Advances in Mechanical Engineering 2016, Vol. 8(3) 1–10
ÓThe Author(s) 2016 DOI: 10.1177/1687814016641569 aime.sagepub.com
Tool path planning and milling surface
simulation for vehicle rear bumper
mold
Peng Wang, Song Zhang, Zhe Li and Jianfeng Li
Abstract
Sculptured surfaces have been widely used in various engineering applications. With the development of five-axis machin-ing technology, five-axis machine tool has played a more and more important role in manufacturmachin-ing sculptured surfaces. This article proposed a complete five-axis machining process planning for vehicle rear bumper mold. First, a numerical model considering tool inclination was developed to predict topography and surface roughness in ball-end milling pro-cesses. And then experimental verification was conducted. Second, according to the simulation results, it can be con-cluded that surface roughness decreases significantly with the tilt angle increasing from 0°to 5°. When the tilt angle is 10°, surface roughness gets the smallest value. What is more, surface roughness increases slightly from 10°to 30°. In the case of the permitted surface roughness, larger feed rate per tooth and the radial cutting depth can be chosen. Finally, a process strategy was made through the analysis of the mold in which the parameters were chosen according to the simulation results above.
Keywords
Five-axis machining, vehicle rear bumper mold, process planning, surface topography, tilt angle
Date received: 28 July 2015; accepted: 7 March 2016
Academic Editor: David R Salgado
Introduction
In the field of free-form machining, computer-aided manufacturing (CAM) software offers various machin-ing strategies dependmachin-ing on the geometry of the surface to be machined. The surface quality depends on process parameters (lead angle, tilt angle, feed per tooth, cut-ting speed, radial depth of cut, cutcut-ting depth) and the choice of the machining strategy. The resulting machin-ing time, productivity, and geometrical surface quality directly depend on these parameters.
Cutting process parameters’ optimization
Wang et al.1 and Zhang and Li2 studied the control and optimization of cutting process and predicted the machining performance. As surface quality is
concerned, the scallop generation mechanism must be well controlled. Erkorkmaz et al.3 presented a new strategy for planning tool trajectories of minimum time in three-axis milling. Process constraint and feed drive control system limitation were both considered. Sun et al.4developed a more comprehensive feed optimiza-tion method with the constraints of geometry error and dynamics. Vickers and Quan5 expressed the
Key Laboratory of High Efficiency and Clean Mechanical Manufacture (Ministry of Education), School of Mechanical Engineering, Shandong University, Jinan, P.R. China
Corresponding author:
Song Zhang, Key Laboratory of High Efficiency and Clean Mechanical Manufacture (Ministry of Education), School of Mechanical Engineering, Shandong University, Jinan 250061, P.R. China.
Email: zhangsong@sdu.edu.cn
path-interval scallop height as a function of the curva-tures of surfaces and path intervals. Kruth and Klewais6took the inclination effect of the cutting axis into the path-interval scallop height model. Due to lim-itation of cutter material, the tooth feed is kept to com-parably less than the path pick, and the previous works only considered the path-interval scallop. In today’s high-speed hard machining technology, however, the tooth feed has been raised to the same level of the path pick. Chen et al.7,8 investigated the scallop height in which the effects of inclination angles were taken into consideration. The researches were conducted by geo-metrical analysis and experimental methods.
The simulation method is more explicit and simple compared with the geometrical analysis and experimen-tal methods. Chen et al.9 studied the generating mechanism of feed and path-interval scallops and the influence of process parameter on surface scallop. However, the model was only constructed with a ball-end cutter with two straight flutes. Quinsat et al.10 developed the calculation model of surface topography characterization parameter taking the feed and path-interval scallops into account. Some research11,12dealt with the prediction of the three-dimensional (3D) sur-face topography obtained in five-axis milling with an end milling cutter. However, the researchers only con-sidered the inclination angle while ignoring the feed rate per tooth and the path interval.
In five-axis machining, more machining strategies are developed for the complex part. It makes the machining process more efficient, and it also brings more machining mistakes such as air travel, overcut, or undercut of the manufacturing. Moreover, as the tool axis orientation generally varies during machining, the resulting surface pattern can be affected.13The predic-tion of the 3D surface topography according to the machining conditions is also an important issue to achieve process planning correctly.
The machining strategy
At present, interactive graphical programming system has been widely used in five-axis milling; CAM systems provide access to a large variety of data such as the pro-grammed feed rate, the tool length, and the number of flutes.14 There are still problems such as collision and interference of tool path. Balasubramaniam et al.15 pre-sented a series of algorithms and heuristics for generat-ing collision-free five-axis computer numerical control (CNC) finishing tool paths automatically. However, the algorithms were tedious. Jun et al.16proposed a search-ing method in the machinsearch-ing configuration space (C-space) to find the optimal tool orientation. It took the local gouging, rear gouging, and global tool collision into account. Zhu et al.17 researched the high-speed milling (HSM) of vehicle front bumper die based on
Unigraphics (UG) CAM. Fan et al.18 developed meth-ods for cutter orientation and tool path generation in five-axis sculptured surface machining without gouging. The famous iso-scallop tool paths’ generation strategy was first proposed by Suresh and Yang.19 Ahmet and Ali20developed a novel iso-scallop tool path generation strategy for the efficient five-axis machining. And the cutter paths were scheduled to make the scallop height formed between two adjacent machining paths con-stant. Moreover, this study also achieved a maximized material removal rate by an optimized tool orientation and curvature matching. Since the iso-scallop and roughness often conflict with each other in the existing iso-scallop path planning methods, Zou et al.21 pro-posed a new framework to plan globally optimal tool path. It views a family of iso-level curves of a scalar function defined over the surface as tool path so that the desired tool path can be generated by finding the function that minimizes certain energy functional and different objectives can be considered simultaneously.
Despite the progress, many challenges still need to be further addressed to improve the quality and effi-ciency of the specific surface machining process plan-ning in industry.
From the literature review, it can be concluded that the current researches were mainly focused on process parameters’ optimization and tool path planning algo-rithms. For the cutting parameters’ selection, the sur-face topography at the five-axis machining is particularly sensitive to the tooth feed, the radial cut-ting depth, and tool axis inclination in the ball-end milling operation. However, there is still a lack of a model to clearly explain the surface scallop generating mechanism for these parameters. Since the lead angle is studied completely, the tilt angle is needed to be studied with deeper insight. As for the tool path planning, few researches were conducted for mold or other specific complex parts.
In this article, parameters’ optimization and tool path planning for specific part rear bumper mold were conducted. A theoretical model of the surface topogra-phy formation in five-axis ball-end milling was pre-sented. The simulation and experimental verification of the model is conducted with tilt angle, feed rate per tooth and radial cutting depth being studied. And then the tool path planning for rear bumper mold based on NX9.0 was proposed. Besides, the cutting parameters were selected according to parameters’ optimization above.
Modeling and controlling of surface
topography in ball-end milling
milling with a filleted-ball-end cutter. Selection of the cutting parameters and their effects on the mold quality are discussed. Meanwhile, the experiment is conducted to validate the model.
Modeling
The machined surface topography is essentially gener-ated by the motion of the cutter edge. So the cutting motion model can be described as equation (1)
P(x,y,z) = x y z 1 2 6 6 4 3 7 7 5
=TwtTtvTv0v
u v w 1 2 6 6 4 3 7 7 5 ð 1Þ
where (x, y, z) is the coordinate of sweeping points’ cloud generated by cutter motion based on workpiece coordinate system.
Twt is the coordinate transformation matrix, as
shown in equation (2), transforming the machine tool coordinate into the workpiece coordinate. b1 is the
rotation angle around the X-axis, b2 is the rotation
angle around theZ-axis. (x0, y0,z0) is the initial
posi-tion of pointP. iis the feed times,vfis the line speed of
the cutting tool, andtis the running time inith feed
Twt=
1 0 0 0
0 cosb1 sinb1 0
0 sinb1 cosb1 0
0 0 0 1
2 6 6 6 4 3 7 7 7 5
cosb2 sinb2 0 0
sinb2 cosb2 0 0
0 0 1 0
0 0 0 1
2 6 6 6 4 3 7 7 7 5
1 0 0 x0+ (i1)ae
0 1 0 y0+vft
0 0 1 z0
0 0 0 1
2 6 6 6 4 3 7 7 7 5
ð2Þ
b1 and b2 are the corresponding angles of the
machine tool principal axis rotation angle. It is always used in matrix transforming since it is convenient and explicit. In five-axis milling, surface topology is affected by lead angle and tilt angle which are not equal tob1
andb2. Lead angle is the included angle between tool
axis andz1-axis in Plane y1z1which is rotated around
they-axis. In addition,y-axis is the feed direction. Tilt angle is the included angle between Plane y1z1 and
Planeyz. In Figure 1, supposing the coordinate ofNis
(ax, ay, az), anda
2
x+a
2
y+a
2
z=1. Apparently, the lead
angle and tilt angle can be established relations withb1
and b2 through geometric transformation. Hence,
equations (3) and (4) are established. (In equations (4)
and (5),Lrepresents the lead angle andTrepresents the tilt angle.) Equation (5) can be obtained by combining equations (3) and (4). Equation (5) enables lead angle and tilt angle to be used in surface topology model directly
b1= arcos(az)
b2= arctanax ay (
ð3Þ
tan (T) =ax az
tan (L) = ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiay a2
x+a
2
z p
a2
x+a
2
y+a
2
z=1
cos (T) cos (L) = ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiaz a2
x+a
2
y+a
2 z q 8 > > > > > > > < > > > > > > > :
ð4Þ
b1= arccos
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1
tan2
T+ tan2
L+ tan2 Ttan2
L+1
r
b2= arctan ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffitanT tan2
L+ tan2 Ttan2
L p 8 > > < > > :
ð5Þ
Ttv is the coordinate transformation matrix, as
shown in equation (6) transforming the tool kinematic coordinate into the machine tool coordinate
Ttv=
1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1
2 6 6 4 3 7 7 5 ð 6Þ
Tv0v is the coordinate transformation matrix, as
coordinate into the tool kinematic coordinate. ui is the
rotation angle of the tool
Tv0v=
cosui sinui 0 0
sinui cosui 0 0
0 1 1 R
0 0 0 1
2
6 6 4
3
7 7
5 ð
7Þ
(u, v, w) is the coordinate of point P (shown in Figure 2) which is on cutter edge based on the tool kinematic coordinate system. As shown in equation (8), a is the included angle betweenOTP andW-axis.g is
the helix angle of the tool
u=R sina cos tang ln cota 2
v=Rsinasin tangln cota 2
w=R(1cosa)
8
<
:
ð8Þ
The workpiece used in this article is the flat surface. And it was divided into (m21)3 (n21) matrix grid (m3n grid points), as shown in Figure 3. The grid points were first valued andHrepresents the scallop of the workpiece inZ-direction. Then calculate and record
the discrete cutting-edge point location when the tool moves to the all-discrete tool path position. The flow chart is shown in Figure 4 to express the calculating procedure of the surface topography.
The simulation result is obtained in MATLAB soft-ware, as shown in Figure 5. Apart from the simulated
surface topology, the surface roughness Ra is also simulated as shown in equation (9). In equation (9), h represents the arbitrary scallop of the point in surface topology, h represents the mean value of all chosen Figure 2. Tool coordinate.
Figure 3. Workpiece model.
points’ scallop.nrepresents the number of the chosen points and is equal to the number of the mash points as simulation setting
Ra=1 n
Xn
i=1 hh
ð9Þ
Experiment validation
A series of experiments were conducted with machining center DMU-70V (shown in Figure 6) under the pro-cess parameters as shown in Table 1. The workpiece material is P20 steel, size of 2503150350 mm. The machined area was 503 50 mm. Each block allowed performing eight different experiments. The cutting tool used in this experiment is a ball-end cutter JH970100-TRIBON. The surface topography is measured by opti-cal profilometer Wyko NT9300.
The experimental and simulated surface topogra-phies are compared in Figure 7. The simulated results remarkably agree with the experimental results. The model is ideal in the geometrical simulation in which
some physical factors were not taken into consider-ation, such as material plasticity, vibrations, and tool flexion.
Simulation results
From the analysis of the surface roughness in Table 5, it can be concluded that the surface roughness is partic-ularly sensitive to the feed rate per tooth and radial cut-ting depth. With the increase in the feed rate per tooth, the surface roughness gets larger. Meanwhile, the radial cutting depth has similar influence on the surface roughness. The experiment results indicate the same tendency, but the experiment results are larger than the simulation results. This is because some physical fac-tors such as tool vibration and material plasticity, which were ignored in the simulation also have some effects on surface roughness. Under the premise of per-mitted surface roughness, we can choose large feed rate per tooth and radial cutting depth.
Moreover, the surface topography is also affected by the inclination of tool axis in the ball-end milling Table 1. Process parameters with same cutting speed 157 mm/s, cutting depth 0.2 mm, and lead angle 0°.
Radial cutting depth,ae(mm) Feed rate per tooth,fz(mm/z) Tilt angle (°) Surface roughness (mm)
Simulation Experiment
0.2 0.2 5 0.515 0.531
0.3 0.2 0.631 0.649
0.4 0.2 0.762 0.795
0.1 0.2 0.453 0.474
0.1 0.15 0.434 0.446
0.1 0.1 0.389 0.407
0.3 0.2 0 1.08 1.37
10 0.628 0.644
15 0.630 0.653
20 0.642 0.661
25 0.644 0.655
30 0.645 0.662
operation. The tool inclination is divided into lead angle and tilt angle. Generally speaking, lead angle is along the feed direction and tilt angle is perpendicular to the feed direction. And the research about lead angle is intensive. However, research about tilt angle is not deep going. Figure 8 shows the effect of the tilt angle on surface roughness. When the tilt angle is 0, the sur-face roughness is large. This is because the velocity of the ball-end top is 0, and material plasticity and mate-rial tear lead to serious surface defect. It can be
indicated from Figure 6 that the surface roughness decreases significantly with the tilt angle increasing from 0°to 5°. When the tilt angle is 10°, surface rough-ness gets the smallest value. What is more, surface roughness increases slightly from 10° to 30°. In other words, the tilt angle can be chosen correspondingly from 5°to 30°. Generally, in order to get a good surface quality, we can choose the best tilt angle 10°. Meanwhile, a certain tilt angle from 5° to 30° can be chosen to satisfy the actual machining process. Figure 7. Simulation and experimental results of surface topology: (a)fz= 0.2 mm/z,ae= 0.3 mm, tilt angle = 15°; (b)fz= 0.2 mm/z,
ae= 0.3 mm, tilt angle = 15°; (c)fz= 0.2 mm/z,ae= 0.3 mm, tilt angle = 0°; (d)fz= 0.2 mm/z,ae= 0.3 mm, tilt angle = 0°; (e)fz= 0.2 mm/
Meanwhile, the conclusion can be used in the NC programming.
In a conclusion, when other processing parameters are identified, the major factors influencing the surface topography are feed rate per tooth (fz) and radial
cut-ting depth (ae). In addition, the surface roughness is
also affected by lead angle and tilt angle.
The rear bumper mold process planning
Machining strategy
The 3D models of rear bumper and rear bumper mold are shown in Figure 9. The workpiece material was P20 steel, which is widely used for plastic mold, extrusion, and hot forging.
The nominal chemical composition and the material properties of P20 steel are given in Tables 2 and 3, respectively. The workpiece material was hardened and tempered to attain a hardness of 30–36 HRc. A 2370 mm3 1130 mm31070 mm rectangular block was used as the sample.
According to the complicated cavity structure, pro-cess strategies were summarized as follows:
1. The mold cavity is very complicated since it has a lot of bulges and grooves. Besides, the machin-ing allowance for steel is not well-distributed. Scrape will appear due to the harsh cutting envi-ronment. The mold manufacturing process should be more reasonable and efficient through optimizing tool paths.
2. Traditional mold manufacturing takes much time on electrical discharge machining (EDM) and manual grinding. In order to remove these traditional processes, the precision of finishing should be kept at a high level. Small step and depth of cut should be adopted to make the tool path more smoothly. Plunge milling should be replaced by spiral feed and circular feed milling. Figure 8. Simulation and experimental results of surface
roughness.
Figure 9. 3D models of (a) the rear bumper and (b) the rear bumper mold.
Table 2. Nominal chemical composition of P20 tool steel (wt%).
C Si Mn Cr Mo Ni Fe
0.28–0.40 0.20–0.80 0.60–1.00 1.40–2.00 0.30–0.55 0.05–0.10 Bal.
Table 3. Material properties of P20 steel at room temperature.
Density (kg/m3) Young’s modulus (GPa) Hardness (HRc) Yield strength (MPa) Thermal conductivity (W/m K)
3. In NC machining, rough machining takes most of the machining time since the size of the work-piece is very large. The tool-life and tool wear need to be considered in the manufacturing.
The mold was divided into three parts according to the functional properties of the bumper mold, for exam-ple, the guide sleeve hole, the cavity surface, and the auxiliary structure. For the convenience of the process planning, the main features are numbered in Figure 10 and described in detail in Table 4.
According to the division of machining features, the processing routes of mold cavity are as follows:
1. Milling of the blank. For the sake of process efficiency and safety of the cutter, an end mill cutter was used to manufacture the blank. The manufacturing could get rid of burrs and defects of the blank, getting ready for the rough milling.
2. Milling of F1 and F2. To improve the effi-ciency, these two features were machined in one time by an end mill cutter.
3. Milling of F3 and F4. These two features were also machined in one time by an end mill cutter.
4. Milling of F5, F6, F7, and F8. These four fea-tures were machined in one time by an end mill cutter.
5. Milling of F9. This feature was also machined by an end mill cutter.
6. Milling of F10. The guide sleeve hole was designed for the installation of the guide sleeve. High precision was needed in the manufactur-ing of the hole.
7. Milling of F11. The cavity auxiliary curved sur-face F11 was steep which meant the errors caused by tool stiffness should be considered. 8. Milling of F12. The cavity auxiliary curved
sur-face F12 was not as steep as F11 which meant the errors caused by tool stiffness can be ignored.
9. Milling of F13 and F14. The groove and the channel were on top side of the feature F12. Relatively speaking, these two features were small compared with F12 which could be machined with F12 together. Besides, the accu-racy was achieved through the clean-up machining.
10. Milling of F15 and F16. The cavity auxiliary curved surface F15 was not as steep as F11 which means that the errors caused by tool stiffness could be ignored. F16 could be machined with F15 in one time, and the accu-racy was achieved through the clean-up machining.
Table 4. Mold cavity surface processing feature.
Feature number Structural feature Specific location
F1 Step surface Upper side of mold
F2 Step surface Upper side of mold
F3 Step surface Below side of F1
F4 Step surface Below side of F2
F5 Step surface Below side of F1
F6 Step surface Below side of F5
F7 Step surface Below side of F6
F8 Step surface Below side of F7
F9 Step surface Below side of F1
F10 Guide sleeve hole Boundary location of mold F11 Cavity auxiliary curved surface Upper side of the cavity surface F12 Cavity auxiliary curved surface Upper side of the cavity surface
F13 Groove Upper side of the auxiliary curved surface
F14 Channel Upper side of the auxiliary curved surface F15 Cavity auxiliary curved surface Upper side of the cavity surface
F16 Convex plate Upper side of F15
F17 Cavity surface Middle position
11. Milling of F17. The cavity surface was the most important part of the mold, so high precision and good surface roughness were required. Since the steep degree changes greatly, the five-axis machining could be used effectively. There were also some small features which could be machined in the same time with the cavity sur-face. And the accuracy was achieved through the clean-up machining.
Selection of cutting tools
Considering good cutting performance, coated cemen-ted carbide tools were seleccemen-ted in the machining. The cutting tools used in the real processing are from Kennametal Company. The recommended cutter para-meters and cutting parapara-meters are shown in Table 5.
NC programming based on NX 9.0
CAM software is usually used to conduct NC program-ming. NX 9.0 (Siemens PLM Software Co., Ltd, Germany) which was widely used in industry was selected in this article. To obtain high-quality surface, the process was divided into roughing, semi-finishing, and finishing processes. The cutting parameters were chosen according to the recommended value from the Kennametal Company tool catalog and the conclusion in section ‘‘Modeling and controlling of surface topo-graphy in ball-end milling’’ of this article.
In rough process, the capacity of the machine tool should be fully used to remove the material. The allow-ance of the subsequent process needs to be well-distrib-uted. Semi-finish process is used to reduce the error left by rough machining process. Moreover, a certain preci-sion can be achieved by semi-finish, preparing for finish processing. In rough process and semi-finishing process, the cutting parameters were mainly chosen according to the recommended value from the tool catalog.
The main purpose of finishing process is to get good machining precision and stable surface topology. Therefore, the tilt angle is set 10° to keep the surface topology consistent according to the conclusion in sec-tion ‘‘The rear bumper mold process planning’’ of this
article. The simulation results of finishing process based on NX 9.0 are shown in Figure 11.
The tool path visualization was achieved after pro-gramming to find if there was air travel, overcut, or undercut of the manufacturing. After the simulation and visualization of the tool path, all collision and over-cut were eliminated.
Conclusion
In this article, parameters’ optimization and tool path planning for specific part rear bumper mold are con-ducted. The novelty of the article is that it studied the tilt angle for the first time with modeling method in detail. And then the tool path planning for the rear bumper mold based on NX9.0 was proposed, with the process parameters chosen accordingly. The completed work is concluded as follows:
1. A numerical model considering tool inclination was developed that predicts topography and surface roughness in ball-end milling processes. Since the simulated results remarkably agree with the experimental results, the model can describe the real surface topology correctly. 2. With the increase in the feed rate per tooth, the
surface roughness gets larger. Meanwhile, the radial cutting depth has similar influence on the surface roughness. In the case of the permitted surface roughness, large feed rate per tooth and radial cutting depth can be chosen to obtain high efficiency.
Table 5. Main cutter parameters and cutting parameters.
Tool type Tool number Diameter (mm) Feed rate per tooth (mm/tooth) Cutting depth (mm)
Flat-end mill T1 66 0.5–1.5 1.5
Ball-end mill T2 32 0.1–1.0 1.5
Flat-end mill T3 32 0.5–1.5 1.5
Ball-end mill T4 16 0.1–0.7 1.0
Flat-end mill T5 20 0.3–0.8 1.0
Ball-end mill T6 8 0.05–0.4 1.0
3. The surface roughness decreases significantly with the tilt angle increasing from 0° to 5°. When the tilt angle is 10°, surface roughness gets the smallest value. What is more, surface roughness increases slightly from 10°to 30°. In other words, the tilt angle can be chosen corre-spondingly from 5° to 30°. Generally, in order to get a good surface quality, we can choose the best tilt angle 10°. Meanwhile, we can also choose a certain tilt angle from 5°to 30°to sat-isfy the actual machining process.
4. The specific rear bumper cutting process was modeled and simulated in NX 9.0. The NC pro-gramming and the tool path planning were conducted.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial sup-port for the research, authorship, and/or publication of this article: This work was supported by the National Natural Science Foundation of China (Grant No. 51575321) and National Major Science and Technology Project: High-end CNC Machine Tools and Basic Manufacturing Equipments (Grant No. 2012ZX04006-011).
References
1. Wang Y-W, Zhang S, Li J-F, et al. Optimal cutting para-meters for desired surface roughness in end milling Inconel 718.Adv Materials Res2010; 126–128: 911–916. 2. Zhang S and Li J-F. Prediction of surface roughness
using back-propagation neural network in end milling Ti-6Al-4V alloy.Adv Materials Res2011; 325: 418–423. 3. Erkorkmaz K, Layegh SE, Lazoglu I, et al. Feedrate
optimization for freeform milling considering constraints from the feed drive system and process mechanics.CIRP Ann: Manuf Techn2013; 62: 395–398.
4. Sun Y, Zhao Y, Bao Y, et al. A novel adaptive-feedrate interpolation method for NURBS tool path with drive constraints.Int J Mach Tool Manu2014; 77: 74–81. 5. Vickers GW and Quan KW. Ball-mills versus end-mills
for curved surface machining.J Eng Ind: T ASME1989; 111: 22–26.
6. Kruth JP and Klewais P. Optimization and dynamic adaptation of the cutter inclination during five-axis
milling of sculptured surfaces.CIRP Ann: Manuf Techn
1994; 43: 443–448.
7. Chen X-X, Zhao J, Dong Y-W, et al. Effects of inclina-tion angles on geometrical features of machined surface in five-axis milling. Int J Adv Manuf Tech 2013; 65: 1721–1733.
8. Chen X-X, Zhao J, Li Y-E, et al. Investigation on ball end milling of P20 die steel with cutter orientation.Int J Adv Manuf Tech2012; 59: 885–898.
9. Chen J-S, Huang Y-K and Chen M-S. A study of the surface scallop generating mechanism in the ball-end milling process. Int J Mach Tool Manu 2005; 45: 1077–1084.
10. Quinsat Y, Sabourin L and Lartigue C. Surface topogra-phy in ball end milling process: description of a 3D sur-face roughness parameter. J Mater Process Tech 2005; 195: 135–143.
11. Lavernhe S, Quinsat Y and Lartigue C. Model for the prediction of 3D surface topography in 5-axis milling.Int J Adv Manuf Tech2010; 51: 915–924.
12. Quinsat Y, Lavernhe S and Lartigue C. Characterization of 3D surface topography in 5-axis milling. Wear2011; 271: 590–595.
13. Lavernhe S, Quinsat Y, Lartigue C, et al. NC-simulation for the prediction of surface finish in 5-axis high-speed machining. In:3rd CIRP international conference on high performance cutting, Dublin, 12–13 June 2008, pp.387– 396. Paris: CIRP.
14. Brecher C and Lohse W. Evaluation of toolpath quality: user-assisted CAM for complex milling processes.CIRP J Manuf Sci Tech2013; 6: 233–245.
15. Balasubramaniam M, Sarma SE and Marciniak K. Colli-sion-free finishing toolpaths from visibility data.Comput Aided Design2003; 35: 359–374.
16. Jun CS, Cha K and Lee YS. Optimizing tool orientations for 5-axis machining by configuration-space search method.Comput Aided Design2003; 35: 549–566. 17. Zhu H, Liu J-N and Zhao J-F. HSM of vehicle front
bumper die based on UG CAM.Manuf Autom2011; 33: 43–45.
18. Fan J-H and Ball A. Flat-end cutter orientation on a quadric in five-axis machining. Comput Aided Design
2014; 53: 126–138.
19. Suresh K and Yang D. Constant scallop-height machin-ing of free-form surfaces.J Eng Ind: T ASME1994; 116: 253–259.
20. Ahmet C and Ali U. A novel iso-scallop tool-path gener-ation for efficient five-axis machining of free-form sur-faces.Int J Adv Manuf Tech2010; 51: 1083–1098. 21. Zou Q, Zhang J-Y, Deng B-L, et al. Iso-level tool path
planning for free-form surfaces. Comput Aided Design