C I R E D 20th International Conference on Electricity Distribution Prague, 8-11 June 2009

Paper 0794

CIRED2009 Session 1 Paper No 0794

Fig. 1 Research field in laboratory

UNIVERSITY APPROACH FOR CIRCUIT BREAKER DEVELOPMENT

Hisatoshi IKEDA Katsuhiko HARADA Shinya OHTSUKA Kyushu Institute of Technology – JapanKyushu Institute of Technology – Japan Kyushu Institute of Technology - Japan ikeda@eke.kyutech.ac.jp harada@ele.kyutech.ac.jp ohtsuka@ele.kyutech.ac.jp Masahiro KOZAKO Masayuki HIKITA Kyushu Institute of Technology – JapanKyushu Institute of Technology - Japan kozako@ele.kyutech.ac.jp hikita@ele.kyutech.ac.jp

ABSTRACT Activities in the university were described for a circuit breaker development. To make experimental facilities minimized several new approaches were adopted. A accurate model investigation was focused in a transient analysis. A MEMS sensor application was aimed for a diagnosis. A small scale facility made an arcing phenomenon research possible. A new material application was achieved successfully.

INTRODUCTION A large scale experimental facility is needed for a circuit breaker (CB) development. The main facility is a high current source. A limited number of universities have a high current source, mainly by a capacitor bank, over the world. For a circuit interrupting test a voltage source is additionally needed. Two sources should be operated in a correct sequence. Furthermore a circuit breaker includes a lot of technical know-how. These aspects have made a university feel difficult. In Japan, the yearly growth rate of an electric power consumption stays at a low level less than 1 per cent. Most of manufacturers for CBs have been forced to reduce a number of research workers due to the rapid decrease of a demand for substation equipments. Manufacturers tend to require fundamental researches out sourced in a university. By considering the circumstances, we, the Funded Laboratory in the university by the Kyushu Electric Power Company, started research activities from April 2007 for fundamental phenomena related to the CB development.

MICRO CIRCUIT INTERRUPTION PHENOMENON LABORATORY We concentrated our research in four special fields which will not need a large scale facility. For minimizing our work we try to make experimental set up in a small size, and call the laboratory by the name of ‘Micro circuit interruption phenomenon’. The four special fields shown in Fig. 1.

Circuit switching phenomenon[1] A transient phenomenon during a circuit switching is a base for a CB development. Well developed software such as Electro-Magnetic Transient Program (EMTP) can

be used for circuit switching phenomena. More correct simulation is sometimes necessary. We investigated a precise transformer model to calculate the transient recovery voltage (TRV) after interrupting the transformer limited fault (TRF) current.

Arcing phenomenon It will be most important to understand arcing phenomena in CB developments. In order to minimize testing facilities a pulsed current is used. The electron density change along the arcing current is measured by the optical method.

MEMS sensor application technology[2] As the most frequent trouble mode of a CB is a mechanical one. As a diagnosis technology we investigated the MEMS (Micro Electro-Mechanical System) to measure acceleration characteristics during contact moving..

New material application technology[3] A circuit breaker has several thousand parts. One of the state-of-arts is a new material application. We investigated the nano-size coating on a polymer surface which is used for an outer insulation.

TRANSFORMER LIMITED FAULT The TRF is defined as a fault where all fault current is supplied through a transformer. To calculate a TRV

C I R E D 20th International Conference on Electricity Distribution Prague, 8-11 June 2009

Paper 0794

CIRED2009 Session 1 Paper No 0794

1

10

100

1000

Impe

danc

e(oh

m)

102

103

104

105

106

Frequency(Hz)

Fig. 2 FRA measurement of Leakage impedance

0.5

0.4

0.3

0.2

0.1

0.0

-0.1

Vol

tage

(V)

350300250200150100500

Time(μs)

(a) Experiment

(file trv95.45khz_gako.pl4; x-var t) v:P 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40[ms]-0.1

0.0

0.1

0.2

0.3

0.4

0.5

[V]

(b) Calculation

Fig. 3 TRV wave shape by current injection

YAG Laser

Pulsegenerator

Power sourceR=100Ω

Photo coupler

Triggercircuit

Photomultiplier

Spectralanalyzer

PLC

YAG LaserYAG Laser

Pulsegenerator

Power sourceR=100Ω

Photo coupler

Triggercircuit

Photomultiplier

Spectralanalyzer

PLCPLC

Fig. 4 Experimental layout

correctly at the TRF condition became one of the recent issues in a high voltage CB. Since the TRV after the TLF interrupting is decided by the transformer impedances, such as a leakage inductance, a stray capacitance and a resister of losses. In the EMTP calculation, a leakage inductance is given by one at the power frequency. It is known that a transformer inductance shows a frequency dependency. We investigated a precise leakage inductance.

FRA analysis of transformer impedance We conducted the FRA (Frequency Response Analyses) for several transformers. Fig. 2 shows the leakage impedance along frequency obtained from the FRA measurement at the 4kVA two winding transformer. It is shown that the leakage impedance decreases with the frequency increase from 100Hz to the resonance frequency. The stray capacitance was obtained from the resonance frequency and the leakage impedance.

TRV measurement by current injection The TRV was obtained by a current injection method where a TRF condition was expressed. The current injection TRV was calculated by EMTP with the leakage inductance and the stray capacitance obtained by the FRA Measurement. Fig.3 (a) and (b) show an experimental and a calculated TRV wave form respectively. By using the leakage impedance at the TRV frequency a good agreement was achieved. This means that the transformer should be expressed by the leakage impedance at the TRV frequency. We tried to get a resistance value from a impedance at the resonance frequency. It did not work well. Further research works is necessary for the TRV oscillation dumping by losses combined with influences of the frequency dependent impedance.

ELECRON DENSITY MEASUREMENT As electron density characteristics will be related with arcing conditions, to obtain an electron density decay along the arcing current decreasing, we constructed an experimental set-up of a high frequency current source and an optical measuring system of the Thomson

Scattering spectroscopy with a YAG laser.

Experimental layout The layout is shown in Fig. 4. The current source is made by 2.5 micro Farad 15kV capacitor, which can supply a current of 1500A peak and 25 micro second width to the arc initiated by a breakdown or a fuse wire between contacts in the chamber.

Sequence control In order to synchronize the laser radiation at the required

point on the arcing current, the arcing current is started by a predetermined delay time from the laser excitation starting. The laser radiation starts about 200 micro second

C I R E D 20th International Conference on Electricity Distribution Prague, 8-11 June 2009

Paper 0794

CIRED2009 Session 1 Paper No 0794

0.4

0.2

0.0

-0.2

Optical Power(a.c.)

Time (10μs / div)

20

15

10

5

0

Applied Voltage,V(kV)

4

2

0

-2

-4

Gate Voltage,V(V)

20

15

10

5

Output Voltage of Laser(V)

(1) Trigger output of Laser (2) Gate voltage of SCR (3) Applied voltage (4) Output of Photo multiplier

Fig. 5 Measured waveform of signals

Oscilloscope

PowerSource

MEMS

圧電型センサ

Amplifier

Oscilloscope

PowerSource

MEMS

圧電型センサ

Amplifier

Fig. 6 CB mechanical model

-1200

-1000

-800

-600

-400

-200

0

200

400

600

800

Acceleration[G]

Time[10ms/div]

圧電型加速度センサMEMS加速度センサ

Fig. 7 Measured acceleration number

Table 1 one cycle of pollution test

after the excitation. Fig. 5 shows measured current and laser radiation signal. The laser radiation is synchronized at the decay point. We need some more works to make a radiation precisely on an arcing. Experiments with different gases and electrode materials are expected.

CONTACT MOVEMENT ACCELERATION MEASUREMENT A MEMS acceleration sensor is widely used in an auto-car. We used the MEMS sensor to measure acceleration characteristics during the contact moving.

CB mechanical model Fig. 6 shows the CB mechanical model on which the MEMS sensor is equipped on the moving contact. The CB mechanical model was designed to operate by charged springs. It moves 40 mm long at maximum and generates about 1000 G acceleration number on the

moving contact.

Measured acceleration characteristics Fig. 7 shows the acceleration number along with the time during the contact opening. The number has a positive peak at the contact movement start and a negative peak at the stop. For the purpose of comparison, the measured characteristic by a conventional standard pieso-electric accelerometer is shown in the same figure. Although the MEMS sensor has a frequency limitation up to 250Hz, it is expected that the use of the MEMS sensor for the purpose will be possible in some limitations.

NANO COATING FOR OUTER INSULATION A polymer outdoor insulator has many advantages to a porcelain insulator, such as a light weight and a thin configuration. The hydro-phobic performance is also good for preventing a leakage current on the surface. It is feared that the hydro-phobic performance is deteriorated by the surface contamination during a long time usage. We investigated the nano-size coating on the polymer surface to prevent the deterioration.

Development of pollution test method We developed a new pollution test method suitable for a polymer insulator, based on an estimation that the polymer surface is deteriorated mainly by un-solvent materials. Table 1 shows a sequence of the pollution test, which is almost one hour sequence with an un-solvent spray, a drying process and a clean water spray.

(2)

(1)

(3)

(4)

C I R E D 20th International Conference on Electricity Distribution Prague, 8-11 June 2009

Paper 0794

CIRED2009 Session 1 Paper No 0794

(a) Non coating (b) Sample A

(c) Sample B (d) Sample C

Fig. 8 Hydrophobic condition on surface after 81 cycles

Improvement by coating The polymer surface was treated by the chemical vapour deposition technology, which gives a nano-size coating. Three types of precursor material were used. Those are samples A; HMDSO=Hexa-methyl-di-siloxane, B; OMCTSO=Octa-methyl-cyclo-tetra-siloxane and C; C4F8=Per-fluoro-cyclo-butane. Fig. 8 shows hydro phobic conditions of sample surfaces after 81 cycles.

CONCLUSIONS Several possibilities are shown to be achieved by universities for the CB development. 1) Switching transient phenomena such as a TRV

need more detailed and accurate simulation models for a numerical calculation.

2) A MEMS sensor application will be one of the topics for diagnosis of CB conditions.

3) An arcing phenomenon can be investigated by a small size testing facilities.

4) A new material application such as a nano technology will be one of the most important field for universities.

REFERENCES [1] M Thein, H. Ikeda, K. Harada, S. Ohtsuka, M.

Hikita, E. Haginomori and T. Koshiduka, 2008, "Investigation of transformer model for TRV calculation by EMTP", the 6th International work shop on high voltage engineering IWHV 2008/Kyoto, Paper No. ED-08-117/SP-08-32/HV-08-46, 49-52

[2] H. Ikeda, K. Makisaka, K. Harada, S. Ohtsuka, M. Hikita, K. Iwamoto and T. Nagata, 2008, "Preliminary investigation for application of MEMS acceleration sensor to SF6 gas circuit breakerr", Proceedings of 2008 International Conference on Condition Monitoring and Diagnosis, vol.2, 873-876

[3] H. Higashikoji, M. Motoyama, S. Ohtsuka, M. Hikita, H. Ikeda, J. Gavillet, G. Lavel, E. Da Silva and M. Frechette, 2008, "Evaluation of pollution characteristic of polymer insulating material with surface nano-coating by a new automatic artificial pollution acceleration test", Proceedings of 2008 IEEE Conference on Electrical Insulation and Dielectric Phenomena-(CEIDP2008), 212-215