FGB32503 – MACHINE TOOL DESIGN Azman Ismail

DYNAMIC OF MACHINE TOOLS

1.0 Introduction

The accuracy of a machined part depends on the precision motion delivered by a machine tool under static,
dynamic, and thermal loads. Two ultimate goals of the machining process are high accuracy and high productivity.
The machining force generated during the material removal process causes elastic deformations in the machine
tool system which includes the workpiece and the cutting tool. In addition, it may cause dynamic instability of the
machining process resulting in machining errors and a decrease in productivity. In order to meet requirements of
high accuracy and productivity, it is necessary to understand well the dynamics of machine tool.

2.0 Machine Tool Elastic System

Any cutting operation must inevitably be accompanied by certain process in motor and the friction process. The
cutting force deflects the machine tool, and therefore, for the purpose of dynamic analysis the machine tool can be
looked upon as an elastic system. The cutting process, the processes in the motor and the friction processes are
known as working processes. The working process and the machine tool elastic system (ES) constitute a closed-
loop system as shown in Fig 1.

Figure 1: blocked diagram of closed-loop machining system

The validity of interaction loops between the elastic system and working processes can be justified by the following
examples:
 The cutting force deflects the elastic system. This deflection changes the relative position between the
cutting tool and workpiece, resulting in a change of the undeformed chip thickness. The change in
undeformed chip thickness in its turn affect the value of the cutting force.
 The friction force component which does not coincide with the axis of the lead screw of the machine results
in a tilting of the carriage. As the carriage tilts, the friction force on the side faces of the slideways changes.
FGB32503 – MACHINE TOOL DESIGN Azman Ismail

In its turn, this change of the friction force on the side faces induces a change in the deformation due to
tilting of carriage.
 A change in speed affects the torque of the electric motor, especially in motors having a soft (dropping)
torque-rpm characteristic. The change in torque again results in a change of the speed of motion.

Besides the interactions between the elastic system and working process, the closed-loop system is acted upon by
two types of disturbances-f(t) acting on the machine tool elastic system and y1(t), y2(t), y3(t) acting on the working
processes.

Two features of the closed-loop system are of great practical significance. These are:

i. The interactions between the working processes and elastic system are simple, i.e. an action of one of the
working processes on the elastic system induces a reverse action of nature which is completely accounted
for by the single closing-loop. This feature simplifies the procedure of studying the dynamic behavior of
the closed-loop system as the latter can be transformed into an open-loop system by disconnecting one of
the loops.

ii. The working processes do not interact directly. A change in one of the working processes can affect the
other working processes only through the elastic system. This feature enables a multiple-loop system to
be transformed into a single-loop system by introducing the concept of equivalent elements.

The multiple-loop system is difficult to analyze, and therefore, for studying vibration problems connected with the
cutting process, the multiple-loop system of Fig. 1 can be replaced by a single-loop system (fig.2) by considering
the elastic system, process in the motor and friction processes as an equivalent elastic system (EES). This equivalent
elastic system represents a machine tool in which the motor is running and the appropriate relative motions
between the workpiece and cutting tool are being executed without actual machining. In other words, it represents
an idle running machine.

Figure 2: Block diagram of simplified single-loop machining system

3.0 General Procedure for Assessing Dynamic Stability of EES

The stability of the first linear approximation of a nonlinear system is an essential and sufficient condition for the
stability of a continuous nonlinear system. Therefore, despite the fact that the closed-loop system discussed above
is, in general, non-linear, its stability can be studied through the linearized system. A transient linear vibratory
process can in general be represented by the characteristic equation:

𝑦 = 𝐶1 𝑒 𝑧1𝑡 + 𝐶2 𝑒 𝑧2𝑡 + ⋯ + 𝐶𝑛 𝑒 𝑧𝑛𝑡 (equation 1)

Where:
𝑍1 , 𝑍2 , … 𝑍𝑛 = 𝑟𝑜𝑜𝑡𝑠 𝑜𝑓 𝑡ℎ𝑒 𝑓𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝑑𝑒𝑡𝑒𝑟𝑚𝑖𝑛𝑎𝑛𝑡
𝐶1 , 𝐶2 … 𝐶𝑛 = 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡𝑠 𝑤ℎ𝑖𝑐ℎ 𝑑𝑒𝑝𝑒𝑛𝑑 𝑢𝑝𝑜𝑛 𝑡ℎ𝑒 𝑖𝑛𝑖𝑡𝑖𝑎𝑙 𝑐𝑜𝑛𝑑𝑖𝑡𝑖𝑜𝑛𝑠

In general, the roots are complex numbers, i.e.

𝑍 = 𝛼 ± 𝑖𝛽 (equation 2)

The vibratory process is stable if the real part of the root is negative, i.e., α is negative. Vice versa, if α is positive,
the vibratory process is unstable. For polynomials of the third order or higher, it is difficult and time-consuming
to find the roots of the characteristic Eq. (3.1), therefore, the stability analysis of such a system is carried out with
the help of stability criteria, such as the Routh-Hurwitz algebraic criterion, Nyquist criterion, etc.

The Nyquist criterion is particularly suitable in studying the dynamic stability of the EES-cutting process closed-
loop system, as it provides information not only about the absolute system stability, but also on how the system
stability can be improved. An important point which must be kept in mind is that the Nyquist criterion is applicable
only when the elements comprising the system are individually stable. This implies that the Nyquist criterion
cannot be applied when the cutting process is unstable, e.g., while machining cast iron, machining with built-up
edge formation, etc.

With the help of the Nyquist criterion the stability of the closed-loop system is assessed through the amplitude-
phase-frequency plot of the open-loop system, which is obtained by disconnecting one of the loops. For instance,
when the y loop of the system shown in Fig. 2 is disconnected, the open-loop system of Fig. 3 is obtained. This
system consists of the EES and cutting proves (CP) elements joined in series. If we disconnect the second loop also,
we can separate the EES and cutting-process elements as shown in Fig. 1.4

Figure 3: block diagram of open-loop system

The amplitude-phase-frequency plot or simply the polar plot of the open-loop system is obtained as the vector
product of the polar plots of EES and cutting process elements. The general procedure of obtaining the polar plot
of an element and the method of obtaining the vector product of two polar plot are explained below.

Figure 4: Block diagram representation of a control system element

Consider an element (fig. 4) in which the input parameter is Xin and the output parameter is Xout. For obtaining the
polar plot of the element, Xin is varied in a frequency range which is of the practical interest for the given element.

At each frequency we record the output of the element. Let the input and output recorded be as shown in Fig. 1.6a.

From these records, we determine:

2. phase between the output and input

Figure 1.6: (a) Record of input and output parameters, (b) amplitude-frequency and phase frequency plots for
an element (c) polar plot of the element

After determining the values of A and Ø at a number of frequencies, we can draw amplitude-frequency and phase-
frequency plots as shown in fig. 1.6b. A combination of the two plots in the complex plane yield the polar plot
FGB32503 – MACHINE TOOL DESIGN Azman Ismail

(corresponding to frequency 𝝎), the radius vector represents the ratio of the amplitudes of Xout to Xin and the angle
which the radius vector makes with the real (Re) axis represents the phase difference between the two.
Now suppose the polar plots of two elements are known. For a certain frequency 𝝎, let the amplitude ration and
phase difference of plot 1 (fig. 1.7a) be A1 and Ø1, while those of plot 2 (fig. 1.7b) be A2 and Ø2. For the
corresponding point of frequency 𝝎 of the result open-loop system, the radius vector is obtained as the algebraic
product of A1 and A2, and the phase difference is obtained as the algebraic sum of Ø1 and Ø2. The graphic
construction and the resultant polar plot is shown in fig. 1.7c.

Figure 1.7: graphic method of obtaining the polar plot of the open-loop system from the polar plots of the
elements

Once the polar plot of the open-loop system is obtained, the stability of closed-loop system can be assessed by
applying Nyquist criterion which states:

1. if the polar plot of the open-loop system intersects the negative side of the real axis (-Re) to the right of
point (-1,0), the closed-loop system is stable,
2. if the polar plot of the open-loop system intersects the –Re axis to the left of point (-1,0), the closed-loop
system is unstable, and
3. If the polar plot of the open-loop system passes through the point (-1,0), the closed-loop system is on the
threshold of stability.

4.0 Stability Analysis

The basic procedure of stability has already been explained. Now with the knowledge of the transfer functions and
dynamic characteristic of the machine tool elastic system and cutting process, we will take up stability analysis of
EES-cutting process closed-loop systems involving different model of EES under various cutting condition. It was
explained that instability of the system manifests itself in the form of self-excited or chatter vibrations. As the
cutting coefficient ᴷs is directly proportional to the undeformed chip width, the chances of the chatter vibrations
in a system increase with an increase of width. In actual practice, therefore, the problem of stability analysis is
often reduced to determining the maximum limiting undeformed chip width ᵇlim at which chatter will not occur.
It is cautioned that in the subsequent discussion cutting is supposed to be associated with machining of smooth
surface with a single point tool, unless otherwise mentioned.
FGB32503 – MACHINE TOOL DESIGN Azman Ismail

4.0.1 single-degree of freedom system – static cutting process characteristic

The polar plot of the open-loop system which is obtained by vector multiplication of the polar plots of the EES and
cutting process will, in this case, be nothing but the EES polar plot multiplied by a magnification factor 𝑘𝑠 if an
arbitrary point of frequency ὼ of the EES polar plot is defined by amplitude A and phase angle ∅, the corresponding
point of frequency ὼ on the open-loop system polar plot will be defined by amplitude A • 𝑘𝑠 and phase angle ∅.
Thus, if the EES of the machine tool can be represented as a single degree of freedom vibratory system and the
cutting conditions are such that they can be represented by a static characteristic, the closed-loop system is always
stable.

Fig. 1.8 (a) Polar plot of single degree of freedom EES, (b) Polar plot of open-loop system

4.1.2 Many degree of freedom system in normal coordinates-static cutting process characteristic

The polar plot of the EES represented as a many degree of freedom system in normal coordinates is shown in Fig.
and the polar plot of the open-loop system using the static cutting process characteristic is shown in Fig. . it is
evident from Fig. that this closed-loop system is also always stable.

1.9 (a) Polar plot of many degrees of freedom EES in normal coordinates, (b) Polar plot of open-loop system

5.0 Forced Vibration of Machine Tools

Forced vibrations are caused by disturbances acting on EES and the cutting process. These vibrations effect the
machinery accuracy, productivity and life of the cutting tool. The effect on productivity is dictated by the
interaction of forced vibrations with chatter and the resulting change in the value of ᵇlim. In this section we will
mainly deal with this interaction as it is essential for proper assessment of the dynamic stability of the EES-cutting
process system. The effect of forced vibration on machining accuracy will also be discussed.
FGB32503 – MACHINE TOOL DESIGN Azman Ismail

The effect of forced vibrations affect the tool life through a change of the effective cutting speed. Let the
amplitude of the vibratory displacement be A. The amplitude of the velocity of vibration will, therefore, be A • ὼ,
where ὼ is the circular frequency of vibrations. If the nominal cutting speed is v, the effective cutting speed will
vary during each half cycle of vibration. Fig. 1.10 shows the vibrations record of a face milling operation. It may be
pointed out that the frequencies 400Hz, 10kHz, and 15kHz are natural frequencies of the spindle unit, milling
cutter blade and cutting bit respectively.

Fig. 1.10 Vibration Spectrum of the Face Milling Operation

6.0 Forced Vibration, Due to Disturbance of the Cutting Process

The sources of disturbance acting on the cutting process include the following:

1. Variable undeformed chip thickness due to inherent characteristic of certain machining operations,
such as milling and broaching.
2. Variable undeformed chip thickness due to machining of a wave formed during a preceding pass or cut.
3. Periodic variation of the undeformed chip thickness while machining an eccentric workpiece
4. Variable undeformed chip thickness at the time when the cutting tool enters the workpiece or comes
out of it
5. Change in the cutting speed during operations such as facing
6. Change in kinematic cutting angles with change in the direction of feed.

7.0 Conclusion

High productivity and quality in machining strongly depend on the process dynamics and stability. Rigidity of
machine tools and selection of process parameters are two main factors in dynamic behavior of cutting operations.
In this paper, methods that can be used for analysis and modeling of machine tool structures and cutting process
stability are reviewed. These methods can be used to analyze as well as improve the dynamic behavior of
machining processes. Perhaps the most important application of these methods is the analysis and suppression of
the chatter which is the most critical vibration type for machining systems. The analytical methods and
suppression techniques presented here can be used to improve stability and productivity of the machining
operations. The selected and presented methods have been applied, and are currently being used in industry.