AGRICULTURAL ENGINEERING

TEMPERATURE PROTECTION METHODS OF INDUCTION MOTOR

Aleksejs Gedzurs
Latvia Universiry of Agriculture
aleksejs.gedzurs@inbox.lv

Abstract
In conditions where induction motors are frequently started, overloaded and used in high inertia applications with
long starting times, supplied from frequency converter, a temperature protection systems are more reliable to protect
induction motor stator winding against thermal overloads. There are different types of temperature protections -
thermostat, PTC thermistor, resistance temperature detector (RTD) and thermocouples, so it is important to know the
properties of each type to choose an adequate protection system. Analyses of temperature sensor properties and their
advantages and disadvantages show, that PTC thermistor is a cost-effective temperature protection solution, but for
medium and high voltage induction motor protection RTD are commonly used. A virtual model has been represented
to simulate the temperature sensor thermal time constant under different thermal conductivity and thickness of
winding magnet wire insulation.
Key words: induction motor, protection, temperature sensor, thermal time constant, heating.

Introduction temperature sensor measurements. Trending the

Today in industrial and domestic environments
moving and rotary mechanisms are mainly driven by
electric motors, 90% of which are induction motors
(IM). Because of simple construction, maintenance
and variable frequency drive (VFD), IM with squirrel
cage rotor are widely used. Three-phase induction
motors (IM) are used as drive motors in pumps, lifts,
cranes, compressors, fans, crushers, mills, cranes,
conveyors, etc. Annual IM failure rate is estimated
at 3-5% per year, and in extreme cases, up to 12%
(Venkataraman et al., 2005), which cause essential
direct and technological losses.
The failures of IM may be classified as following:
electrical faults;
mechanically related;
environmental impact;
other reason.
Statistics show that overheating of IM parts during
operation is one of the common reasons of an IM
failure (Venkataraman et al., 2005). Stator windings
insulation is most sensitive to thermal overheating.
Exceeding the thermal limit of the insulation will
result in acceleration of the oxidation process that
will decrease the insulation life or even cause IM
failure. It is well known that the duration of insulation
life is decreased by half for each increase of 10 °C
temperature above its thermal limit temperature.
In the past decades improved materials with better
electrical, mechanical and thermal properties allowed
manufactures to reduce the materials within motor and
production costs of the IM (Glew, 1999). For example,
reduced cross section of magnet wire (winding wire)
increases the copper losses in IM winding thus
increasing winding heating. Therefore, it is important
to select an adequate protection system to protect
critical IM parts from overheating and prevent failure.
One of effective techniques to detect contamination
is direct winding temperature monitoring to archive
stator winding temperature over time, and if there is
a gradual increase in temperature, it may be cooling
problems caused by contamination (Culbert, 2008).
An operation review (Barbinyagra, 2007) from
ammonia factory shows, that majority of IM failures
are caused by vibration, overload and hazardous
environment. To increase cycled loaded IM reliability,
a temperature monitoring system with temperature
sensors was installed in the factory. Experimental
study on heating of 2.2 kW IM fed by pulse width
modulation inverter (Benhaddadi et al., 1997) shows,
that at rated load stator winding temperature is 7
°C higher than if IM is fed by sinusoidal voltage.
Inverter fed IM stator winding insulation also needs
to be resistant to partial discharge (PD). Therefore,
magnet wires of IM contain nano sized metal oxides
to impart PD resistance (Lynn et al., 1985). This
very thin layer of metal oxide material is heat shock
and thermal cycling sensitive and when the layer is
cracked, the PD resistance property is reduced (Stone
and Braswell, 2004). In high inertia application and
application, where the rotor can lock or stall, the
heating process of IM parts is very rapid and direct
temperature sensing is considered to be more reliable
than stator or armature current sensing. Studies (Staton
et al., 2009; Gedzurs et al., 2014) show, that for low
power IM (2.5 kW and 1.1 kW) the temperature rise
speed is 6.2 °C.s-1 and 5 °C.s-1 respectively under
locked rotor conditions. These studies show that in the
mentioned operation conditions a direct temperature
measurement of IM winding is more reliable for an
adequate IM protection.
The objective of the study is to analyze temperature
sensor properties for temperature protection of
induction motors under extreme overloads and
hazardous operation conditions.

258 RESEARCH FOR RURAL DEV ELOPMEN T 2015, V OLUME 1

TEMPERATURE PROTECTION
METHODS OF INDUCTION MOTOR
a b
R, Ω
4000
1300
550
250
100
20
- 20 0 T, oC RRT - 20
RRT - 5
RRT
Figure 1. Temperature sensors resistance - temperature characteristics:
a - PTC thermistor; b - RTD PT100 sensor.
Materials and Methods
Induction motor temperature protection monitors
the stator winding temperature by directly measuring
the temperature using embedded temperature
detectors. There are different types of temperature
detectors (sensors), that are used for electric motor
protection - thermostat, PTC thermistor, resistance
temperature detector (RTD), thermocouples.
Winding thermostat is a temperature dependent
device that uses bi-metallic strip to change the
position of a pair of contacts at the preset rated
response temperature. When temperature preset
level is exceeded, the contacts switch a control
device, relay or contactor. The control voltage can
be applied directly to the thermostat, which makes a
tripping mechanism unnecessary. The disadvantages
of the thermostat are - long thermal delay, tripping
temperature can be affected by careless fitting, large
size compared to modern sensors.
Thermistor is a small resistance sensor with non-
linear resistance - temperature relationship. There
are two types of thermistors - negative temperature
coefficient (NTC) and positive temperature coefficient
(PTC) thermistors. The PTC thermistors are commonly
used in IM motor protection systems. The resistance
at normal temperature is relatively low and remains
nearly constant up to the rated response temperature
(RRT). As the RRT is exceeded, the gradient of the
resistance increases sharply, giving the PTC thermistor
a high sensitivity to small changes of temperature. At
the set point, a temperature rise of a few degrees results
in a large increase in resistance. The PTC thermistor
resistance characteristic is shown in Figure 1(a). The
resistance change of the PTC thermistor is monitored
by a thermistor protection relay. The advantages of the
PTC thermistor are small size, which allows them to be
installed in direct contact with the stator winding and
a low thermal inertia, which gives rapid and accurate
response to the winding temperature changes.
RESEARCH FOR RURAL DEV ELOPMEN T 2015, V OLUME 1
Aleksejs Gedzurs
R, Ω
180
160
140
120
100
RRT +20 o
RRT +5 0 50 100 150 200 T, C
Resistance temperature detectors (RTD)
monitors temperature by measuring the change of
resistance of an accurately calibrated resistive sensor,
usually made of copper, platinum or nickel, but
platinum RTD PT-100 is commonly used for stator
winding temperature monitoring. RTD sensors can
be of the wire wound type, or can be of the metal
film type, which are lower cost with faster response
but their characteristics can deteriorate over a time.
A RTD has a linear resistance and temperature trend
(Figure 1(b)), usually 0.4 Ω.°C-1 for PT-100 sensor.
A very sensitive instrument, based on Wheatstone
bridge, is required to measure the small changes in
the resistance of RTD. The instrument passes a small
excitation current through the resistive sensor.
Thermocouple is a temperature sensor that
consists of two dissimilar metals, joined together
at one end and for a junction (a couple). There are
different types of thermocouples depending on the
combination of metals used to make the sensor. Type
J (iron and constantan), K (Chromel and Alumel), T
(copper and constantan) and E (Chromel and Alumel)
thermocouples are commonly used. The voltage-
versus-temperature relationship of most types of the
thermocouple is not linear. Thermocouple measurement
devices need a cold junction compensation. The size
of thermocouple is small so it can be placed between
windings in small power IM.
To simulate changes of the thermal time constant of
the temperature sensor at different IM stator winding
insulation parameters, thermal conductivity and
thickness, a virtual model is represented in MATLAB
SIMULINK environment. The mathematical model is
obtained from the following mathematical operations.
Assuming that at start time there is no heat transfer
from the sensor to environment, the heat transferred
from the winding to the temperature sensor is
described by the following equation:
259
 ost types
erent types of the thermocouple
of thermocouples is not on
depending linear. Thermocouple
the combination measurement devices need a cold junction
of metals
ompensation.
constantan), KThe size of thermocouple
(Chromel and Alumel), T is (copper
small soand it canconstantan)
be placedand between windings in small power IM.
To simulate
are commonly used.changes of the thermal time constant
The voltage-versus-temperature of the of
relationship temperature sensor at different IM stator
inding
linear. insulation
Thermocouple parameters, thermal devices
measurement conductivity needand thickness,
a cold junctiona virtual model is represented in MATLAB
TEMPERATURE PROTECTION
IMULINK
small so it can environment.
be placed betweenThe mathematical
windings inmodel is obtained
small power IM. from the following mathematical operations.
Aleksejs Gedzurs METHODS OF INDUCTION MOTOR
ssuming that at start time there is no heat transfer
mal time constant of the temperature sensor at different IM stator from the sensor to environment, the heat transferred from the
inding to the
nductivity andtemperature
thickness, asensor virtualis model
described by the following
is represented in MATLAB equation:
cal model is obtained from the following mathematical operations.
transfer from the sensor to environment, the heat transferred dT from the effect, therefore,
dT only two lead wires are required. A
bed by the following equation: h S (Tw T ), a c m , h S (Tw TPTC ) cthermistor
m , (1)
dt can be used either for alarm or trip
t
dt
here t – increment heat from windings to sensor, W;
function. If both functions are needed, then the one
dT dT thermistor is used for alarm and the second one for
T ), a c m , h S (Tw T ) c m , , (1) (1)
dt dt trip. Like RTD, thermistor needs a constant current/
to sensor, W; voltage source, but one controller can been used for
where θt – increment heat from windings to sensor, several thermistors connected in series. This makes a
W; PTC thermistor a cost-effective temperature protection
θa – sensors accumulated heat flow, W; method. RTD among above mentioned temperature
h – heat transfer coefficient, W.(m2 .°C)-1; sensors has the best accuracy due to linear resistance-
lated heat flow, W; S – heat transfer surface, m2;
fficient, .
W (m W; 2.
°C) ;
-1 temperature relationship and the highest stability. Lead
ted heat2 flow, Tw – winding temperature, °C; wire resistance and self-heating errors affect the RTD
face, m ; . 2 . -1
icient, W (m °C) ; T – sensor temperature, °C; measurements. Three lead wire configurations will
rature,2 °C;
ce, m ;
ure, °C; c – sensor specific heat, J.(kg .°C)-1; been used to decrease the lead wire resistance effect
ture, °C; m - mass of the sensor, kg.
eat, J.(kg .°C)-1;
e, °C;
and four lead wire configurations for very precise
or, kg. t - time, s. measurement needs. A Wheatstone bridge based
at, J (kg °C) ;
. . -1

r, kg. instrument needs to measure small resistance changes

According to expressions (1) a differential equation of RTD. These facts make RTD a more expensive
sions (1) a differential of equation
sensor`s of sensor`s
heating heating form
in normal in normalis as form is as follows:temperature protection system. Thermocouples are
follows:
ons (1) a differential equation of sensor`s heating in normal form is as follows: simple, self-powered and have wide temperature
dT
s Tw T , (2) measurement
(2) range, but for temperature protection
dTdt of medium and high voltage induction motors the
Tw T , (2)
ime constant, s. where s
dt τs – sensor thermal time constant, s. RTD are more conformable, because of their better
me constant, s. accuracy and stability.
and (2) it is possible to get thermal time constant s of the temperature sensor:
From equations, (1) and (2) it is possible to get
nd (2) it is possible tothermal get thermaltime time
constant τ of the
constant s of temperature
the temperature sensor: Thermal time constant of temperature sensor.
sensor:
c m c ms R , (3)As mentioned in (Venkataraman et al., 2005),
s
c hmS c m S R , (3) due (3) to thermal inertia of temperature sensors, the
sistance, (m2 .°C) .W -1. s temperature protection system cannot react to fast
h S S
stance, (m2 .°C) .W -1. heating process of the IM stator winding, such as
where R = 1.h-1 - thermal resistance, (m2 .°C) .W -1. locked rotor or prolonged starting of IM modes. It
parison of the temperature sensors. PTC thermistors have high sensitivity to small
ated response temperature range and due to higher resistance of the thermistor, islead important to know the thermal time constant of
arison of the temperature Results and Discussion
sensors. PTC thermistors have high sensitivity to small temperature sensor to choose properly settings.
ery small effect, therefore, only two lead wires are required. A PTC thermistorthe can
ed response temperatureTable range and due to comparison
1 shows higher resistance of the of the thermistor, lead
temperature Heating process model of the temperature sensor,
ip function. If both functions are needed, then the one thermistor is used for alarm
y small effect, therefore, only two lead wires are required.
sensors. PTC thermistors have high sensitivity A PTC thermistor
to can placed on the surface of end winding, shown in
p function. If both functions are needed, then
small temperature the one
changes in thermistor
the ratedis response
used for alarm
figure 2 (Fraden, 2010).
temperature range and due to higher resistance of the Table 1To evaluate the thermal time constant of the
Comparisonthermistor, of temperature lead wiresensors
resistance error has a very small temperature sensor, the technical data of a Minco
Table 1
Comparison of temperature sensors
mistor Platinum RTD Thermocouple Table 1
y;
stor - accuracy;
Platinum RTD -
Comparison
self-powered;
Thermocouple of temperature sensors
res required; - linearity; - less expensive than RTD;
- accuracy; - self-powered;
ect several - stability. - small size and thermal
s required; - linearity;
Sensor PTC Thermistor- less expensive than RTD; Platinum RTD Thermocouple
s in series to one mass;
t several - stability.
Advantages - sensitivity; - small size
- and thermal
accuracy; - self-powered;
- wide temperature range
- 2 lead wires required;mass;
- simple.- linearity; - less expensive than RTD;
n series to one
ctive;
- can connect several - wide temperature - stability. range - small size and thermal
mal mass. thermistors in series to -one mass;
ve; simple.
alrity;
mass. - lead wire resistance relay;error (3 or 4 - cold-junction - wide temperature range
mistor has pre-set lead wire required); - cost-effective; compensation; - simple.
ty; - lead wire resistance error (3 or 4 - cold-junction
nt ( can be used - generally high -responselow thermal timemass.
for - non - linearity for most
istor has pre-set lead wire required); compensation;
arm or for trip);
( can be used
wire-wound RTD;
Disadvantages
- generally - non-linearity;
high response time for
types; - lead wire resistance error (3 or 4 lead - cold-junction
- non - linearity for most
oltage excitation. - low vibration resistance;
- each thermistor has pre-set - least sensitive;
wire required); compensation;
m or for trip); wire-wound RTD;switch point ( can be used types; - generally high response time for wire- - non - linearity for most
- self-heating; - least stable;
only
tage excitation. - low vibration resistance; alarm or for trip); - least sensitive;
- current/voltagefor excitation; - low outputwound RTD;
signal. types;
- self-heating; - current/voltage - least stable;
excitation. - low vibration resistance; - least sensitive;
- requires very sensitive instrument
- current/voltage excitation; - low output - self-heating;
signal. - least stable;
to measure.
- requires very sensitive instrument - current/voltage excitation; - low output signal.
to measure. - requires very sensitive instrument to
D, thermistor needs a constant current/voltage source, but one controller measure.can been
connected in series. This makes a PTC thermistor a cost-effective temperature
, thermistor needs a constant current/voltage source, but one controller can been
mong above mentioned temperature sensors has the best accuracy due to linear
connected in series. This makes a PTC thermistor a cost-effective temperature
onship and the highest stability. Lead wire resistance and self-heating errors affect
ng above mentioned temperature sensors has the best accuracy due to linear
ee lead wire configurations will been used to decrease the lead wire resistance effect FOR RURAL DEV ELOPMEN T 2015, V OLUME 1
ship and the highest260 stability. Lead wire resistance and self-heating errors RESEARCH
affect
tions for very precise measurement needs. A Wheatstone bridge based instrument
lead wire configurations will been used to decrease the lead wire resistance effect
stance changes of RTD. These facts make RTD a more expensive temperature
ons for very precise measurement needs. A Wheatstone bridge based instrument
uples are simple, self-powered and have wide temperature measurement range, but
ance changes of RTD. These facts make RTD a more expensive temperature
of medium and high voltage induction motors the RTD are more conformable,
 TEMPERATURE PROTECTION
METHODS OF INDUCTION MOTOR Aleksejs Gedzurs

1 θt 3
2

Rw-s T Rs-e
Tw Te

m,c

Figure 2. Heating process model of the temperature sensor:

1 - Stator winding of IM; 2 - temperature sensor; 3 - air gap between stator winding and IM case;
Rw-s - thermal resistance between winding and temperature sensor; Rs-e - thermal resistance between
temperature sensor and air gap.

1 2 3 4 5
Venkataraman et al., 2005), due to thermal inertia of temperature sensors, the
θt rotor
m cannot react to fast heating process of the IM stator winding, such as locked
odes. It is important to know the thermal time constant of the temperature sensor to
ing process model of the temperature sensor, placed on the surface of end winding,T
S
10). Tw k1 k2 k3
m, c
t 3
l1 l2 l3 l4
2
Figure 3. Simplified heating process models of the temperature sensor:
Rw-s T
Tw 1R- s-e
stator winding
Te of IM; 2 - magnet wire insulation; 3 - winding impregnation;
4 - sensor element insulation; 5 - sensor; l1, l2, l3 - thickness of insulation layer;
l4 - sensor thickness; k1, k2, k3 - thermal conductivity of insulation layers.
m,c
thin-film RTD stator temperature sensor (model and k1 = k2 = 0.2 W.(m °C)-1 for high performance
S200050PD) have been chosen as an example. insulation material. Thermal conductivity of platinum
gure 2. Heating processThemodel of the temperature
temperature sensor - PT100 sensor: has the following k = 71.6 W.(m °C)-1. According to (Stone and Braswell,
IM; 2 - temperature sensor; 3 - air gap between stator winding and IM case; 4
parameters: sensing element dimensions - 2 mm x 2004), the insulation thickness of magnet wire l1 =
een winding and temperature sensor; Rs-e - thermal resistance between temperature
2.3 mm,
sensor and airthickness
gap. - 2 mm, insulation material of the 0.05 - 0.15 mm and thickness of winding impregnation
sensing element (epoxy glass). If the sensor has been usually is l2 = 0.025 - 0.05 mm and rarely l2 = 0.38
placed on the end winding surface and the contact - 0.5 mm. Since the thickness of insulation layer of
mal time constant of istheideal, then there
temperature is the
sensor, a layer of insulation
technical data of a Minco between the temperature sensor is not given in specifications,
thin-film
r (model S200050PD) stator
havewinding and the
been chosen as ansensor.
example. TheThe insulation
temperature layersensorit was
- assumed to be equal to the average thickness of
meters: sensing elementconsists - 2 mmmagnet
of a winding
dimensions x 2.3 mm, wirethickness
insulation, - 2winding the magnet wire insulation, i.e., l3 = 0.1 mm.
mm, insulation
impregnation
nt (epoxy glass). If the sensor has and beensensor
placedinsulation
on the endlayer. windingA simplified
surface and theUsing equation (3) and parameters of insulation
a layer of insulationheating
between model
statorofwinding
the sensor andisthe shown
sensor. in Figure 3. The layer
The insulation materials and the temperature sensor, a virtual model
wire insulation, winding impregnation and sensor insulation
thermal resistance of the each insulation layer can layer. be in MATLAB SIMULINK has been developed (Figure
A simplified
is shown in Figure 3. The thermal
calculated usingresistance
equation of 4. the each insulation layer can4)beto simulate effects of different insulation layer
thickness and thermal conductivity on the thermal
l, (4) time (4) constant of the temperature sensor. Thermal time
R
k constant τo = 4.69 s at the following parameters l1 = l2
vity, W.(m .°C)-1; = 0.05 mm, l3 = 0.1 mm, k1 = k2 = 0.2 W.(m °C)-1.
ickness, m; where k – thermal conductivity, W (m °C) ; . . -1
Figure 5 represents simulation results of the
l – insulation layer thickness, m;
thermal time constant of the temperature sensor
y of sensor epoxy glass insulation layer is k3 = 0.343 W (m °C) (Sarvar et al., . -1

on the quality depending on the thermal conductivity k1 (0.2 - 1 W

of magnet wire and winding
Thermal impregnation
conductivity depends of sensor epoxy ofglass insulation.
006) thermal conductivity k1 = klayer
insulation 2 = 0.2
is W
.
k3(m
.
°C) for
= 0.343
-1
Wnormal
.
(m °C)raisin -1 and ket1 = k(m
(Sarvar 2 =
°C)-1) of magnet wire insulation. At k1 = 0.2 W.
mance insulation material. Thermal conductivity of platinum 4 = 71.6 wireW.(m °C) (m-1.°C)-1 thermal time constant is τo = 4.69 s, at k1 = 1
al., 1990). Thermal conductivity of kmagnet
mm W and(m °C) - τs = 3.55 s, so the thermal time constant
-1
aswell, 2004), the insulation
and winding thickness of magnet
impregnation wire l1on
depends = the
0.05quality- 0.15 of
nation usually is l2 =insulation.
0.025 - 0.05 mm andtorarely
According (Dorrel l2 =
et 0.38
al., 2006) mm. Sincedecreased
- 0.5 thermal the by ∆τs = 1.14 s. As Figure 5 shows the trend
of the temperature sensor is not given in specifications, it-1was assumed to be equal is not linear and from k1 0.2 - 0.4 W.(m °C)-1 thermal
conductivity k1 = k2 = 0.2 W (m °C) for normal raisin
. .
magnet wire insulation, i.e., l3 = 0.1 mm. time constant change is the biggest ∆τs = 0.71 s.

1 2 3 4 5
RESEARCH FOR RURAL DEV ELOPMEN T 2015, V OLUME 1 261
t

T S
T k k k
 TEMPERATURE PROTECTION
Aleksejs Gedzurs METHODS OF INDUCTION MOTOR

l3, m

R3, m2.oC.W-1

k3, W.(m.oC)-1
R2, m2.oC.W-1

τs(k1)
R, m2.oC.W-1
l2, m

k1/k2/l1/l2
c, J.(kg.oC)-1
τs, s
k2, W.(m.oC)-1

l1, m

R1, m2.oC.W-1

k1, W.(m.oC)-1
R4, m2.oC.W-1
S, m2

k4, W.(m.oC)-1 l5, m

m, kg
ρ, kg.m-3

l6, m
l4, m

Figure 4. Simulation block diagram of thermal time constant of temperature sensor.

τs, s
4.9
τo
4.7

4.5

4.3
∆τs
4.1
4.1

3.9
3.9

3.7
3.7

3.5
3.5
0.2
0 0.3
0.1 0.4
0.2 0.5
0.3 0.6
0.4 0.7
0.5 0.8
0.6 0.9
0.7 1
0.8
k1, W.(m.oC)-1

Figure 5. Simulation results of thermal time constant of temperature sensor at different

thermal conductivity k1 values of magnet wire insulation.

τs, s
88
7.5
7.5

77
∆τs
6.5
6.5

6
6

5.5
5
τo
4.5

4
0.05
0 0.06
0.01 0.07
0.02 0.08
0.03 0.09
0.04 0.1
0.05 0.11
0.06 0.12
0.07 0.13
0.08 0.14
0.09 0.15
0.1
l1, mm
Figure 6. Simulation results of thermal time constant of temperature sensor at
different insulation layer l1 values.

262 RESEARCH FOR RURAL DEV ELOPMEN T 2015, V OLUME 1

TEMPERATURE PROTECTION
METHODS OF INDUCTION MOTOR Aleksejs Gedzurs

Figure 6 represents simulations results thermal
time constant of the temperature sensor τs and magnet
wire insulation wire thickness l1 relationship. Thermal
time constant increase is linear from τo = 4.69 s
at l1 = 0.05 mm τs = 7.54 s at l1 = 0.15 mm. The
thermal time constant increased by ∆τs = 2.85 s or
by 64%.
Simulation results show that, the thickness changes
of magnet wire insulation have a bigger influence
on the thermal time constant than that of thermal
conductivity.
Conclusions
1. Analyses of PTC thermistor sensor properties
show that it requires mostly two wire leads, is very
sensitive to temperature changes at rated response
temperature range: several thermistors can be
connected to one controller, small size of the
sensor and low thermal mass, which makes PTC
thermistor a cost-effective solution for induction
motor temperature protection.
2. Analyses show that to measure small resistance
changes of RTD a sensitive instrument and
three or four lead wires are required for precise
measurements, but accuracy, linearity and a long-
term stability of the RTD makes it a conformable
solution for temperature protection of medium and
high voltage induction motors.
3. Thermal time constant simulation results of the
thin-film RTD PT100 temperature sensor show
that at stator winding magnet wire insulation
thickness l1 = 0.05 mm the thermal time constant
is τo = 4.69 s and at insulation thickness l1 = 0.15
mm thermal time constant increases to τs = 7.54 s.
Therefore, insulation material thickness between
the stator winding and temperature sensor has an
essential impact on the thermal time constant of
small size temperature sensors.

References
1. Barbinyagra M.P. (2007) Опыт эксплуатации сети мониторинга температуры асинхронных
электродвигателей (Operation experience of temperature monitoring network of induction motors).
Ukraine, pp. 56-60.
2. Benhaddadi M., Khaldi R., Benghanem M. (1997) Experimental Stufy of heating in Induction Motor fed
by PWM Inverter. In: Instrumentation and Measurement Technology Conference. IEEE, Ottawa, Canada,
pp. 335-338.
3. Culbert I.M. (2008) A review of cleaning methods for motor windings. In: Cement Industry Technical
Conhference Record. IEEE, Miami, FL, USA, pp. 291-296.
4. Dorrel D.G., Staton D.A., McGlip M.I. (2006) Design of Brushless Permanent Magnet Motors - A
Combined Elekctromagnetic and Thermal Approach to High Performance Specification. In: IEEE
Industrial Electronics. 32nd Annual Conference, IEEE, Paris, France, pp. 4853-4858.
5. Fraden J. (2010) Handbook of modern sensors, Springer, London, UK, 663 p.
6. Gedzurs A., Sniders A. (2014) Induction Motor Stator Winding Thermal Process Research and Modelling
under Locked Rotor Mode. In: Proceedings of 13th International Scientific Conference. Engineering for
Rural Development, May 29-30, Jelgava, Latvia, pp. 265-270.
7. Glew N. (1999) Design and Manufacture of Energy Efficient and Environmentally Friendly Large
Machines. In: IEE Power Division Colloquium 99/178, IET, Glasgow, UK, pp. 1-5.
8. Sarvar F., Poole N.J., Witting P.A. (1990) PCB glass-fibre laminates: Thermal conductivity measurements
and their effect on simulation. Journal of electronic Materials, 12, pp. 1345-1350.
9. Staton D., Šušnjič L. (2009) Induction Motors Thermal Analysis. Strojarstvo, 51, pp. 623-631.
10. Stone G.C., Braswell L.E. (2004) Motor insulation Systems.In: Toliyat H.A. and Hliman G.B. (eds)
Handbook of electric motors, CRC Press, Florida, USA, 824 p.
11. Venkataraman B., Godsey B., Premerlani W., Shulman E., Thakur M., Midence R. (2005) Fundamentals
of a Motor Thermal Model and its Applications in Motor Protection. In: Protective Relay Engineers. 58th
Annual Conference, Black & Veatch Corporation, Kansas City, USA, pp. 127-144.

RESEARCH FOR RURAL DEV ELOPMEN T 2015, V OLUME 1 263