Finite Element Modeling and Simulation of Orthogonal Cutting With Multi Layer Coated Tools

This paper focuses on the development of Finite Element Method (FEM) in modeling and simulation of coated cutting tools with multi-layer coats. A special Finite Element code called (MSC.MARC mentat) is used in the numerical tests, the results are then compared with experimental work. The paper studied the effect of number of coats of cutting tools on the following field parameters; tool-chip contact length, chip contraction coefficient and shear angle at similar machining conditions.The metal being machined is (AISI 1045 steel) with orthogonal machining conditions. The three cutting tools and models are coated with (TiN, TiN/TiC, TiN/Al 2 O 3 /TiC), while the fourth one is uncoated.The results show good agreement between the experimental and numerical tests. Some of the results are compared with other published papers. The comparison of the predicted results shows good agreement with experimental tests with maximum relative difference of (18%) for the chip contraction coefficient and contact length, and (10 %) for the shear angle. The insert with double coats shows excellent result, compared to others from point of view of chip contraction coefficient, contact length and shear angle.


1-Introduction
The development of new cutting tool materials during the last few years is based mainly on sub micron hard metal substrates as well as new coatings techniques.Coatings improve wear resistance, increase tool life, broaden the application range of a given grade and enable use at higher speeds.By improving performance, coatings are helping cutting tool manufacturers respond to change workpiece materials and process requirements [1].The cutting tool is the most critical part of the machining system today, approximately (85%)of carbide tools are coated, almost exclusively by the chemical vapor deposition (CVD) method [2].Numerical methods and Finite Element modeling in particular have become increasingly popular due to the advancement in computers and the development of complex codes.Some of the models model orthogonal cutting using the Eulerian formulation.However, a majority has relied on the Lagrangian formulation, which allows the chip to be modeled from incipient to steady state [3].The majority of inserts presently used in various metal cutting operations are cemented carbide tools coated with a material consisting of nitrides(TiN, CrN, etc.), carbides (TiC, CrC, W 2 C, WC/C, etc.), oxides (e.g.Alumina Al 2 O 3 ) or combinations of these [4,5].Reif (1995) [6]studied the effect of the addition of boron (B) on surface roughness and hardness in cutting tools coated with Ti (B, N) HSSdrills and cemented carbide inserts, he compared the results with commercially available reference tools coated with TiN and (Ti,Al)N.The results of his research concluded that the hardness of the produced nitride coating increases slightly, and also a decrease in the roughness value (R a ) of the value of pure TiN with increasing boron content.W. Grzesik (2000) experiments, less average surface roughness was obtained by using a 3layer coated tool coated outermost with TiN.The reduction of cutting speed by about (33%), will improve the surface roughness by about (26%), while when increasing the cutting speed by (310%) an improvement by (69%) in surface finish is achieved.

2-Numerical Work
In order to establish the Finite Element model, the important stages in developing the FE model are summarised as follows: • Determination of FE appropriate model geometry.
• Generation of FE mesh.
• Application of boundary conditions and loading parameters.
• Determination of material properties.

2-1 Model Geometry
The Finite Element model is composed of a deformable workpiece and a rigid tool.The tool penetrates through the workpiece at a constant cutting speed and feed rate.The initial arrangement of both the workpiece and the tool in the simulation model are done using the Cartesian coordinate, 2-D model as shown in Fig. (1).The length of the workpiece is assumed to be (100 mm); four models are suggested, the cutting tool is modeled as multiple coating (Fig. 2-a, b&c), in order to study in detail the behavior of these layers under different conditions of layers number, the cutting conditions are assumed to be constant values as shown in (Table 1).

2-2 Mesh Generation
The Finite Element models used for the plane-strain orthogonal metal cutting simulation are based on the Lagrangian techniques and explicit dynamic, mechanical modeling software with adaptive remeshing.This means that the initial mesh becomes distorted after a certain length of cut.The workpiece discretized by bilinear four-node quadratic.The initial geometry and mesh is presented in Figure (1)  The number of nodes differ from one model to another due to the change in the coated layer number, the upper part of mesh, which constitutes the removed workpiece material, is finer, to enable the stresses, strains in the chip and tip region to be accurately distingueshed.

2-3Loading and Boundary Conditions
For the boundary conditions of our models, the nodes of the workpiece are fixed in both X and Y directions by gluing them to a rigid curve, cutting tool is moving horizontally towards -X diretion, while it is constrained in the vertical displacement.Friction is another boundary condition that could have significant consequence to the solution of the model.Due to the difficulty in measuring the friction experimentally, the values of coefficient of friction between chip and tool is assumed to be constant (µ p =0.5) for uncoated tool, and assumed to be (µ p =0.3) for coated tool, depending on the values found in the literatures[12, 13]

2-4 Material Properties
To shows the mechanical properties of (WC), table (5) [15] shows the mechanical properties of coating layers used in the simulation tests.

3-Numerical Tests and governoring relations
The following field variables were chosen to be predicted for the simulated models at constant values of feed rate, depth of cut,cutting speed and multiple coated tools: where: t c = deformed chip thickness (mm).t =undeformed chip thickness (mm).γ = rake angle.Φ = shear angle.

• Chip Thickness (t c )[16]
In practical tests, the mean chip thickness can be obtained by measuring the length and weight of a piece of chip, then t c = ω / ρ w l …… 2 where: ω =weight of a piece of chip (gm).ρ=density of work material (kg/m 3 ).w = width of the chip (mm).l = length of a piece of chip (mm).
interface region is measured directly using outside micrometer, its accuracy is 0.05mm.The numerical tests were conducted with a Pentium 4 personal computer for the following specifications: processor 2.4 GHz.D, Ram 512 MB.The time for each Numerical test varied from (8092 to 51152 sec.)depending on number of coated layers.

4-Experimental Tests
To imitate orthogonal metal cutting process, axial turning of the end of a tube is selected, as shown in  6).The Chip thickness on each cutting was measured by using external micrometer having measuring range of (0 -25 mm); scale value (0.01mm).The contact length on each cutting tool was measured by using ESE way-hardness tester, type SPVRB.2.M.

4-2 Cutting Tools Inserts
Four types of commercially available tungsten based cemented carbide inserts were tested.The cutting inserts tested were uncoated -insert 1, TiN coated -insert 2, TiN/TiC coatedinsert 3 and TiN/Al2O3/TiC coatedinsert 4, respectively.All the inserts are suitable for machining different kinds of steels at high speeds and high feed rates.All the inserts have identical geometry designated by the American National Standard Institute (ANSI) as [CNMM 120404].The inserts were rigidly mounted on a right hand style tool holder with a cutting rake and a back rake of (-5º).The tool holder is designated by ANSI as [PCLNR 2020 K 12].[17,18].The tool geometry which is used for the orthogonal and experimental tests is shown in table (7).

4-3 Workpiece Material
To achieve orthogonal cutting the experiments were done using a hollow cylinder from the end on a lathe.In this experiment, the diameter of the workpiece should be relatively larger than the depth of the cut (wall thickness of the tube) to satisfy orthogonal cutting condition.Workpiece  ) with the outer diameters of (40 mm), depth of cut was (0.5 mm).Cutting tools are made of tungsten based cemented carbide and had rake angles of ( -5°) and (5°) clearance angle.

5-1 Effect of Coated Layers on Chip Contraction Coefficient
The relationship between the coated layers and the corresponding chip contraction coefficient (λ) is represented in Figure ( 6) that shows the Finite Element Analysis, and experimental work.It can be seen that the TiN coated tool insert shows the maximum value of chip contraction coefficient with respect to other tools, while the TiN/TiC coated tool insert has the lowest chip contraction coefficient.The value of chip contraction coefficient increases from minimum value of (2.38) for TiN/TiC coated tool insert, to reach maximum value of (2.9) for TiN coated tool insert.The maximum difference between chip contraction coefficients from numerical and experimental work does not exceed (18 %).It can be concluded that the two layer coatings insert produces minimum deformation in the chip thickness resulting from many parameters relating to the TiN/TiC layer wear specifications.

Contact Length (L c )
It can be seen from most values in Figure (7) that, the tool insert with two layers coating has the minimum value of contact length reaching to (0.27 mm), which is the optimum one.While the (TiN/Al 2 O 3 /TiC) coated tool insert has the maximum contact length value that reaches (0.357 mm), which may be due to the existence of medium layer (Al 2 O 3 ) that has excellent thermal-insulation properties, so that it protects the heat transfer from the chip to the tool during chip formation process.For other cases of inserts, the uncoated insert and three layer coated insert show little difference in value of contact length reaching to (0.7 %).Also from the results we see that the maximum difference between contact length value from numerical and experimental work occurs at the TiN/TiC coated tool insert, that reaches (2 %).In general the results being obtained in numerical tests show greater values than experimental one within arange of (2-18 %).This phenomenon agrees well with paper being published at the literature [19].

5-3 Effect of Coated Layers on Shear Angle (Ф)
The relationship between the coated layers and the corresponding shear angle (Ф) is represented in Figure ( 8), as predicted by the Finite Element Analysis and experimental work.It can be seen that the maximum shear angle is found in the TiN/TiC coated tool insert reaching (28º) and the minimum shear angle is found in the TiN coated tool insert reaching (13º) with respect to the other tools.The maximum difference between shear angle value from numerical and experimental work reaches (6 %).

6-conclusions
The main conclusions that can be deduced from the present paper can be summarized as follows: 1.The FE software MSC Marc is successfully accomplished in modeling and simulation of our models relating to tool coating layers.

The chip contraction coefficient
reaches maximum value of (2.9) at tools coated with three layers, and minimum value of ( Fig (3) shows a representative successful simulation test for chip formation process through the Marc mentat code.

Fig ( 3 )Finite
Fig (3):A representative successful simulation test for chip formation process through the MARC code.

Figure
Figure (7): Comparison between the FEM and exp.results for contact length contact length.
The Al 2 O 3 coated tool showed superior wear-resistance over the TiN/Al 2 O 3 coated tool due to the TiN coating that deteriorated the effect of the Al 2 O 3 outer layer.

Table (
1):Constant cutting conditions for both numerical and experimental tests.Material type AISI 1045 St.

Table ( 6
): Specifications of centre lathe machine used in experimental tests.