- Email: [email protected]
Prediction of Residual Stresses after Laser-assisted Machining of Inconel 718 Using SPH
Available online at www.sciencedirect.com
ScienceDirect Procedia CIRP 31 (2015) 19 – 23
15th CIRP Conference on Modelling of Machining Operations
Prediction of Residual Stresses after Laser-assisted Machining of Inconel 718 Using SPH M. A. Balbaaa, Mohamed N.A. Nasra,b,* a Dept. of Mechanical Engineering, Faculty of Engineering, Alexandria University, Alshatby, Alexandria 21544, Egypt Dept. of Materials Science & Engineering, Egypt-Japan University of Science & Technology, New Borg El-Arab, Alexandria, 21934, Egypt
b
* Corresponding author. Tel.: +20-3-591-5848; fax: +20-3-590-2715. E-mail address: [email protected].
Abstract Laser-assisted machining (LAM) has been proven to improve the machinability of hard-to-cut materials, by lowering cutting forces, increasing tool life and improving surface finish. Yet, not enough work has been done so far on how LAM would affect residual stresses (RS). The current study investigates the effects of LAM on RS, using finite element modelling, after dry orthogonal cutting of Inconel 718. First, the cutting process was modelled using the smoothed-particles hydrodynamics (SPH) method; secondly, an implicit Lagrangian model was used to simulate the relaxation process. Predicted LAM results were compared to conventional machining. The model was validated using previously published experimental results.
© 2015 2015 The The Authors. Authors.Published Publishedby byElsevier ElsevierB.V. B.V. This is an open access article under the CC BY-NC-ND license © (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of The International Scientific Committee of the “15th Conference on Modelling of Machining Operations”. Peer-review under responsibility of the International Scientific Committee of the “15th Conference on Modelling of Machining Operations Keywords: Residual stress; Finite element method (FEM); Laser-assisted machining (LAM); Turning; Smoothed-particles hydrodynamics (SPH).
1. Introduction Demanding operating conditions in nuclear and aerospace sectors required drastic development of high strength alloys; thus appeared the importance of Ti-based and Ni-based alloys. Inconel 718 represents a vastly used Ni-based alloy that is characterized by high strength, high wear and corrosion resistance, but met the challenge of being hard to machine. Hard-to-machine materials typically have low thermal conductivity and diffusivity, which cause a steep thermal gradient and high tendency to strain hardening during machining [1]. Also, the presence of abrasive carbides and intermetallic phases as well as adhesion to tool material adds to the complexity of machining these alloys. All these factors cause excessive tool wear, built up edge and premature tool cracking, high cutting forces, which all result in lower metal removal rates and high costs [1-3,6]. Several approaches have been adopted to solve the issues with hard-to-machine materials; mainly, cryogenic machining and laser-assisted machining (LAM). In cryogenic machining, liquid nitrogen is typically used to lower cutting temperatures
and improve productivity. On the other hand, LAM depends on pre-heating the workpiece with a laser beam ahead of the tool. This increase in temperature causes the material to soften; thus lowering the cutting forces and tool wear. 1.1. Literature review Surface integrity, especially machining-induced residual stresses (RS), is very crucial to controlling part performance. However, very little attention has been paid to the effect of LAM on RS in the available literature; even less work is available on modeling the LAM process itself. By reviewing the available literature, one can find that studies are focused on tool wear, surface roughness, surface hardness, cutting forces and chip formation [1-7]. A study by Navas et al.[2] focused on age-hardened and solution-annealed Inconel 718, where the machinability of both alloys was improved, especially the former. Cutting forces were found to drop with higher laser power densities and cutting speeds. Machinability also improved when the distance between the laser spot and cutting tool was decreased; however, a minimum distance was identified to avoid tool degradation by overheating. Attia et al.
2212-8271 © 2015 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the International Scientific Committee of the “15th Conference on Modelling of Machining Operations doi:10.1016/j.procir.2015.03.034
20
M.A. Balbaa and Mohamed N.A. Nasr / Procedia CIRP 31 (2015) 19 – 23
[1] performed an experimental study on the effect of LAM on finish turning of Inconel 718, and reported a significant drop in cutting forces compared to conventional machining (CM). Material removal rate was found to increase by almost 800% and surface finish improved by 25%. Tool wear was the focus of the work by Anderson et al. [3], and they reported a decrease in flank wear under LAM conditions compared to CM. Also, LAM specific cutting energy was significantly lower, and surface roughness was improved. separately. Below is an example which the authors may find useful. Ding et al. [4] performed an experimental investigation on LAM when turning AISI 4130 steel. LAM resulted in lower cutting forces, higher surface tensile RS, but no effect was found on surface hardness; however, no explanation was provided. Dumitrescu et al. [5] also reported a reduction in cutting forces after LAM of hardened AISI D2 tool steel. Finite element modeling (FEM) of LAM effects on machining parameters is very minimal in the literature. An example is the study by Shi et al. [6], where experimental and FEM work was done on the use of LAM under different cutting conditions, and recorded lower cutting forces versus those in CM. Also, Bouchnak et al. [7] modeled LAM of 42CrMo4, where cutting forces also dropped at different cutting speeds. Nasr et al. [8] performed FE simulations to study the effect of LAM on surface RS in AISI 4340 at different feed rates. Higher surface tensile RS were induced after LAM. Nasr et al. [9] studied the effect of laser power on surface RS through FE modeling. It was concluded that, surface tensile RS decrease and the tendency to induce compressive RS increases with the increase in laser power. Based on the above, and to the authors’ best of knowledge, very minimal focus was given to studying the effects of LAM on RS. Accordingly, the current work investigates the effects of LAM on surface RS when orthogonal dry cutting Inconel 718. 1.2. SPH modelling FEM of metal cutting faces some challenges when using the Lagrangian approach, due to high element distortion that could lead to termination of the process. Also, the Lagrangian approach requires a parting line (sacrificial layer) along which elements are deleted as they fail; thus leading to material volume loss. Accordingly, it is mainly limited to simulating sharp-edged tools, which is not always the case. The solution to these problems first came in the form of the Lagrangian adaptive re-meshing and arbitrary-Lagrangian-Eulerian (ALE) techniques [10]. SPH is a meshless method developed by Monaghan et al. [11], which has the advantage of being distortion free. It is a relatively new technique when it comes to modeling metal cutting; accordingly, it is not yet fully developed when compared to the FE method. The use of SPH in modelling the cutting process has drawn some attention in the past few years, with the main target being focused on exploring its
capabilities and making use of its advantages. Limido et al. [12] demonstrated the capabilities of SPH in modelling high speed machining and compared the results to experimental and FE modeling data. They found that SPH closely predicted continuous and serrated chips, and cutting forces were found to be within a small range of the experimental results. The authors also introduced the concept of time scaling, where the tool velocity was imposed ten times higher the real value, in order to reduce the computational time. Madaj et al. [13] confirmed the same results, as they constructed an SPH model for orthogonal cutting of Al2024-T351 alloy. The friction coefficient was varied to observe its effects on chip formation and cutting forces. They concluded that the predicted forces correlate well with the experimental results. Continuous chip formation was successfully modelled; however, the formation of serrated chips required the use of Johnson-Cook failure damage parameters, and the minimum strain for failure was required. The work of Villumsen et al. [14] on SPH modeling of 3D orthogonal cutting of Al6082-T6 alloy represents a significant contribution to the field, as the authors defined the material properties, equation of state (EOS), contact settings and boundary conditions (BCs) required for SPH. They also performed a sensitivity analysis on the effect of particle pitch, mass scaling, time scaling and friction coefficient on cutting forces. It can be shown from the previous review that there is not enough work done on SPH modeling of metal cutting. That fact coupled with the absence of understanding the effects of LAM on surface RS makes it an interesting point for investigation. Accordingly, a set of SPH models was built in order to compare RS after LAM and CM, when orthogonal dry cutting Inconel 718, as mentioned earlier. Cutting conditions were chosen similar to those of Shi et al. [6] for model validation. 2. Model Description In order to predict RS; first, the cutting process was modelled using the smoothed-particles hydrodynamics (SPH) method; secondly, an implicit Lagrangian model was used to simulate stress relaxation and cooling down. Even though the current work focuses on orthogonal cutting, a 3D model was built as SPH is not available for 2D analyses in the commercial software ABAQUS/Explicit, which was used for the current study. Two models were built to simulate CM and LAM effects on surface RS when orthogonal cry cutting Inconel 718. It is worth mention that, SPH was only used for the workpiece, while the tool was modeled using a Lagrangian mesh. To the authors’ best of knowledge, the current work is the first to present an SPH cutting model using the commercial software ABAQUS. After the cutting process was modelled, a standard implicit Lagrangian model was used to simulate stress relaxation and cooling down to room temperature for RS prediction.
M.A. Balbaa and Mohamed N.A. Nasr / Procedia CIRP 31 (2015) 19 – 23
2.1. Workpiece properties
2.2. Contacts definition
The cutting model can be seen in Fig.1 with the applied boundary conditions. The dimensions of the workpiece are 1.5 mm x 1 mm x 0.02 mm. Coupled temperature-displacement explicit analysis was used, and the workpiece was modeled with a total of 35511 PC3D particles. The workpiece is fixed at the bottom surface, and a velocity BC was imposed to the back side of the tool.
The contact between the workpiece and tool was defined using node-to-surface contact algorithm, with a Coulomb friction model assumed along with a friction coefficient of 0.2. Despite being a simple model, the Coulomb friction model is not limited to SPH; however, it has been used widely with Lagrangian and ALE models, where acceptable results were achieved [10,18, 23]. 2.3. Heat transfer
Cutting tool Workpiece
In the current work, it was assumed that 90% of plastic deformation energy is converted to heat. Also, the materials’ properties allowed 33.5 % of the generated friction heat to be conveyed to the workpiece. During cutting, heat convection and radiation were neglected. On the other hand, convection to the environment (air) was modelled during the stress relaxation (RS) step, where a convection coefficient of 10 W/ (m2oC) and a sink temperature of 20 oC were used.
Cutting direction
Fig. 1 SPH 3D cutting model (front view)
Inconel 718 was defined for the workpiece material, and temperature dependent physical properties were defined based on those supplied in [15] The well-known Johnson-Cook (J-C) plasticity model (Eq. 1) [16] was used to model the workpiece material behavior, with its parameters listed in Table 1 [17]. The J-C model has been widely used for metal cutting simulations, including machining of Inconel 718 [10,17-20]. ߪ = ( ܣ+ ߝ ܤ
)
ఌሶ
்ି்ೝ
ఌሶ బ
் ି்ೝ
ቀ1 + ܥln ቀ ቁቁ ቀ1 െ ቀ
ቁ ቁ
(1)
Table 1. J-C constitutive model parameters [16]. Parameter
Value
Initial yield strength A (MPa)
1241
Strain hardening coefficient B (MPa)
622
Strain hardening exponent n
0.6522
Strain rate coefficient C
0.0134
Thermal softening exponent m
1.3
Reference plastic strain rate ߝሶ (s-1)
1
The Mie-Gruneisen EOS was used to describe the pressure-volume relationship of the workpiece material, which is essential when using SPH. However, it is important to note that, the parameters of Monel were used as those of Inconel 718 are not currently available [21]. This was also done in a recent study by Oak Ridge National Laboratory (for the US Department of Energy), where the Monel EOS was used for Inconel 718 during impact of space reactor cores, due to the non-availability of the exact Gruneisen parameters [22].
2.4. Cutting conditions Cutting conditions were adopted from the work of Shi et al. [6]; whereas cutting speed of 100 m/min and uncut chip thickness (t1) of 0.125 mm (feed rate of 0.125 mm/rev) were used for CM and LAM conditions. A K313 carbide insert with a sharp edge, as assumed by Shi et al. [6], and a zero rake angle was used, whose properties can be found in [24]. 2.5. LAM preheating phase The laser beam heating effect was modeled in the form of a preheating phase, before modelling the cutting process. The workpiece temperature distribution was sought to match that presented by Shi et al. [6]. Accordingly, a transient thermal analysis was run for a period equal to the lead time of the laser beam ahead of the tool. The analysis started by the same surface temperature measured experimentally by Shi et al. [6], which was 800 oC, as a thermal boundary condition. Such temperature was reported to be generated by a 3000 W laser. 3. Results and Discussion The results below show the effect of LAM on cutting forces and maximum chip thickness. These results are validated by comparing them to the results of Shi et al. [6], from which the current cutting conditions are derived. Also, the in-depth temperature gradient and plastic strain are plotted to demonstrate their effect on surface RS, under LAM and CM conditions. Fig. 2 shows the chip formation process during cutting, where t1 represents the uncut chip thickness and t2 represents the deformed chip thickness.
21
22
M.A. Balbaa and Mohamed N.A. Nasr / Procedia CIRP 31 (2015) 19 – 23
Since RS are affected by plastic strain, temperature gradient and phase transformation, which was not present in the current study (as reported in [6]); therefore, in order to explain the reasons behind the effects of LAM on surface RS, the plastic strain in cutting direction and temperature gradients are shown in Fig. 4 and Fig. 5, respectively.
t2
Shear angle
0.10
Fig. 2 Chip generation in SPH
3.1. Cutting forces Cutting forces were predicted for CM and LAM, and the results are compared to those of Shi et al. [6] as shown in Table 2.
Plastic Strain PE11
t1
CM
0.08
LAM
0.06 0.04 0.02 0.00 -0.02
10
30
50
70
Depth from machined surface (microns)
Table 2. Cutting forces. Exp.
SPH
CM cutting force (N)
170
156
LAM cutting force (N)
130
126
Difference “LAM - CM” (%)
-23.2
-19.2
It can be seen that the SPH model closely predicted the cutting forces. Also, the effect of LAM on cutting forces is depicted clearly and nearly with the same trend for experimental and SPH results. LAM caused the cutting force component to drop. This can be traced back to the thermal effect of LAM, as the material softens with the rising temperature thus lowering the resistance head of the tool. 3.2. Surface residual stresses in cutting direction Surface RS in cutting direction are presented in Fig. 3, where LAM resulted in lowering the tensile surface RS; actually, RS turned to be compressive instead of being tensile. Unfortunately, no experimental RS are currently available to confirm such results. However, the effects of LAM on other process parameters (forces and temperature) support the validity of its effects on RS.
Residual stresses (MPa)
600 400 200
By looking at the plastic strain in cutting direction (PE11), we can observe that the SPH model predicted a higher tensile PE11 in case of LAM compared to CM. Higher tensile PE11 under LAM conditions can be traced back to the thermal softening effect of the material ahead of the tool. Therefore, the compressive stresses generated ahead of the tool tip decreases, compared to CM. This tends to results in an overall higher tensile stresses behind the tool tip. Accordingly, higher tensile PE11 is generated under LAM conditions, compared to CM. Since surface tensile plastic strain results in more compressive or less tensile RS [25], this explains why LAM resulted in less tensile (actually, compressive) surface RS as presented in Fig.3. As shown in Fig. 5, it was found that the difference in workpiece temperature underneath the tool tip, during cutting, between CM and LAM is limited to the nearsurface layer (about 11 microns). Conv
LAM
500 400 300 200 100 0
0 -200
Fig. 4 In-depth plastic strain in cutting direction (PE11).
Temperature (oC)
Parameter
CM
LAM
0
50 100 Depth from machined surface (microns)
-400 -600
Fig. 3 Predicted surface RS in cutting direction.
Fig. 5 Temperature gradient for machined workpiece.
M.A. Balbaa and Mohamed N.A. Nasr / Procedia CIRP 31 (2015) 19 – 23
4. Conclusions The current study focused on SPH modeling of LAM process and its effect on surface residual stresses in cutting direction. Also, the effects of LAM on cutting forces, temperatures and plastic strain in cutting direction were also examined. Based on the presented results, the following conclusions were drawn. 1.
The model closely predicted the cutting force component, and captured the trend reported experimentally where LAM resulted in lower forces.
2.
For the studied conditions, LAM resulted in surface compressive residual stresses in cutting direction; however, conventional machining resulted in surface tensile residual stresses.
3.
The effect of LAM on residual stresses are attributed to the thermal softening effects of the laser beam ahead of the tool tip, which resulted in overall more tensile plastic strain in cutting direction. Accordingly, less tensile (more compressive) residual stresses.
References [1] Attia H, Tavakoli S, Vargas R, Thomson V. Laser-assisted high-speed finish turning of superalloy Inconel 718 under dry conditions. CIRP Annals-Manufacturing Technology. 2010; 59(1): 83-8. [2] Garcí V, Arriola I, Gonzalo O, Leunda J. Mechanisms involved in the improvement of Inconel 718 machinability by laser assisted machining (LAM). International journal of machine tools and manufacture. 2013;74: 19-28. [3] Anderson M, Patwa R, Shin YC. Laser-assisted machining of Inconel 718 with an economic analysis. International Journal of Machine Tools and Manufacture. 2006; 46(14): 1879-91. [4] Ding H, Shin YC. Laser-assisted machining of hardened steel parts with surface integrity analysis. International Journal of Machine tools and manufacture. 2010; 50(1): 106-14. [5] Dumitrescu P, Koshy P, Stenekes J, Elbestawi M. High-power diode laser assisted hard turning of AISI D2 tool steel. International Journal of Machine Tools and Manufacture. 2006; 46(15): 2009-16. [6] Shi B, Attia H, Vargas R, Tavakoli S. Numerical and experimental investigation of laser-assisted machining of Inconel 718. Machining Science and Technology. 2008; 12(4): 498-513. [7] Braham-Bouchnak T, Germain G, Morel A, Lebrun J-L. The influence of laser assistance on the machinability of the titanium alloy Ti555-3. The International Journal of Advanced Manufacturing Technology. 2013; 68(9-12): 2471-81. [8] Nasr MN, Balbaa M, Elgamal H. Modelling Machining-induced Residual Stresses after Laser-assisted Turning of Steels. Advanced Materials Research; 2014: Trans Tech Publ. pp. 622-627. 2014 [9] Nasr MN, Balbaa M, Effect of Laser Power on Residual Stresses when Laser-assisted Turning of AISI 4340 Steel, Canadian Society for Mechanical Engineering (CSME) Int. Congress, 2014,Toronto, Canada, June 1-4. [10] Nasr MN, Ng E-G, Elbestawi M. Modelling the effects of tool-edge radius on residual stresses when orthogonal cutting AISI 316L. International Journal of Machine Tools and Manufacture. 2007; 47(2): 401-11. [11] Gingold RA, Monaghan JJ. Smoothed particle hydrodynamics: theory and application to non-spherical stars. Monthly notices of the royal astronomical society. 1977; 181(3): 375-89.
[12] Limido J, Espinosa C, Salaün M, Lacome J-L. SPH method applied to high speed cutting modelling. International journal of mechanical sciences. 2007; 49(7): 898-908. [13] Madaj M, Píška M. On the SPH orthogonal cutting simulation of A2024T351 alloy. Procedia CIRP. 2013; 8: 152-7. [14] Villumsen, M. F., Fauerholdt, T. G., Simulation of Metal Cutting using Smoothed Particle Hydrodynamics, 7. LS-DYNA Anwenderforum, Bamberg 2008, Germany.C-III-17–C-III-36. [15] Special Metals: Inconel 718 properties datasheet. [16] Johnson GR, Cook WH. A constitutive model and data for metals subjected to large strains, high strain rates and high temperatures. Proceedings of the 7th International Symposium on Ballistics; 1983: The Netherlands. [17] Kortabarri A, Madariag A, Fernandez E, Esnaol J, Arrazola P. A comparative study of residual stress profiles on Inconel 718 induced by dry face turning. Procedia Engineering. 2011; 19: 228-34. [18] Joyot P, Rakotomalala R, Pantale O, Touratier M, Hakem N. A numerical simulation of steady state metal cutting. Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science. 1998; 212(5): 331-41. [19] Outeiro J, Pina J, M'saoubi R, Pusavec F, Jawahir I. Analysis of residual stresses induced by dry turning of difficult-to-machine materials. CIRP Annals-Manufacturing Technology. 2008; 57(1): 77-80. [20] Chandrasekaran H, M'saoubi R, Chazal H. Modelling of material flow stress in chip formation process from orthogonal milling and split Hopkinson bar tests. Machine Science and Technology. 2005; 9(1): 13145. [21] Steinberg D. Equation of state and strength properties of selected materials: Lawrence Livermore National Laboratory Livermore, CA; 1996. [22] Kim, Seokho H. and Flanagan, George F. Impact analysis for candidate space reactor core concept designs for potential critical study. Oak Ridge, TN: Oak Ridge National Laboratory. [23] Shet C, Deng X. Finite element analysis of the orthogonal metal cutting process. Journal of Materials Processing Technology. 2000; 105(1): 95109. [24] Nasr M, Ng E, Elbestawi M. Effects of workpiece thermal properties on machining-induced residual stresses-thermal softening and conductivity. Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture. 2007; 221(9): 1387-400. [25] Nasr MN, Ng E-G, Elbestawi M. Effects of strain hardening and initial yield strength on machining-induced residual stresses. Journal of Engineering Materials and Technology. 2007; 129(4): 567-79.
23
ScienceDirect Procedia CIRP 31 (2015) 19 – 23
15th CIRP Conference on Modelling of Machining Operations
Prediction of Residual Stresses after Laser-assisted Machining of Inconel 718 Using SPH M. A. Balbaaa, Mohamed N.A. Nasra,b,* a Dept. of Mechanical Engineering, Faculty of Engineering, Alexandria University, Alshatby, Alexandria 21544, Egypt Dept. of Materials Science & Engineering, Egypt-Japan University of Science & Technology, New Borg El-Arab, Alexandria, 21934, Egypt
b
* Corresponding author. Tel.: +20-3-591-5848; fax: +20-3-590-2715. E-mail address: [email protected].
Abstract Laser-assisted machining (LAM) has been proven to improve the machinability of hard-to-cut materials, by lowering cutting forces, increasing tool life and improving surface finish. Yet, not enough work has been done so far on how LAM would affect residual stresses (RS). The current study investigates the effects of LAM on RS, using finite element modelling, after dry orthogonal cutting of Inconel 718. First, the cutting process was modelled using the smoothed-particles hydrodynamics (SPH) method; secondly, an implicit Lagrangian model was used to simulate the relaxation process. Predicted LAM results were compared to conventional machining. The model was validated using previously published experimental results.
© 2015 2015 The The Authors. Authors.Published Publishedby byElsevier ElsevierB.V. B.V. This is an open access article under the CC BY-NC-ND license © (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of The International Scientific Committee of the “15th Conference on Modelling of Machining Operations”. Peer-review under responsibility of the International Scientific Committee of the “15th Conference on Modelling of Machining Operations Keywords: Residual stress; Finite element method (FEM); Laser-assisted machining (LAM); Turning; Smoothed-particles hydrodynamics (SPH).
1. Introduction Demanding operating conditions in nuclear and aerospace sectors required drastic development of high strength alloys; thus appeared the importance of Ti-based and Ni-based alloys. Inconel 718 represents a vastly used Ni-based alloy that is characterized by high strength, high wear and corrosion resistance, but met the challenge of being hard to machine. Hard-to-machine materials typically have low thermal conductivity and diffusivity, which cause a steep thermal gradient and high tendency to strain hardening during machining [1]. Also, the presence of abrasive carbides and intermetallic phases as well as adhesion to tool material adds to the complexity of machining these alloys. All these factors cause excessive tool wear, built up edge and premature tool cracking, high cutting forces, which all result in lower metal removal rates and high costs [1-3,6]. Several approaches have been adopted to solve the issues with hard-to-machine materials; mainly, cryogenic machining and laser-assisted machining (LAM). In cryogenic machining, liquid nitrogen is typically used to lower cutting temperatures
and improve productivity. On the other hand, LAM depends on pre-heating the workpiece with a laser beam ahead of the tool. This increase in temperature causes the material to soften; thus lowering the cutting forces and tool wear. 1.1. Literature review Surface integrity, especially machining-induced residual stresses (RS), is very crucial to controlling part performance. However, very little attention has been paid to the effect of LAM on RS in the available literature; even less work is available on modeling the LAM process itself. By reviewing the available literature, one can find that studies are focused on tool wear, surface roughness, surface hardness, cutting forces and chip formation [1-7]. A study by Navas et al.[2] focused on age-hardened and solution-annealed Inconel 718, where the machinability of both alloys was improved, especially the former. Cutting forces were found to drop with higher laser power densities and cutting speeds. Machinability also improved when the distance between the laser spot and cutting tool was decreased; however, a minimum distance was identified to avoid tool degradation by overheating. Attia et al.
2212-8271 © 2015 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the International Scientific Committee of the “15th Conference on Modelling of Machining Operations doi:10.1016/j.procir.2015.03.034
20
M.A. Balbaa and Mohamed N.A. Nasr / Procedia CIRP 31 (2015) 19 – 23
[1] performed an experimental study on the effect of LAM on finish turning of Inconel 718, and reported a significant drop in cutting forces compared to conventional machining (CM). Material removal rate was found to increase by almost 800% and surface finish improved by 25%. Tool wear was the focus of the work by Anderson et al. [3], and they reported a decrease in flank wear under LAM conditions compared to CM. Also, LAM specific cutting energy was significantly lower, and surface roughness was improved. separately. Below is an example which the authors may find useful. Ding et al. [4] performed an experimental investigation on LAM when turning AISI 4130 steel. LAM resulted in lower cutting forces, higher surface tensile RS, but no effect was found on surface hardness; however, no explanation was provided. Dumitrescu et al. [5] also reported a reduction in cutting forces after LAM of hardened AISI D2 tool steel. Finite element modeling (FEM) of LAM effects on machining parameters is very minimal in the literature. An example is the study by Shi et al. [6], where experimental and FEM work was done on the use of LAM under different cutting conditions, and recorded lower cutting forces versus those in CM. Also, Bouchnak et al. [7] modeled LAM of 42CrMo4, where cutting forces also dropped at different cutting speeds. Nasr et al. [8] performed FE simulations to study the effect of LAM on surface RS in AISI 4340 at different feed rates. Higher surface tensile RS were induced after LAM. Nasr et al. [9] studied the effect of laser power on surface RS through FE modeling. It was concluded that, surface tensile RS decrease and the tendency to induce compressive RS increases with the increase in laser power. Based on the above, and to the authors’ best of knowledge, very minimal focus was given to studying the effects of LAM on RS. Accordingly, the current work investigates the effects of LAM on surface RS when orthogonal dry cutting Inconel 718. 1.2. SPH modelling FEM of metal cutting faces some challenges when using the Lagrangian approach, due to high element distortion that could lead to termination of the process. Also, the Lagrangian approach requires a parting line (sacrificial layer) along which elements are deleted as they fail; thus leading to material volume loss. Accordingly, it is mainly limited to simulating sharp-edged tools, which is not always the case. The solution to these problems first came in the form of the Lagrangian adaptive re-meshing and arbitrary-Lagrangian-Eulerian (ALE) techniques [10]. SPH is a meshless method developed by Monaghan et al. [11], which has the advantage of being distortion free. It is a relatively new technique when it comes to modeling metal cutting; accordingly, it is not yet fully developed when compared to the FE method. The use of SPH in modelling the cutting process has drawn some attention in the past few years, with the main target being focused on exploring its
capabilities and making use of its advantages. Limido et al. [12] demonstrated the capabilities of SPH in modelling high speed machining and compared the results to experimental and FE modeling data. They found that SPH closely predicted continuous and serrated chips, and cutting forces were found to be within a small range of the experimental results. The authors also introduced the concept of time scaling, where the tool velocity was imposed ten times higher the real value, in order to reduce the computational time. Madaj et al. [13] confirmed the same results, as they constructed an SPH model for orthogonal cutting of Al2024-T351 alloy. The friction coefficient was varied to observe its effects on chip formation and cutting forces. They concluded that the predicted forces correlate well with the experimental results. Continuous chip formation was successfully modelled; however, the formation of serrated chips required the use of Johnson-Cook failure damage parameters, and the minimum strain for failure was required. The work of Villumsen et al. [14] on SPH modeling of 3D orthogonal cutting of Al6082-T6 alloy represents a significant contribution to the field, as the authors defined the material properties, equation of state (EOS), contact settings and boundary conditions (BCs) required for SPH. They also performed a sensitivity analysis on the effect of particle pitch, mass scaling, time scaling and friction coefficient on cutting forces. It can be shown from the previous review that there is not enough work done on SPH modeling of metal cutting. That fact coupled with the absence of understanding the effects of LAM on surface RS makes it an interesting point for investigation. Accordingly, a set of SPH models was built in order to compare RS after LAM and CM, when orthogonal dry cutting Inconel 718, as mentioned earlier. Cutting conditions were chosen similar to those of Shi et al. [6] for model validation. 2. Model Description In order to predict RS; first, the cutting process was modelled using the smoothed-particles hydrodynamics (SPH) method; secondly, an implicit Lagrangian model was used to simulate stress relaxation and cooling down. Even though the current work focuses on orthogonal cutting, a 3D model was built as SPH is not available for 2D analyses in the commercial software ABAQUS/Explicit, which was used for the current study. Two models were built to simulate CM and LAM effects on surface RS when orthogonal cry cutting Inconel 718. It is worth mention that, SPH was only used for the workpiece, while the tool was modeled using a Lagrangian mesh. To the authors’ best of knowledge, the current work is the first to present an SPH cutting model using the commercial software ABAQUS. After the cutting process was modelled, a standard implicit Lagrangian model was used to simulate stress relaxation and cooling down to room temperature for RS prediction.
M.A. Balbaa and Mohamed N.A. Nasr / Procedia CIRP 31 (2015) 19 – 23
2.1. Workpiece properties
2.2. Contacts definition
The cutting model can be seen in Fig.1 with the applied boundary conditions. The dimensions of the workpiece are 1.5 mm x 1 mm x 0.02 mm. Coupled temperature-displacement explicit analysis was used, and the workpiece was modeled with a total of 35511 PC3D particles. The workpiece is fixed at the bottom surface, and a velocity BC was imposed to the back side of the tool.
The contact between the workpiece and tool was defined using node-to-surface contact algorithm, with a Coulomb friction model assumed along with a friction coefficient of 0.2. Despite being a simple model, the Coulomb friction model is not limited to SPH; however, it has been used widely with Lagrangian and ALE models, where acceptable results were achieved [10,18, 23]. 2.3. Heat transfer
Cutting tool Workpiece
In the current work, it was assumed that 90% of plastic deformation energy is converted to heat. Also, the materials’ properties allowed 33.5 % of the generated friction heat to be conveyed to the workpiece. During cutting, heat convection and radiation were neglected. On the other hand, convection to the environment (air) was modelled during the stress relaxation (RS) step, where a convection coefficient of 10 W/ (m2oC) and a sink temperature of 20 oC were used.
Cutting direction
Fig. 1 SPH 3D cutting model (front view)
Inconel 718 was defined for the workpiece material, and temperature dependent physical properties were defined based on those supplied in [15] The well-known Johnson-Cook (J-C) plasticity model (Eq. 1) [16] was used to model the workpiece material behavior, with its parameters listed in Table 1 [17]. The J-C model has been widely used for metal cutting simulations, including machining of Inconel 718 [10,17-20]. ߪ = ( ܣ+ ߝ ܤ
)
ఌሶ
்ି்ೝ
ఌሶ బ
் ି்ೝ
ቀ1 + ܥln ቀ ቁቁ ቀ1 െ ቀ
ቁ ቁ
(1)
Table 1. J-C constitutive model parameters [16]. Parameter
Value
Initial yield strength A (MPa)
1241
Strain hardening coefficient B (MPa)
622
Strain hardening exponent n
0.6522
Strain rate coefficient C
0.0134
Thermal softening exponent m
1.3
Reference plastic strain rate ߝሶ (s-1)
1
The Mie-Gruneisen EOS was used to describe the pressure-volume relationship of the workpiece material, which is essential when using SPH. However, it is important to note that, the parameters of Monel were used as those of Inconel 718 are not currently available [21]. This was also done in a recent study by Oak Ridge National Laboratory (for the US Department of Energy), where the Monel EOS was used for Inconel 718 during impact of space reactor cores, due to the non-availability of the exact Gruneisen parameters [22].
2.4. Cutting conditions Cutting conditions were adopted from the work of Shi et al. [6]; whereas cutting speed of 100 m/min and uncut chip thickness (t1) of 0.125 mm (feed rate of 0.125 mm/rev) were used for CM and LAM conditions. A K313 carbide insert with a sharp edge, as assumed by Shi et al. [6], and a zero rake angle was used, whose properties can be found in [24]. 2.5. LAM preheating phase The laser beam heating effect was modeled in the form of a preheating phase, before modelling the cutting process. The workpiece temperature distribution was sought to match that presented by Shi et al. [6]. Accordingly, a transient thermal analysis was run for a period equal to the lead time of the laser beam ahead of the tool. The analysis started by the same surface temperature measured experimentally by Shi et al. [6], which was 800 oC, as a thermal boundary condition. Such temperature was reported to be generated by a 3000 W laser. 3. Results and Discussion The results below show the effect of LAM on cutting forces and maximum chip thickness. These results are validated by comparing them to the results of Shi et al. [6], from which the current cutting conditions are derived. Also, the in-depth temperature gradient and plastic strain are plotted to demonstrate their effect on surface RS, under LAM and CM conditions. Fig. 2 shows the chip formation process during cutting, where t1 represents the uncut chip thickness and t2 represents the deformed chip thickness.
21
22
M.A. Balbaa and Mohamed N.A. Nasr / Procedia CIRP 31 (2015) 19 – 23
Since RS are affected by plastic strain, temperature gradient and phase transformation, which was not present in the current study (as reported in [6]); therefore, in order to explain the reasons behind the effects of LAM on surface RS, the plastic strain in cutting direction and temperature gradients are shown in Fig. 4 and Fig. 5, respectively.
t2
Shear angle
0.10
Fig. 2 Chip generation in SPH
3.1. Cutting forces Cutting forces were predicted for CM and LAM, and the results are compared to those of Shi et al. [6] as shown in Table 2.
Plastic Strain PE11
t1
CM
0.08
LAM
0.06 0.04 0.02 0.00 -0.02
10
30
50
70
Depth from machined surface (microns)
Table 2. Cutting forces. Exp.
SPH
CM cutting force (N)
170
156
LAM cutting force (N)
130
126
Difference “LAM - CM” (%)
-23.2
-19.2
It can be seen that the SPH model closely predicted the cutting forces. Also, the effect of LAM on cutting forces is depicted clearly and nearly with the same trend for experimental and SPH results. LAM caused the cutting force component to drop. This can be traced back to the thermal effect of LAM, as the material softens with the rising temperature thus lowering the resistance head of the tool. 3.2. Surface residual stresses in cutting direction Surface RS in cutting direction are presented in Fig. 3, where LAM resulted in lowering the tensile surface RS; actually, RS turned to be compressive instead of being tensile. Unfortunately, no experimental RS are currently available to confirm such results. However, the effects of LAM on other process parameters (forces and temperature) support the validity of its effects on RS.
Residual stresses (MPa)
600 400 200
By looking at the plastic strain in cutting direction (PE11), we can observe that the SPH model predicted a higher tensile PE11 in case of LAM compared to CM. Higher tensile PE11 under LAM conditions can be traced back to the thermal softening effect of the material ahead of the tool. Therefore, the compressive stresses generated ahead of the tool tip decreases, compared to CM. This tends to results in an overall higher tensile stresses behind the tool tip. Accordingly, higher tensile PE11 is generated under LAM conditions, compared to CM. Since surface tensile plastic strain results in more compressive or less tensile RS [25], this explains why LAM resulted in less tensile (actually, compressive) surface RS as presented in Fig.3. As shown in Fig. 5, it was found that the difference in workpiece temperature underneath the tool tip, during cutting, between CM and LAM is limited to the nearsurface layer (about 11 microns). Conv
LAM
500 400 300 200 100 0
0 -200
Fig. 4 In-depth plastic strain in cutting direction (PE11).
Temperature (oC)
Parameter
CM
LAM
0
50 100 Depth from machined surface (microns)
-400 -600
Fig. 3 Predicted surface RS in cutting direction.
Fig. 5 Temperature gradient for machined workpiece.
M.A. Balbaa and Mohamed N.A. Nasr / Procedia CIRP 31 (2015) 19 – 23
4. Conclusions The current study focused on SPH modeling of LAM process and its effect on surface residual stresses in cutting direction. Also, the effects of LAM on cutting forces, temperatures and plastic strain in cutting direction were also examined. Based on the presented results, the following conclusions were drawn. 1.
The model closely predicted the cutting force component, and captured the trend reported experimentally where LAM resulted in lower forces.
2.
For the studied conditions, LAM resulted in surface compressive residual stresses in cutting direction; however, conventional machining resulted in surface tensile residual stresses.
3.
The effect of LAM on residual stresses are attributed to the thermal softening effects of the laser beam ahead of the tool tip, which resulted in overall more tensile plastic strain in cutting direction. Accordingly, less tensile (more compressive) residual stresses.
References [1] Attia H, Tavakoli S, Vargas R, Thomson V. Laser-assisted high-speed finish turning of superalloy Inconel 718 under dry conditions. CIRP Annals-Manufacturing Technology. 2010; 59(1): 83-8. [2] Garcí V, Arriola I, Gonzalo O, Leunda J. Mechanisms involved in the improvement of Inconel 718 machinability by laser assisted machining (LAM). International journal of machine tools and manufacture. 2013;74: 19-28. [3] Anderson M, Patwa R, Shin YC. Laser-assisted machining of Inconel 718 with an economic analysis. International Journal of Machine Tools and Manufacture. 2006; 46(14): 1879-91. [4] Ding H, Shin YC. Laser-assisted machining of hardened steel parts with surface integrity analysis. International Journal of Machine tools and manufacture. 2010; 50(1): 106-14. [5] Dumitrescu P, Koshy P, Stenekes J, Elbestawi M. High-power diode laser assisted hard turning of AISI D2 tool steel. International Journal of Machine Tools and Manufacture. 2006; 46(15): 2009-16. [6] Shi B, Attia H, Vargas R, Tavakoli S. Numerical and experimental investigation of laser-assisted machining of Inconel 718. Machining Science and Technology. 2008; 12(4): 498-513. [7] Braham-Bouchnak T, Germain G, Morel A, Lebrun J-L. The influence of laser assistance on the machinability of the titanium alloy Ti555-3. The International Journal of Advanced Manufacturing Technology. 2013; 68(9-12): 2471-81. [8] Nasr MN, Balbaa M, Elgamal H. Modelling Machining-induced Residual Stresses after Laser-assisted Turning of Steels. Advanced Materials Research; 2014: Trans Tech Publ. pp. 622-627. 2014 [9] Nasr MN, Balbaa M, Effect of Laser Power on Residual Stresses when Laser-assisted Turning of AISI 4340 Steel, Canadian Society for Mechanical Engineering (CSME) Int. Congress, 2014,Toronto, Canada, June 1-4. [10] Nasr MN, Ng E-G, Elbestawi M. Modelling the effects of tool-edge radius on residual stresses when orthogonal cutting AISI 316L. International Journal of Machine Tools and Manufacture. 2007; 47(2): 401-11. [11] Gingold RA, Monaghan JJ. Smoothed particle hydrodynamics: theory and application to non-spherical stars. Monthly notices of the royal astronomical society. 1977; 181(3): 375-89.
[12] Limido J, Espinosa C, Salaün M, Lacome J-L. SPH method applied to high speed cutting modelling. International journal of mechanical sciences. 2007; 49(7): 898-908. [13] Madaj M, Píška M. On the SPH orthogonal cutting simulation of A2024T351 alloy. Procedia CIRP. 2013; 8: 152-7. [14] Villumsen, M. F., Fauerholdt, T. G., Simulation of Metal Cutting using Smoothed Particle Hydrodynamics, 7. LS-DYNA Anwenderforum, Bamberg 2008, Germany.C-III-17–C-III-36. [15] Special Metals: Inconel 718 properties datasheet. [16] Johnson GR, Cook WH. A constitutive model and data for metals subjected to large strains, high strain rates and high temperatures. Proceedings of the 7th International Symposium on Ballistics; 1983: The Netherlands. [17] Kortabarri A, Madariag A, Fernandez E, Esnaol J, Arrazola P. A comparative study of residual stress profiles on Inconel 718 induced by dry face turning. Procedia Engineering. 2011; 19: 228-34. [18] Joyot P, Rakotomalala R, Pantale O, Touratier M, Hakem N. A numerical simulation of steady state metal cutting. Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science. 1998; 212(5): 331-41. [19] Outeiro J, Pina J, M'saoubi R, Pusavec F, Jawahir I. Analysis of residual stresses induced by dry turning of difficult-to-machine materials. CIRP Annals-Manufacturing Technology. 2008; 57(1): 77-80. [20] Chandrasekaran H, M'saoubi R, Chazal H. Modelling of material flow stress in chip formation process from orthogonal milling and split Hopkinson bar tests. Machine Science and Technology. 2005; 9(1): 13145. [21] Steinberg D. Equation of state and strength properties of selected materials: Lawrence Livermore National Laboratory Livermore, CA; 1996. [22] Kim, Seokho H. and Flanagan, George F. Impact analysis for candidate space reactor core concept designs for potential critical study. Oak Ridge, TN: Oak Ridge National Laboratory. [23] Shet C, Deng X. Finite element analysis of the orthogonal metal cutting process. Journal of Materials Processing Technology. 2000; 105(1): 95109. [24] Nasr M, Ng E, Elbestawi M. Effects of workpiece thermal properties on machining-induced residual stresses-thermal softening and conductivity. Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture. 2007; 221(9): 1387-400. [25] Nasr MN, Ng E-G, Elbestawi M. Effects of strain hardening and initial yield strength on machining-induced residual stresses. Journal of Engineering Materials and Technology. 2007; 129(4): 567-79.
23