Energy and Power Engineering, 2013, 5, 838-841
doi:10.4236/epe.2013.54B161 Published Online July 2013 (http://www.scirp.org/journal/epe)
Analysis for the Defect of Chromatogram in 35kv
Oil-immersed Electric Reactor
Yongdong Li, Qingda Meng, Zheyuan Zhao, Po Yang, Yuzheng Wang
Maintenance of northern Hebei Province Electric Power Co. Ltd, Substation maintenance center, Beijing, China
Received April, 2013
ABSTRACT
A defect of chromatogram in a 35 kV oil-immersed three-phase one-piece electric reactor has been found in a 500kV
substation. The initial analysis suggests that the defect is caused by the loose connection of the iron core magnet shield
(magnet shield layer of iron core or earthling top). Examination of winding DC resistance, DGA and other relevant
examinations were carried out to investigate the defect reasons, which is demonstrated by the dismantlement. In-depth
study is taken and corresponding prevention measures are put forward in the paper.
Keywords: Oil-immersed Electric Reactor; Chromatogram Defect; Magnet Shield Layer; Partial Discharge
1. Introduction
35 kV shunt reactor is suitable for the high voltage paral-
lel circuit of 35 kV power system and high-speed rail
power supply system. It transfers perceptual reactive
power to system through the main transformer, prevents
the voltage of light load from increasing and maintains
the voltage stability of the power transmission system. It
increases the saving energy and power factor of the elec-
tric power system. As the additional compensation
equipment for the power system stabilization，it plays an
important role in adjusting the power grid, affecting the
safe and stable operation of power grid [1]. Oil-immersed
reactor has gradually been widely used because of its
running stability, low dissipation, small floor, no leakage
pollution, free maintenance function and so on.
2. Failure Analysis
On April 20, 2012,Reactor insulation oil chroma-
tographic experiments of the 500 kV substation of chang-
ping 321 shows the acetylene content is 128.1 μL/L, the
hydrogen content is 259.8 μL/L, beyond the rules, rapid
growth in acetylene content. On April 21, 2012 this re-
actor was out of service. This reactor is Oil-immersed
self-cooled reactor and its factors: Japan's nissin produc-
tion motor co., LTD. Nominal voltage: 35 kV, Rated ca-
pacity: 30 MVA, Model: RMO-VCA, Date of production:
June 1995, Put into operation date: December 10th, 1995;
Equipment running time: nearly 17 years.
Since 2005, as the aging problems, the total hydrocar-
bon content of this reactor is always over proof. (regula-
tion value is 150 ul/l), the data of the test oil chroma-
togram shows in Table 1.
On 20th April, 2012, in the experiment of oil chroma-
tographic, there is a large amount of H2 and C2H2 in the
oil. According to use of the three-ratio method in stan-
dards, it plays an important role which is elaborated in
the diagnosis of actual failures of the transformer, elec-
tric reactor, oil-filled electrical equipments and related
equipments. The result of oil chromatographic analysis
and calculation is shown below in the Table 2[3,4].
Table 1. Oil chromatographic data.
H2CH4C2H6C2H4C2H2CO CO2 hydrocarbon
content
Test
date
1 32.5113.33 0 78.2 768.9 148.8 2005.7.7
1.141.4118.34 0 115.4 801.4 163.7 2006.1.5
2.747.8127.94.10 119.6 732.1 179.7 2007.4.3
1.450143.84.50 109.9 865.9 198.3 2008.4.10
1.551.8169.94.90 111.4 1035.8 226.6 2009.4.10
1.456.5150.64.80 113.6 902.8 211.9 2010.2.2
1.461.5163.15.50 117.9 826.2 230.1 2011.1.20
1.154.8150 5.50 107.3 874.8 210.3 2011.7.27
2.462.3183.86.50 113.9 986.6 252.6 2011.10.20
259.8 71.918123.9128.1109.1 886.5 404.9 2012.4.20

Y. D. LI ET AL. 839
Table 2. Calculated data with three-radio method.
C2H2/C2H4 CH4/H2 C
2H4/C2H6
5.360 0.277 0.132
The encoding of the three-ratio method is 2, 0 and 0.
According to preliminary inspection, it is low energy
discharging which may be caused by spark discharge in
oil between potential difference or spark discharge be-
tween floating potential [5].
Nissin electric co., LTD and we made an initial evalu-
ated depending on standards values which are stipulated
by Electrical collaborative research, gas anomaly diag-
nosis diagrams and characteristic of the gas analysis as
shown below:
1) As the absolute value of the CO content is below
the standards value of 300 ppm, CO2/CO= 8.1>3 and the
phenomenon of increasing of CO content caused by
overheating of the insulating is never happened, the
possibility is very low which is the wind and insulating is
overheating in sudden. In addition, it takes a high risk
that the failure is may happened around the insulating.
2) In general, when the equipment is overheating sud-
denly (C2H2/C2H4≦0.2),but the defect C2H2/C2H4＝5.3
> 0.2 here, this reason can be considered as partial dis-
charge in the low energy area.
3) The value of acetylene(C2H2)content (128.1 ppm)is
far beyond the evaluation abnormal value(5 ppm) which
is written by the book of “maintenance management of
oil immersed transformer” in Japan. The partial dis-
charge leads to the increasing of（C2H2）.
In accordance with the analysis of the results above, it
can be estimated that the rise of C2H2 content and H2
content is caused by that partial discharge is happened in
the part of bare metal of surrounding insulating (Core or
the magnetic shielding grounding side).
3. Inspection of the Equipment
Disintegration
Electric reactor should be dismasted and disintegrated to
check in the factory.
1) Check and test in the factory
When the equipment returning to the factory, it should
take a test and measure in Reactor winding dc resistance
of ontology, insulation, reactance, consume, spilt ratio
and so on. Compared with the result of acceptance test of
elector equipment, the experimental result is normal.
2) Introduction of disintegration
As this reactor is integrated structure, the component is
not removed. Bottom of reactor can't be cut. The test of
disintegration should be taken by cutting the bottom of
the reactor. The structure is shown below as Figure 1.
When testing the A core magnetic shielding, we find
out that around the Aluminum foil stickers, it appeared
the discharge stains on the top of the seventh layer and
the eighth layer, the bottom of the eighth layer and the
ninth layer of A core magnetic shielding. The flowchart
of dismantling the reactor is shown below as Figure 2.
The aluminum foil and the grounding potential separa-
tion of copper is shown as below in Figure 3.
4. Failure Confirm
According to the state of equipment disintegration, we
know that the cause of failure is:
1) The shape of shielding copper in the core magnetic
shielding is banding, both sides of it is fixed on the alu-
minum foil. In the long running process, as the repeating
operation of reactor, the winding vibration results in the
contact area of aluminum foil and copper board endures
the stress. The contact area of aluminum foil and copper
board adopts the manual welding, affecting by the human
factors. When the stress is applied to this area, the con-
tact of aluminum foil and copper board will be broken
and produce the floating electrode.
Figrue 1. Diagram of inner structure of the reactor.
Figure 2. Flowchart of dismantling the reactor.
Figure 3. Photo of discharge trace of magnet shield layer.

Y. D. LI ET AL.
840
The potential difference between floating electrode
and grounding copper plate will discharge, results in a
large number of characteristic gas(H2, acetylene).[7]
2) According to the site situation，we use the simulated
calculation to describe the electric field distribution of
this area. (Figure 4) as we can see from the Figure 4, the
electric field intensity of the broken area between the
aluminum foil and copper board is the highest. So the
characteristic gas is produced by the electric discharge.
5. Process Improvement and
Troubleshooting
5.1. Process Improvement
Now we improve the original welding process from the
manual welding to the mechanical pressure welding. This
method decreases the manual factors of the production
process. The comparison of production process shows
below:(Figure 5)
5.2. Fault Handling
5.2.1. Core Magneti c Sh ield Operation Process
1) Shearing the aluminum foil according to the design
value.
(2) Welding Braided wire on the cooper board. (Fig-
ure 6)
3) Pressure welding the aluminum foil and cooper
board by the welding spot.(Figure 7 )
Figure 4. Suspension electric field strength of aluminum
foil.
Figure 5. Photo of contrast between before and after the
improvement.
Figure 6. Process of Core magnet shielding.
Figure 7. Process of Core magnet shielding.
Figure 8. Process of Core magnet shielding.
4) Making sure there is no space between the pressure
welding spots. Both five gears between the spots must be
corresponding. Using the aluminum foil stickers to roll
on the insulating cylinder. This prevents it from the loop
current. The whole process shows below:(Figure 8)
5.2.2. The Spare Parts Replacemen t and Factory Test
We replace the old reactor seal ring, valve, thermometer
and gas capsule by the new one and then spray new
paint.
According to the factory test, the experiment required
standard is: measurement of winding resistance, meas-
urement of dielectric loss as well as electrostatic capacity,
reactance test, loss measurement, noise test, measure-
ment of insulation resistance, withstand voltage test, par-
tial discharge test, BCT test, temperature rise test, gas
analysis and all the experiment data is in the normal
range.
6. Conclusions
The main reason that this equipment is on failure is
caused by aluminum foil stickers on core magnetic
shielding and open wild of contact area of copper board.
This reactor should be repaired well. The issue also
should be paid attention as such problems. What's more,
the ways to improve it should be summed up as well. It's
essential to take preventive actions when such problems
happen. There some suggestions to improve it as shown
below:
1) We can quickly make a judgment about failure rea-
son if the same type of this kind of products is in the such
situation. There is a better way to improve the efficiency
of maintenance.
2) When such product is on the design period, we

Y. D. LI ET AL.
841
should consider sufficiently that processing craft of the
part which need to be connected tightly and take the right
measurement of soldering.
3) For such same type products in the same factory
which have been already put into operation, the ways
which are core magnetic shielding of manual welding
process. we should pay a high attention about the status
of operation. When such problems are happened, we
should report without delay and return it for repairing in
factory to avoid increasing the risk if necessary.
REFERENCES
[1] Joon-Ho Choi, Jae-Chul Kim. Advanced Voltage
Regulation Method at the Power Distribution
Systems Interconnected with Dispersed Storage
and Generation Systems. IEEE Transactions on
Power Delivery, Vol. 15, No. 2, 2000, pp.
691-696.
[2] M. A. Farrag, M. M. El-Metwally and M. S.
El-Bages, “New Model for Distribution System
Planning,” International Journal of Electrical
Power and Energy System, Vol. 21, No. 7, 1999, pp.
523-531.
[3] G. Celli and F. Pilo, “MV Network Planning Un-
der Uncertainties on Distributed Generation
Penetration,” IEEE power Engineering Society
Summer Meeting. 2001.
[4] B. F. Hampton and R. J. Meats, “Diagnostic
Measurement at Pt. C, in gas insulated substa-
tions, IEE Proceedings, Vol. 135, No. 2, 1988, pp.
137-145
[5] IEC60071-2, Insulation Coordination, Part 2,
Application Guide, IEC Publication, 1996.
[6] R. Bozzo, F. Guastavina, G. guema, “PD Detec-
tion and Localization by Means of Acoustic
Measurements on Hydrogenerator Stator Bars,”
IEEE Transactions on Dielectrics and Electrical
Insulation, Vol. 2, No. 4, 1995, pp. 660-666.
[7] G. B. Nicholas, “Papadopoulos Michael P. Optimal
Feeder Routing in Distribution System Planning
Using Dynamic Programming Technique and Gis
Facilities,” IEEE Transactions on Power Delivery,
2002, Vol. 17, No. 1, pp. 242-247.