# Problems During Milling and Roughness Registration of Free-form Surfaces

# Problems During Milling and Roughness Registration of Free-form Surfaces

# M Rybicki1

Published under licence by IOP Publishing Ltd

, ,

Abstract:

Finishing machining of free-form surfaces is usually performed using ball mills. Constant rotational speed n and feed rate per wedge in axis of the mill *fz *are applied most often for the whole surface. It causes long machining time and theoretical machined surface roughness *Rzt *heterogeneity. In the paper we propose a milling method with variable parameters n and *fz *which gives greater productivity and more homogenous roughness of free-form surfaces. Based on roughness parameter measurement of the samples, many problems have been enumerated: necessity of the surface curvature pre-filtering from primary profile, necessity of 2D roughness assessment in two perpendicular directions, requirement for a high vertical measurement range and special long tips in contact profilometers most often, roughness variability and necessity of milling technology knowledge, to find a region with maximal roughness.

I. Introduction

Free-form machined surfaces have more and more participation in machining. Applying of the HSM strategy in finish machining has allowed e.g. to decrease machining time and to improve of surface quality. For the sake of small passes interval and small federate, time of finish machining is still amounted to almost or over 50% of total machining time of the parts, and surfaces can required lapping in some places.

Constant allowance q0 for finish machining are applied. Nowadays the worle machined surface are milled with constant freedrate vf in axis of the mill and constant rotational speed n most often (in whole machining cycle). It causes increasing of such quantities likevf0, ψ, hm(hzsr) on concave surface and decreasing of the quantities on convex surface (figure. 1). When milling free-form surface with constant rotational speed n by ball mill cutting speed vc is changed additionally due to change of effective diameter Def. I causes local changes of tool’s load (form errors) and kinematic-geometrical mapping (roughness) on milling length. Besides, constant values of the vf and n are selected for the most difficult to cutting region of the surface, what causes ineffective machining most time (in easier regions) and long machining time.

Figure 1. Kinematics and dimensions of cut a) as well as impact of machined surface tilt angle α on

effective ball mill diameter Def with recommended speeds n and vf b) during milling of curvilinear

profile

During milling by ball mill roughness can be considered in feed and pick direction. In finish

machining tool path assuring constant theoretical roughness in pick direction are applied mainly

[5,19]. Many sources suggest that in order to keep constant theoretical roughness feedrate per

revolution f or per wedge fz should be equal to distance between paths (pick fp) in finish machining. It

explained by possibility to keep constant 3D roughness and high material removal rate, assuming that

ball maps in machined material. In reality wedge path is epicyclical and when p=f component of

roughness coming from feedrate is 3 to 4 times higher (depending on the mill diameter d and angle α

of its axis til relative to machined surface) that component coming from pick [2].

In the paper is proposed original method of simultaneous controlling of feedrate vf and rotational speed

n on milling length giving more homogenous roughness and shorter machining time.

Strategy of changeable rotational speed, in order to assure constant cutting speed vc=const., is known

in turning. It causes shorter machining time and improvement of machined surface roughness. But the

strategy is not applied during milling by ball mill. Producers of cutting tools recommend cutting speed

vc in narrow range (or rotational speed n when diameter Def of the mill is known). For various angles

of machined surface tilt α various rotational speeds n are recommended usually. During milling with

constant rotational speed n=const its smallest recommended value are selected. It causes underrating

of the n value in regions it should be higher. And then feedrate vfn is also small and machining time

ts1/vf is long in a given toll path. Small cutting speed (for α=0°) can be avoided by numerical control

of machine tool’s rotational axes but during milling of deep cavities it is possible in very narrow

range.

Also during milling with variable feedrate vf is connected with considerable time saving, what results

from increasing of its value on small curvature segments of milling path. If concave surface has the

highest roughness Rzmax and force Fmax, feedrate vf in axis of the mill can be increased in small

curvature regions to obtain the same roughness and force as in the concave one. Then maximal

roughness and force will not be higher but machining time will be shorter. Feedrate can be changed in

way assuring constant deflection of the mill [11,18] or roughness [3], what can minimize or eliminate

manual lapping of cavities in dies and moulds.

The most often applied in literature method of the feedrate changes gives constant feedrate vf0 of

wedges, which is a real rate of work material forming. In order to obtain the vf0 value feedrate vf pr

should be programmed in axis, often given in in form of relationship (1) in which vf is feedrate in axis

when tool path is straight. Quantity R0 is path radius of the mill’s axis, which can be write as R0=R+r

for convex surface and R0=R-r for concave one.

2. Range, conditions and technique of research

The experimental research will be carried out onto the Cincinnati VMC-500 Arrow milling machine.

The Mastercam X3 and option Surface Finish Blend was used for making NC program. In the

machining strategy pick is kept constant along curvilinear surface if possible.

Machining allowance q0=0,06mm, rotational speed n=8000rev/min and constant feedrate in axis of the

mill fz=0,06mm/wedge (variable vf0) as well as pick fp=0,12mm equal to feedrate per revolution f=fz·z

were applied.

Sample was made from hardened hot work tool steel 55NiCrMoV of hardness 57 HRC and milled

down without cooling.

Ball mill made of Fraisa company with designation D15140.300, diameter d=6mm, number of wedges

z=2, angles λs=30° and = -10 was used.

Roughness measurements were done on to the HOMMEL roundscan 535 parallel to the vf vector.

In earlier measurements on the T500 profilometer made of Hommewerke measurements were not

possible in all places due to too short measuring tip and vertical range of the device. Traverse length

lt=4,8mm and double filtration was used. First radius of machined surface R=9mm was filtered from

primary profile. Measuremet tests without the filtration had gave another values of roughness

parameters Next filter ISO 11562(M1) with wavelength of cut-off λc=0,25 mm was used. Value of the

c>f was applied intentionally to assessment kinematic-geometrical mapping of the mills wedges.

3. Results and analysis of research

3.1. Machined surface roughness

It can be concluded from figure 1 that during milling with constant feedrate in mill’s axis roughness

should be the highest on concave surface due to the highest value of the vf0. Besides the feedrate value,

roughness is also resulted from the same machined surface curvature [22], chatter etc. Roughness is

the highest when milling surface perpendicular to axis of the mill, when effective diameter Def and

cutting speed vc are small. Theoretical roughness height shown at the figure, assumes that in work

material maps a ball and do not gives consideration to tilt of axis of the mill.

Theoretical space RSmt between roughness elements was assumed to be equal to feed per revolution of

the mill’s wedges (figure 4). From the figure follows that for high α angle micro irregularities arises at

space equal to the f0, like during cylindrical milling, and roughness profile has no visible mapping

space equal to feed per wedge fz0. Frequency spectrum of roughness profile when milling with angle

α=0° is very scattered.

14th International Conference on Metrology and Properties of Engineering Surfaces IOP Publishing

Journal of Physics: Conference Series 483 (2014) 012007 doi:10.1088/1742-6596/483/1/012007

3.2. Proposal of variable cutting parameters applying

For ball mill roughness and form errors are affected by angle α of the mill’s axis tilt relative to

machined surface [3,7,11,18] and local radius R of the surface. So it has been proposed that changes of

machining parameters vf and n took place when the α and R values had change. Changeable

(depending on machined surface geometry) machining parameters vf and n were calculated from

equations (3) and (4). Equation (3) consist feed per wedge fzpr calculated by formula (1), but it can also

be used formula from [3] which assure constant three dimensional theoretical roughness height Rmax.

Equation (4) consists diameter Def connected with angle α.

At figure 5 is shown theoretical example how many percent of time can be save thanks to changeable

feedrate and rotational speed in various places of machined surface applying. Except time saving more

homogenous roughness is expected. It was assumed that machining time during milling with constant

parameters is a 100%. For analysis was assumed machined surface model with radius R=13mm, length

of straight feed intervals in in single path 180mm and maximal angle α=90° as well as diameter

d=10mm and wedges number z=2. Recommended machining parameters were set fz=0,035mm/wedge

and vc=500m/min. It was assumed that maximal rotational speed of machine tool nmax=40000rev/min

and maximal feedrate vfmax=5000mm/min.

It is proposed modification of NC program for machining with constant parameters vf and n. Blocs of

the program can be divided in places where exact changes of radius R of angle α has occurred, and

new parameters calculated from equations (3) and (4) can be loaded. The places can be marked basis

on coordinates X, Y and Z of characteristic tool point and vector defining contact place of the tool

with workpiece.

4. Conclusions

Basis on roughness parameters measurement of sample, many problems have been enumerated, as:

necessity of the surface curvature pre-filtering from primary profile, necessity of 2D roughness

assessment in two perpendicular directions, required high vertical measurement range and special long

tips in contact profilometers most often, roughness variability and necessity of milling technology

knowledge to find region with maximal roughness. The regions are concave surfaces and surfaces

cutting by edges of the mill near of its axis.

During ball milling roughness is affected by angle α of the mill’s axis tilt relative to machined surface

and local radius R of the surface in feed direction. For high angle α micro irregularities arises at space

RSm equal to feedrate per revolution f0, like during cylindrical milling

By applying of variable cutting parameters on curvilinear tool path machining time can be reduced and

constant roughness or constant load can be obtained.

Thanks to Division of Metrology and Measurement Systems of Poznan University of Technology for

making measurements of machined surface texture parameters.

[1] Bouzakis K D, Aichouh P and Efstathiou K 2003 Determination of the chip geometry, cutting

force and roughness in free form surfaces finishing milling, with ball end tools International

Journal of Machine Tools & Manufacture 43 499

[2] Chen J S, Huang Y K and Chen M S 2005 A study of the surface scallop generating mechanism

in the ball-end milling process International Journal of Machine Tools and Manufacture 45

1077

[3] Choi I H, Yang M Y, Hong W P and Jung T S 2005 Curve interpolation with variable feedrate

for surface requirement International Journal of Advanced Manufacturing Technology 25 325

[4] Chu C N, Kim S Y and Lee J M 1997 Feed-rate optimization of ball end milling considering

local shape features Annals of the CIRP 46 433

[5] Docker S and Dickin P J 2007 Modern machining techniques for mouldmaking. 4th

International Conference and Exhibition on Design and Production of MACHINES and

DIES/MOULDS, Cesme, TURKEY, 21-23/6/2007

[6] Heo E Y, Merdol D and Altintas Y 2010 High speed pocketing strategy CIRP Journal of

Manufacturing Science and Technology 3 1

[7] Kecelj B, Kopač J, Kampuš Z and Kuzman K 2004 Speciality of HSC in manufacturing of

forging dies Journal of Materials Processing Technology 157-158 536

[8] Kim G M, Chub C N 2004 Mean cutting force prediction in ball-end milling using force map

method Journal of Materials Processing Technology 146 303

[9] Kim K K, Kang M C, Kim J S, Jung Y H and Kim N K 2002 A study on the precision

machinability of ball end milling by cutting speed optimization Journal of Materials Processing

Technology 130-131 357

0

30

60

90

120

0 10 20 30 40 50 60 70 80 90

ts [%]

a [o]

n,vf=const.

n,vf=var.

time saving: 18,66%

14th International Conference on Metrology and Properties of Engineering Surfaces IOP Publishing

Journal of Physics: Conference Series 483 (2014) 012007 doi:10.1088/1742-6596/483/1/012007

[10] Kloypayan J and Lee Y S 2002 Material engagement analysis of different end mills for adaptive feedrate control in milling processes Computers in Industry 47 55

[11] Lin J C and Tai C C 1999 Accuracy optimization for mould surface profile milling The International Journal of Advanced Manufacturing Technology 15 15

[12] Lotfi B, Zhong Z W and Khoo L P 2009 Variable feed rates and variable machine forces for a constant material removal rate and constant cutting force along Pythagorean-hodograph curves International Journal of Advanced Manufacturing Technology 40 171

[13] Pateloup V, Duc E and Ray P 2000 Corner optimization for pocket machining. International Journal of Machine Tools & Manufacture 44 1343

[14] Rybicki M 2012 Wybrane problemy dokładnego kształtowania powierzchni swobodnych na obrabiarkach CNC Mechanik 1 31 (in CD)

[15] Schuett T J 2002 Look-ahead for faster machining. Rapid acceleration/deceleration key to high-speed machining Cutting Tool Engineering 54

[16] Sencer B, Altintas Y and Croft E 2008 Feed optimization for five axis CNC machine tools with drive constraints International Journal of Machine Tools & Manufacture 48 733

[17] Sun G and Wright P 2005 Simulation-based cutting parameter selection for ball end milling Journal of Manufacturing Systems 24 352

[18] Sun Y, Jia Z, Ren F and Guo D 2008 Adaptive feedrate scheduling for NC machining along curvilinear paths with improved kinematic and geometric properties International Journal of Advanced Manufacturing Technology 36 60

[19] Toh C K 2005 Design, evaluation and optimization of cutter path strategies when high speed machining hardened mould and die materials Materials & Design 26 517

[20] Weinert K, Enselmann A, Friedhoff J 1997 Milling simulation for process optimization in the field of die and mould manufacturing. Annals of the CIRP 46 325

[21] Choi I H, Yang M Y, Hong W P and Jung T S 2005 Curve interpolation with variable feedrate for surface requirement Int J Adv Manuf Technol 25 325

[22] Lin R and Koren Y 1996 Efficient Tool-Path Planning for Machining Free-Form Surfaces Journal of Engineering for Industry 118 20

[23] Quinstat Y, Sabourin L and Lartigue C 2008 Surface topography in ball end milling process: Description of a 3D surface roughness parameter Journal of Materislas Processing Technology 195 135

[24] Jung T S, Yang M Y and Lee K J 2005 A new approach to analysing machined surfaces by ball-end milling, part I: Formulation of characteristic lines of cut reminder Int J Adv Manuf Technol 25 833

[25] Chen J S, Huang Y K and Chen M S Feedrate optimization and tool profile modification for the high-efficiency ball-end milling process International Journal of Machine tools & Manufacture 45 1070

[26] Suresh K and