position pH of the highest damage equivalent stress with cutting speed in continuous turning. It can be seen in Fig. 13(a) that Sh became larger as the cutting speed increased, meaning that tool life decreased with the increment of cutting speed. It was found from Fig. 13(b) that the increase of cutting speed led to lower value of pH. However, pH remained positive, which indicated that tool fracture arose on the tool rake face in spite of the change of cutting speed. It can be deduced from Fig. 13(b) that the position where tool fracture started became closer to the cutting edge as the cutting speed grew larger. Fig. 14 shows the evolution of the value Sh and position pH of the highest damage equivalent stress with cutting speed in intermittent turning. Comparisons between Figs. 13 and 14 indicate that the developing trends of Sh and pH with cutting speed were similar for continuous and intermittent turning. However, when the cutting condition stayed the same, larger values of Sh and pH appeared in intermittent turning, which correspond to shorter tool life and longer distance from the cutting edge to the position where tool fracture began, respectively. Figs. 15 and 16 show the values of Sh and experimental tool life

9

Fig. 13. The development of the value Sh and position pH of the highest damage equivalent stress with cutting speed in continuous turning (γo ¼  6°).

L obtained at different combinations of tool rake angle and cutting speed for continuous turning and intermittent turning, respectively. It can be found from Figs. 15 and 16 that relatively long tool life can be obtained when Sh was relatively low. This verified the correctness of the use of damage equivalent stress for analyzing tool failure mechanism. It can be found from Fig. 15 that when continuous turning was applied, the optimum combination of tool rake angle and cutting speed was γo ¼  6° and v¼ 60 m/min. The lowest value of Sh and the longest tool life L can be acquired simultaneously at this optimum combination. The optimum combination changed when intermittent turning was used. It can be deduced from Fig. 16 that the optimum combination became γo ¼ 8° and v ¼60 m/min in intermittent turning.

4. Conclusions On the basis of theoretical and experimental investigations, the influences of tool rake angle and cutting speed on ceramic tool failure in continuous and intermittent hard turning were revealed in the present work. The integrated effects induced by the initial state of the tool material microstructure and the external loads were considered in this study. The following conclusions can be

Please cite this article as: X. Cui, et al., Influences of tool rake angle and cutting speed on ceramic tool failure in continuous and intermittent turning of hardened steel, Ceramics International (2016), http://dx.doi.org/10.1016/j.ceramint.2016.05.015i

10

X. Cui et al. / Ceramics International ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Fig. 14. The evolution of the value Sh and position pH of the highest damage equivalent stress with cutting speed in intermittent turning (γo ¼  6°).

drawn from this work: 1. A two-dimensional model for the microstructure of the ceramic tool material was constructed. On the basis of the microstructure model, the size of the representative element was determined as 21  21 grains or 47 mm  47 mm by means of analyzing the statistical properties and average values of the grain diameters. Damage mechanics analysis of the representative element was conducted and the initial damage of the ceramic tool material was obtained as 0.0066. 2. The maximum damage equivalent stress on the tool body exhibited similar evolving trend in continuous and intermittent turning. The maximum damage equivalent stress changed cyclically as the cutting time increased, which was caused by periodic change of the interaction between the cutting tool and the workpiece. However, in intermittent turning, peak values of damage equivalent stress arose in the cutting process, which was not encountered in continuous turning. It was deduced that there existed greater possibility of tool fracture in intermittent turning when the ceramic tool cut into and was about to leave the workpiece. In continuous turning, the position where the maximum damage equivalent stress on the tool body arose varied little in the cutting process. However, the position changed substantially in intermittent turning. It was found that

Fig. 15. The values of Sh and experimental tool life L obtained at different combinations of tool rake angle and cutting speed in continuous turning.

tool fracture was likely to begin on the tool rake face close to the cutting edge in continuous turning. Taking the evolution of maximum damage equivalent stress into consideration, it can be concluded that in intermittent turning, there was greater possibility of tool fracture on the tool rake face as the tool cut into or was about to leave the workpiece. 3. The highest value of damage equivalent stress in the whole cutting process was utilized as an indicator in order to study the influences of tool rake angle and cutting speed on tool failure. It was found that the highest damage equivalent stress decreased first and then increased as the tool rake angle increased. Relatively long tool life can be acquired at the tool rake angle of 6° when continuous turning was adopted. In intermittent turning, relatively long tool life will arise at the rake angle of  8°. When the cutting speed increased, the value of the highest damage equivalent stress became larger for both continuous and intermittent turning, indicating that tool life decreased with the increment of cutting speed. The position where the highest damage equivalent stress appeared was on the tool rake face and became closer to the cutting edge when the tool rake angle

Please cite this article as: X. Cui, et al., Influences of tool rake angle and cutting speed on ceramic tool failure in continuous and intermittent turning of hardened steel, Ceramics International (2016), http://dx.doi.org/10.1016/j.ceramint.2016.05.015i

X. Cui et al. / Ceramics International ∎ (∎∎∎∎) ∎∎∎–∎∎∎

11

References

Fig. 16. The values of Sh and experimental tool life L obtained at different combinations of tool rake angle and cutting speed in intermittent turning.

or cutting speed increased. However, it should be noted that tool fracture in intermittent turning was more likely to arise at a position further away from the cutting edge compared to continuous turning. When the cutting condition was the same, longer tool life can be obtained in continuous turning. When continuous turning was used, the optimum combination of tool rake angle γo and cutting speed v was γo ¼  6° and v ¼60 m/ min. The optimum combination changed to be γo ¼ 8° and v ¼ 60 m/min in intermittent turning.

Acknowledgements This project is supported by National Natural Science Foundation of China (Grant No. 51505132) and China Postdoctoral Science Foundation (Grant No. 2015M580628).

[1] A.S. Kumar, A.R. Durai, T. Sornakumar, Machinability of hardened steel using alumina based ceramic cutting tools, Int. J. Refract. Met. Hard Mater. 21 (2003) 109–117. [2] A.S. Kumar, M.A. Khan, R. Thiraviam, T. Sornakumar, Machining parameters optimization for alumina based ceramic cutting tools using genetic algorithm, Mach. Sci. Technol. 10 (4) (2006) 471–489. [3] X. Cui, J. Guo, J. Zheng, Optimization of geometry parameters for ceramic cutting tools in intermittent turning of hardened steel, Mater. Des. 92 (2016) 424–437. [4] Z. Liu, X. Ai, H. Zhang, Z. Wang, Y. Wan, Wear patterns and mechanisms of cutting tools in high-speed face milling, J. Mater. Process. Technol. 129 (2002) 222–226. [5] Y. Karpat, T. Özel, 3-D FEA of hard turning: investigation of PCBN cutting tool micro-geometry effects, Trans. N. Am. Manuf. Res. Inst. SME 35 (2007) 9–16. [6] T. Özel, Y. Karpat, A. Srivastava, Hard turning with variable micro-geometry PcBN tools, CIRP Ann. Manuf. Technol. 57 (1) (2008) 73–76. [7] T. Özel, Y. Karpat, L. Figueira, J.P. Davim, Modelling of surface finish and tool flank wear in turning of AISI D2 steel with ceramic wiper inserts, J. Mater. Process. Technol. 189 (1) (2007) 192–198. [8] V.N. Gaitonde, S.R. Karnik, L. Figueira, J.P. Davim, Machinability investigations in hard turning of AISI D2 cold work tool steel with conventional and wiper ceramic inserts, Int. J. Refract. Met. Hard Mater. 27 (4) (2009) 754–763. [9] Z. Yin, C. Huang, J. Yuan, B. Zou, H. Liu, H. Zhu, Cutting performance and life prediction of an Al2O3/TiC micro–nano-composite ceramic tool when machining austenitic stainless steel, Ceram. Int. 41 (5) (2015) 7059–7065. [10] E. Aslan, N. Camuşcu, B. Birgören, Design optimization of cutting parameters when turning hardened AISI 4140 steel (63HRC) with Al2O3 þ TiCN mixed ceramic tool, Mater. Des. 28 (5) (2007) 1618–1622. [11] S. Chinchanikar, S.K. Choudhury, Effect of work material hardness and cutting parameters on performance of coated carbide tool when turning hardened steel: An optimization approach, Measurement 46 (4) (2013) 1572–1584. [12] J. Zhao, X. Ai, Fabrication and cutting performance of an Al2O3-(W, Ti)C functionally gradient ceramic tools, Int. J. Mach. Mach. Mater. 1 (3) (2006) 277–286. [13] X. Cui, J. Zhao, Y. Zhou, Z. Pei, Cutting forces and tool wear in intermittent turning processes with Al2O3-based ceramic tools, Key Eng. Mater. 499 (2012) 205–210. [14] J. Zhao, X. Yuan, Y. Zhou, Cutting performance and failure mechanisms of an Al2O3/WC/TiC micro- nano-composite ceramic tool, Int. J. Refract. Met. Hard Mater. 28 (3) (2010) 330–337. [15] D. Krajcinovic, Damage mechanics: accomplishments, trends and needs, Int. J. Solids Struct. 37 (2000) 267–277. [16] J. Lemaitre, R. Desmorat, Engineering Damage Mechanics: Ductile, Creep, Fatigue and Brittle Failures, Springer, Berlin, 2005. [17] G. Su, Evolution and mechanisms of saw-tooth chip formation in high-speed machining (pH.D. thesis), Shandong University, Jinan, 2011. [18] G. Zheng, J. Zhao, Z. Gao, Q. Cao, Cutting performance and wear mechanisms of Sialon-Si3N4 graded nano-composite ceramic cutting tools, Int. J. Adv. Manuf. Technol. 58 (1–4) (2012) 19–28. [19] Y. Wu, W. Zhou, B. Wang, F. Yang, Modeling and characterization of two-phase composites by Voronoi diagram in the Laguerre geometry based on random close packing of spheres, Comput. Mater. Sci. 47 (2010) 951–961. [20] Z. Fan, Y. Wu, X. Zhao, Y. Lu, Simulation of polycrystalline structure with Voronoi diagram in Laguerre geometry based on random closed packing of spheres, Comput. Mater. Sci. 29 (2004) 301–308. [21] D. Wang, J. Zhao, J. Zhao, A. Li, X. Chen, Microstructure-level modeling and simulation of the flexural behavior of ceramic tool materials, Comput. Mater. Sci. 83 (2014) 434–442. [22] H. Horii, S. Nemat-Nasser, Brittle failure in compression: splitting, faulting, and brittle-ductile transition, Philos. Trans. R. Soc. Lond. 319 (1986) 337–374. [23] H. Li, J. Zhao, T. Li, Micromechanical modelling of the mechanical properties of a granite under dynamic uniaxial compressive loads, Int. J. Rock. Mech. Min. Sci. 37 (6) (2000) 923–935. [24] J. Kemeny, A model for non-linear rock deformation under compression due to sub-critical growth, Int. J. Rock. Mech. Min. Sci. Geomech. Abstr. 28 (6) (1991) 459–467. [25] M.F. Ashby, S.D. Hallam, The failure of brittle solids containing small cracks under compressive stress states, Acta Mater. 34 (3) (1986) 497–510. [26] J.T. Fredrich, B. Evans, T.F. Wong, Effect of grain size on brittle and semibrittle strength: implications for micromechanical modelling of failure in compression, J. Geophys. Res. 95 (1990) 10907–10920. [27] G. Ravichandran, G. Subhash, A micromechanical model for high strain rate behavior of ceramics, Int. J. Solids Struct. 32 (1995) 2627–2646. [28] S. Nemat-Nasser, M. Obata, A microcrack model of dilatancy in brittle materials, J. Appl. Mech. 55 (1988) 24–35. [29] H. Tada, P.C. Paris, G.R. Irwin, Stress Analysis of Cracks Handbook, Del Research Corporation, St Louis, 1973. [30] J. Lemaitre, H. Lippmann, A Course on Damage Mechanics, Springer, Berlin, 1992.

Please cite this article as: X. Cui, et al., Influences of tool rake angle and cutting speed on ceramic tool failure in continuous and intermittent turning of hardened steel, Ceramics International (2016), http://dx.doi.org/10.1016/j.ceramint.2016.05.015i