XXVIIth Int. Symp.

on Discharges and Electrical Insulation in Vacuum – Suzhou - 2016

Vacuum Arc and Characteristics of Arc Voltage

in Curved Surface Contact
Wenlei Huo, Jianwen Wu and Bowen Jia
School of Automation Science and Electrical Engineering, Beihang University, Beijing 100191, China

Abstract- In this paper, we proposed two novel contact only through the effective area [8].
structures for vacuum interrupter. The sine surface In order to increase the area of diffusion and
contacts (sin-ruled surface contact and sin-revolution absorption, two novel contact structures are proposed in
surface contact) were experimental studied in the this paper. Vacuum arc appearance and voltage
condition of 3mm opening gap. Experiments were carried characteristics are investigated.
out in the oscillating circuit with the contact diameter
being 40mm, the anode contact material being CuW80 II. EXPERIMENT CIRCUIT
alloy and cathode contact being CuCr25. The vacuum arc
appearance and voltage characteristics of butt contact, A. The Experimental system
sine-ruled surface contact and sine-revolution surface
The experiment system is shown in Fig.1. A
contact were analyzed contrastively. Experimental results
single-frequency oscillation circuit is used to study the
showed that the diffusion path of cathode spots was along
vacuum arc. Before triggering the thyristor, charge
the curved surface. The movement of cathode spots did not
voltage of capacitor C1 and capacitor C2. The voltage of
be inhibited by the curved surface. Compared to butt
C1 is higher than that of C2. Trigger VT2 and open the
contact, arc voltage and voltage noise of curved surface
vacuum interrupter to ignite a direct-current arc. Then
contacts were relatively low. The losing of plasmas was a
the current is injected by triggering thyristor VT 1, when
main reason why voltage noise generated. In arc stability
the contact is fully open. The VT2 is switched off due to
and diffusion stage, plasmas and metal vapor were mainly
negative voltage. By adjusting capacitor C1 and inductor
absorbed by curved surface between contacts but a few of
L1, we can achieve current frequency. Capacitor C0 and
plasmas spread outside of arc gap, for which the service
resistor R0 are used to adjust the rate of rise of the
life of the vacuum interrupter chamber was effectively
recovery voltage parameters.
improved.
A high-speed video camera with a speed of 500,000
frames/s and an exposure time of 2 μs is used to record
I. INTRODUCTION
arc appearance. A multi-channel isolated oscilloscope
The vacuum interrupter (VI) has the advantageous YOGOKAWA DL750 is used to record the arc voltage
characteristics: small volume, light weight, and large and current. All the arc voltage data used in the
current interrupting ability. Due to their super volt-ampere characteristics are the voltage drop on VI.
characteristics, vacuum switches have been applied VT1 L1

widely in power distribution system [1]. There are two Voltage

VI
kinds of electrodes in vacuum interrupters. One VT2 R0
Measure
Anode High Speed

generates transverse magnetic fields [2], [3] and the Cathode Camara

other axial magnetic fields [4], [5]. Charge

R1

Electrode diameter and material are important fields Unit C1

C0 Operation
PC

in vacuum arc research. Liu [6] found the interruption +

Mechanism

capacity of vacuum interrupters with the slot-type AMF VD1 C2 R2 Current

Measrue
electrodes is related to the ratio of the anode diameter to
the gap length. It is indicated by the experiment results Measure and Control System

that breaking capacity of vacuum interrupter and

properties of intermediate-frequency vacuum arc are Fig. 1. Magnetization as a function of applied field
closely related to the electrode diameter and material.
B. Curved Surface Contact
With the electrode diameter increasing, arc voltage
Two kinds of sine surface contacts are used in the
decreases gradually, and diffuse arc tends to be stable,
experimental system to study the characteristics of
for which the breaking capacity is better [7].
vacuum arc, as sine-ruled surface contact shown in
Big electrode diameter and large area of high AMFs
Fig.2 and sine-revolution surface contact shown in Fig.3.
ensure better breaking performance. A new design
The amplitude of sinusoidal curve is 3mm. Experiments
criterion for diffuse arcs is established by using
are carried out with specially sealed VI in the
concepts such as the effective area and the effective
experimental system. The electrode parameters are
current density when arc currents are assumed to flow
This work was supported by the National Natural Science Foundation
of China (51377007), the Specialized Research Fund for the Doctoral
Program of Higher Education (20131102130006) and the
Fundamental Research Funds for the Central University.

978-1-4673-9780-3/16/$31.00 ©2016 IEEE

listed in Table1. The normally present metal shield of The parameters fitting graph is shown in Fig.4 at
VI is removed in order to observe the vacuum-arc capacitance of C1 5712 μF and inductance of L1 2.55 μH.
appearance, and this may have an effect on the arc The experimental curve used in fitting is obtained under
motion. However, as the distance between the glass and contact closure. Accordingly, the frequency is
contact is relatively long, it should have little impact on calculated with ignoring the arc resistance. The
the arc motion. experimental frequency nearly 1.2 kHz with contact
interrupting is recorded by oscilloscope.
12

10
Experimental curve
8

Current (kA)
Fitting curve
4

-2

-4
0 0.1 0.2 0.3 0.4 0.5
Fig. 2. Structure of sine-ruled surface contact Time (ms)

Fig. 4. Parameters fitting of experimental system

III. RESULTS AND DISSCUSSION

A. Arc volt-ampere characteristics of butt contact

Arc voltage and arc current are recorded at frequency
1.09 kHz and peak current 13.54 kA, as shown in Fig.5.
Typical vacuum arc appearance in butt contact is shown
in Fig.6,
Fig. 3. Structure of sine-revolution surface contact
20 300

18
t1 t2 t3 t4 t5 t6 t7 t8 200
TABLE 1. ELECTRODE PARAMETERS OF THE VI 16 100

Contact Cathode Anode Gap Air 14 0

Arc Current (kA)

12 -100

Arc Voltage (V)

diameter material material length pressure 10
142.2μs
Arc Voltage -200

8 205.0μs -300
40mm CuCr25 CuW80 3mm 5×10-5Pa 6 57.4μs
Arc Current -400
254.2μs
4 -500
311.6μs
Contact resistances are tested separately before and 2
16.4μs 360.8μs
-600

0 -700
after experiments, the results are listed in Table2. From -2
442.8μs
-800
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5
the contact resistance, we can find that the resistance of Time (ms)

the three kinds contact change a little after almost 30 Fig. 5. Current and arc voltage of butt contact (Ipeak=13.54kA)
times separate experiments. The resistances of curved
surface contacts are close to that of butt contact. Besides,
the results are obtained at contact pressure 1800N.
TABLE 2. CONTACT RESISTANCE
Contact Contact resistance(μΩ)
pressure Before After
(N) experiment experiment

Butt contact 1800 10 10

Sine-ruled surface
1800 11 12
contact

Sine-revolution
1800 10 12
surface contact

C. Experimental system parameter fitting Fig. 6. vacuum arc appearance in butt contact

Due to the actual measurements of system parameters As can be seen from figure 5, voltage noise occurs
inaccuracy, ant colony algorithm is applied to fit the from t3 to t5, and its time span is about 112 μs. The noise
experimental results, and the inductance and voltage fluctuates from 60V to 224V obviously.
capacitance in system can also be calculated correctly. According to Fig.6, the arc concentrates at the site of its
initiation. With the increase of the current, the arc According to the vacuum arc appearance, the arc
diameter increases at t2=57.4μs. The current reached to concentrates at the site of its initiation at t1. With the
peak at t3=142.2μs when the voltage noise starts. Arc increasing of the current, the arc diameter increases and
column even scatters at the perimeter of the contact. the arc column starts to move along the surface at t2.
Then cathode spots move and diffuse between the The arc continue concentration is not observed, so the
contacts, a large number of plasmas collide with each arc movement and diffusion are not been restricted by
other and spread outside of the gap. The bright spots on bending of the surface. When the current reaches to
the exterior side surface of the cathode can also be peak at t3, the plasmas and cathode spots do not spread
observed. So metal vapor and plasma loss during t3 to t7, outside of the arc gap. Most of plasmas and metal vapor
which has an effect on the electrical endurance of the are absorbed by contact surface, and only a few move to
contact. After t6=311.6μs, arc voltage becomes out of the gap. Furthermore, the burning of the arc is
smoothness by the reducing of arc current. During this more stable in sin-ruled surface contact than that in butt
period, arc burns stably and the plasma colliding is contact. Meanwhile, the loss of plasmas is fewer in
restricted by reduction of energy between the arc gap. sine-ruled surface contact.
At this stage, a few of plasmas and liquid droplets also
C. Arc volt-ampere characteristics of sine-revolution
spread outside of the gap.
surface contact
From the perspective of the noise of arc voltage, with
The interrupting experiment with sin-revolution
continuously regeneration and split of the cathode spots,
surface contact is completed at the same condition as
a high-frequency oscillation is superimposed on the arc
forward. The parameters are measured, such as
voltage wave form. The loss of arc plasma by its spread
Ipeak=14.38kA and f=1.09 kHz. The results are showed
and movement is a main reason for generating the
in Fig.9 and Fig.10. As can be seen form the arc
voltage noise.
volt-ampere characteristics of sin-revolution contact, the
B. Arc volt-ampere characteristics of sine-ruled noise of arc voltage starts from t2 and ends at t3. The
surface contact time lasts about 57.4μs. Accordingly, the arc is bright
Arc volt-ampere characteristics of sine-ruled during this period, and a few of plasmas spread out of
surface contact at Ipeak 15.26kA, frequency 1.16 kHz is the gap.
shown in Fig.7. Fig.8 is vacuum arc appearance.
20 200
Compared with butt contact, arc voltage of sin-ruled 18
t1 t2 t3 t4 t5 t6 t7 t 8 100

surface contact is smooth and range between 50V and 16 0

90V. There is no voltage noise during interrupting. 14 -100

Arc Current (kA)

12 -200

Arc Voltage (V)

196.8μs
10
Arc Voltage -300
20 200 131.2μs
18
t1 t2 t3 t4 t5 t6 t7 t8 100
8
319.8μs
-400

6
Arc Current -500
16 0
73.8μs
4 -600
14 -100 377.2μs
147.6μs 2 410.0μs -700
Arc Current (kA)

12 -200 16.4μs
Arc Voltage (V)

0 -800
10 90.2μs Arc Voltage -300
434.6μs
-2 -900
213.2μs 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5
8 -400 Time (ms)
6 Arc Current -500
41.0μs
4 295.2μs -600
Fig. 9. Arc volt-ampere characteristics of sine-ruled surface
2 -700
16.4μs 360.8μs contact (I peak=14.38kA)
0 -800
418.2μs
-2 -900
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5
Time (ms)

Fig. 7. Arc volt-ampere characteristics of sine-ruled surface

contact (I peak=15.26kA)

Fig. 10. vacuum arc appearance in sine-revolution surface contact

IV. Analysis of voltage noise with different contacts

Fig. 8. vacuum arc appearance in sine-ruled surface contact

From what have been discussed, it can be seen that
arc voltage characteristics of three contacts are different.
The comparison chat is shown in Fig.11, and the in sin surface contact increased and the arc column
parameters are listed in Table III. Peak voltages of sin started to move along the surface. The arc continue
surface contacts are obviously lower, and floating range concentration was not observed, so the arc movement
is smaller. In Table3, the Δt, which is longest in butt and diffusion were not restricted by bending of the
contact, means the lasting time of voltage noise. surface.
3) The loss of arc plasma by its spread and movement
250

200 Sine-revolution surface contact was a main reason for generating the voltage noise. The
150
Butt contact
peak voltage of noise in butt contact was highest, with
100 plasma colliding and plasma loss.
Arc Voltage (V)

50
4) As seen in the appearance of sin surface contact,
0

-50
Sin-ruled surface contact the plasmas, cathode spots and metal vapor did not
-100 spread outside of the arc gap. Most of plasmas were
-150
absorbed by contact surface, and only a few moved to
-200

-250
out of the gap, for which the service life of the vacuum
0 0.1 0.2
Time (ms)
0.3 0.4
interrupter chamber was effectively improved.
Fig. 11. Arc voltage comparison of three contacts
REFERENCES
TABLE 3. ARC VOLTAGE COMPARISON
[1] Z. Liying, W. Jianwen and Z. Xueming, "Arc Movement of
Ipeak Upeak ΔU Δt Intermediate-Frequency Vacuum Arc on TMF Contacts," in
(A) (V) (V) (μs) IEEE Transactions on Power Delivery, vol. 28, no. 4, Oct.
2013, pp. 2014-2021.
Butt contact 13.54 224 164 112 [2] D. Feng, S. Xiu, Y. Wang, G. Liu, Y. Zhang and N. Li,
"Experimental Investigation of Arc Ignition Modes of
Vacuum Arc Under Transverse Magnetic Field," in IEEE
Sine-ruled surface
15.26 90 40 23.6 Transactions on Plasma Science, vol. 43, no. 5, May 2015, pp.
contact 1806-1814.
Sine-revolution [3] C. Wolf, M. Kurrat, M. Lindmayer, and D. Gentsch, “Arcing
14.38 190 140 57.4 behavior on different TMF contacts at high-current
surface contact interrupting operations,” IEEE Trans. Plasma Sci., vol. 39, no.
Thus, during the arc burning and diffusion stage, 6, pt. 1, pp. 1284–1290, Jun. 2011.
[4] J. Wang, J. Wu, L. Zhu and Y. Shi, "Arc modes of
many plasmas spread out of the butt contact gap. The intermediate-frequency vacuum arc with axial magnetic
loss of the plasmas makes arc voltage high and noise field," Discharges and Electrical Insulation in Vacuum
lasting time long. The proposed sin surface contacts (ISDEIV), 2010 24th International Symposium on,
Braunschweig, 2010, pp. 312-315.
hinder the diffusion of arc plasma outside. On condition
[5] L. Wang et al., "Experimental Investigation on Vacuum Arc
of same diameter, the surface area is larger than plain Behaviors Subjected to Larger Diameter Cup-Shaped and
area. Increasing of area makes more plasmas absorbed Coil-Shaped Axial Magnetic Field Electrode," in IEEE
by contact. At the same time, electrical life of sine Transactions on Plasma Science, vol. 43, no. 3, pp. 884-891,
March 2015.
surface contact increases.
[6] Z. Liu, S. Cheng, Y. Zheng, M. Rong and J. Wang,
“Comparison of vacuum arc behaviors between
V. CONCLUSSION axial-magnetic-field contacts,” IEEE Trans. Plasma Sci., vol.
36, pp. 200-207, Febuary 2008.
In this paper, two new contact structures, named [7] J. Yuan, W. Jianwen and T. Wei, "Properties of
sin-ruled surface contact and sin-revolution surface intermediate-frequency vacuum arc influenced by electrode
diameter and material in axial magnetic field," Electric
contact, were proposed. The voltage characteristics were Power Equipment-Switching Technology (ICEPE-ST), 2015
experimental analyzed. Compare with butt contact, the 3rd International Conference on, Busan, 2015, pp. 85-89.
results were obtained as follows: [8] Hyeong Goo Lee, Jong Sung Kang, IL Cheol Ahn, Jung Se
1) Compared with butt contact, the arc voltage of Kim. Study on design of axial magnetic field electrodes
depending on electrode diameter and current [J].2012 25th
sin-ruled and sin-revolution surface contacts were International Symposium on Discharge and Electrical
smoother, and the peak voltage of noise was lower. The Insulation in Vacuum (ISDEIV), 2012, pp. 489-492.
noise voltage in butt contact fluctuated greatly and
lasted longest time. E-mail of the author(s): ranfengleilei@[Link]
2) With the increasing of the current, the arc diameter