Yoshino T. et al.: Experimental Study on Rotary Arcs in CO2 Model Gas Circuit Breaker

Plasma Physics and Technology

2015, 2, 2, 211-214

Experimental Study on Rotary Arcs in CO2 Model Gas Circuit Breaker

Yoshino T.1, Kawano H.1, Mori T.1, Miyazaki K.1, Fujino T.2

1TOSHIBA Co., 2-1 Ukishima-cho Kawasaki 210-0862, Japan, tomoyuki1.yoshino@toshiba.co.jp 2University of Tsukuba, 1-1-1 Tennnodai Tsukuba 305-8577, Japan, tfujino@kz.tsukuba.ac.jp

The present study has examined the influence of rotary arcs with a permanent magnet in CO2 gas circuit

breaker under conditions of low and high current interruption. The rotary arc extremely increases extinc-

tion peak and decreases arc conductance of 200ns before current zero in the case of low current condition,

because the rotary arc generates the rotation gas velocity and enhances turbulent energy transport, when

there is no pressure rise of thermal puffer in the case of low current.

Keywords: rotary arc, CO2, GCB

1 INTRODUCTION

SF6 gas has been used in most high-voltage

transmission and distribution network because

of its remarkable arc quenching properties,

dielectric insulation and ease in handling.

However, the SF6 gas has a potential of sig-

nificant environmental impacts if it leaks into

the atmosphere. CO2 gas is expected as a one

of the alternative solutions to SF6 gas [1].

The arc quenching properties of CO2 gas is

inferior to that of SF6 gas. One of the ideas for

the improvement of current interruption capa-

bility in Gas Circuit Breakers (GCBs) is to

rotate arcs by electromagnetic force [2-5]. By

the rotation, the arc is expected to be cooled

by fresh gas, and hot gas is expanded rapidly

and raises gas pressure of a thermal puffer.

But rotary arcs in CO2 gas have not been in-

vestigated in detail.

Authors have been conducted current interrup-

tion test with and without a magnet in CO2

gas, and examine the feasibility of application

of rotary arcs to CO2 gas circuit breaker.

2 EXPERIMENTAL SETUP

Figure 1 shows a schematic figure of a model

chamber used in this study. The interruption

chamber has a movable arcing contact and a

stationary arcing finger. The stationary arcing

finger is set in a thermal puffer which does not

have a mechanical puffer. A permanent mag-

net is set around the stationary arcing finger

concentrically. A disc electrode, called an arc-

runner, is set at the front of the stationary arc-

ing fingers with which the arc-runner is con-

nected in parallel electrically.

When the movable arcing contact is driven

and detached from the stationary arcing fin-

gers, an arc ignites between the stationary arc-

ing finger and the movable arcing contact, and

then the arc is expected to transfer from the

arcing finger to the arc-runner.

The permanent magnet is of Nd-Fe-B. The

magnet is coated with aluminum, and the

coated magnet is set in a case made of PTFE

to protect the magnet from hot gas or arc itself.

The maximum magnetic flux density of radial

component on the surface of the arc-runner is

about 0.14T.

Interruption current is supplied by a short-

circuit generator, and its magnitude is set to

values of 4.5 and 40kA peak. The arc voltage

is measured by a voltage divider, and the cur-

rent is measured by Rogowski coil [6]. Pres-

sure in the thermal puffer is measured by a

pressure transducer as shown in Fig. 1.

The tank is filled with CO2 gas. The gas pres-

sure is 0.5MPa-abs. Experiments have also

performed for a chamber without a magnet for

comparisons.

Pressuretransducer

Thermal puffer(About 500cc)

Arcingfinger

Arc-runner

Nozzle

Movablearcing contact

PermanentMagnet

Tank

f16

Fig.1: Schematic figure of model chamber

Yoshino T. et al.: Experimental Study on Rotary Arcs in CO2 Model Gas Circuit Breaker

212

3 RESULTS AND DISCUSSIONS

3.1 LOW CURRENT CONDITION

(4.5kA peak)

Figure 2 shows arc voltage waveforms with

and without the magnet in low current condi-

tions of about 4.5kA peak. The arcing time is

about 26ms, and current zeros exist at around

8ms and 17ms. High frequency oscillation is

observed on the arc voltage in the case with

the magnet, while it cannot be observed with-

out the magnet. Arc voltage with the magnet is

higher than that without in almost all time dur-

ing arc discharge. In particular, the extinction

peaks increase drastically at every current ze-

ros by applying the magnetic field.

Figure 3 shows the pressure rise in thermal

puffer with and without the magnet. The pres-

sure rise is no more than several tens of kPa

for both cases. It seems that there is little gas

flow between contacts in low current condi-

tions.

-10

-8

-6

-4

-2

0

2

4

6

8

10

-1

-0.8

-0.6

-0.4

-0.2

0

0.2

0.4

0.6

0.8

1

-6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28

Cu

rre

nt,

kA

Arc

vo

ltag

e, k

V

Time, ms

1

0.8

0.6

0.4

0.2

0

-0.2

-0.4

-0.6

-0.8

-1

Arc

vo

ltag

e [k

V]

10

8

6

4

2

0

-2

-4

-6

-8

-10

Cu

rren

t [k

A]

-4 0 4 8 12 16 20 24 28Time [ms]

With magnet (26.6ms)

Without magnet

(25.8ms)

Current

Fig.2: Example of arc voltage waveforms with and

without magnet (4.5kA)

100

50

0

-50

Pre

ssu

re r

ise

[kP

a]

Cu

rren

t [k

A]

-10 0 10 20 30 40Time [ms]

With magnet (26.6ms)

Without magnet(25.8ms)

Current

10

5

0

-5

Fig.3: Pressure rise in thermal puffer with and

without magnet (4.5kA)

Figure 4 depicts current-voltage characteristics

with and without the magnet under conditions

of various arcing times. The extinction peaks

are clearly observed in every arcing time con-

ditions when the magnetic field is applied.

However, it can be found from Fig. 4 (a) that

the extinction peak cannot be observed with-

out the magnet. This means that turbulent en-

ergy transport is weak around the current zero.

It seems that the gas flow keeps laminar and

does not become turbulent without magnet.

The rotary arc generates the rotation gas ve-

locity, and increases turbulent energy transport.

Therefore, arc cooling around current zero is

enhanced drastically, and the extinction peak

arises by applying the magnetic field, as

shown in Fig. 4 (b).

1

0.8

0.6

0.4

0.2

0

Arc

vo

ltag

e [k

V]

10-3 10-2 10-1 100 101 102 103 104

Current [A]

11.0ms

16.6ms

25.8ms

19.6ms

8.2ms

7.5ms

1

0.8

0.6

0.4

0.2

0

Arc

vo

ltag

e [k

V]

10-3 10-2 10-1 100 101 102 103 104

Current [A]

9.7ms

9.4ms

14.2ms

17.6ms

26.6ms

(a): without magnet

(b): with magnet

Fig.4: Current-voltage characteristics under

various arcing time conditions (4.5kA)

Yoshino T. et al.: Experimental Study on Rotary Arcs in CO2 Model Gas Circuit Breaker

213

3.2 HIGH CURRENT CONDITION

(40kA peak)

Figure 5 shows arc voltage waveforms with

and without the magnet in high current condi-

tions of about 40kA peak. The arcing time is

about 21ms. There are few frequency oscilla-

tions in arc voltage for both cases of with and

without the magnet. The voltage rising by ap-

plying the magnetic field is observed in first

half-cycle before 12ms. The extinction peak

with the magnet also increases around 12ms.

On the other hand, the voltage rising including

extinction peak by applying the magnetic field

cannot be observed in second half-cycle after

12ms.

Figure 6 shows the pressure rise in thermal

puffer with and without the magnet. The max-

imum value of pressure rise with the magnet

increases to the same extent as that without.

There is the influence of magnet for the pres-

sure rise at shorter arcing time before 12ms.

1.2

0.8

0.4

0

-0.4

-0.8

-1.2

Arc

vo

ltag

e [k

V]

60

40

20

0

-20

-40

-60

Cu

rren

t [k

A]

0 4 8 12 16 20 24Time [ms]

With magnet (19.6ms)

Without magnet(21.2ms)

Current

Fig.5: Example of arc voltage waveforms with and

without magnet (40kA)

1.2

1.0

0.8

0.6

0.4

0.2

0

-0.2

Pre

ssu

re r

ise

[P.U

.]

Cu

rren

t [k

A]

-10 0 10 20 30 40Time [ms]

With magnet(19.6ms)

Without magnet(21.2ms)

Current

100

50

0

-50

Fig.6: Pressure rise in thermal puffer with and

without magnet (40kA)

Figure 7 illustrates the current-voltage charac-

teristics with and without the magnet under

conditions of various arcing times. The extinc-

tion peak at the shorter arcing time of 10 to

15ms becomes higher by applying magnetic

field. However, there is no effect of the mag-

net at the longer arcing time over 17ms.

It can be found from Fig. 7(b) that voltage os-

cillations occur in arc discharge below several

kA of current by applying magnetic field, but

there are no oscillations above 10kA. This is

likely due to the difference in arc structure. No

oscillation of arc voltage occurs in high cur-

rent duration above 10kA because the cylin-

drical stable arc structure is formed. The arc

discharge becomes spiral structure and spiral-

shaped arc yields the destabilization of arc

voltage when the arc current is reduced below

several kA.

1.4

1.2

1

0.8

0.6

0.4

0.2

0

Arc

vo

ltag

e [k

V]

10-3 10-2 10-1 100 101 102 103 104 105

Current [A]

12.0ms

16.8ms

21.8ms

11.6ms

6.7ms

21.2ms

Arc

vo

ltag

e [k

V]

Current [A]

10.4ms

15.0ms

8.4ms

17.5ms

6.5ms

10-3 10-2 10-1 100 101 102 103 104 105

1.4

1.2

1

0.8

0.6

0.4

0.2

0

19.6ms

destabilization

stabili-zation

(a): without magnet

(b): with magnet

Fig.7: Current-voltage characteristics under

various arcing time conditions (40kA)

Yoshino T. et al.: Experimental Study on Rotary Arcs in CO2 Model Gas Circuit Breaker

214

3.3 COMPARISON OF ARC

CONDUCTANCE

Figure 8 shows the arc conductance of 200ns

before current zero (G-200) with and without

the magnet under low and high current condi-

tions. The G-200 is reduced by applying mag-

netic field for every current and arcing time

conditions. In the case of low current condi-

tion, the G-200 is significantly reduced at

shorter arcing time around 10ms. G-200 with

the magnet becomes also lower in the case of

high current conditions.

Arc

co

nd

uct

ance

[m

S]

Arcing time [ms]

5 10 15 20 25 30

100

10

1

0.1

0.01

With magnet

Without magnet

(a): Low current condition (4.5kA)

Arc

co

nd

uct

ance

[m

S]

Arcing time [ms]

5 10 15 20 25 30

With magnet

Without magnet

(b): High current condition (40kA)

Fig.8: Arc conductance of 200ns before current

zero in high and low current conditions

As mentioned above, the CO2 gas flow is dif-

ficult to become turbulent because CO2 gas

has the small mass number compared to the

SF6 gas. The rotation gas velocity by applying

the magnetic field increases the turbulent en-

ergy transport. Therefore, the influence of ro-

tary arcs is greatly enhanced under the condi-

tion of no pressure rise in thermal puffer such

as low current condition in gas circuit breaker

with CO2 gas.

4 CONCLUSIONS

The present study has examined the influence

of rotary arcs with a magnet in CO2 gas under

conditions of low and high current interruption.

Conclusions are as follows.

The rotary arc extremely increases extinction

peak and decreases arc conductance of 200ns

before current zero in the case of low current

condition in CO2 gas. The CO2 gas flow is

difficult to become turbulent because of the

small mass number. The rotary arc by apply-

ing magnetic field generates the rotation gas

velocity and turbulent energy transport, when

there is no pressure rise in thermal puffer.

Therefore, it is expected that the current inter-

ruption capability of CO2 gas is drastically

improved by the magnet in low current condi-

tion of CO2 gas circuit breaker.

REFERENCES

[1] Uchii T, Majima A, Koshizuka T, Kawano H,

Thermal interruption capabilities of CO2 gas and

CO2 based gas mixtures, In: Proc. of 18th Int. Conf.

on Gas Discharges and their Applications, 2010,

78-81.

[2] Sasao H, Hamano S, Wada Y, Hasegawa H,

Kobayashi N, Development of a magnet-assisted

autopuffer GCB, IEEE 90 WM 158-6 PWRD,

1990.

[3] Young K A, Spencer J W, Optical diagnostics

for studying a rotating arc in SF6 gas, In: 14th Int.

Conf. on Gas Discharges and their Applications,

Vol.II, 2002, 353-356.

[4] Ennis M G, Investigation of Fundamental Pro-

cesses Affecting the Behaviour of Electric Arcs in

Electromagnetic Interrupters, Thesis for PhD at

Univ. of Liverpool, 1996.

[5] Mori T, Spencer J W, Humphries J, Jones G R,

IEEE Trans. Power Delivery 20 (2005) 765-771.

[6] Smeets R P P, Kertesz V, Nishiwaki S, Koshi-

zuka T, Suzuki K, Short-line fault interruption as-

sessment of high-voltage circuit breakers by

means of current zero analysis, In: CIGRE A3

Colloquium, Sep. 15-16, 2003.