Jestr Journal of Engineering Science and Technology Review 8 (5) (2015) 19-23

JOURNAL OF
Engineering Science and
Technology Review

www.jestr.org

Research Article

The Temperature Measurement in a Three-Phase Power Transformer under Different

Conditions

K. Karakoulidis1,* J. G. Fantidis1,2 and V. Kontakos2

1
Department of Electrical Engineering, Technological Education Institute of Eastern Macedonia and Thrace, Saint Loucas, Kavala, Greece
2
MSc in Innovation in Technology & Entrepreneurship, Technological Education Institute of Eastern Macedonia and Thrace, Saint Loucas,
Kavala, Greece

Received 23 September 2015; Accepted 21 December 2015

__________________________________________________________________________________________

Abstract

Infrared thermography is a powerful non contact method with the ability to fast inspection of abnormal situations in many
electrical systems and equipments. With the aim of a high resolution thermal camera a laboratory power transformer was
checked under different scenarios. These scenarios include thermal measurements for 58%, 87% and 116% of rated load
conditions, problems in primary or secondary phases and an asymmetric charge. The thermograpic system illustrate fast
and reliable the changes in the windings of the power transformer.

Keywords: Thermography; Power transformer; temperature measurement;

__________________________________________________________________________________________

1. Introduction supply and continued overvoltage for a long period of time.

The major goal of this work is to study the thermal behavior
Power transformers is one of the most important and of a transformer under different operations conditions. All
expensive elements in power generation and transmission the measurements realized at the Kavala Institute of
systems. Considering the importance of the power Technology in the electrical machine laboratory [5-6].
transformers in the electric system their correct functioning
or maintenance is vital to system operation. Commonly
reasons for failure include external factors such as lightning 2. Background of infrared thermography physics
strikes, short circuits, system overload, and internal factors
such as insulation deterioration, winding failure, and The principle of infrared thermography is based on the fact
overheating. For these reasons fault diagnosis power that any object that has a temperature above absolute zero (-
transformers is necessary. Even though a number of 273.15 °C) radiates energy at a wavelength corresponding to
industrial methods exist (Visual Inspection, Insulation its surface temperature. In physics, a blackbody is defined as
Resistance Test, Transformer Turns Ratio Test, Dissolved an object which absorbs all radiation that falls on it at any
Gas Analysis, Magnetic Balance Test, Tan Delta Test, wavelength. Infrared thermography is the science of
Transformer Oil Break Down Test, thermography) both for transforming infrared measurements to construct a
on-line and off-line monitoring of transformers, most of radiometric image. In physics, a blackbody is defined as an
them are costly and complex [1-3]. object which absorbs all incident radiations it receives
Infrared thermography is a non-destructive and whatever the wavelength and re-emits all these absorbed
visualizing technique which commonly used in preventive radiations.
maintenance for the advantage of carrying out rapid, precise, Electromagnetic radiation emitted from a blackbody
and broad area inspections. They can identify heat patterns (Wλb) can be calculated using Planck’s law, as in Equation
or temperature changes in materials and structures. With (1), where c is the speed of light, h is the Planck’s constant,
thermography technique, defective parts can be identified k is Boltzmann’s constant, T is absolute temperature of the
through simple observation without disrupting the blackbody in Kelvin and λ is wavelength.
transformer's operation. According to Mady and Attia [4] the
primary reasons for transformer overheating includes
2π hc3
excessive transformer loading, excess current in the neutral Wλb = 5 hc / λ kT
10−6 [ watt / m 2 µ m] (1)
of the transformer, high harmonic content in the power λ (e − 1)
______________
By integrating Planck’s formula from λ = 0 to λ = ∞ it
* E-mail address: [email protected]
allows calculating the total amount of radiation (Wb) emitted
ISSN: 1791-2377 © 2015 Kavala Institute of Technology.
by the blackbody at a certain temperature T (in every
All rights reserved.
direction and over all wavelength)
 K. Karakoulidis, G. Fantidis and V. Kontakos/Journal of Engineering Science and Technology Review 8 (5) (2015) 19- 23

regulation. The current in the resistor is limited by tubular

wire fuses in each phase. Last but not least the Terco MV
Wb = σ ⋅Τ 4
(2) 1300 power pack, which is suitable for laboratory
experiments, was used as power supply unit. All outputs are
which states that the total emissive power of a blackbody is fused by MCB´s and have load switches. The Power Pack
proportional to the fourth power of its absolute temperature has also Earth Leakages Circuit Breaker (ELCB). The
while σ is the Stefan-Boltzmann constant, equals to technical specifications both for load and power supply unit
5.6704·10-8 (W/m²K4). era also presented in Table 2 while the Figure 1 is a picture
A more general case is of a non-blackbody emitters for of the experimental test bed [12].
which the emissivity is constant regardless wavelength are
called grey bodies. In this case the total amount of radiation
emitted by the grey body characterized by its emissivity, ε Table 2. Specifications of the used electrical equipment
and can be calculated using the Stefan-Boltzmann law: MV 1100 Load Resistor
3-phase 3.3 kW,
E = ε ⋅σ ⋅Τ4 (3) Type
continuously adjustable

where E is the electromagnetic radiation (W/m2), ε is the

Star connection 400/230 V 0,8 - 5 A
object surface emissivity after transmissivity (t) and
reflectivity (ρ) are taken into account [7-9]. Star connection 230/133 V 0.5-5 A
Delta connection 400/230 V 2.4-8.7 A
ε = 1 - (t+ρ) (4) Delta connection 230/133 V 1.03 - 8.07 Α
DC parallel connection 220 V 2,3 έως 15 A
Overload capacity, brief
3. Case study approx. 20%.
duration
Dimensions 630 x 250 x 890 mm
The Jenoptik VarioCAM® 7800 thermographic system was Weight 46 kg
used for the purposes of this work. It is a high-resolution, MV 1972 Transformer 3-phase
portable, digital color infrared and visual camera with a Power rating: 2 kVA
noncooled Focal Plane Array microbolometer which is used
400 V ± 5% or 230 V ± 5%
as an infrared radiation sensor. The thermographic system Primary voltage:
per phase
during the measurement has a standard 30mm lens with
minimum focus 0.3m IFOV 0.8 mrad and FOV (30×23)° 2 x 66,5 V ± 5% per phase
and communicates with the PC via FireWire (IEEE 1394) Secondary voltage
2,5 kV
which are later processed in the suitable analysing software
IRBIS® 3 professional. Table 1 shows the main Efficiency 92%
specifications of the thermographic system Vario CAM®
7800 [10-11]. Dimensions 350x 165x260 mm
Weight 30kg
Table 1. Technical characteristics of the VarioCAM® 7800 Terco MV 1300 power supply unit
thermographic system. MV 1300-405 Supply
Parameter Value Type voltage 380-400 / 220-230
V 50 / 60 Hz 3-ph.
Spectral range 7.5 ‒ 14 µm

Resolution 640 × 480 pixels (resolution DC fixed 220 V 3.5 A

enhancement onto 1280 × 960 DC variable 0-220V 16A
pixels) AC fixed 230/133V 10A 3-
Output voltage
Temperature measuring -40 ‒ 1200 °C ph
range AC variable 3 x 0-230V,
Temperature resolution at Better than 0.08 K 10A 3- ph
30°C
Measurement accuracy ± 1.5 K (0 ‒ 100 °C), ± 2% (< Standard AC fixed 230V 10A
0 and >100) °C Dimensions 660x435x790 mm
Emissivity Adjustable from 0.1 to 1, in
increments of 0.01 Weight 103 Kg
Spatial resolution/IFOV 0.8 mrad

Field of view/FOV 30° (H) ×23° (V) 4 Results and discussion

In order to study the thermal behavior of transformer the

Terco MV 1972 was tested under different operations
In this article a 3-phase, 2 kVA transformer (Terco MV conditions. In the first scenario the transformer was operated
1972) with E- type core was used (Figure 1). Table-2 at 58% load for two hours. Figure 2 illustrates the relevant
provides the specifications of the transformer. The necessary results. It is clear from the measurements that for 58% load
load was supplied by the Terco MV 1100 load resistor which the temperature increases slowly during the first 90 minutes
contains three ganged resistors with continuous spindle and is nearly steady after this time. For 58% of rated load
20
 K. Karakoulidis, G. Fantidis and V. Kontakos/Journal of Engineering Science and Technology Review 8 (5) (2015) 19- 23

conditions the maximum temperature in the transformer

remained in low levels reaching up to approximately 43°C.
Figure 3 illustrates the real thermal images for 58% load
after 30, 60, 90 and 120 minutes of continues operation.

Fig. 3 Thermal images of the transformer after a. 30 minutes, b. 60

minutes, c. 90 minutes and d. 120 minutes of operation for 58% load.

In the second scenario the rated transformer load was

87% and the relative results are shown in Figure 4. As
expected the temperature rise faster and after 2 hours is
above than 60°C. In this case the temperature rises rapidly in
the first 40 minutes, with slower speed for the next 60
minutes and is almost fixed after 100 minutes continuous
operation. The relatively thermographs after 30, 60, 90 and
120 minutes are given in Figure 5. The third scenario
includes the case in which the transformer operates in
overload conditions (116% of rated load). As a sequel to this
event there is a steep increment in temperature during the
first 60 minutes, with lower rates for the next 35 minutes and
almost stops to increase after 95 minutes of continuous
operations. The maximum temperature is nearly 89.5°C (see
Figure 6). Figure 7 provides the thermal images with steps
every 30 minutes for total 2 hours operation.
65
L1
60 L2
L3
55

50

45
θmax (oC)

Fig. 1. Experimental bench. 40

35

30

25
45
L1 20
L2
L3 15
40 0 20 40 60 80 100 120
Time of operation (min)
Fig. 4 The measured temperatures in the 3 phases for 87% of rated load
35 conditions.
θmax (oC)

30

25

20

15
0 20 40 60 80 100 120
Time of operation (min)
Fig. 2. The measured temperatures in the 3 phases for 58% of rated load
conditions.
Fig. 5 Thermal images of the transformer after a. 30 minutes, b. 60
minutes, c. 90 minutes and d. 120 minutes of operation for 87% load.

21
 K. Karakoulidis, G. Fantidis and V. Kontakos/Journal of Engineering Science and Technology Review 8 (5) (2015) 19- 23

90 65
L1 L1
L2 60 L2
80 L3 L3
55
70
50

60 45
θmax (oC)

θmax (oC)
40
50
35
40 30

25
30
20
20
15

10 10
0 20 40 60 80 100 120 0 20 40 60 80 100 120
Time of operation (min) Time of operation (min)
Fig. 6 The measured temperatures in the 3 phases for 116% of rated Fig. 9 The measured temperatures with break circuit in the L3 phase on
load conditions. the primary.

In the last experiment the transformer supplied with

unbalanced load in every phase. L1 has 116% of rated load,
L2 has 87% of rated load and L3 has 58% of rated load. The
real thermal images after 0, 30, 60, 90 and 120 minutes of
operation are shown in Figure 10 while the Figure 11
illustrates the results. The difference in temperature between
the phases is noticeable instantly. After 1 hour operation L1,
L2, and L3 have temperature approximately equal to 70, 53
and 41°C correspondingly while after 2 hours the relevant
values are 83, 64, and 51°C. Common deduction in all cases
is the ability of the thermal camera to detect very fast the
changes on the operation conditions.
Fig. 7 Thermographs of the transformer after a. 30 minutes, b. 60
minutes, c. 90 minutes and d. 120 minutes for overload (116%)
conditions.

The next testing setups comprise theoretical problems in

the primary or secondary windings. Specifically for 87% of
rated load conditions after 30 minutes normal operation we
cutoff the transformer in L1 phase of the primary windings.
Immediately the temperature in the other two phases (L2 and
L3) drops down while the temperature in the L1 increases
normally. Figure 8 shows the related results. In the next step
after again 30 minutes normal operation we cutoff the L3 in
the secondary winding. The results were plotted in the
Figure 9 and are obvious that the "problem" transpires very
quickly.

60
L1
55 L2
L3
50

45

40
θmax (oC)

35

30

25

20

15
Fig. 10 The real thermographs in the 3 phases for asymmetric load a. at
10 the begin of the experiment and after b. 30 minutes, c. 60 minutes, d. 90
0 20 40 60 80 100 120 minutes and e. 120 minutes after continues operation.
Time of operation (min)
Fig. 8 The measured temperatures with break circuit in the L1 phase on
the primary.

22
 K. Karakoulidis, G. Fantidis and V. Kontakos/Journal of Engineering Science and Technology Review 8 (5) (2015) 19- 23

L1
80
L2
L3 5. Conclusions
70

60 Power transformers play a vital role in every electric system.

Owing to this fact a numerous diagnosis methods have been
θmax (oC)

50
developed. Infrared thermography is a helpful non-
40 destructive method for the conditioned monitoring and
diagnosis of transformers. In this work a small power
30
transformer tested under different scenarios with the aid of a
20 portable and high-resolution infrared thermographic system.
In any case the system is suitable to detect rapidly the
10
0 20 40 60 80 100 120 problems on the transformer operation.
Time of operation (min)
Fig. 11 The measured temperatures in the 3 phases for asymmetric
charge (L1 116%, L2 87%, L3 58%).

______________________________
References

1. Ladani Dhaval H., Sandeep A. Mehta, Pallav Gandhi, “Predictive 7. Bortoni E. C., Siniscalchi R. T., Jardini J. A., “Hydro generator
maintenance and modeling of Transformer”, International Journal efficiency assessment using infrared thermal imaging techniques”,
of Engineering Trends and Technology (IJETT) 4 (4), (2013), In: Power and Energy Society General Meeting, 2010 IEEE. IEEE,
1186-1189 2010. p. 1-6.
2. Kudelcik, Jozef, Miroslav Gutten, and Martin Brandt. 8. Barreira E., de Freitas V.P., Delgado J.M.P.Q. and Ramos N.M.M.,
“Development of electrical breakdown in transformer oil”, “Thermography Applications in the Study of Buildings
Advances in Electrical and Electronic Engineering 5, 1-2 (2011), Hygrothermal Behaviour, Infrared Thermography”, Dr. Raghu V
277-280. Prakash (Ed.), ISBN: 978-953-51-0242-7 (2012).
3. Cabanas M. F., Pedrayes González F., Melero M. G., Rojas García 9. Stipetic S., Kovacic M., Hanic Z., Vrazic M., “Measurement of
C. H., Orcajo G. A., Cano Rodríguez J. M., & Norniella J. G., Excitation Winding Temperature on Synchronous Generator in
“Insulation fault diagnosis in high voltage power transformers by Rotation Using Infrared Thermography”, IEEE Transactions on
means of leakage flux analysis”, Progress In Electromagnetics Industrial Electronics, 59 (5): 2288-2298 (2012).
Research, 114, (2011), 211-234. 10. Fantidis J. G., Karakoulidis K., Lazidis G., Potolias C., Bandekas
4. Mady Ibrahim, Attia Amani, “Infrared Thermography and D. V., “The study of the thermal profile of a three-phase motor
Distribution System Maintenance in Alexandria Electricity under different conditions”, ARPN Journal of Engineering and
Distribution Company”, 21st International Conference on Applied Sciences, 8 (11): 892 – 899 (2013).
Electricity Distribution, Frankfurt, 6-9 June 2011. 11. Fantidis J. G., Bandekas D. V., Karakoulidis K., Lazidis G.,
5. Bagavathiappan S., Lahiri B. B., Saravanan T., Philip J., & Potolias C., “The temperature measurement of the windings in a
Jayakumar T. “Infrared thermography for condition monitoring–a three-phase electrical motor under different conditions”, Gazi
review”, Infrared Physics & Technology, 60 (2013) 35-55. University Journal of Science Part A: Engineering and Innovation,
6. Chou Ying-Chieh, Yao Leehter. “Automatic diagnostic system of (2015), 3.2: 39-44.
electrical equipment using infrared thermography”. In: Soft 12. http://www.tercosweden.com (Transmission Line, Transformer &
Computing and Pattern Recognition, 2009. SOCPAR'09. Protection Laboratory)
International Conference of. IEEE, 2009. p. 155-160.

23