Journal of International Council on Electrical Engineering

ISSN: (Print) 2234-8972 (Online) Journal homepage: www.tandfonline.com/journals/tjee20

Reliability Improvement of Machine Tool changing

Servo Motor

Baek-Ju Sung & Jong-Bae Lee

To cite this article: Baek-Ju Sung & Jong-Bae Lee (2011) Reliability Improvement of Machine
Tool changing Servo Motor, Journal of International Council on Electrical Engineering, 1:1,
28-32, DOI: 10.5370/JICEE.2011.1.1.028

To link to this article: https://doi.org/10.5370/JICEE.2011.1.1.028

Published online: 10 Sep 2014.

Submit your article to this journal

Article views: 1040

View related articles

Citing articles: 2 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=tjee20
28 Journal of International Council on Electrical Engineering Vol. 1, No. 1, pp. 28~32, 2011

Reliability Improvement of Machine Tool changing Servo Motor

Baek-Ju Sung* and Jong-Bae Lee†

Abstract - First of all, it is the most important to know the cause and effect of major failures to
increase the reliability of servo motor. All reliability parameters can be achieved after analysis of
them and experiments based on the analysis data. So, we firstly made out the failure analysis data
of servo motor such as FMMA, FMECA, FTA, and 2-stage QFD considering the reference
materials, field failure data, and the opinion of experts of servo motor. And the test items were
decided which are needed to evaluate the performance characteristics of servo motor as analysis
results of failure analysis data. And then, the major reliability parameters are deduced by
experiments and MINITAB analysis. In addition, according to prove a major failure mode of servo
motor by above analysis results, we produced FEA (Finite Element Analysis) model about shape
of stator and rotor of servo motor to find out the reducing method of cogging torque. This paper
presents the useful major reliability parameters, detailed failure mode analysis data, experimental
results using 3 samples, and a cogging torque reducing method of servo motor, and also verified
the validity of the deduced parameters and analysis data.

Keywords: Reliability, Servo motor, Shape parameter, Scale parameter, Weibull, Finite Element
Analysis

1. Introduction servo motor by consultant of field users, reference materials,

and experts in this field. This step explains the components
The servo motor is a core component for mechanism of servo motor and their each functions and present failure
control of electrical machinery equipments. And it is and failure mechanisms of each components. Table 1
requested for higher product reliability is required for describes this step.
environment-friendly use in the various operational
situations. The servo motor reliability is closely related with Table 1. Failure modes and mechanisms analysis
the reliability of overall system which is used it. This paper
presents the failure analysis for machine tool changing Primary
Function Failure modes Failure mechanisms
components
servo motor and reliability parameters for its reliability test,
Wear 1-1 Spalling
and a cogging torque reducing method. The FMMA Bearing
Reduce friction
torque Corrosion,
(Failure Modes Analysis), FMECA (Failure Modes, Effects, Deformation 1-2
fatigue
and Criticality Analysis), FTA (Fault Tree Analysis), and 2- Not proper Disconnection,
Produce 2-1
stage QFD (Quality Function Deployment) are introduced Stator windings motion short
magnetic field
as the result of failure analysis [1]. The major reliability Unusual noise 2-2 External impact
parameters are produced by life test and MINITAB Non-uniform
Not proper
3-1 magnetic flux
software analysis. And, a cogging torque reducing method Rotor Rotary motion motion
density
is proposed, which uses a new FEA model for detailed Unusual noise 3-2 External impact
description of servo motor.
Twist 4-1 Fatigue
Shaft Transmit torque Plastic
Bending 4-2
deformation
2. Failure Analysis of Servo Motor Frame
Structure for
Unusual noise 5-1
Loosening,
motor deformation
2.1 FMMA (Failure Modes and Mechanisms Analysis)
2.2 FMECA (Failure Modes, Effects, and Criticality
This step is to find out and analyze the major failures of Analysis)
† Corresponding Author: Dept. of Reliability Research Center,
Korea Institute of Machinery & Materials ([email protected])
This step explains the major failures of primary components
* Dept. of Reliability Research Center, Korea Institute of Machinery
& Materials ([email protected]) of servo motor and influence of their effects on the motor,
Received: May 1, 2010; Accepted: October 16, 2010 and criticality for each items. Especially, criticality analysis
 Baek-Ju Sung and Jong-Bae Lee 29

is expressed the severity and failure frequency according to important level is given to the each failure modes. The
the procedures of MIL-STD-882D for showing the severity result is shown in the Table 3. By the results of the first
distribution in viewpoint of quality [2]. stage of QFD, the main test items are decided in the second
stage of QFD. The main test items require the caution for
Table 2. Failure modes, effects, and criticality analysis decision because the performance of servo motor is directly
evaluated by them. Table 4 represents the analysis results.
Primary Failure Failure Criticality
components causes effects Frequency Severity Criticality
Table 3. 2-stage QFD level 1
Reduce Wear High High 9
Bearing friction
torque Deformation Medium High 7 Primary Stator
Bearing Rotor Shaft
Not proper components windings
Produce

Unusual noise

Unusual noise
Stator Low High 5 Failure mode

Deformation
motion

Not proper

Not proper
magnetic

Bending
windings

motion

motion

Twist
Wear
field Unusual noise Low Medium 3
Not proper
Rotary Low High 5
Rotor motion Requirement
motion
Unusual noise Low Medium 3 Encoder error
● ● ● ● ●
Transmit Twist Low Medium 3 measurement
Shaft Line resistance ● ● ● ● ● ● ●
torque Bending Low Medium 3
Line inductance ● ● ● ● ● ● ●
Structure
Frame Unusual noise Low Medium 3 Counter electromotive
for motor ◎ ◎ ● ● ◎ ●
force constant
Cogging torque ● ● ● ▲ ◎ ● ●
2.3 FTA (Fault Tree Analysis) Axial vibration ◎ ◎ ▲ ● ● ● ◎
Low temperature
▲ ▲
operation
TA expresses the cause and generation process of a High temperature
failure as the type of tree branch in order to reduce the ● ▲ ● ● ◎ ●
operation
examination time of failure cause. Fig. 1 [3] shows a FTA Life test ◎ ◎ ◎ ● ◎ ● ●
of servo motor. The contents marked circle are causes of the Score 31 29 23 17 32 19 5 24
Notice importance : ◎very important (5), ●important (3), ▲normal (1)
failure about each primary components and contents
marked rectangular are phenomena of failure which
Table 4. 2-stage QFD level 2-1
generated by causes of the failure and the primary
components happened fault. The causes are combined by
Test items
OR Gate in case of component having multiple causes. We
Primary components

Counter electromotive

can know that failure was occurred by the shock, fatigue,

Failure mode

High temperature
Low temperature
Line inductance

Cogging torque
Axial vibration
Line resistance

force constant

overload, and wear, and finally these partial faults in

Encoder error
measurement
Score

Life test
primary components make a servo motor out order.

● ● ● ◎ ▲ ● ◎
Bearing

Wear 31

Deformation 29 ● ● ● ◎ ▲ ● ◎

Not proper
windings

23 ● ◎ ◎ ◎ ● ▲ ●
Stator

motion

Unusual noise 17 ◎ ● ◎ ● ● ●
Not proper
32 ● ◎ ◎ ● ◎ ▲ ●
Rotor

motion
Unusual noise 19 ● ◎ ● ● ● ●
Fig. 1. Fault Tree Analysis. Twist 5 ● ● ● ▲ ▲
Shaft

2.4 2-stage QFD (Quality Function Deployment) Bending 24 ▲ ● ● ◎ ● ●

345

417
383

620

604
543

312
650
60

Test effectiveness score

The first stage of QFD expresses the required function at and rank
each primary components and failure modes corresponding 7 5 6 2 3 4 9 8 1
to them. At that time, the proper score which is assigned by Notice importance : ◎very important (5), ●important (3), ▲normal (1)
30 Reliability Improvement of Machine Tool changing Servo Motor

3. MINITAB Analysis for Decision of Reliability goodness-of-fit examination about four kind of typical life
Parameter distribution such as Weibull distribution, Lognormal
distribution, 3-parameter Weibull distribution, Normal
3.1 Life Data distribution. The result is shown in Fig. 4.

Some life test results are needed to estimate the life

distribution or to deduce parameters of life distribution by
statistical analysis for them. In this paper, we analyzed the
life trend of servo motor by use of 4 unit linear model
which are selected in accordance with proper degradation
measured period, as it were, elapsed life test time-0 hour,
80 hours, 170 hours, and 238 hours. And the predicted life
time for these degradation data can be achieved by ALTA
Pro 7 software. Fig. 3 is the life prediction graph for
degradation data, and table 5 shows the predicted life time.

Fig. 4. Goodness-of-fit analysis of life data distribution.

The correlation coefficient is used to express the reference

of distribution’s goodness-of-fit, and the more value of
correlation coefficient is nearly 1.0, the more goodness-of-fit
is suitable to the distribution [4]. By analysis result of Fig. 4,
the correlation coefficient value of Weibull distribution is
Fig. 2. Servo motor and drives for test.
0.980, Lognormal distribution is 0.993, 3-parameter Weibull
distribution is 1.000 and Normal distribution is 0.991.
Therefore, we can know that the correlation coefficient
value of 3-parameter Weibull distribution is the most
suitable as servo motor’s life distribution.

3.3 Deduction of Major Reliability Parameters

The major reliability parameters can be predicted by

marking of the life data of 3-parameter Weibull distribution
in Fig. 5 on the Weibull probability paper. Fig. 5 represents

Fig. 3. Life prediction.

Table 5. Predicted life time

No. Life Data (hour)

1 190.8
2 177.4
3 169.0

3.2 Goodness-of-fit Analysis of Life Data Distibution

To decide the most suitable life distribution for life data Fig. 5. Marking on Weibull probability paper and deduction
as the first step of statistical analysis, we performed of major parameters.
 Baek-Ju Sung and Jong-Bae Lee 31

the pointing results on Weibull probability paper. In Fig. 5,

we can deduce the values of reliability parameters, as it
were, the shape parameter β as 1.25, scale parameter θ as
19.28, location parameter γ as 163.02. And also, the mean
time to failure (MTTF) is 180.98. the prediction results of
major reliability parameters obtained from this study are
shown in Table 6. Model 3 Model 4
(changing shape of rotor) (applying assistance slot, changing
Table 6. Predicted life time shape of rotor)

95% confidence level Fig. 6. Finite Element model for cogging torque analysis of
Item Value servo motor.
lower upper
Shape parameter (β) 1.25155 0.154057 10.1676
Scale parameter (θ) 163.024 163.024 163.024
Location parameter (γ) 19.2794 6.95155 53.4695
MTTF 180.975 163.024 204.363
B10 life 166.217 163.024 178.313

cogging torque positive torque

4. FEA (Finite Element Analysis)

From the analysis results of major failure mode in

previous clause, we know the cogging torque effect on
servo motor as the important failure mode. So, we describe
a cogging torque reducing method and various experimental
results in this clause. Cogging torque causes torque ripple
total harmonic distortion counter electromotive force
which is influence on location control of servo motor, and
become the cause of vibration. In order to reduce these
cogging torque, we produced FEA (Finite element Fig. 7. Results of Finite Element Analysis.
Analysis) model about shape of stator and rotor of servo Table 8. Comparison result of cogging torque simulation
motor, and compared the results of the FEA. Table 7
represents the basic data for FEA model, and Fig. 6 shows Model 1 Model 2 Model 3 Model 4
FEA model for analysis of cogging torque of servo motor.
Max. 0.88903 0.55763 0.05738 0.01614
Min. -0.87803 -0.53241 -0.0575 -0.0163
Table 7. Basic data for FEA
Maximum cogging torque of model 1 which is basic
Capacity 1.8 kW
model is 0.88, but it is reduced to 0.55 as applying notch,
Rotor core SCM440
and it is similar with applying a notch as changing the
Stator core S18
shape of permanent magnet. It was largely reduced to
Residual magnetic flux density of
1.25 [T] 98.2% comparing basic model as applying a notch and
permanent magnet
changing a shape of permanent magnet.
Permeability of permanent magnet 1.05
Shaft All-in-one type

5. Conclusion

This paper presents the failure analysis results and major

reliability parameters for machine tool changing servo
motor.
1) For reliability assessment of machine tool changing
Model 1 Model 2
servo motor, we draw out the failure analysis data and
(applying assistance slot) determined main performance test items for reliability
32 Reliability Improvement of Machine Tool changing Servo Motor

test of servo motor by composition of 2-stage QFD. [3] Gray s. Wasserman, “Reliability Verification, Testing,
2) We can predict the life trend by analysis of and Analysis in Engineering Design”, Marcel Dekker,
performance degradation data of servo motor, and can Ins. New work, 2003.
know that 3-parameter Weibull distribution is the [4] MIL-STD-690C, “Guide for Life Test”, Agencies of
most suitable for life distribution of servo motor the Department of Defense, U.S.A, 1999.
through the goodness-of-fit analysis for representative
four kinds of life distribution based on these data.
3) This paper proposed a cogging torque reduction Baek-Ju Sung received the Ph.D
method which is a major failure modes by analysis of degree in electrical engineering from
FEA model. Chungnam National university in 2006.
4) As used servo motor in the test is recently And also received the M.S. and B.S.
development step, 3-samples are performed on the degrees in electrical engineering for
test and analysis. Although it is guaranteed accuracy Busan National University in 1992 and
of reliability parameters, we will perform the study 1990, respectively. His research
that is added number which is able to guaranteed interests are electrical actuator and reliability research about
accuracy of reliability parameters. machinery components.

Jong-Bae Lee received the M.S. and

References B.S. degrees in electrical and
electronics engineering for Kyungsung
[1] Logistics Engineering Technology Branch University in 2006 and 2004,
Carederock div, “Hand book of Reliability Prediction
respectively. His research interests are
Procedures for Mechanical Equipment”, Naval
electrical actuator and reliability
Surface Warfare Center Carderock Division, 1998. 9.
research about machinery components.
[2] Roy Billinton and Easin Khan, “A Security Based
Approach to Composite Power System Reliability
Evaluation”, IEEE Trans. Power Systems, vol.PS-7,
no.1, pp.65-72, Feb. 1992.