Author’s Accepted Manuscript

Effect of cutting conditions and tool geometry on

process damping in machining
L.T. Tunc- , E. Budak

PII: S0890-6955(12)00020-X
DOI: doi:10.1016/j.ijmachtools.2012.01.009
Reference: MTM2734
www.elsevier.com/locate/ijmactool
To appear in: International Journal of Machine Tools & Manufacture

Received date: 12 September 2011

Revised date: 17 January 2012
Accepted date: 19 January 2012

Cite this article as: L.T. Tunc- and E. Budak, Effect of cutting conditions and tool
geometry on process damping in machining, International Journal of Machine Tools &
Manufacture, doi:10.1016/j.ijmachtools.2012.01.009

This is a PDF file of an unedited manuscript that has been accepted for publication. As a
service to our customers we are providing this early version of the manuscript. The
manuscript will undergo copyediting, typesetting, and review of the resulting galley proof
before it is published in its final citable form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply
to the journal pertain.
 EFFECT OF CUTTING CONDITIONS AND TOOL GEOMETRY
ON PROCESS DAMPING IN MACHINING

L.T. Tunç and E. Budak*

Manufacturing Research Lab., Sabanci University, Istanbul Turkey

*Corresponding author
+90 216 483 95 19
ebudak@sabanciuniv.edu
ABSTRACT
Process damping can be a significant source of enhanced stability in metal
cutting operations especially at low cutting speeds. However, it is usually
ignored in stability analysis since models and methods on prediction and
identification of process damping are limited. In this study, the effects of
cutting conditions and tool geometry on process stability in turning and
milling are investigated through simulations. The previously developed
models by the authors are used in simulations to demonstrate conditions for
enhanced process damping, and thus chatter stability. Some representative
cases are presented and verified by experimental data and conclusions are
derived. Cutting conditions and tool geometry for enhanced chatter stability
due to increased process damping are presented.

KEYWORDS: Process damping, Chatter stability, Tool Geometry, Cutting Conditions

Nomenclature
α: Slope of the generated surface
αij: Directional coefficients.
γ: Clearance angle
k: Modal stiffness
ζ: Modal damping ratio
ζt: Total damping ratio
Λ: Analytical eigenvalue of the end milling system
Λexp: Experimental eigenvalue of the end milling system
ΛR: Real part of the eigenvalue
c: Structural damping coefficient
cp: Process damping coefficient
c : Average specific process damping coefficient
Kd: Indentation constant
A: Vibration amplitude
U: Indentation area
Ed: Regeneration energy dissipated by average process damping coefficient
Ecd : Regeneration energy dissipated by average process damping coefficient
Pd: Damping power due to the average process damping coefficient.
Gij: Transfer function of the end milling system.
Kf: Cutting force coefficient in feed direction
Kt: Cutting force coefficient in tangential direction
Kr: Cutting force coefficient in radial direction
: Stability limit at low cutting speeds
: Stability limit at high cutting speeds

1. INTRODUCTION
In machining research and practice it is well known that process damping caused by
cutting tool and workpiece interaction may lead to increased stability especially at low
cutting speeds [1]-[6]. Tlusty and Ismail [3] showed that stability limit increases with
decreasing cutting speed due to higher process damping. However, Altintas and Weck [6]
stated that identification and modeling of process damping is one of the unsolved
problems in metal cutting. Process damping is ignored in most of the studies [7]-[9],
leading to significant errors in estimation of the stability limit at low cutting speeds.
There have been several attempts for identification and modeling of process damping. In
one of these attempts, Lin et al. [4] proposed a dynamic data system (DDS), where the
amount of process damping was identified from tool vibration measurements under
working conditions. Dynamic cutting force coefficients (DCFC) have used as one of the
most common approaches for modeling of process damping [2]. The problems such as
measurement accuracy and repeatability of tests, related to this approach are summarized
by Tlusty [2]. Sisson and Kegg [1], Wu [10], Elbestawi et al. [11], Shawky et al. [12],
and Lee et al. [13] showed that the indentation forces due to flank-wave contact against
tool vibration contribute to the process dynamics by increasing the overall damping of the
system. Sisson et al. [1] concluded that the edge hone on the tool, cutting speed and
clearance angle are the most important factors for process damping. In these studies, the
effect of process damping is only considered in time domain simulations. In this respect,
the indentation forces are related to the indentation volume between tool flank and
workpiece through a material constant [10] then, indentation forces are involved in the
time domain simulations. However, an analytical approach for estimation of process
damping coefficients has not been proposed in these studies. In a recent study by Altintas
et al. [14], dynamic cutting force coefficients are identified from series of dynamic
cutting tests, where the cutting tool is oscillated by a fast tool servo at the desired
frequency and amplitude. These tests have to be performed for different cutting and
dynamic conditions as well as tool geometries to demonstrate their effects on process
damping and enhanced stability conditions.

In this paper, the effects of tool geometry (hone radius, clearance angle, flank type),
cutting conditions (speed, radial cutting depth) and dynamic conditions (vibration
frequency and vibration amplitude) cutting conditions on process damping are
investigated using the previously developed models for process damping in orthogonal
cutting [15] and end milling [16] by the authors. This is done with very limited number of
tests using the analytical approach developed for identification and prediction of process
damping. The generated data for various cases are presented and used in stability analysis.
The paper is organized as follows; in the next section the approach for identification and
modeling of process damping is briefly explained. Then, the effects of cutting conditions
and tool geometry on process damping and stability limits are shown in Section 3 and
Section 4, respectively. The use of the models in selection of cutting conditions and tool
geometry for enhanced stability is demonstrated through representative examples in
Section 5. Finally, conclusive remarks and discussions are given.

2. IDENTIFICATION AND MODELING OF PROCESS DAMPING

Undulations develop on the machined surfaces due to tool vibration during dynamic
cutting as shown in Figure 1. For cutting depths above the stability limit the slope of the
undulated surface increases due to increased vibration amplitude. The flank face starts to
indent into the workpiece as the surface slope approaches to the clearance angle, γ.
Contact forces proportional to the indentation volume arise opposite to the tool vibration
resulting in process damping which stabilizes the process.

Figure 1: Indentation between the flank face and the undulated surface.

In this section, identification and modeling of process damping methods developed

earlier [15], [16] are summarized. Inverse solution of the stability formulation is used for
identification whereas energy analysis is used together with the indentation volume
calculations for modeling.

2.1. Inverse Stability Solution for Identification of Process Damping

In this study, process damping is identified from experimental data, i.e. chatter limits, as
proposed by Budak and Tunc [15], [16]. The process damping effect due to the
indentation forces on the flank face of the tool is represented as an average process
damping coefficient, cp, and treated in the formulation accordingly.

2.1.1. Turning
For a single degree of freedom (SDOF) orthogonal cutting system the absolute stability
limit is written in terms of the total transfer function and the cutting force coefficient as
follows [6]:
1
alim   (1)
2 Kf Re G t 
min
In the above equation, Re [Gt]min represents the minimum real part of the transfer function
with total damping, i.e. structural and process damping, which can be expressed in terms
of modal stiffness, k, and total damping ratio, t:
1
Re G t   (2)
min 4k ζ t
In equation (2) the total damping ratio can be expressed in terms of the total damping
coefficient, ct, modal stiffness and modal mass, m.
ct
t  (3)
2 km
Equation (1) is re-arranged [15] using equation (2) and equation (3) to express stability
limit in terms of the feed cutting force coefficient, Kf, the modal stiffness and damping
ratio, ζ [15] as follows:
2k ζ t hs 2kζ
alim   and alim   (4)
Kf Kf
After using equation (3) in equation (4) and some mathematical manipulations the
process damping coefficient is written in terms of the experimental stability limit, alim ,
which includes the effect of the process damping, and the theoretical stability limit, alim hs

which excludes the effect of process damping, as follows:

a 
c p  c lim
hs
 1 (5)
 alim 
Equation (5) provides a very practical way for identification of the process damping
coefficient by knowing the stability limit at any cutting speed. Moreover, the average
specific process damping coefficient per unit cutting depth can also be defined by
dividing cp to alimhs
, which is further used in estimation of stability limit through modeling.
 1 1 
c p  c hs   (6)
 lim
a a lim 

2.1.2. Milling

In end milling, the derivation of analytical stability limit at high cutting speeds is well
detailed and proposed by Altintas and Budak [8] as below; however the reader is directed
to reference [8] for detailed investigation:
2 R
alim  
NK t
1  2 
where
  R  I i  
1
2a 0
a1  a12  4a 0  (7)


a 0  Gxx G yy  xx  yy   xy  yx 
a1   xxGxx   yy G yy
The inverse solution of this analytical stability limit expression is used for identification
of process damping in end milling. It is obvious that the stability limit is related to the
total transfer functions of the system through the eigenvalue, Λ, which provides an
opportunity to identify the process damping from the experimental data [16]. The
eigenvalue of the end milling system with total damping, Λexp, is calculated by
rearranging equation (7) and substituting the experimental stability limit, alim, and chatter
frequency, ωc, [16]:
a NK 1  cos cT  alim NKt sin cT
 exp   lim t  i (8)
4 4
Then, the expression for the eigenvalue of the system, Λ, is equated to Λexp, and this
equality is solved iteratively [16] in order to identify the total damping ratios in -x and -y
directions,  xt and  yt , which are the summations of the structural and process damping in
the system.

2.2. Modeling of Process Damping

Process damping coefficients are specific to tool geometry and cutting conditions, which
require them to be identified for each case under interest. However, if the process
damping coefficients are related to the indentation volume and an indentation coefficient
such a requirement can be eliminated. In this section, energy balance analysis used for
modeling of process damping is briefly discussed. The details of the approach can be
found in references [15] and [16].

2.2.1. Calculation of specific process damping coefficients

The indentation force due to flank-workpiece contact is modeled as a function of the

indentation volume and the indentation constant, Kd. The energy analysis is used to
determine the indentation coefficient from the experimentally identified process damping
coefficient by equating the energy dissipated by them to each other [15]. The
regeneration energy dissipated by the average process damping coefficient, Ed , and the
regeneration energy dissipated by the indentation forces, Edc , can be written as follows
[15]:

Ed 

T
Pd dt 
 T
c p A2c2 cos 2 c t  dt  c p A2c (9)

Edc 

T
Pdc dt  K d Aac
 T
U  t  cos c t  dt

The average specific process damping coefficient is written in terms of the indentation
(10)

volume and indentation constant as follows [15]:


T
K
cp  d U (t ) cos x t  dt (11)
A
0
In equation (11) A is the vibration amplitude at the border of stability, ωx is the vibration
frequency, and T is the period of a single vibration wave. The same approach is used for
calculation of the average specific process damping coefficients for milling, as well.
However, the averaging is done for one revolution of the cutter as the damping force in
chip thickness direction is oriented in two orthogonal directions as the tool rotates [16].

2.2.2. Estimation of stability limit

The calculated specific process damping coefficients are multiplied with the average
contact length and used in estimation of the stability limit for the case under
consideration. In orthogonal cutting, where the contact length is equal to the cutting depth,
the average specific process damping is substituted back into equation (6) and the
stability limit is expressed in terms of modal damping coefficient, average specific
process damping coefficient and the experimental stability limit obtained at high cutting
speeds, by rearranging equation (6) as follows:
hs
calim
alim  (12)
c  c p alim
hs

However, in milling there is not any explicit relation between the stability limit and the
average specific process damping coefficients as two orthogonal degrees of freedom are
involved in process dynamics. The contact length is defined as the length of the flute in
cut with the material and it is averaged over one revolution of the tool. Therefore,
calculation of the average contact length requires the cutting depth to be known, which
can be calculated only if the process damping coefficients are known. This creates a
recursive relation between the stability limit and the process damping coefficients.
Therefore, an iterative procedure is followed to estimate the stability limit in milling as
detailed in reference [16].
 (a) Variation of indentation volume. (b) Cutting edge geometry. (c) Flank
length in milling.

Figure 2: Indentation volume and cutting edge geometry.

3. EFFECT OF CUTTING CONDITIONS AND TOOL GEOMETRY ON PROCESS

DAMPING

At low cutting speeds the stability limit directly depends on the average process damping
coefficient, which is a function of the indentation volume. The nonlinear relation between
the indentation volume and several parameters is illustrated in Figure 2a. In this section,
the effects of tool geometry and cutting conditions on average specific process damping
coefficients in turning and milling are discussed through representative simulations.
Cutting speed, vibration frequency and radial depth of cut are considered as the most
important parameters affecting process damping. In addition, hone radius, flank geometry
and clearance angle, i.e. tool geometry parameters, (see Figure 2b) also have significant
effect on process damping.

In general, milling tools are manufactured with either planar or cylindrical flank faces.
The former type is divided into two segments with clearance angles γ1, γ2 and lengths l1
and l2 as shown in Figure 3a. Therefore, the length of the clearance face (see Figure 2c) is
also an important parameter for planar flank geometries as it affects the indentation
volume. The cylindrical flank geometry is defined by an amount of drop, d1, at angle, Φ1,
measured from the radial axis of the cutting edge (see Figure 3b). The clearance between
the cutting edge and the workpiece is governed by the drop distance, whose effect is also
discussed.

(a) Planar (b) Cylindrical

Figure 3: Types and definitions of flank geometries.

Representative simulations and experimental data related to the effect of edge geometry
on process damping in orthogonal cutting are presented for the cases given in Table 1,
where indentation force coefficients of 10,000 and 70,000 N/mm3 [15] are used for
AL7075 and AISI105, respectively. Milling analysis is also performed for half immersion
down milling of Ti6Al4V for the cases given in Table 1, where the indentation force
coefficient of 30,000 N/mm3 [16] is used.
Table 1: Analysis cases for process damping.

Orthogonal Cutting
Hone Modal Parameters
o
Case Material Radius Clearance Angle( ) Stiffness Frequency Damping
(μm) (N/μm) (Hz) Ratio (%)
o
1 AISI1050 60 3,5,7 ( ) 15.9 1267 2.62
o
2 AISI1050 10,30,60 3( ) 16.7 1181 2.34
End Milling
Clearance Angle(o) / Frequency (Hz) Amplitude
Drop Distance(μm) x y x, y
3 10 [6,8,..,20](o) 2500 2500 10,10
4 10 [50,75, …,250](μm) 3000 3000 10,10
5 Ti6Al4V [10, 15,.,50] 20(o) 3000 3000 10,10
o
6 10 6( ) 3000 3000 10,10
o
7 10 10( ) / 110(μm) 2000 2000 15,15

In cases 1 and 2 orthogonal cutting is analyzed where the rest of the cases deal with end
milling. The effect of clearance angle on process damping in orthogonal cutting is
investigated in case 1, where hone radius is set to 60 microns. In case 2, clearance angle
is fixed at 3 degrees and comparison is performed for hone radii of 10, 30 and 60 microns.
In case 3, the effect of clearance angle on process damping in end milling is investigated,
where the hone radius is 10 microns. The effect of the drop distance for end mills with
cylindrical flank geometry is analyzed in case 4, where the hone radius is set at 10
microns and the drop is measured at 10 degrees. In case 5, the effect of hone radius is
considered, the cutting edge has clearance angle of 20 and 70 degrees at the first and
second segments, respectively. In case 6, different values for length of the flank face are
used in the simulations at several cutting speeds. The clearance angle at the first and
second segments is 6 and 50 degrees, respectively, and the hone radius is 10 microns. In
case 7, comparison between cutting edges with linear and cylindrical flank face is
performed, where the details are given in Table 1.
 (a) Indentation area (b) Absolute stability

Figure 4: Variation of absolute stability limit and indentation area with speed.

The wavelength decreases with decreasing cutting speed, causing surface slope to
increase which in turn increases the indentation volume. As a result the average process
damping coefficient increases with decreasing cutting speed. The absolute stability limit
decreases to ahs
lim at higher speeds where the effect of process damping diminishes.
However, the stability limit tends to increase towards alim at low cutting speeds due to
increased process damping (see Figure 4). In Figure 5 it is seen that the process damping
drastically decreases with increasing cutting speed, where the results of cases 1 and 2 are
plotted. A similar trend for end milling is also observed as shown in Figure 6, where the
results of cases 3, 4 and 5 are presented. In the rest of this section effects of other
parameters are investigated.

(a) Case 1, effect of clearance angle. (b) Case 2, effect of hone

radius.

Figure 5: Effect cutting speed on process damping in orthogonal cutting.

3.1. Effect of Clearance Angle, Drop Distance and Hone Radius

The clearance between the cutting edge and workpiece is governed by the clearance angle
or drop distance for the cutting edges having planar flank face and cylindrical flank face,
respectively. The clearance decreases with both decreasing clearance angle and drop
distance, resulting in increased indentation volume. In this respect, clearance angle and
drop distance affect the process damping in a way similar to the cutting speed. The
variation of average specific process damping coefficient with clearance angle in
orthogonal cutting, i.e. case 1, is plotted in Figure 5. The effect of clearance angle, i.e.
case 3, and drop distance, i.e. case 4, on process damping in end milling is presented in
Figure 6a and Figure 6b. It can be said that average specific process damping
substantially decreases with increasing clearance angle and drop distance.
In the literature it is shown that the existence of hone radius causes the workpiece
material to be extruded against the cutting tool [10], which causes the tool penetrate into
the workpiece. If the tool is sharp enough, i.e. the hone radius is very small this effect is
negligible whereas larger hone radius causes increased indentation volume. The effect of
hone radius on process damping is analyzed in case 2 and case 5, as presented Figure 5b
and Figure 6c, for orthogonal cutting and end milling, respectively. It can be concluded
that larger hone radius values result in increased specific process damping.

(a) Clearance angle (b) Drop distance

(c) Hone radius (d) Flank length

Figure 6: Effect of tool parameters on process damping in milling.

3.2. Effect of Flank Length and Type

For cutting tools having planar flank face either only the first or both segments may
penetrate into the workpiece depending on the ratio between the segment lengths and the
wavelength, as illustrated in Figure 2c. Therefore, under some circumstances process
damping may be affected by the length of the segments for planar flank geometry. In
Figure 6d, such a case is analyzed, where hone radius and clearance angle are 10 microns
and 6o, respectively. The average specific process damping coefficient increases with
segment length and then it saturates, where the saturation point varies with the cutting
speed.

The slope of the planar flank face is constant along the edge. However, the slope starts
from zero and continuously increases for cylindrical flank faces due to the circular cross
section. Therefore, the variation of indentation volume and hence process damping
directly depends on the flank type. For instance, when the surface slope is smaller than
the clearance angle the indentation volume is very small if the cutting tool has planar
flank face. However, even for very small surface slopes, the indentation volume may be
significant for tools with cylindrical flank face. Comparison between planar and
cylindrical flank face is given in Figure 7a, where it is seen that cylindrical flank face
results in higher process damping even at higher cutting speeds.

(a) Effect of flank type (b) Effect of vibration frequency

(c) Amplitude vs. damping force (d) Effect of vibration amplitude.

Figure 7: Effect of flank type, vibration frequency and amplitude.

3.3. Effect of Vibration Frequency and Amplitude

For a given cutting speed the surface slope increases with vibration frequency (see Figure
4a), resulting in more indentation volume and damping force. The increased damping
force dissipates more regeneration energy resulting in higher average process damping
coefficient. The effect of vibration frequency on the average specific process damping
coefficient in –x direction is shown in Figure 7b. Process damping increases as the
vibration frequency increases, especially at low cutting speeds.

The vibration amplitude is another important parameter for stability as the indentation
volume and the damping force are nonlinearly related to the vibration amplitude.
Therefore, limit cycle oscillations, which is a well-known phenomenon in nonlinear
dynamics, can be observed under certain circumstances [15]. In this respect, the stability
limit depends on the vibration amplitude, i.e. acceptable maximum stable oscillation
amplitude. In other words, at cutting depths of a1 and a2 the system may converge to
stability at vibration amplitudes of A1 and A2. For a half immersion-down milling of
Ti6Al4V the variation of the indentation (damping) force in –u direction and resulting
average process damping coefficients in –x and –y directions for different vibration
amplitudes are plotted in Figure 7c and Figure 7d, respectively, where the nonlinear
effect of the vibration amplitude on the damping force and the resulting process damping
are clearly seen.

4. EFFECTS OF TOOL GEOMETRY AND CUTTING CONDITIONS ON STABILITY

LIMIT UNDER THE EFFECT OF PROCESS DAMPING

Stability limit directly depends on the amount of process damping acting on the system,
which is affected by indentation forces. In this section, the variation of stability limit with
cutting edge geometry and cutting conditions is discussed through representative cases,
where the conditions and modal parameters are listed in Table 2 and Table 3, respectively.
Table 2: Cases for simulation of stability limit.

Case 1 2 3 4 5 6 7 8
Cutting Radial (%) 25 50 50 50 50 50 40
Condition Dow
Mode Down Down Down Down Down Down
s n
Plana Plana Plana Plana Plana Plana
Flank Type Cylin. Planar
r r r r r r
Tool Hone (μm) 20 10 20 40 20 20 15 60
o
Geometry Clearance(
10 10 10 10 NA 5 10 3
)
Drop (mm) NA NA NA NA 0.1 NA NA NA
AL707
Material Ti6Al4V
5
 Table 3: Modal parameters for simulation of stability limit.

Milling
Modal Natural Damping
Direction
Mass (kg) Frequency (Hz) Ratio (%)
Tool 1 X 0.225 3154 1.74
Y 0.228 3151 2.03
Tool 2 X 0.453 1921 2.75
Y 0.501 1896 1.58
Orthogonal Cutting
X 0.284, 0.851, 3.229 1195, 681, 343 1.49, 1.25, 1.04

(a) alim / ahslim for 25% and 50% immersion (b) Cases 1-4
 (c) Cases 3- 6 (d) Case 7
(e) Case 8

Figure 8: Simulation of stability limit.

The simulated stability limits for Tool 1 are presented in Figure 8. In classical stability
analysis [8], [9] it is well known that the increased radial cutting depth results in
decreased absolute stability. However, inclusion of the process damping effect into the
stability analysis may lead to different observations. The contact time between the cutting
tool and the workpiece per one tool rotation increases with the radial cutting depth,
resulting in more indentation volume. When case 1 and case 3 are compared the effect of
radial immersion on stability limit can be observed. The normalized stability limit for
quarter and half radial immersion cases is plotted in Figure 8a by dividing the stability
limit to the high speed absolute stability limit. It is observed that for half immersion case,
normalized stability limit is higher relative to quarter immersion case. Comparison of
cases 2, 3 and 4, in Figure 8b, illustrates the effect of hone radius on stability limit. It is
seen that the stability limit substantially increases with the hone radius. In addition, the
critical cutting speed at which the process damping becomes effective also depends on
the hone radius. For instance, process damping starts to be effective at 600 rpm (22
m/min), 900 rpm (34 m/min) and 1600 rpm (60 m/min) for hone radius of 10, 20 and 40
microns, respectively. Case 3 is simulated for stable vibration amplitude of 10 microns
and 15 microns in order to see the effect of vibration amplitude on the stability limit (see
Figure 8b). It is observed that if the stable vibration amplitude increases larger stability
region can be obtained. Comparison of cases 3, 5 and 6 is presented in Figure 8c, where
the effect of clearance angle and the type of flank geometry can be observed. Comparison
of cases 3 and 5 clearly illustrates the significant effect of flank type on the stability limit
at low cutting speeds. When the clearance angle and drop distance for planar and
cylindrical flank geometries lead to similar clearance geometry, cylindrical flank results
in higher stability border. As the results of the simulations performed for 5o (case 6) and
10o (case 3) of clearance angles it can be concluded that, clearance angle has a significant
effect on stability limits at low cutting speeds. As clearance angle decreases, stability
limit increases. In case 7, the simulations are compared with experiments for
representative verification. In Figure 8d, it is seen that the estimated stability limits agree
well with experimental results. In case 8 the relation between stability limit and natural
frequency is considered in orthogonal cutting, where hone radius is 60 microns and
clearance angle is 3o. According to the results given in Figure 8e, it can be said that
increased vibration frequency results in increased stability limit for the conditions
considered in this example.

(a) Stability limit without process damping (b) Stability limit with process damping

Figure 9: Effect of number of flutes.

Simulations for case 2 are repeated with the modal parameters of Tool 2 in Table 3 for
cutting flutes count of 2, 4 and 8 and 16 to discuss the effect of the number of cutting
flutes on chatter free material removal at low cutting speeds. Stability limits are
calculated both ignoring [8] and including [16] the effect of process damping, which are
plotted in Figure 9a and Figure 9b, respectively.

In Figure 9a it is observed that higher stable cutting depth can be achieved at higher
spindle speeds due to lobing effect when process damping is ignored. As the cutting
speed decreases, the stable cutting depth converges to different limits for 2, 4, 8 and 16
fluted cutting tools which implies a disadvantage of using multi fluted cutting tools in
terms of stable cutting depth. As the effect of process damping is included into the
analysis it is observed that using multi fluted cutting tools may become advantageous
especially at lower cutting speeds as higher stability limits can be achieved. This is
basically due to the fact that while the gain of the dynamic cutting system increases with
number of cutting flutes the amount of process damping acting on the system increases,
as well. It is also seen that as the cutting speed decreases cutting tools with 2, 4, 8 and 16
flutes converges to the same stability limit.

5. SELECTION OF CUTTING CONDITIONS AND TOOL GEOMETRY FOR

INCREASED CHATTER-FREE MATERIAL REMOVAL RATES

Process models are developed and used to reach increased productivity in industrial
applications. Selection of cutting conditions and tool geometry for some representative
examples is demonstrated and discussed for such a purpose in this section. The modal
parameters of Tool 2, given in Table 3, are used in the simulations.

(a) Deep Slotting (b) Pocketing operation

Figure 10: Illustration of the operations considered in the examples.

5.1. Example 1 – Selection of hone radius & cutting speed

In the first example, selection of cutting speed and hone radius value to achieve a desired
cutting time below 10 minutes for a specific operation is demonstrated. 20 mm of cutting
depth is required for deep slotting of Ti6Al4V as illustrated in Figure 10a, where the total
toolpath length is 4000 mm and 0.1 mm/rev per tooth feed rate is selected. The cutting
tool is a 12 mm diameter end mill with 4 flutes and the clearance angle on the flank face
is 10 degrees.

The stability limit diagrams for hone radius values of 10, 15 and 20 µm were determined
through simulations and are presented in Figure 11, where it is seen that the stability limit
increases with increasing hone radius, as emphasized in the previous section. It is found
that 20 mm of cutting depth can be reached at 680 rpm, 935 rpm and 1150 rpm for 10, 15
and 20 µm of hone radius, respectively. At 680 rpm 10 µm of hone radius provides a
feasible machining time of 15 minutes. With 15 µm of hone radius 10.6 minutes of
feasible machining time can be achieved at 940 rpm. . 20 µm of hone radius provides
machining time of 8.7 minutes at 1150 rpm. Thus, 20 µm of hone radius is selected as the
minimum hone radius value, providing the required cutting depth and machining time.
However, it should be noted that consideration of tool life would alter the solution as the
tool life behaviour of the cutting tool would depend on both the hone radius and the
cutting speed.
 (a) Example 1 (b) Example 2

Figure 11: Stability diagrams for example 1 and example 2.

5.2. Example 2 – Selection of cutting depth and step over

In the second example it is required to select both cutting depth and step over to achieve
minimum cutting time for roughing of an open pocket as illustrated in Figure 10b. The
length, l, the width w and, the depth, d, of the pocket are 260, 90, 25 mm, respectively.
The geometry and modal parameters of the cutting tool are same as the one used in the
first example, where the hone radius is 15 µm and the flute length is 30 mm.

Stability diagrams with the effect of process damping are generated for six different step
over values as plotted in Figure 11b. It is seen that the absolute stability at high cutting
speeds is higher for 25% of step over and it decreases as step over increases. However,
the step over value providing the highest stability limit changes with cutting speed as the
effect of process damping becomes significant, where the stability limit converges to the
flute length with decreasing cutting speed. Considering that the total machining time
depends on the total number of passes in axial and radial directions it is hard to identify
the most preferable step over and spindle speed directly from the stability diagrams. For
each step over value, the machining time is calculated for every spindle speed knowing
the corresponding stability limit, and plotted in Figure 12. It is found that 85% of step
over provides the minimum cutting time of 5.5 minutes at 1060 rpm and 25 mm cutting
depth. However, the tool life at the selected step over and cutting speed should meet the
calculated machining time. Therefore, this conclusion should be combined with the tool
life information, as well. It is expected that the tool life curve changes with radial cutting
depth as the cutting tool is subjected to longer heating time and shorter cooling time as
radial cutting depth increases. Inclusion of the tool life information may change the
conclusion on the feasible minimum machining time among different step over values.

(a) Machining time in example 2 (b) Stability diagram in example 3

Figure 12: Machining time and stability diagram for example 2 and example 3.

5.3. Example 3 – Selection of Flank Type

Selection of tool geometry is an important step in process design. It is previously stated

that the geometry of the tool flank face has a significant effect on process damping and
hence the stability limits. In this example, two alternative cutting tools having planar
flank face and cylindrical flank face are compared using simulations in terms of the
stability limits considering the effect of process damping. Both cutting tools are identical
in terms of diameter, helix angle and number of cutting flutes. The tool with planar flank
geometry has 10o of clearance angle, and the tool having cylindrical flank geometry has
0.09 mm of radial drop at 10o which corresponds to 10 degrees of clearance angle on the
average.

The stability diagram for each tool is plotted in Figure 12b, where it is observed that for
the same average clearance angle cylindrical flank results in higher stability limits even at
higher cutting speeds. So, it can be concluded that using cylindrical flank geometry
would be beneficial to reduce machining time.

6. CONCLUSIONS

In this paper, effects of cutting conditions and tool geometry on process damping are
investigated through simulations and experimental results. Previously developed models
for identification and modeling of process damping [15], [16] are used in the simulations,
and the following conclusive remarks are derived;

- When cutting speed decreases process damping and hence stability increases
substantially. The amount of stability increase, and the critical speed where it starts,
depends on the tool geometry and chatter frequency.
- Flank face geometry, clearance angle and drop distance have also strong effects on the
process damping and chatter stability at low cutting speeds. It is observed that cylindrical
flank geometry results in increased process damping when compared to planar flank
geometry even at higher cutting speeds. Under certain circumstances, length of the flank
face for planar flank geometry may become important for process damping.
- Higher hone radius causes more process damping and stability.
- Vibration frequency is another important parameter where higher vibration frequency
leads to increased indentation volume and process damping.
- Vibration amplitude is an important criterion for simulation of process damping. It is
observed that process damping coefficient nonlinearly changes with vibration amplitude
and behaves like a softening spring.
- In milling, increase in total contact length between the tool and the material results in
higher process damping, thus higher stability. Due to this effect, increased radial depth of
cut may not result in a significant decrease in the stability limits unlike the situation for
high cutting speeds where the effect of process damping diminishes. Similar effects
should be expected for increased number of cutting teeth and helix angle on milling tools.

Selection of cutting tool geometry and cutting conditions is also illustrated on

representative example cases. It is observed that inclusion of process damping into the
dynamics of cutting operations significantly affects the preferable cutting conditions and
tool geometry.

7. ACKNOWLEDGEMENTS
The authors acknowledge the support of Turkish National Science Foundation under
grant 108M340 and Pratt & Whitney Canada in completion of this study.

REFERENCES

[1] Sisson, T.R., Kegg R.L., An Explanation of Low-speed Chatter Effects, ASME Journal of
Engineering for Industry, 91 (1969) 558–951.
[2] Tlusty, J., Analysis of the State of Research in Cutting Dynamics, Annals of the CIRP, 27/2
(1978) 583-589.
[3] Tlusty, J., Ismail, F., Special Aspects of Chatter in Milling, ASME, J. Eng. Ind, 105 (1983) 24-
32.
[4] Lin, S.C., Devor, R.E., Kapoor, S.G., Sutherland J.W., A New Approach to Estimating the
Cutting Process Damping Under Working Conditions, Transactions of NAMRI/SME, (1990)
154-160.
[5] Kals H.J.J., On the Calculation of Stability Charts on the Basis of the Damping and Stiffness
of the Cutting Process, Annals of the CIRP, 19 (1971) 297-303.
[6] Altintas Y., Weck M., Chatter Stability in Metal Cutting and Grinding, Annals of the CIRP,
53/2 (2004) 619–642.
[7] Minis, I., Yanushevsky R., Tembo A, Hocken, R., Analysis of Linear and Nonlinear Chatter in
Milling, Annals of the CIRP 39 (1990) 459-462.
[8] Altintas Y., Budak E., Analytical Prediction of Stability Lobes in Milling, Annals of the CIRP,
44/1 (1995) 357-362.
[9] Budak E., Analytical Models for High Performance Milling. Part II: Process Dynamics and
Stability, International Journal of Machine Tool & Manufacture, 46 (2006) 1489-1499.
[10] Wu, D. W., A New Approach of Formulating the Transfer Function for Dynamic Cutting
Process, ASME Journal of Engineering for Industry 111 (1989) 37-47.
[11] Elbestawi, M.A., Ismail F., Du, R., Ullagaddi, B.C., Modeling Machining Dynamics Including
Damping in the Tool-Workpiece Interface, Trans. of ASME, Journal of Engineering for
Industry 116 (1994) 435-439.
[12] Shawky, A.M., Elbestawi, M.A., An Enhanced Dynamic Model in Turning Including the Effect
of Ploughing Forces, Trans. of ASME, Journal of Manufacturing Science and Engineering,
119 (1997) 10-20.
[13] Lee, B.Y., Tarng, Y.S., MA, S.C., Modeling of the Process Damping Force in Chatter
Vibration, International Journal of Machine Tool & Manufacture, 35/7 (1995) 951-962.
[14] Altintas, Y., Eynian, M., Onozuka, H., Identification of dynamic cutting force coefficients and
chatter stability with process damping, Annals of the CIRP, 57/1 (2008) 371-374.
[15] Budak, E., Tunc, L.T., A New Method for Identification and Modeling of Process Damping in
Machining, Trans. of the ASME, Journal of Manufacturing Science and Engineering,
131/051019 (2009) 1-10.
[16] Budak, E., Tunc, L.T., Identification and modeling of process damping in turning and milling
using a new approach, CIRP Annals-Manufacturing Technology, 59 (2010) 403-408.
 Table 4: Analysis cases for process damping.

Orthogonal Cutting
Hone Modal Parameters
Case Material Radius Clearance Angle(o) Stiffness Frequency Damping
(μm) (N/μm) (Hz) Ratio (%)
1 AISI1050 60 3,5,7 (o) 15.9 1267 2.62
2 AISI1050 10,30,60 3 (o) 16.7 1181 2.34
End Milling
Clearance Angle(o) / Frequency (Hz) Amplitude
Drop Distance(μm) x y x, y
3 10 [6,8,..,20](o) 2500 2500 10,10
4 10 [50,75, …,250](μm) 3000 3000 10,10
5 Ti6Al4V [10, 15,.,50] 20(o) 3000 3000 10,10
6 10 6(o) 3000 3000 10,10
o
7 10 10( ) / 110(μm) 2000 2000 15,15

Table 5: Cases for simulation of stability limit.

Case 1 2 3 4 5 6 7 8
Cutting Radial (%) 25 50 50 50 50 50 40
Condition Dow
s Mode Down Down Down Down Down Down
n
Plana Plana Plana Plana Plana Plana
Flank Type Cylin. Planar
r r r r r r
Tool Hone (μm) 20 10 20 40 20 20 15 60
Geometry Clearance(o
10 10 10 10 NA 5 10 3
)
Drop (mm) NA NA NA NA 0.1 NA NA NA
AL707
Material Ti6Al4V
5

Table 6: Modal parameters for simulation of stability limit.

Milling
Modal Natural Damping
Direction
Mass (kg) Frequency (Hz) Ratio (%)
Tool 1 X 0.225 3154 1.74
Y 0.228 3151 2.03
Tool 2 X 0.453 1921 2.75
Y 0.501 1896 1.58
Orthogonal Cutting
X 0.284, 0.851, 3.229 1195, 681, 343 1.49, 1.25, 1.04
 Figure 13: Indentation between the flank face and the undulated surface.

Figure 14: Indentation volume and cutting edge geometry.

(a) Variation of indentation volume. (b) Cutting edge geometry. (c) Flank length in
milling.

Figure 15: Types and definitions of flank geometries.

(a) Planar (b) Cylindrical

Figure 16: Variation of absolute stability limit and indentation area with speed.
(a) Indentation area (b) Absolute stability

Figure 17: Effect cutting speed on process damping in orthogonal cutting.

(a) Case 1, effect of clearance angle. (b) Case 2, effect of hone
radius.

Figure 18: Effect of tool parameters on process damping in milling.

(a) Clearance angle (b) Drop distance
(c) Hone radius (d) Flank length

Figure 19: Effect of flank type, vibration frequency and amplitude.

(a) Effect of flank type (b) Effect of vibration frequency
(c) Amplitude vs. damping force (d) Effect of vibration amplitude.

Figure 20: Simulation of stability limit.

(a) alim / ahslim for 25% and 50% immersion (b) Cases 1-4
(c) Cases 3- 6 (d) Case 7 (e) Case 8

Figure 21: Effect of number of flutes.

(a) Stability limit without process damping (b) Stability limit with process damping

Figure 22: Illustration of the operations considered in the examples.

(a) Deep Slotting (b) Pocketing operation

Figure 23: Stability diagrams for example 1 and example 2.

(a)Example 1 (b) Example 2

Figure 24: Machining time and stability diagram for example 2 and example 3.
(a) Machining time in example 2 (b) Stability diagram in example 3
 Effects of cutting conditions and tool geometry on process damping are examined
 Effective parameters on process damping are identified by modeling and experiments
 Process damping coefficients can be generalized considering the contact geometry
 Models can be used to design cutting tools for enhanced process damping
 High radial depth or number of teeth may not lower stability with process damping