An Overview on Induction Machine's Diagnosis Methods

Abstract

Several methods for diagnosing the faults of the squirrel-cage induction motors are cited in the literature. The authors focused their researches on the rotor faults of the squirrel cage induction machines. The aim of the paper is to compare these fault diagnosismethods by means of data processing in LabVIEW the results obtained by measurements. It will be emphasized that beside the current signature analysis also the threephase current vector, the instantaneous torque, respectively the outer magnetic filed can be used for diagnosing the rotor faults.

Full text

An Overview on Induction Machine's Diagnosis Methods
Szabó L.*, Tóth F.**, Kovács E.** and Fekete G.**
* Department of Electrical Machines, Marketing and Management
Technical University of Cluj
RO-400750 Cluj, P.O. Box 358, Romania
e-mail: Lorand.Szabo@mae.utcluj.ro
** Department of Electrical and Electronic Engineering
University of Miskolc
H-3515 Miskolc-Egyetemváros, Hungary
e-mail: elktoth@gold.uni-miskolc.hu
Abstract – Several methods for diagnosing the faults of
the squirrel-cage induction motors are cited in the
literature. The authors focused their researches on the
rotor faults of the squirrel cage induction machines. The
aim of the paper is to compare these fault diagnosis
methods by means of data processing in LabVIEW the
results obtained by measurements. It will be emphasized
that beside the current signature analysis also the three-
phase current vector, the instantaneous torque,
respectively the outer magnetic filed can be used for
diagnosing the rotor faults.
Keywords: induction machine, rotor faults, electrical
machine's faults detections and monitoring
I. INTRODUCTION
Due to their reliability, robustness and low price the
squirrel cage induction motors are widely used as energy
converters. But also these reliable electrical machines can
get faulty.
The statistical data show that 10% of the induction
machine's faults are the rotor faults [1]. The typical rotor
faults are the damaged rotor bars, discontinuities in
between the bars and the end rings, rotor eccentricity, etc.
In order to prevent more serious damages, and due to this
long time plant shutdowns, it is recommended to detect
these faults as soon as possible. The electrical machines
can be diagnosed via several methods.
The rotor bar faults can be detected using the following
methods:
• Sensing the machine's vibrations and using frequency
analysis of the vibrations [2, 3]
• Measuring the line currents and analysing its harmonic
content [3, 4, 5]
• Studying the three-phase current vector (Park vector)
[6, 7, 8]
• Sensing the torque's variation and applying its
frequency analysis [9, 10]
• Maesuring the machine's external magnetic fields and
analysisng its harmonic content [11, 12, 13]
Since the 80's the most widely used diagnosis method was
based on the vibration analysis. In the 90's the motor
current signature based fault detection methods (MCSA –
Motor Current Signature Analysis) got more widespread.
In this paper the main methods used for the broken rotor
bars detection will de overviewed. The measurements were
performed by using advanced data acquisition boards and
methods. The measured data were processed using the
LabView program.
II. THE EFFECTS AND DIAGNOSIS OF BROKEN
ROTOR BARS
During the startup of the squirrel cage induction motors
their current is 5-8 times greater than the rated current. Due
to this a great quantity of heat is produced in the rotor. The
temperature of the bars is rapidly increasing and the bars
are suspected to significant mechanical stresses. In
steady-state regime the strong thermal and mechanical
stresses are diminished. If the induction machine is rather
old or is frequently started the connection between the bars
and the end rings can be damaged. Hence the resistance of
the of a part of the rotor circuit is increased, which leads to
an asymmetry in the machine's field.
A. The line current's spectrum analysis
In the three-phase induction motor under perfectly
balanced conditions (healthy motor) only a forward
rotating (direct sequence) magnetic field is produced,
which rotates at synchronous speed, pfn 11 =, where 1
f
is the supply frequency and p the pole-pairs of the stator
windings. The rotor of the induction motor always rotates
at a speed (n) less than the synchronous speed. The slip,
11 )( nnns
−
=
, is the measure of the slipping back of the
rotor regarding to the rotating field. The slip speed
(112 nsnnn
=
−
=
) is the actual difference between the
speed of the rotating magnetic field and the actual speed of
the rotor.
The frequency of the rotor currents is called the slip
frequency and is given by pnspnf 122 == . The speed of
the rotating magnetic field produced by the current carrying

rotor conductors with respect to the stationary stator
winding is 112 nnnnnn
=
−+=+ . With respect to a
stationary observer on the fixed stator winding, the speed
of the rotating magnetic field from the rotor equals the
speed of the stator rotating magnetic field, namely, the
synchronous speed. Both fields are locked together to give
a steady torque production by the induction motor.
With broken rotor bars in the motor there is an additional,
backward rotating magnetic field produced. This is rotating
at the slip speed with respect to the rotor. The backward
rotating magnetic field speed produced by the rotor due to
broken bars and with respect to the rotor is:
)21(2)1( 111112 snsnnnssnnnnb−=−=−−=−= (1)
The stationary stator winding now sees a rotating field at
)21(
1snnb−= or )21(
1sffb−= (expressed in terms of
frequency). This means that a rotating magnetic field at that
frequency cuts the stator windings and induces a current at
that frequency (fb). This in fact means that fb is a twice slip
frequency component spaced 1
2fs down from 1
f. Thus
speed and torque oscillations occur at 1
2fs , and this
induces an upper sideband at 1
2fs above 1
f.
Classical twice slip frequency sidebands therefore occur
around the supply frequency:
)21(
1ksffb±= (2)
where k = 1,2,3…
An estimate of the number of the broken rotor bars can be
analytically determined [4]. The ratio of the linear
magnitude of the lower sideband at 1
)21( fs− to the
magnitude of the supply current 1
f.
`
)2(2
sin
απ
α
−
≈p
Rs, (3)
where:
R
pn
π
α
2
=, (4)
and n is the number of the broken bars, p the number of
pole pairs, respectively R the rotor's slot number.
B. The spectrum analysis of the vibrations
In electrical machines the torque is produced upon the
interaction of the stator and rotor fields. The radial
(vibration generating) forces are balanced by the specific
design of the induction machines (pair number of slots and
uniform air-gap). If one or more rotor bars are broken, the
force balance inside the machine is damaged, hence the
unbalanced forces generate significant vibrations. These
vibrations can be measured with adequate sensors and
analyzed. The specific terms used in vibration analysis are:
• rotational frequency: 60
n
ff= [Hz]; (5)
• slip frequency: s
p
f
nn
fslip 1
0
60 =
−
=; (6)
• pole pass frequency: slipp fpf ⋅= 2 (7)
From (6)÷(7) results the pole frequency:
1
2fsf p⋅
⋅
=
(8)
The spectrum of the vibration's amplitude versus the
frequency is given in Fig. 1.
C. Plotting the three-phase current vector
It is known that the m.m.f. generated by the symmetrical
induction machines stator winding is given by:
⎥
⎦
⎤
⎢
⎣
⎡++−+⋅
⋅
⋅
=Θ
)
3
2
cos()()
3
2
cos()(cos)(
2
),( 1
π
ϑ
π
ϑϑ
ξ
π
ϑ
tititi
p
N
t
scsbsa
s
s
(9)
using the complex versor 3/2
π
j
ea = equation (9) can be
written as:
⎭
⎬
⎫
⎩
⎨
⎧⎥
⎦
⎤
⎢
⎣
⎡+⋅+⋅
⋅
⋅
=Θ
−
ϑ
ξ
π
ϑ
j
scsbsa
s
s
etiatiati
p
N
t
)()()(
3
2
Re
3
),(
2
1
(10)
where
⎥
⎦
⎤
⎢
⎣
⎡+⋅+= )()()(
3
2
)( 2tiatiatiti scsbsa
s (11)
is the so-called three-phased current vector (Park vector, or
space vector). If the stator current varies in a sinusoid way,
)cos(
ˆ
)( issa tIti
ϕω
+= , then the three-phase current vector
can be given as:
)(
ˆ
)( i
tj
s
seIti
ϕω
+
= (12)
where s
I
ˆ is the peak value of the line current.
Relation (12) shows that the symmetric three-phase
winding and the sinusoidally varying three-phase currents
generate a vector rotating at constant frequency ω. If the
currents in the windings are not equal, this change will be
observed also in the three-phase vector (12). The former
circular shape will be modified in an elliptical shape. The
difference from the circular shape indicate the fault's rate.
The stator current (12) has two (a real and an imaginary)
component:
Fig. 1. The vibration's spectrum of amplitude

{}
)(
3
1
3
2
Re scsbsa
s
xiiiii +−⋅== (13)
{}
)(
3
1
Im scsb
s
yiiii −== (14)
The voltage signals proportional to these two components,
ix and iy, can be plotted.
D. Potting the torque during startup
The flux vector of the three-phase windings can be
expressed in a similar way as the current vector [14]:
)(
3
22
cba
saa Ψ+Ψ⋅+Ψ=Ψ (15)
Knowing the flux and current vectors the torque can be
computed as [15]:
{}
s
sim ⋅Ψ= ∗
Im
2
3 (16)
When asymmetry occurs the torque will be compound of a
direct and an inverse component and a balancing
component. The last component has double frequency and
generates noises and vibrations [14].
E. Sensing the external magnetic flux
The external magnetic filed of an electrical machine can be
sensed by a coils with great number of turns. The so-called
search coil can be placed at the end of the machine (see
Fig. 2a). In this case the axial components of the flux will
be detected. Also it can be placed in a radial direction for
detecting the radial external magnetic flux of the machine
(see Fig. 2b). The e.m.f in the coil depends on the distance
from the machine, but as we observed the waveform is not
influenced by the distance.
The external magnetic flux is due to the unbalance between
the m.m.f. of the stator and the rotor, which are not in
equilibrium neither in the case of a healthy machine. When
faults occur in the spectrum of the external flux, typical
harmonic components can be detected.
The voltages induced by the radial external fluxes are given
in Fig. 3 (measured at no-load condition).
III. THE SAMPLED VALUES
The measurements were performed using an induction
machine with the following nameplate data: Pn = 1.5 kW,
Un = 400/230 V (Y/Δ), f1 = 50 Hz, In = 4.2/7.3 A,
n = 1361 1/min, cos φn=0.69. The slot numbers were:
Zs = 24, Zr =18. The pictures of the stator and the rotor of
the sample machine are given in Fig. 4. In Fig. 4.b. both the
healthy and the damaged (right side) rotor can be seen.
a)
b)
Fig. 2. Laboratory setup for mesuring the external flux
a) healthy machine
b) machine with broken rotor bars
Fig. 3. The voltage induced by the external radial flux
a)
b)
Fig. 4. The stator and rotor of the sample machine

A. The time plots of the sampled values
The data acquisition was performed using the LabVIEW
program. The time plots of the sampled values are shown
in the next figures: in Fig. 5 the plots for the healthy
machine, respectively in Fig. 6 those for the machine
having broken rotor bars:
• in figures a) the torque;
• in figures b) the phase current;
• in figures c) the voltage induced by the external
radial flux;
• in figures d) the vibrations of the machine mesured
by a sensor placed on the macine's housing
B. The spectrum of the sampled values
The Fourier analysis was performed using the same
LabVIEW program. The spectrum of the signals plotted in
Fig. 5 and 6 are given in Fig. 7 and 8 (those corresponding
to the healthy machine in Fig. 7, respectively those for the
faulty machine in Fig. 8). The plots correspond for:
• in figures a) the spectrum of the torque;
• in figures b) the phase current's spectrum;
• in figures c) the spectrum of the voltage induced by
the external radial flux;
• in figures d) the spectrum of the machine's
vibrations
Fig. 5. Plots versus time of the sampled values
(healthy machine) Fig. 6. Plots versus time of the sampled values
(machine having broken rotor bars)

C. The time plots of the sampled values during startup
During startup the differences between a healthy and faulty
machine can be observed much easy. These differences are
highlighted in Fig. 9 and 10. In a same manner as in the
previous figures the plotted values are:
• in figures a) the torque;
• in figures b) the phase current;
• in figures c) the voltage induced by the external
radial flux;
• in figures d) the vibrations of the machine
Fig. 7. The spectrum of the sampled values (healthy
machine Fig. 8. The spectrum of the sampled values (machine
having broken rotor bars)
Fig. 9. Plots versus time of the sampled values during startup
(healthy machine) Fig. 10. Plots versus time of the sampled values during
startup (machine having broken rotor bars)

D. The plots versus time of the phase currents and the
three-phase current vectors
The rotor bar faults of a squirrel cage induction machine
can be also diagnosed by plotting the three-phase current
vector. A suggestive example is given in Fig. 11 (Fig 11a
for the healthy machine, respectively Fig. 11b for that
having broken rotor bars).
As it can be seen very clearly in Fig. 11b the pulsations of
the phase current are highlighted also in the plot of the
current vector.
IV. CONCLUSIONS
As it could be observed the broken bars of a squirrel cage
induction machine can be diagnosed with all of the method
above described. In the last time, due to its simplicity the
motor current signature analysis (MCSA) method was the
mostly used in industrial environment.
It can be predicted that in the near future diagnosis methods
based on the detection of the external magnetic fields of the
electric machines, a more simple and cheap method, could
be more widespread.
ACKNOWLEDGMENT
The work was possible due to the support given by the
Romanian Ministry of Education and Research, National
Authority for Scientific Research (CNCSIS) and the
Hungarian National Office for Research and Technology
(NKTH) in the framework of the "Romanian-Hungarian
Intergovernmental S&T Cooperation Program 2008-2009".
The authors should like to sincerely thank this way for the
financial support.
REFERENCES
[1] Tian Han, Bo-Suk Yang, Won-Ho choi and Jae-Sik Kim, "Fault
Diagnosis System of Induction Motor Based on Neutral
Network and Genetic Algorithm Using Stator Current Signal",
International Journal of Rotating Machinery, Vol. 2006,
pp. 1-13.
[2] Mori D., Ishikawa T., Force and vibration analysis of induction
motors", IEEE Transaction on Magnetics, vol. 41, no. 5 (May
2005), pp. 1948-1951.
[3] Weidong Li: "Detection of Induction Motor Faults: A
Comparison of Stator Current, Vibration and Acoustic
Methods," Journal of Vibration and Control, Vol. 12, No. 2,
pp. 165-188 (2006).
[4] Ming Xu, Tom Alfod, "Motor current analysis and its
applications in induction motor fault diagnostics", ENTERACT,
98, pp. 1-21.
[5] Mark Fenger, Blake A., "Case histories of current signature
analysis to detect fault in induction motor drives", Electric
Machines and Drives Conference, IEMD '2003, vol. 3,
pp. 1459-1465.
[6] Cruz S. M. A.; Cardoso A.J.M., "Rotor Cage Fault Diagnosis in
Three-Phase Induction Motors by Extended Park"s Vector
Approach", Electric Machines and Power Systems, Volume
28, Number 4, pp. 289-299.
[7] M. Arkan, H. Calis, M.E. Taguk "Bearing and misalignment
fault detection in induction motors by using the space vector
angular fluctuation signal", Electrical Engineering (2005),
no. 87, pp. 197-206.
[8] Izzet y Önel, K Burak Dalci and Ibrahim Senol, "Detection of
bearing defects in three-phase induction motors using Park's
transform and radial basis function neutral networks", Sadhana,
Vol. 31, Part 3, pp. 235-244.
[9] Rastko Fiser, Stanislav Ferkolj, "Development of Steady-State
Mathematical Model of Induction Motor with Rotor Cage
Faults", Proceeings of the ICEM '1998, Vol. 3, pp. 2188-2191.
[10] P. Bajec, R. Fiser, J. Nastran, "Detection of Induction Motor
Squirrel Cage Asymmetry Using Run-Up Test", Proceedings of
the ICEM '2004, pp. 837-838.
[11] Voitto Kokko, "Conditiom Monitoring of Squirrel-Cage Motors
by Axial Magnetic Flux Measurements", Dissertation, Oulu,
2003.
[12] R. Romary, R Corton,D. Thailly, and J.F. Brudny, "Induction
machine fault diagnosis using an external radial flux sensor",
The European Physical Journal, Applied Physics, 32 (2005),
pp. 125-132.
[13] Negrea, M.D. "Electromagnetic flux monitoring for detecting
fault in electrical machines", TKK Dissertations 51, Espoo,
2006.
[14] K.P. Kovács, "Symmetrische Komponenten in Wechsel
strommaschinen", Birkhäuser Verlag Basel und Stutgart, 1962.
[15] Dr. Retter Gyula: "Villamosenergia átalakítók" 2. kötet, MK,
Bp.1987.
a) healthy machine
b) machine with broken bars
Fig. 11. The phase current's plots versus time and the
plots of the three-phase current vector