# 29th International Symposium on Discharges and Electrical Insulation in Vacuum (ISDEIV)

Europe/Rome

Via Orto Botanico, 15 – 35123 Padova, Italy GPS: 45.399444, 11.879807 Orto Botanico - http://www.ortobotanicopd.it/en/
Description

The International Symposium on Discharge and Electrical Insulation in Vacuum (ISDEIV) is a non-profit, international organization whose purpose is to encourage the advancement of the science and application of electrical insulation and discharges in vacuum, primarily by conducting symposia for the exchange of scientific information.

The Symposia are held biennially. They are interdisciplinary meetings for the exchange of results, presentation of progress, and discussion of ideas and challenges for the future in the field of electrical discharges and insulation in vacuum. Both fundamental and applied aspects are covered. Symposia program consist of invited talks, invited oral contributions, and posters. Minicourses and informal discussions on relevant topics may also be offered in addition to the regular Symposium schedule.

This Symposium edition, hosted by Consorzio RFX, was initially scheduled for June 2020, but due to the COVID-19 pandemic it has been rescheduled to run from September 27 to September 30, 2021.

It is our strong commitment to organize this Symposium as a face-to-face event, but the pandemic is far from being under control and it is impossible to predict the situation in September 2021, though vaccination processes have started worldwide.

For this reason, in agreement with the Permanent International Scientific Committee (PISC) of the ISDEIV, it has been decided to organize the Symposium with an hybrid format that allows both in presence - with a maximum number of 120 attendees - and remote participation, provided the Pandemic Emergency Status in Italy is ceased at the date of 31 May 2021.

In case of Symposium in presence, it will run at the Orto Botanico Venue, located in the city center of Padova, Italy. Registrations will be opened on Sunday 26 September afternoon at the Venue. The attendees will be welcomed during the opening reception planned for Monday 27evening.

The PISC has taken also the decision that the Symposium will be held only in remote mode if in Italy the Pandemic Emergency Status will be still in force at the date of 31 May 2021. Please keep yourselves updated, checking frequently the website at the IMPORTANT NEWS page.

In any case, we do rely on your support to make this very particular edition of the PISC a success, whatever the format will be!

If you need an Invitation letter for your VISA, please, make the request as soon as possible through the Workshop Registration Form.

Art and culture treasures are located in Padova such as the Cappella degli Scrovegni (Giotto’s frescos), Palazzo della Ragione (Medioeval justice palace), Palazzo Bo (Historical palace of the University of Padova) and Basilica of Sant’Antonio. The town hosts one of the largest and oldest universities in Italy, founded in 1222.

Prior Conferences:

This conference is supported by our sponsors and industry exhibitors:

Registration
ISDEIV Registration Form
Participants
• Akira TONEGAWA
• Aleksandr Russkikh
• Alexander Batrakov
• Alexander Logachev
• Alexander Medovnik
• Andrea Pisent
• Andreas Kyritsakis
• Andrey Kozyrev
• Andrey Ushakov
• Anton Schneider
• Atsushi Kojima
• Baipeng Song
• Benjamin Weber
• Bruno Balhan
• Chen Guan
• DAN WANG
• Daniil Zuza
• Decun Dai
• Denis Zolotukhin
• Dietmar Gentsch
• Dirk Uhrlandt
• Dmitry Shmelev
• Dmitry Sinelnikov
• Edgar Dullni
• Efim Oks
• Evgeny Nefedtsev
• Evgeny Oreshkin
• Fei Kong
• giuseppe Chitarin
• Grigory Ozur
• h craig miller
• Hai Chen
• Hao Yu
• haoyu Wang
• Heinz Pursch
• Helena Kaufmann
• Hiroki Kojima
• Horst Leng
• Ilya Lvovich Muzyukin
• Irina Schweigert
• Jan Borburgh
• Jiagang Li
• Jingyu Shen
• Jinjin Li
• JinWang Yang
• Kai Yin
• Konstantin Savkin
• Konstantin Zabello
• kyunghwan oh
• Leslie Falkingham
• li peng
• Lijun Wang
• Liyan Zhang
• Luigi Cordaro
• Lukasz Bachorz
• Marco Cavenago
• Martin Leusenkamp
• matteo zuin
• Mattia Dan
• Mikhail Tsventoukh
• ming zhang
• Naoki Kita
• Nicola Pilan
• Nicolò Marconato
• Nuno Ferreira
• Pavel Mikhailov
• pengfei xu
• Peter Siemroth
• Qiang Tang
• Ralf Janczak
• Ralf Methling
• Ralph Hollinger
• Richard Forbes
• Rustam Berezov
• Ryota Konagi
• Sandeep Kulkarni
• Sergey Barengolts
• Sergey Shkol'nik
• Shangwen Xia
• Shenli Jia
• Shu ZHANG
• silvia spagnolo
• Siyuan LIU
• Sreeram Vengayil
• sun wenhao
• Thierry Delachaux
• Timo Meyer
• Tommaso Patton
• Xiaofei Yao
• Xiaoming Liu
• Xinggui Chen
• Xinming Ma
• Yan Wenlong
• Yasunori Tanaka
• Yifan Fu
• Yimeng LI
• Yingyao Zhang
• Yinon Ashkenazy
• Yongpeng Mo
• Yun Geng
• Yuriy Mamontov
• Yury Zemskov
• Yuto Hatanaka
• Ze Yang
• Zhenxing Wang
• Zhisong Long
• Zhiyuan Cao
• ziang tong
• Zongqian Shi
ISDEIV 2020 Contact
• Sunday, 26 September
• Registration

Attendees Registration at the conference desk at Orto Botanico

• Workshop
• 10:30 AM
Coffe break
• Workshop
• 1:00 PM
Lunch break
• Workshop
• 4:00 PM
Coffe break
• PISC meeting
• Monday, 27 September
• Registration

Attendees Registration at the conference desk at Orto Botanico

• 8:15 AM
Opening remarks
• 10:30 AM
Coffe break
• ORAL: SESSION 01

Topic B: Vacuum Arcs

• 1
INFLUENCE OF IGNITION POSITION ON THE PROPERTIES OF VACUUM ARC GENERATED BY SWITCHING RMF CONTACTS

Radial magnetic field (RMF) contacts are widely used for the control of vacuum arcs at high currents. The self-induced magnetic field from the flow of the current through the contact structure drives the arc motion, reducing the thermal load on the contact surfaces. The initiation behavior of the drawn arc can affect the arc motion, and thereby the performance of the contacts. This contribution reports on the influence of the arc ignition position on the behavior and properties of switching arc between the RMF contacts. An AC current pulse with a peak value of about 28 kA and frequency of 50 Hz is used. Electrodes were made of Cu-Cr. Defined ignition positions were mechanically created on the cathode surface, with optical confirmation. Besides the arc current and voltage measurements, various optical diagnostics were used. Two high-speed cameras equipped with narrow-band filters characterized the arc behavior, which distinguished the emission of atomic, and ion copper lines. The anode activity was observed by an additional high-speed camera. Near infrared radiation (NIR) spectroscopy determined the anode surface temperature after current zero crossing. In addition, the density of neutral chromium vapor close to the current zero crossing was measured by means of broadband optical absorption spectroscopy. Three Cr I resonance lines at 425.43 nm, 427.78 nm, and 428.97 nm are used for the analysis. The results show clear influence of initial arc position on anode temperature and Cr density.

Speaker: Ralf Methling (Leibniz Institute for Plasma Science and Technology (INP Greifswald))
• 2
Investigation of vacuum arcs between with varying gaps up to 40 mm

As one of the most common circuit breaker technology, the vacuum interrupter (VI) is widely used in medium grid voltages. A main task is the interruption of high short circuit currents, where the VI outperforms other approaches due to low arc voltages during breaking operations. These low voltages are the result of the special kind of plasma produced as the contacts are separated. The applied current evaporates the contact surface and with the lack of a gaseous atmosphere, a metal vapor plasma is ignited. To avoid evaporation on a stationary point axial or transversal magnetic fields are applied to move the plasma over the contact surface. This plasma constricts due to the pinch effect with rising currents. The constriction behavior for gaps in the range of 20 mm gaps is well known and allows constructing VI for middle voltage level. Higher voltage levels require lager gaps to withstand dielectric stress. With growing gaps, the question of the constriction behavior of vacuum arcs arises.
To investigate this task two approaches are combined for this paper. Pin electrodes are used to create a reproducible reference for different gaps. To achieve long contacts gaps a programmable frequency converter with servo drive is used. Over a three-stage control loop, a predefined opening curve is implemented. Different opening curves with final gaps up to 35 mm are performed. Furthermore, the frequency of the synthetic test circuit is adjusted to 35 Hz to achieve longer arcing times. At this frequency, short circuit breaking operations with up to 7 kA peak currents are performed and observed using a high-speed camera in a rectangular arrangement. This test setup is combined with a software-based evaluation tool for reproducible analysis of arcing behavior. The description of the behavior with rising gaps shows the point where constricted arc is too long to be stable between the electrodes. The optimum of the arc length is essential of vacuum interrupters. This contributes to the feasibility of VIs in higher voltage levels.

Speaker: Benjamin Weber (TU Braunschweig)
• 3
Neutral atom density at current zero in interruption of intense-mode vacuum arc

A large number of vacuum circuit breakers (VCBs) are installed in medium-voltage-class AC networks. Improvement of interruption capacities of the VCBs is required for the application to high-voltage (HV) AC networks as SF6-free current breaking device. The current interruption capabilities of the VCBs are highly dependent on plasma parameters such as a neutral atom density. Measuring such plasma parameters and elucidating the vacuum arc phenomena occurring in the vacuum gap is indispensable for the development of HV-VCBs. In the interruption phase of an intense-mode vacuum arc discharge, interruption failure at current zero occurs due to the ionization of neutral atoms evaporating from the anode surface, and clarifying the critical density for current interruption is essential. However, the measurement of the neutral atom density distributions is extremely difficult for conventional sensors due to the spatial non-reproducibility of the vacuum arc. This paper aims to find the critical neutral atom density for the success and failure of the current interruption. The electrodes are axial magnetic field types with a 6-mm diameter and their compositions are CuCr (Cr: 50wt%). The 50-Hz half-wave sinusoidal current is injected to a 4.5 mm-gap and the transient recovery voltage with a peak value of 25 kV is applied to the vacuum gap after current zero. The success or failure of the current interruption is controlled by varying the interruption current value; the current values for interruption success and failure are 2 kArms and 4 kArms, respectively. The neutral atom density distributions are measured by the Shack-Hartmann type laser wavefront sensors at about 20 μs before the current zero and the instantaneous current value is about 300 A. The critical neutral atom density as a determinant factor for the interruption success or failure is discussed based on the comprehensive experimental results.

Speaker: Mr Ryo Kikuchi (The University of Tokyo)
• 4
The turbulent nature of ion flow of the vacuum arc.

The new approach to explaining of ions and electrons acceleration in the vacuum arc is presented. The ion acoustic plasma shock waves can produce the well-known shape of ion energy distribution as well as the dependence of that on the ion charge state. Experimentally measured mean ion velocities dependence on cathode material properties can be explained by ion sound shock wave dependence on “hot” electron temperature and ion mass. Experimentally measured energy distribution can be formed statistically by ion acoustic waves with different amplitudes and velocities. In addition, the plasma turbulence can be the cause of several unexplained effects: The correspondence of ion flow peaks to strong discharge current falls can be explained by plasma turbulence rising which in turn causes abnormal plasma resistance. The Fourier spectra of ion flow law 1/f can be explained by Kolmogorov’s established turbulence spectra.

Speaker: Dr Ilya Muzyukin (Institute of Eectrophysics UB RAS)
• 1:00 PM
Lunch break
• 2:00 PM
Walk to Aula Magna

Participants has to reach Palazzo Bo,Via VIII Febbraio, 2, for the afternoon session

• Opening session @ Aula Magna Aula Magna (Palazzo Bo)

### Aula Magna

#### Palazzo Bo

Via VIII Febbraio, 2, 35122 Padova PD
• 4:00 PM
Coffe break Palazzo Bo

#### Palazzo Bo

Via VIII Febbraio, 2, 35122 Padova PD
• ORAL: SESSION 02

Topic B: Vacuum Arcs

• 5
Influence of Floating Shield Wall on Dynamic Behavior of Metal Vapor during Arc Sustaining by Moving Particle Semi-implicit (MPS) Simulation

For improvement of arc quenching performance in vacuum circuit breakers (VCBs), it is necessary to well-understand vacuum arc phenomena. Control techniques of the vacuum arc are important for enhancing its current interruption performance. In addition to experimental verification, to understand arc behaviors by numerical simulation is one of the important challenges. A Novel approach of a hybrid model using moving particle semi-implicit (MPS) method and finite volume method (FVM) [1], simulating heavy particle behavior in arcs, has been developed to investigate dynamic behavior of metal vapor in vacuum arcs. In our previous study using the above unique method, the influence of presence of floating metal shield wall was investigated on arc characteristics including the distributions of heavy particle temperature, ionization degree and current density and so on [2]. In case with floating metal shield wall, ionization degree of heavy particles between contacts was found to decrease compared to those without the wall. This is because all the particles colliding with the shield wall were neutralized and reflected toward the center. After the above calculation results, the present work focuses on the role of floating metal shield wall against arcs between the electrodes. Influence of its boundary condition was also studied on the deposition rate of metal vapor on the shield wall during arc sustaining.
The developed model can calculate temporal evolutions in particle density, heavy particle temperature and ionization degree between the electrodes during vacuum arc sustaining. Boundary conditions of floating shield, such as reflection velocity, deposition ratio of heavy particles, were varied as a parameter to study dynamic behavior of metallic vapor between the electrodes. Calculation results show that the heavy particle temperature reached 7000 K between the electrodes, while it partly around cathode surface reached 12,000 K. Such heavy particle temperature increase is due to energy transfer by collisions with energetic electrons. The heavy particle temperature is also influenced by the boundary condition on floating metal shield wall because the heat exchange between vapor and the floating shield wall can contribute to decrease the heavy particle temperature. A change in heavy particle behavior during collision on floating shield wall can influence the temperatures of heavy particles and electron and ionization degree between the electrodes. This influence on ionization degree and the number of existing heavy particle would also affect the electrical conductivity of arc decaying process.

[1] Y. Tanaka, et al., ICEPE-ST 2017, E-O-3, 10, 2017, Xi’an, China
[2] T. Miyazaki, et al., ICEPE-ST 2019, C-3-5, 10, 2019, Kitakyushu, Japan

Speaker: Dr Yusuke Nakano (Kanazawa University)
• 6
Characteristics of Vacuum Arc Voltage and Material Transfer with Different Contact Materials

In this paper, the contact material transfer characteristics and contact surface micro morphology characteristics of 41mm contact diameter vacuum interrupter are analyzed. The electrode material of one contact of the vacuum interrupter is CuW80, and the electrode material of the other contact is CuCr25. When the vacuum interrupter interrupts the intermediate frequency arc of 5-15kA at a small gap distance, the morphology of different positions of the contact surface changes to different degrees, and the electrode components at different positions transfer to different degrees. In this paper, the micro morphology of contact surface in flat, crack, hole and droplet splash and the composition characteristics of contact material are analyzed, and the influence of ablation degree on contact micro morphology is obtained. The results show that for the Vacuum Interrupters with CuW80 and CuCr25 as contact materials, the transfer degree of Cu with lower melting point is the slightest, while Cr and W have different degrees of transfer after breaking large current.

Speaker: Mr Ziang Tong (Beihang University)
• 7
Observations of the Distortion of The Energy Distributions of Cathode Material Ions in Vacuum Arc Plasma Generated by the Microsecond LC-line Discharge

Low current vacuum arc discharge was investigated. Pulse source were the LC-lines with quasi-rectangular pulse shape. Pulse durations were 2 and 4 microseconds. Discharge electrodes were made of copper. Ion flow composition and ion energy spectra were obtained via the Thomson spectrometer with automated image recording and digital data processing.
Entering of the ion flow to the spectrometer was partially limited by the electrostatic gate with variable duration and delay of the closing pulse. Therefore the detection of ions depended on the ion velocity and the generation moment of the ion. It was found that the ion spectra have low energy (tens of eV) and high energy (hundreds of eV)parts. High energy ion fraction were sufficiently large; and the high energy ions were apparently generated at the discharge end at the given discharge parameters.

Speaker: Yury Zemskov (Institute of Electrophysics, Ural Branch, Russian Academy of Sciences)
• 8
PROPERTIES OF DROPLETS EMITTED FROM VACUUM ARC CATHODES MADE OF COPPER, TUNGSTEN AND TITANIUM

Molten droplets emitted from the cathode of vacuum arcs have been investigated by using a time-of-flight set-up along a vertical flight tube equipped with two light barriers at different distances from the cathode surface. The light scattered by particles passing the light barriers cross section was registered. The velocity was determined by the flight-time. Simultaneously, from the amount of the scattered light, the size of the passing particle was inferred applying the Mie-scattering theory. As the arrangement of the cathode could be oriented differently with respect to the flight tube, a coarse angular distribution of the parameters was obtained covering the range from almost parallel to the cathode surface (2…5°) to perpendicular.
This paper extends the results presented in our former publication on copper cathodes [1] with new investigation of particles emitted by arcs on tungsten and titanium cathodes. The multi-spot arcs carry currents from 0.1kA up to 1kA maximum for a fixed duration of about 1ms. Several ten-thousand discharges have been carried out on the various cathodes at some ten different angular positions. Because of the large amount of data, electronic processing by newly developed algorithms was required to identify and characterize the several tens of thousands of particle peaks.
As expected, the behavior of the particles emitted from the cathodes made of different materials exhibits similarities as well as differences. The number of particles per discharge depends strongly on the cathode material and varies remarkably with the angle between flight direction and cathode surface. Generally, smaller particles were found to have higher velocities and vice versa. However, a few large particles were recorded having unexpectedly high velocities. Regarding all cathode materials, several particles were observed flying close to the normal direction with respect to the cathode surface. The distribution of the particle sizes follows an exponential law restricted at small and large values by a detection limit and the statistics, respectively. The distribution of the emission velocities exhibits more than one local maximum pointing to distinct particle populations having distinct typical velocities. These populations contribute differently to the overall amount of particles at various angles. The observations could be attributed to a complex size-dependent re-organization of the particle velocities possibly induced by a supplemental interaction of the flying particles with the embedding cloud of arc plasma.
[1] Siemroth P., Laux M., Pursch H., Sachtleben J., Balden M., Rohde V., Neu R., IEEE Trans. Plasma Sci., 47 (2019), pp. 3470-3477

Speaker: Dr Peter Siemroth (retired from Arc Precision GmbH)
• Tuesday, 28 September
• Registration

Attendees Registration at the conference desk at Orto Botanico

• ORAL: SESSION 03

Topic C: Applications

• 9
Precision high voltage platform for accelerators and ion mass analysis

High voltage platform are of widespread use in nuclear physics accelerators in order to provide the initial acceleration of ions. The ion source environment is particular challenging for high voltage (HV) holding, since the discharge can be energized from source radiation or secondary particles, generated by the ion beam.
Moreover, the accelerator spectrometers for charge to mass ratio selection of exotic nuclei require a well defined beam energy, which fact poses challenging requests to high voltage design, rising the question of which electrode design rules are necessary for a quiet voltage operation, with tolerable energy rms relative fluctuation in the $10^{-5}$ order, including contributions from all devices in a beamline (spanning a 10 m size). Need for a thick plate ground is discussed and compared to existing installations. Several components must be optimized for achieving and measuring with the required precision: the transformer and transmission line system must not couple 50 Hz ripple and harmonics larger than HV generator ones; the cooling system may offer a convenient bleeder (for safety and for proper load of the HV generator); corona free operation of large area surfaces is necessary. The acceleration and deceleration tubes (each one formed by many electrodes) should stand vacuum arcs, possibly triggered by beam accidental loss, without any damage for long periods.
The parameters required for stability of this nonlinear system with many degrees of freedom are discussed, with particular attention to the case
of resistive voltage dividers. Such dividers, provided that inter-electrode capacitance and resistances are balanced and calibrated, also offer a convenient method to measure ripple voltage, to be compared with other breakdown diagnostics. The ion mass spectrometer itself offers another method, if calibrated by extensive ion orbit simulations, which can be compared with previous methods.

Speaker: Marco Cavenago (INFN-LNL)
• 10
The Onset of the Magnetized Pulsed Vacuum Arc: a Promising Way for High Efficiency Propulsion

Vacuum arcs nowadays are used in the wide range of applications including high current electronics, thin films deposition, particle accelerators and advanced plasma propulsion. Small, light-weight low-power micro-cathode arc thrusters are well-suitable for altitude control of small satellites like cubesats, and the thrust of such thrusters can be remarkably enhanced by implementing a magnetic field. However, for such applications as orbit raising or interplanetary missions, their thrust level needs to be improved even further by the adding a second acceleration stage. Recently, we proposed a possible approach for two-staged thruster – a micro cathode thruster with magneto plasma dynamical (MPD) stage with external magnetic field created by permanent magnet or magnetic coil. The concept of two-stage micro-cathode arc thruster implies the preliminary production of initial fully-ionized metal plasma by the first stage based on pulsed vacuum arc, and acceleration of this plasma by the j x B force created in the second stage. MPD approach looks promising since it allows accelerating quasi-neutral plasma without using low-transparent accelerating grids, high voltages, and additional power-consumable cathodes-neutralizers with limited lifetime.
Investigating this new plasma thruster, we found that being sufficiently magnetized by a dc magnetic field having both axial and radial components, such two-stage magnetized pulsed vacuum arc discharge demonstrates a threshold behavior: such single-pulse parameters as thrust, total charge of expelling ions, and ion-to-arc charge ratio rapidly jump after certain threshold dc voltage applied between the cathode of the first-stage micro-cathode arc thruster and the accelerating (MPD) electrode. Using independent experimental methods (electrical measurements and dynamic measurements with torsional thrust stand), we observed that the mentioned effect allows increasing the thrust (in several tens of times) together with thrust-to-power ratio (in several times). This opens up the perspectives of creation of new type of miniature and low-power plasma thrusters for small satellites propulsion.
The work was supported by NASA DC Space Grant Consortium and Vector Launch Inc.

Speaker: Dr Denis Zolotukhin (The George Washington University)
• 11
Developing of compact plasma thruster based on flashover discharge

Currently, the space industry is actively developing, and with it, the demand for different space thrusters is growing. In this work, a prototype of a small size pulsed plasma engine is described and investigated.
The main idea was to make a small electric propulsion system that could be used on CubуeSat and other small satellites. It follows that the mass of the propulsion system should not exceed 250g and the power consumption of 5W. The design of the thruster is simple; it consists of a metal cathode and anode and a solid insulator between them as a propellant. Thrust is produced by plasma flow generated by pulsed vacuum flashover discharge. In this prototype planar electrode design with a gap about 1mm was used. As the material for the electrodes, Cu or Al can be used. In addition to the thruster, it was necessary to develop a Power processing unit (PPU) which could generate high voltage impulses with amplitude up to 10kV and frequency from 1 to 10кHz. With the right pulse parameters, it is possible to achieve a working mode with minimal cathode erosion, to ensure high system stability.
The following system parameters were studied: thrust, reliability, ion flow velocity and their distribution. To determine ion speed and spatial structure of the ion flow we used small-sized ion detectors installed at various angles and distances from thruster. We used torsional pendulum to measure trust.

Speaker: Pavel Mikhailov (The Institute of Electrophysics of the Ural Division of the Rus)
• 12
Plasma Deposition Films to Improve Surface Insulation Properties in Vacuum

Ceramic as an insulation material is widely used in the field of pulse power technology, due to the good mechanical and electrical properties. However, the surface flashover at the interface of vacuum-dielectric is observed frequently, which limits the development of pulse power technology. In order to satisfy the demand of industry development, the improvement of the surface insulation performance of ceramic is urgent. Therefore, in this paper, the functional films are deposited on the ceramic surface by plasma modification technology in order to suppress the surface flashover. The experimental system includes pulse power supply, reactor and electrical measurement system. The pulse power supply (CMPC-40, Institute of Electrical Engineering, CAS) has a pulse rise time of 0.5 μs, a pulse width of 8 μs, and a pulse recurrence frequency (PRF) up to 3 kHz. The ceramic samples are treated with precursor assisted by plasma for several minutes. The precursor is carried out by argon gas. Before and after the plasma modification, surface charge accumulation and dissipation characteristics are discussed, respectively. The results shows the withstanding voltage increased to 29% after plasma modification. The physical and chemical characteristics before and after modification are also discussed.

Speaker: Dr Fei Kong (Chinese Academy of Sciences)
• 10:30 AM
Coffe break
• ORAL: SESSION 04

Topic C: Applications

• 13
A GEM-BASED HIGH-RATE X-RAYS DIAGNOSTIC FOR FLUX MEASUREMENT DURING HIGH VOLTAGE CONDITIONING IN VACUUM INSULATED SYSTEMS

The MITICA experiment at Consorzio RFX Padova, within the framework of the European Magnetic Confinement Fusion effort, is a full-scale prototype of the Neutral Beam Injector (NBI) heating system envisioned for ITER, which will be required to produce 1MeV beams of deuterium with an ion current of 40 A for up to 1 hour. These specifications far exceed any current NBI devices with a compact electrostatic accelerator to operate at 1MV being double the voltage reached up to now. To achieve these conditions the entire beam source and accelerator of MITICA are insulated by a vacuum instead of the more commonly used insulating gas. This is to avoid radiation induced conductivity in the gas by radiation produced from fusion reactions. However, this serves to be an overall more complex system and vacuum insulation has its own substantial hurdles regards stability and voltage holding. The High Voltage Padova Test Facility (HVPTF) is a concurrent R&D project focused on understanding the physical processes behind breakdown and micro-discharges in high-voltage insulation, as well as developing new diagnostics, models, and operational modes to address these challenges and aid in the stable operation of MITICA. The high-resolution characterization of the X-ray energy spectrum during the high voltage conditioning of a multi electrode vacuum insulated system such as HVPTF could be vital to understand the fundamentals behind voltage breakdowns and which stage is causing the micro discharge onset. This technique seems particularly promising for measuring small dark currents (< 10μA), which are generally difficult to be measured with standard techniques. The X-rays produced during HV experiments are mainly due to bremsstrahlung interaction between the electrons escaping from the HV electrodes (field emission effect) and the background gas or the chamber walls. The X-rays spectrum extends from low energy ( ̴ keV) up to several hundreds of keV depending on the maximum potential difference applied to the electrodes (up to 800 kV). During these events high photon flux (>106 photons/cm2s) is produced. This paper shows the development and preliminary results of a newly designed X-Ray diagnostic based on Gas Electron Multiplier (XR-GEM). This detector is able to stand very high rate (> MHz) in single photon counting mode and can cover the energy range from 3 - 50 keV. The XR GEM detector is equipped with anodic pads (256 pads 6x6 mm2) readout with a new data acquisition system called GEMINI, which gives the possibility to obtain a counting rate of several MHz and, thus, sub-ms time resolution together with mm spatial resolution.

Speaker: Gabriele Croci (Universita & INFN, Milano-Bicocca (IT))
• 14
Ceramic coatings for arc prevention between plasma facing components

The design of the RFX-mod2 fusion experimental device requires a 3 mm copper shell close to the plasma to achieve improved plasma confinement properties. Such conductive structure, located around the plasma and aimed to passively stabilize the plasma, shall have an electrical discontinuity in both the poloidal and toroidal directions, in order to allow the penetration of electromagnetic fields into the plasma region. These gaps avoid the formation of net poloidal and toroidal eddy currents during transient phases of the plasma confinement experiment. Moreover, the shell has been designed with an overlapping region at the poloidal gap in order to reduce the induced field errors.
During operations in Reversed Field Pinch magnetic configuration, the loop voltage, that is the externally induced electromotive force sustaining the plasma current, can reach values up to 400 V. These values can rapidly step up to 1.5 kV during fast plasma current terminations. Therefore, intense electric fields can generate between the shell flaps, only a few millimetres apart, along the overlapping region. Furthermore, taking into account that the stabilizing shell, being located inside the vacuum chamber, is exposed to low temperature plasma, the formation probability of harmful electric arcs is high.
In order to avoid arc formation, different kind of insulation coating on the copper, able to withstand the applied electric fields in the presence of plasma, are under investigation.
The electrical insulation performances of ceramic coating on a copper samples are investigated. Several deposition methods have been evaluated, including magnetron sputtering, atmospheric plasma spray and detonation gun spray, which differ in the adhesion to the substrate, compactness, porosity and mechanical strength of the deposited material. The last two methods have the advantage of being conducted in air at atmospheric pressure, without the aid of complex vacuum systems.
In order to validate the process, an experimental apparatus was prepared in laboratory, aimed to reproduce the expected conditions at the shell gap. It consists of a vacuum chamber in which a helium plasma was generated by means of a hot tungsten filament and a DC power supply. A bias voltage was applied between a copper plate and a cylindrical electrode (Ø 4 mm). The plate side facing the electrode was covered with alumina. The two electrodes were floating and biased by a small capacitor bank (0.3÷2 μF). The voltage on the electrodes was applied for 200ms, with a repetition rate of 1Hz.
In this contribution, the experimental results, aiming to study the conditions for the arc formation in presence of weakly ionized plasma (ne ~ 1016 m-3), are presented. In particular, voltage pulses up to 2.7 kV were applied, with a background gas pressure between 10-3 and 10 mbar. Furthermore, the electrodes were kept both in contact and spaced up to 5 mm. Furthermore, the results of similar experiments performed on a mini mock-up, simulating the RFX-mod2 overlapping shell, are presented.

Speaker: Dr Luigi Cordaro (Consorzio RFX)
• 15
RF induced discharges in vacuum in the SPIDER experiment

SPIDER is the full-size prototype of ITER Neutral Beam Injector ion source, in operation since June 2018 as part of the ITER Neutral Beam Test Facility located in Padova, Italy. The ion source of SPIDER is composed of eight Radio Frequency drivers operated in low hydrogen gas pressure, 0.3 Pa, driven at 1 MHz with a total nominal power of 800 kW, an extraction and acceleration system is used to produce a negative ion beam with an energy up to 100 keV.
RF breakdowns outside the ion source beset the SPIDER experimentation since the beginning of the operations. Such breakdowns could be sustained by the power supplies but in many cases cause the plasma pulses to stop and might cause source damages. The scope of this work is to summarize the experimental pieces of evidence, classify the type of breakdowns and describe the electrical model for these events.
Three classes of diagnostics have been used to investigate the RF breakdowns: electrical measurements from the Ion Source and Extraction Power Supply, visible cameras (both fast and slow) and optical fibers which collect lights from the rear of the SPIDER experiment. Each of these signals is somehow affected by the RF breakdowns; their comparative analysis allows improving the understanding of this issue and helped the identifications of different types of breakdowns that can occur outside the SPIDER Ion Source.
A discussion of the reasons behind the occurrence of the discharges, induced by the Radio Frequency, concludes the work.

Speaker: Mattia Dan (RFX)
• 16
Preparation ofthe vacuum insulation tests onthe MITICA 1 MV electrostatic Accelerator

MITICA is the prototype of the ITER Heating Neutral Beam Injector (NBI), which is designed to generate a 40 A beam of Hydrogen/Deuterium negative ions, to accelerate the ions up to an energy of 1MeV and then to neutralize them, producing a neutral beam of about 16.6 MW power. The design of MITICA is as far as pos- sible identical to the design of the ITER HNB, except that in the ITER NBI, the neutral beam will be directed to the plasma in the Tokamak vessel via a duct, whereas in MITICA the neutral beam will be directed to a water-cooled calorimeter. The NBI system will consist of a Vacuum Vessel (AISI304L, overall dimensions 15 × 5 × 5 m3), where the negative ion Beam Source, which is the most complex component of the NBI system, will be installed, together with other components (i.e. gas-cell Neutralizer, residual Ion Dump and Cryo-Pumps). The negative ion Beam Source includes a Plasma Chamber (with a gas injection system and auxiliaries for the production of negative Hydrogen/Deuterium ions) and a multi-stage Electrostatic Accelerator, constituted by a series of 7 metallic grids each having 1280 apertures. The Plasma Chamber and the metallic grids of the multistage electrostatic Accelerator will be kept at different electric potentials ranging from about - 1MV (Plasma Chamber, Plasma Grid and Extraction Grid) to ground potential (Grounded Grid) in steps of 200 kV. The Vacuum Vessel will be grounded. The insulation between electrodes having different electric potential will be provided by gaps in vacuum (filled with low pressure Hydrogen during operation) and by alumina post insulators. The voltage holding capability of the MITICA Beam Source at 1 MV is a very challenging issue, which could not be fully addressed so far on the basis of the theoretical models and experimental results available in literature. This paper describes a specific HV test campaign to be implemented in the MITICA Vacuum Vessel using a mock-up of the Beam Source. The tests have been designed with the aim of incorporating the essential features both of the single-gap and of the multi-stage insulation, so as to obtain reliable data on voltage holding at 1 MV and, if necessary, to focus on the most effective solutions. The tests shall be performed in the MITICA Vessel (already available in the NBTF site) both in vacuum and in low-pressure gas, using a specific Test Power Supply, in parallel with the construction and assembly of the MITICA Beam Source.
In the paper, the test objectives and requirements are first introduced. Then the strategy and test configurations are defined. Finally, some design solutions for the electrode realization are described and preliminary plan together with a list of test equipment is given.

Speaker: Dr Giuseppe Chitarin (University of Padova and Consorzio RFX)
• 1:00 PM
Lunch break
• POSTER 1: Session #1 P.A. Chatterton
• ORAL: SESSION 05
• 17
Discrimination of Discharge Pattern in Cut Model of Vacuum Interrupter

Electrical insulation of a vacuum interrupter (VI) is a composite system composed of vacuum, metal, and solid insulator. Detection of weak points in insulation where breakdown may occur is important for insulation design of VI. In order to grasp insulation conditions inside VI, it is important to identify discharge sites and discharge patterns. On the other hand, solid insulators of VI are mostly made of opaque ceramics, and it is very difficult to directly visually observe discharge sites and patterns inside VI. Therefore, a method is needed to identify discharge patterns from outside, based on discharge waveforms and physical mechanisms of particular discharge patterns.
Previously, we have proposed a discrimination method of discharge patterns in VI by extracting features and parameters of applied voltage, discharge current, and shield potential change, based on the relation between discharge type (gap breakdown or surface flashover), discharge voltage, developing time, and discharge path length. And, we confirmed the effectiveness of this method by using the simplified electrode system simulating the structure inside VI. In this study, we aim to verify that our proposing method can discriminate discharge patterns even in an actual structure of VI.
We prepared a cut model of VI, whose structure is similar to that of actual VI. Its shield electrode and alumina ceramic insulators have been partly cut for observing the discharge site. In this VI cut model, we simulated the weak points in insulation by additional needle electrodes. The VI cut model was set in a vacuum chamber at ${10}^{−5}$ Pa. We applied a negative standard lightning impulse voltage between contact electrodes in the VI cut model. We measured the applied voltage and the shield potential with voltage dividers (50 MHz) and the anode current with two high frequency current transformers (20 MHz and 200 MHz). Still images of discharge were captured with a digital camera to check the discharge patterns.
In various settings, e.g. with/without needle electrodes, different positions and lengths of needle electrodes, different gap lengths between contact electrodes, and with/without electric field relaxing electrode at the edge of alumina insulator, we verified the effectiveness of our proposing discharge discrimination method. In all settings, we almost succeeded to discriminate the initiation pattern of discharge. Even in the case that the discharge image could not be captured because the discharge would occur behind electrodes or insulators, it was estimated that the method enabled us to discriminate the discharge pattern.
In the case of complex discharge pattern with gap and/or surface discharge, basically we could also discriminate the discharge pattern as well as in our previous electrode model simulating VI. In the present VI cut model similar to actual VI, we found the new discharge pattern, in which the first gap discharge induced another gap discharge due to the diffusion of particles from the first discharge. However, by modifying the discrimination method based on the above physical mechanism of new discharge pattern, we got the prospects that the method can be applied to actual VI structure.

Speaker: Dr Hiroki Kojima (Nagoya University)
• 18
Lightning impulse conditioning of a combined field grading and shielding arrangement for vacuum double break

Advantages as environmentally friendly operation and relative low maintenance costs of vacuum circuit breakers (VCB) in comparison to other interrupters have established the VCB in the medium voltage range. In order to enlarge the operational voltage level the approach of double break is common. To symmetrize the voltage distribution over the series gaps the use of external grading capacitors is one possible solution.
In the current work a double breaking vacuum chamber with additional field grading and shielding electrodes in a separate surrounding vacuum chamber to symmetrize the voltage distribution is presented. The purpose of this work is to investigate the shielding behaviour of the arrangement in a first step. Therefore a grounded surrounding electrode is used to disturb the symmetric field distribution over the arrangement. In a second step the dielectric behaviour in dependence of the shield preparation before and during conditioning is investigated. Complete breakdowns of the arrangement are detected using a common voltage divider. To detect partial breakdowns, the shields own capacitances are determined and used to generate a voltage divider. This approach also allows the estimation of the floating middle potential. All tests were performed with lightning impulse voltage stress (LIV). The present study combines the topic of conditioning with the topic of field shielding and delivers new approaches to symmetrize electric fields over vacuum double break arrangements for high voltage use.

Speaker: Mr Timo Meyer (TU Braunschweig)
• 19
Small inductive current switching with high-voltage vacuum circuit breakers

Circuit breakers are appropriate devices to energize and de-energize electrical circuits. The requirements for standard switching applications are typically covered by the IEC 62271-100 and IEC 62271-1. In addition to these standard applications other switching cases requires performance which is not automatically qualified by type tests in accordance to IEC 62271-100 and -1.
This is explicitly the case for switching of small inductive currents. The range of requirements for these applications is extremely wide -from switching of unloaded transformers, to the quite high current of tertiary winding applications at transformers. Nowadays, gas circuit breakers are the preferred solution to cover all high-voltage applications for these inductive loads. Typically, individual pole operated circuit breakers with a point on wave switching device will be used to operate the circuit breaker in the reignition free window to ensure a safe and sustainable operation. Compared to standard switching application these special configurations require a reduction in service interval for the contact system.
The application of vacuum circuit breakers in high-voltage systems would cause the need, that the technology also covers these requirements.
In this paper, the subject is explored to show that the application of a common operated vacuum circuit breaker in combination with a filter circuit is a suitable instrument to avoid switching overvoltages and internal winding resonance overvoltages when small inductive loads are de-energized.
The basis of this approach is the simulation of these circuits with an electromagnetic transient program with high-frequency models of the relevant grid components to evaluate appropriate dimensions and locations for filter circuits. This simulation study comprises a methodology to determine optimized parameters for the filter circuits and a subsequent verification by means of simulation of breaking operations with a vacuum circuit breaker model. The methodology is based on the long-term experience in the medium voltage range will also be applied to the high-voltage applications. On this basis also the acceptance criteria are derived from medium voltage knowhow.
The simulation activities are accompanied by appropriate tests to evaluate and verify the simulation results.

Speaker: Dr Stefan Giere (Siemens AG)
• 20
Dielectric Strength Improvement by Higher Frequency AC Conditioning

To increase the dielectric strength of vacuum interrupters (VIs), spark conditioning with repetitive breakdowns (BDs) by impulse or AC voltage application is one of the effective methods. AC conditioning with the frequency of 50 Hz or 60 Hz has been widely used owing to the advantage of lower cost for conditioning completion with hundreds of BDs. However, the dielectric strength after AC conditioning is usually lower than that after impulse conditioning. This might be attributed to harmful BDs during AC conditioning, i.e. some BDs would damage the electrode surface. We confirmed that multiple BDs occurred successively in a half cycle of AC voltage application. Multiple BDs would not be caused by weak points on electrode surface, but triggered by metal vapor due to the previous BD in the same half cycle. In order to suppress such multiple or harmful BDs during AC conditioning, in this paper, AC conditioning with the higher frequencies is proposed. When a BD occurs in a half cycle with the higher frequencies, the instantaneous voltage can be quickly decreased and the subsequent or multiple BDs can be suppressed. The effects of AC conditioning with the higher frequencies on the dielectric strength are discussed in this paper.

First, we measured the increase of BD voltage by AC conditioning at two different frequencies, 120 Hz and 550 Hz. We set sphere-plane electrodes made of OFHC Cu in a vacuum chamber at 10-6 MPa. AC conditioning was conducted at each frequency with the gap distance of 1 mm under a fixed applied voltage of 32 kVrms for 10 minutes. The BD voltages before and after conditioning were evaluated by negative standard lightning impulse voltage with up/down method with the enlarged gap distance of 3 mm.

Experimental results revealed that the evaluated BD voltages before and after AC conditioning at 120 Hz were 125 kV and 181 kV, respectively, which means 45 % increase. On the other hand, those at 550 Hz were 144 kV and 240 kV, corresponding to 67 % increase, which is higher than that at 120 Hz. This improvement of evaluated BD voltage and its increase ratio at 550 Hz can be regarded as the effect of higher frequency AC conditioning. The total numbers of BDs during AC conditioning were 465 times at 120 Hz and 356 times at 550 Hz, respectively, including the intermittent BDs with a long time interval (ms order) beyond the half cycle and the multiple BDs in the same half cycle. The multiple BDs were classified into the first BD and the subsequent BDs in the same half cycle. The number ratio of subsequent BDs at 550 Hz was 28.4%, which is lower than 49.9% at 120 Hz. These results suggest the feasibility of higher frequency AC conditioning for the dielectric strength improvement of VIs.

Speaker: Mr Ryota Konagi (Nagoya University)
• 4:00 PM
Coffe break
• 4:30 PM
Bus to RFX
• 5:00 PM
VISIT TO NBTF-RFX CNR research area

#### CNR research area

Corso Stati Uniti 4, 35127, Padova
• 7:00 PM
Bus to Orto Botanico
• Wednesday, 29 September
• Registration

Attendees Registration at the conference desk at Orto Botanico

• ORAL: SESSION 06

Topic A: Breakdown and Flashover

• 21
A method to analyze particle induced vacuum breakdown with obvious field emission current

To reveal vacuum breakdown mechanism is important for insulation enhancement between vacuum gap. In this experiment, impulse voltage was applied across pure copper plane electrode gap in ultra-high vacuum condition (ranging from 10-6 to 10-7 Pa). Surface analysis with XPS (X-ray photoelectron spectroscopy) was carried out in the surface analysis chamber. The circuit current and applied voltage were measured during the breakdown test. Two cameras were fixed with a right angel of 90° to capture the breakdown pictures and the breakdown location on the cathode surface was obtained. The anode and cathode electrode surfaces were both scanned under SEM. The electric field flux was also simulated. The results showed a typical kind of measured breakdown waveform: obvious field emission current appeared but no breakdown occurred at the peak value of field emission current. When the field emission current decreased from the peak, a sudden breakdown happened. Through surface analysis, the electrode surface was pure copper was confirmed. Through two cameras were fixed with a right angel of 90°, breakdown site was located. The SEM results showed no micro-protrusions existed around breakdown site and many particles covered on the center of electrode. So, the breakdown waveform was viewed as particle induced breakdown. A model how the particle induced breakdown with obvious field emission current was built to explain this phenomenon.

Speaker: Dr Shimin Li ( Jiangsu Key Laboratory of New Energy Generation and Power Conversion, Nanjing University of Aeronautics and Astronautics)
• 22
Influence of Middle Gap Distance on Breakdown Characteristics of Triple Vacuum Gaps in Series

As vacuum circuit breakers develop to high voltage level, the multiple floating shields are used in vacuum interrupters. The dielectric strength of the shield gaps needs to be evaluated during insulation design inside the vacuum interrupters. And the shield gaps can be regarded as multiple vacuum gaps in series. The objective of this paper is to determine the influence of gap distance on the breakdown characteristics of triple vacuum gaps in series with asymmetrical distribution under lightning impulse voltage. In the experiments, a negative standard lightning impulse voltage was applied. A sphere-plane electrodes system was selected in the experiments, which was conducted in the vacuum interrupter with the glass bulb. The material of the electrodes was Cu. The diameter of sphere was 18 mm. The diameter of plane electrode was 50 mm. The gap distance of the sphere-plane electrodes could be adjusted. The triple vacuum gaps in our experiments was composed of three asymmetrical sphere-plane electrodes systems. The experimental researches can help to further understand the breakdown characteristics of multiple floating shields in vacuum interrupters.
The experimental results will be presented in the full paper.

Speaker: Mr Jingyu Shen (Xi’an Jiaotong University)
• 23
Prebreakdown Current Characteristics of Vacuum Gaps with Various Gap Distances

Mechanical shocks are long considered to be responsible for late breakdowns or non-sustained disruptive discharges of a vacuum circuit breaker, during the switching process. However, there still remains no quantitative research on this phenomenon. The objective of this paper is to determine a quantitatively relationship between the mechanical shock strength and the DC breakdown voltage under different vacuum gaps. An elaborate test equipment was developed to provide a mechanical impact force on the transmission part of the test vacuum interrupter (VI). The mechanical vibration imposed on the movable contact of the VI can be quantitatively evaluated by effect of the impact velocity on the vibration, which ranger from 1.0 m/s to 3.0 m/s, stepped by 0.5 m/s. A 40.5 kV vacuum interrupter (VI), with a couple of butt-type contacts, was used. The contact material was CuCr25 (Cu 75%wt, Cr 25%wt). The contact gap was set to range from 1 mm to 10 mm, stepped by 3 mm. DC voltages were applied across the test VI, with an up and down method. The 50% DC breakdown voltages U50 were determined, under each mechanical shock case. In each experimental case, the DC voltage stressed on the test VI lasts for one second. Details of the experimental result and discussion will be presented in a full paper version.

Speaker: Dr Jian'gang Ding (State Key Lab. of Electrical Insulation and Power Equipment, Xi’an Jiaotong University)
• 24
Investigation of Vacuum Breakdown in Pulsed DC Systems

Vacuum breakdown in the pulsed DC experiment at CERN was investigated by means of electrical measurements, SEM imaging, Focused Ion Beam technique (cross sectioning), and numerical modelling. The breakdown sites comprise one or more craters. There is plastic deformation beneath the craters. The craters are surrounded by a ‘wave pool’, which could be from 70 nm up to 1 μm thick.

The crater(s) represent the epicenter of the breakdown, where electron emission and vaporization of the metal occur, and the plasma ball is formed. The only possible mechanism of ionization of neutral vapor seems to be ionization by the emitted electrons. The pressure exerted by the plasma ball is responsible for the formation of the crater. This physical picture is generally similar to what was seen in simulations of formation of cathode spots in vacuum arcs under conditions typical for, e.g., circuit breakers and unipolar arcs in fusion devices. There is, however, a very important difference. There are external factors provoking the formation of cathode spots in vacuum arcs (the plasma cloud left over from an (extinct) spot that previously existed in the vicinity of the (new) spot being ignited), and the ignition of unipolar arcs (plasma instabilities which deliver high energy and particle fluxes to the plasma-facing components). There is no such external agent which would facilitate the breakdown in the CERN experiment. Therefore, the breakdown voltage is by orders of magnitude higher than in vacuum and unipolar arcs. Moreover, the breakdown is dominated by field electron emission, which is irrelevant in vacuum and unipolar arcs and enhancement of the field on cathode microprotrusions must be a decisive effect.

Simulations have been conducted with the aim to describe the mechanisms of vacuum breakdown, from the initiation of field emission and heating of the protrusion, to vaporization of the cathode material, production of plasma by ionization of the metal vapor and destruction of the protrusion by melting. The model comprises the Navier-Stokes equations describing the motion of the molten metal of the cathode, in conjunction with the heat transfer equation and current continuity equations in the cathode (including both the molten and solid parts). The Laplace equation is used to compute the electric potential and field in the gap between electrodes in the initial phase of the breakdown; switching to a model with a thin near-cathode plasma layer will occur in the course of the development of the breakdown.

First results show that the heating due to electron emission (i.e., Nottingham effect) is a very important mechanism, resulting in rapid increase of the protrusion temperatures; this may be indicative of a microexplosion. Further results will be presented on the evolution of voltage and current in the gap, destruction of the microprotrusion and crater formation.

Acknowledgments The work at Universidade da Madeira was supported by FCT - Fundação para a Ciência e a Tecnologia of Portugal (the projects Pest-OE/UID/FIS/50010/2019 and UIDP/50010/2020) and by European Regional Development Fund through the program Madeira 2014-2020 under project PlasMa-M1420-01-0145-FEDER- 000016 (UMa).

Speaker: Dr Helena Kaufmann (University of Madeira)
• 10:30 AM
Coffe break
• PISC meeting
• ORAL: SESSION 07

Topic A: Breakdown and Flashover

• 25
Pre-breakdown markers on the surface of copper cathodes of vacuum gaps

Recently, a new view on the problem of initiation of vacuum breakdown is forming, which is, the factor of limitation of electrical strength of vacuum gaps should be seek not only on the surface, but also in the bulk of electrode material. Recent studies indicate that the initiators of the vacuum breakdown may be defects of crystal structure of cathode material [1, 2], most likely the dislocations [2–5]. In works [6–8] we suggested that the initiation of vacuum breakdown can be associated not only with dislocations arising under electric field action in the bulk of cathode material near inhomogeneities, but also with dislocations that initially exist in the material and have crops into the surface. However, so far there has been no direct experimental evidence that dislocation phenomena or dislocations themselves may be the initiators of vacuum breakdown. In this paper, a purposeful study the correlations of dislocation outcrops on the surface of single-crystal and coarse-grained particularly pure copper with positions of explosion-emission marks left subsequently by a short cathode vacuum spark was carried out. The coincidence of the cathode micro-explosive erosion sites with the positions of dislocation outcrops that existed before the breakdown was revealed. The appearance of extensive zones with altered crystal structure around the explosive centers is detected. Data on the nearsurface dislocation structure arising after a short-pulse spark on initially dislocation-free silicon single crystals are also presented.

1. A.S. Pohjonen, F. Djurabekova, K. Nordlund, A. Kuronen, and S.P. Fitzgerald, J. Appl. Phys., vol. 110, 023509 (pp. 1–6), 2011.
2. K. Nordlund, and F. Djurabekova, Phys. Rev. Accel. Beams., vol.15., 071002, 2012.
3. S. Calatroni, A. Descoeudres, J.W. Kovermann, M. Taborelli, H. Timko, W. Wuensch, F. Djurabekova, K. Nordlund, A. Pohjonen, and A. Kuronen, Proc. of Linear Accelerator Conference LINAC-2010, Tsukuba, Japan MOP070, pp. 217–219.
4. A. S. Pohjonen , S. Parviainen, T. Muranaka, and F. Djurabekova, J. Appl. Phys.,vol. 114, 033519 (pp. 1–8), 2013.
5. S. Vigonski , F. Djurabekova, M. Veske, A. Aabloo , and V. Zadin. Modelling Simul. Mater. Sci. Eng. 2015. – V.23. – 025009 (P. 1–17).
6. E.V. Nefedtsev, S.A. Onischenko , and A.V.Batrakov. Russian Physics Journal. – V. 60. – No. 8. – P. 1437-1448.
7. S.A. Onischenko, E.V. Nefedtsev , and D.I. Proskurovsky Proc. 27 Intern. Symp. on Discharges and Electrical Insulation in Vacuum (ISDEIV-2016). – Suzhou, China. – 2016.– V. 1.– P. 13-16.
8. S.A. Onischenko , E.V. Nefyodtsev, A.V. Batrakov, and D.I. Proskurovsky. Proc. 26 In-tern. Symp. on Discharges and Electrical Insulation in Vacuum (ISDEIV-2014). – Mumbai, India, 2014. – V.1. – Р. 5 – 8.
Speaker: Dr Evgeny Nefedtsev (Institute of High Current Electronics SB RAS)
• 26
Temperature Distribution of Cathode Micro-protrusion Based on Smoothed Particle Hydrodynamics Method

In this paper, Smoothed Particle Hydrodynamics (SPH) method is introduced into the study of the temperature distribution of the cathode micro-protrusion with the resistive heating caused by the emission current. Firstly, the partial differential governing equations for hydrodynamics with material strength are established. Then, the kernel approximation is used to approximate the field function in the partial differential equations, and the particle approximation is used to discretize the kernel approximation equations in order to obtain a series of SPH governing equations. At the same time, the thermodynamic equations are introduced into the SPH method. Based on the source code of SPH method, the mathematical model of cathode micro-protrusion is established. The SPH equations are solved by using the explicit integration method. The temperature distribution of cathode micro-protrusion with different field emission current Joule heat are simulated. The results can provide theoretical guidance for the quantitative study of the physical mechanism of vacuum discharge induced by field emission.

Speaker: Ms Decun Dai (Tongji University)
• 27
Pulsed Breakdown Behaviors across Microgaps at Various Atmospheres

With the rapid miniaturization and integration of electronic and electromechanical devices, especially for vacuum micro devices, it is urgent to study the electrical breakdown behaviors of pulsed breakdown at microscale and explore its underlying principle. In this paper, the influence of atmospheric pressure and electrode geometries on pulsed breakdown at microscale was studied by the home-built electro-optical measuring system. Results show that when the gap width is less than 5 μm, the role of gas molecules density or gas pressure inside the gap could be eliminated which means that the breakdown process is similar as the vacuum breakdown. For different electrode geometries, when the gap width is larger than 15 μm, the breakdown voltage remains the same regardless of the electrode geometries, whereas difference behaviors can be observed when the gap width is less than 15 μm, and these results can be analyzed from different breakdown mechanism such as field electron emission, ion-enhanced field emission and Townsend avalanche. The results presented in this paper would be of a great help for better understanding the physical mechanism of pulsed breakdown at microscale.

Speaker: Yimeng LI (Xi'an jiaotong university)
• 28
New Development of BIRD model

The possibility that the flow of electrons emitted by a cathode at high dc voltage is essentially due to local emission of a covering dielectric layer is at the basis of the BIRD (Breakdown Induced by Rupture of Dielectric) model. This model assumes that, in presence of sufficient electric field, the electrons trapped in polarization structures of the dielectric layer are extracted by quantum tunneling effect. As a consequence of the layer electron depletion, the electric field inside the dielectric layer increases and the rupture (breakdown) of the layer itself can occur, provided certain conditions are met. To investigate experimentally the features of this model the High Voltage Short Gap Test Facility (HVSGTF) has recently been built in Padua. The experimental dark current measured at different electrode configurations permits us to test the correctness of the model predictions. In particular, we consider the trend of the current as a function of time and its dependence on the characteristic properties of the dielectric layer. From the theoretical side, we investigate the consistency between a semi-classical model and a simple quantum model

Speaker: Dr Emanuele Spada (Consorzio RFX)
• 1:00 PM
Lunch break
• 1:30 PM
EXCURSION TO VENICE
• ORAL: SESSION 08

Topic B: Vacuum Arcs

• 29
STUDY OF HIGH-CURRENT ANODE MODES FOR VARIOUS ELECTRODE MATERIALS

The anode activity in switching vacuum arcs deserves special attention since this electrode is the main source of neutral metal vapour when the arc current is sufficiently high. The vapour distribution between the electrodes is arbitrary, can’t be controlled by electromagnetic fields, and promotes the arc re-ignition in case of sufficient high density. However, the desired surface properties can be achieved by the choice of appropriate electrode material, which provide certain possibilities to influence the anode activity.
In the frame of the present work electrodes made from CuCr alloy with the same composition, but different thermal processing has been studied under the AC current load of up to 6.5 kA. Especially, the appearance of various anode modes has been analyzed. The anode modes were distinguished by optical observation using a high-speed camera techniques (diffuse, footpoint mode, anode spot type 1), as well as by arc voltage measurements (anode spot type 2). The existence range of various anode modes have been determined. Simultaneously to the optical observations, the anode surface temperature was determined by means of NIR spectroscopy. In addition, the density of neutral chromium vapor after current zero crossing was measured by means of broadband optical absorption spectroscopy. Three Cr I resonance lines in the region 425-429 nm are used for the analysis. The morphology of the anode surface after appearance of various anode modes have been analyzed by microscopy. The results show a clear influence of material properties on the anode activity and electrode erosion. The results of anode surface temperature and Cr I density measurements for various electrode materials will be presented and discussed.

Speaker: Prof. Dirk Uhrlandt (Leibniz Institute for Plasma Science and Technology (INP))
• 30
Time-Dependent 3D Reconstruction of Arc Shape in a TMF contact system

Abstract:

It is well known that the transverse magnetic field (TMF) can drive the vacuum arc to move continuously on the contact, resulting in a non-axisymmetric characteristic of the vacuum arc. Therefore, the particle distribution of the vacuum arc is also non-uniform. And the various particles will directly affect the interaction between the vacuum arc and electrodes, which in turn influence the burning of arc and the successful interruption of interrupter. The two-dimensional information from a single direction is not sufficient to show the particle distribution of the arc.

In this paper, the CuI ions were observed by a 3D imaging system with optical filters from multiple directions, and the distribution was reconstructed by filter back-projection (FBP) and Algebra Reconstruction Technique (ART) method based on the record from those directions. A series of metal contacts with different copper and chromium contents were used in our experiment.

Through the above method, we obtained a series of 3D distribution of Cu ion under different copper and chromium contents, according to the 3D distribution, the pattern of particle generation in a vacuum arc was analyzed.

Key words—Vacuum arc, Transverse magnetic field, three-dimensional reconstruction, particle distribution

Speaker: Dr Zhenxing Wang (Xi'an Jiaotong University)
• 31
Two-Dimensional Spectroscopy for Vacuum Arc in Steady State and Decaying State under Free- Recovery Condition

Vacuum circuit breaker (VCB) is attracting attention recently from the viewpoint of environmental friendliness and ease of maintenance, etc. In the VCB, a vacuum arc discharge can be maintained by metal vapor supplied from the electrodes during high current interruption process. This vacuum arc is extinguished mainly due to natural diffusion of metal vapor around current zero. To improve the arc extinction performance in VCBs and the controllability of vacuum arc behavior, it is necessary to understand the complex phenomena in the vacuum arcs. For this purpose, we have developed a numerical model with a particle method and a fluid method for vacuum arcs[1]. But, the present research focuses on experimental approach to measure dynamic behavior of metal vapor in steady state and in transient free-recovery state vacuum arcs by using two-dimensional spectroscopy.
In the present paper, first, a fundamental experimental system was developed, which consists of a DC current source, a vacuum chamber with electrodes and an IGBT switch. The electrode used was made of an alloy of copper and chromium. Between the electrodes, a vacuum arc discharge was formed at currents of 30 A and 50 A. The IGBT was connected in parallel with the electrodes to commutate the arc current between the electrodes to the IGBT, which makes free-recovery state between the electrodes. Free-recovery condition is the most fundamental transient state for decaying arcs. High speed imaging was conducted for a steady state and transient decaying arcs with a high speed video camera with frame rates of 10,000 fps for steady state and 200,000 fps for free-recovery state. Furthermore, two-dimensional spectroscopic observation was also carried out using a high speed video camera with an imaging spectrometer, especially focusing on copper and chromium atomic spectral lines and their ion lines.
As the result, there were several cathode spots, which were repeatedly coupled and split. Near the cathode spots, a bright region is present involving both copper and chromium atomic spectral lines and copper ion lines. No chromium ion lines could be detected. This indicates that copper and chromium were provided from the electrode to the arc almost uniformly and the electrons originates from the ionization of copper. From the bright region, a cone-shaped light emitting area is expanded toward the anode. After current down for free-recovery condition, this bright region near the cathode spot decayed gradually with time, while the cone-shaped light emitting area disappeared rapidly within 20 $\mu$s. This result implies that vapor in the cone-region decays due to rapid diffusion.
[1] Y.Tanaka, et al., ICEPE-ST2017, E-O-3, 2017.10,Xian, China

Speaker: Mr Yuto Hatanaka (Kanazawa University)
• 32
Streak image observations of vacuum arc spots in a magnetically steered arc plasma source

The physics of vacuum arc spot operation remains subject of research due to difficulties associated with short characteristic times (ns to µs) and short length scales (µm or even smaller). The advancement of modern plasma diagnostic technologies may allow us to probe arc spot plasma with higher spatial and temporal resolution than before. In this contribution, we report about of the ignition and evolution of arc spots using a steered arc plasma source which is in its construction similar to a linear sputtering magnetron: the E × B field is utilized to have some control of spot locations while performing the study. Optical emission from arc spot plasma is investigated for different conditions including type of cathode material, background pressure, pulse width, and current. The main plasma diagnostics tool we employ is a streak camera with an image intensifier which allows us to observe the dynamics of arc spots and sub-cells, if present. We report on preliminary results, focus on the concept but also stress the limitations of the approach which are, apart from technical issues related to the stochastic nature of the phenomena, routed in the finite probability of spontaneous emission. This suggest that future methods may have to use external light sources.

Speaker: Mr Kyunghwan Oh (Leibniz-Institute of Surface Engineering)
• Thursday, 30 September
• Registration

Attendees Registration at the conference desk at Orto Botanico

• ORAL: SESSION 09

Topic A: Breakdown and Flashover

• 33
Dynamic coupling between particle-in-cell and atomistic simulations

We propose a method to directly couple molecular dynamics, finite element method and particle-in-cell techniques, to simulate response of a metal surface to high electric fields. We use this method to simulate the evolution of a field emitting tip under thermal runaway, by fully including the 3D space-charge effects. We also present a comparison of the runaway process between two tip geometries of different widths. The results show with high statistical significance, that in case of sufficiently narrow field emitters, the thermal runaway occurs in cycles where intensive neutral evaporation alternates with cooling periods. The comparison with previous works shows, that the evaporation rate in the regime of intensive evaporation is sufficient to ignite a plasma arc above the simulated field emitters.

The proposed method, which is under constant development, forms the basis for fully simulating the processes that lead from thermal runaway of intensively field emitting nanotips tips to full arc plasma onset. By introducing and handling new particle species such as neutrals and ions, we aspire to bridge the gap of understanding between the dark current and the vacuum arc ignition. This understanding is becoming increasingly important, in view of recent developments in the analysis of the dependence of the breakdown rate to the available electromagnetic power.

Speaker: Andreas Kyritsakis
• 34
Investigation of Electron Emission from Metal/Insulator Interface in Vacuum and Electric Charge Distribution on the Insulator

Flashover along a surface of solid insulator (surface flashover) is one of the problems in improving insulation performance in gas, liquid and vacuum. It has been regarded that the emission of electrons from the cathode triple junction (CTJ) is a first step of initiating the surface flashover. Therefore, it is important to investigate field electron emission phenomenon around the CTJ region.
In order to investigate the electron emission characteristics from the boundary region between the insulator and the metallized layer, an alumina insulator sample with a metallized layer formed by the Mo-Mn method was used. In addition, since the positive charge of alumina is possible to be related to the electron emission from CTJ, an experiment was also conducted on a sample with Cr2O3 coating on the surface to suppress the charge. In the experiment, the precursor current was measured and the discharge position was photographed.
The experimental results showed that a several numbers of breakdown are observed around the boundary between the insulator and the metallized layer. After such experiments repeat several times, the prebreakdown pico-ampere order current was measured. The value of prebreakdown current increased slowly with the elapsed time even if the applied voltage was kept constant. It increased more than two orders of magnitude during about 3-5 minutes, finally lead to a breakdown event or saturated to a certain value. In addition, for the coated sample, the electron emission current decreased and the breakdown voltage improved. Furthermore, no current rise phenomenon was observed when the voltage was constant. Based on the above results, this phenomenon is probably due to the effect of the alumina being charged by the electrons emitted from the CTJ. Therefore, We conducted a charge measurement experiment in the next step.
In the experiment for charge distribution measurement, a sample in which CTJ was formed by alumina ceramics located on the center of a stainless steel plate was used. This sample was also experimented with a sample additionally coated with Cr2O3. In the experiment, a voltage was applied before the breakdown occurred, and immediately after that, the surface potential was measured.
As a result of the experiment, it was observed that the alumina was negatively charged on the almost whole area while a part was positively charged. On the other hand, for the coated sample, no positive charge was observed. From the above results, it was confirmed that the electric field was locally amplified by locally positive charging, leading to an increase in current and discharge in the previous study.

Speaker: Chihiro Tateyama (Toshiba Infrastructure Systems & Solutions Corporation)
• 35
Progress in Vacuum Flashover Mechanism: Applicability of ETPR Theory

I. INTRODUCTION
Surface flashover is a ubiquitous insulation failure that exhibits a far lower breakdown voltage than that of volume breakdown for the same dimension[1]. It remains the primary restriction with respect to the development of high-temperature superconductivity (HTS) equipment, spacecraft solar arrays, and pulse-power devices. There is little consensus regarding the intermediate discharge mechanism of surface flashover. In recent years, interfacial secondary electron emission derived from field emission has been continuously developed and achieved general acceptance[2]. Nevertheless, electron triggered polarization relaxation (ETPR) theory, a below-surface mechanism proposed in 1991[1], is still controversial. Less progress on flashover mechanism from the perspective of dielectrics restricts innovations in insulating materials and prevents a better understanding of flashover.
We report the flashover voltages of 5% mol Bi0.95Y0.05FeO3/epoxy composites (BY5FO/EP) with a wide range of filler contents. The polarization of composites was simulated, and the relationship with the flashover variations was established. Meanwhile, whether the ETPR theory is applicable for nanosecond flashover is discussed. Frequency and air molecular concentration are proposed which can lead the dominant flashover mechanism transition.
II. EXPERIMENTAL
A. Sample preparation
The details about how to prepare Y doped BiFeO3 and composites.
B. Characterization and Experimental test
Flashover test and detail parameter setting.
III. AVAILABILITY OF ETPR THEORY FOR DC FLASHOVER
A. Flashover sine shaped variation
The flashover voltages show sine-shaped variations. In order to better understand this variation, we define it as ‘X’ curve which explains the interaction between filler and matrix. The ‘X’ curve can be divided into a variable-frequency sine Eq. (1) f1(x) and a double exponential function Eq. (2) f2(x). Bi-exponential f2(x) can be used as a extended model to illustrate other properties which show different trends before and after percolation.
B. Simulation of polarization
We believe the variations in flashover voltage related to polarization of composites and computed the local polarization of the composites based on a well-established multi-layer core model and filler-matrix phase reversal.
C. Trap parameters
We also calculate the trap depth of composite as filler increasing which also show similar variation. Simultaneous sine-variation of dielectric constant and trap depth leads to regular variation of flashover. Meanwhile, the voltage conditioning can further prove the contribution of polarization energy and trap stability.
IV. PREDICT OF APPLICATION SCOPE OF ETPR THEORY
Through dielectric property as a function of frequency, we analyze the reason for limitation of ETPR theory for nanosecond flashover. However, we proposed some methods to verify whether the ETPR theory is applicable for nanosecond impulse flashover.
V. TRANSITION OF DOMINANT MECHANISMS
We report the transition of dominant flashover mechanisms from two dimension: air pressure and frequency, and discuss two extremely cases applicable for SEEA and ETPR theory.
VI. INTERPRETATION OF CONTROVERSIAL PHENOMENA
Based on our experimental results and analysis, the controversial phenomena including the effect of impurity, surface conductivity and dielectric constant on flashover performance are explained.
[1] G. Blaise and C. L. Gressus, J. Appl. Phys. 69, 6334 (1991).
[2] R. A Anderson and J. P Brainard, J. Appl. Phys. 51, 1414 (1980).

Speaker: Dr Kai Yin (Aalborg University)
• 36
Role of Interfacial Molecular Structure on Surface Flashover in Vacuum

To investigate the influence of surface molecular structure on vacuum flashover, specimens with different interfacial fluorine-containing chains are prepared through UV curing process. Surface elemental composition and roughness are characterized by X-ray photoelectron spectra and mobile roughness measuring instrument. Surface charge decay behavior in air and charge accumulation characteristics in vacuum are further investigated. Experimental results show that molecular structure has significant influence on electrical properties after eliminating the influence of surface roughness. The introduction of oxide in C-F chains can deteriorate insulating strength. Also, the surface charge decay and accumulation results indicate that surface chemical state is the fundamental causes for surface charge behavior rather than surface conductivity. This work offers an opportunity for the studies of physical essence in the field of fluorination for dielectric.

Speaker: Mr Chao Wang (Xi'an Jiaotong University)
• 10:30 AM
Coffe break
• ORAL: SESSION 010

Topic B: Vacuum Arcs

• 37
Hybrid Simulation of the Sputtering Phenomena between the Vacuum Arc Plasma Jet and the Anode in an Axial Magnetic Field

• During the forced extinguish period of the DC vacuum arc, as the current is reduced to zero in tens of microseconds, the arc becomes unstable and exhibits more transient characteristics compared with the AC vacuum arc, which will bring a lot of difficulties to the dielectric recovery process after current zero. The objective of this paper is to investigate the transient behavior of the plasma during the DC vacuum arc forced extinguish period. In this paper, for the first time, a 3D vacuum arc model based on hybrid method coupling the external circuit is developed. The hybrid method treats the ions kinetically and electrons as a massless fluid. In addition, an external electric circuit is incorporated into this model to achieve the calculation of the arc current in a self-consistent manner. In this way, the vacuum arc model can contains the plasma jets ejected from the separate cathode spots and the cathode spots can be turned off individually during the simulation. From the results, it can be seen that the current drops and the total amount of the ions decreases during the extinguish period. Besides, the ions density distribution becomes asymmetric due to the closure of the cathode spots and the average velocity of ions gradually increase. Finally, the interruption will be more difficult when the ramp down time of the arc current becomes shorter.

Speaker: Dr Liqiong Sun (State Key Laboratory of Electrical Insulation and Power Equipment, Xi’an Jiaotong University)
• 38
Development of a Numerical Model on Metallic Vapor Behaviour in Vacuum Arc based on Moving Particle Semi-implicit (MPS) Method and Finite Volume Method

A high voltage rated vacuum interrupter (VI) for vacuum circuit breakers (VCB) has recently been developed, because vacuum is an environmentally friendly high-voltage withstand medium compared to fluoride gas. Therefore, investigating arc behavior under reduced pressure is important for high voltage VCB design. In a vacuum circuit breaker, the arc plasma is maintained only by the metal vapor itself supplied from the electrodes. Metal vapor is generated by evaporation from the cathode and anode at high current densities. The emitted metal atom vapor is ionized to generate ions and electrons, forming an arc plasma.
In order to simulate such an arc plasma under reduced pressure, several numerical models have been developed. One is a fluid model, which is used to simulate arc plasma dynamics under reduced pressure during the high current phase. However, it is difficult to simulate low pressure, low current phase arc dynamics using a fluid model. Another calculation method is the particle method. Several particle methods have been developed, such as the Direct Simulation Monte Carlo (DSMC) method and the Particle-in-Cell (PIC) method. The DSMC method is often used to solve Boltzmann equations and simulate particle transport, for example, in vacuum arc deposition. The PIC method is also widely used for dynamic simulation of charged particles in low-pressure plasma. However, these methods require a long computation time to solve because of the very short time steps.
In this study, we developed a unique hybrid model for the dynamic behavior of metal vapor in a vacuum arc [1]. This model is based on the Moving Particle Semi-implicit (MPS) method for hydrodynamics of metal vapors, treated as a combination of many fluid elements with free boundaries. The original MPS method was developed to simulate incompressible flow dynamics. On the other hand, in this study, the original MPS method was modified to handle both compressible and incompressible flows of vapor, taking into account the vapor sound speed. This MPS method solves the Navier-Stokes equation, the Poisson’s equation for pressure and the energy transport equation for dynamics of vapor elements. On the other hand, electron energy conservation for electron temperature and Possion’s equations for electrostatic and vector potentials for electro-magnetic field were solved by Finite Volume Method (FVM). Thus, the resultant hybrid model can treat ions and atoms as heavy particles by the MPS model, whereas electrons as a fluid by the FVM method.
The developed simulation model can calculate temporal evolutions in particle density for each of species, heavy particle temperature, ionization degree, current density distributions between the electrodes during vacuum arc with electromagnetic field. The calculation can be done in several hours, which is relatively shorter than those required for conventional particle models like DSMC and PIC.
[1] Y.Tanaka, et al., ICEPE-ST2017, E-O-3, 2017.10,Xian, China

Speaker: Prof. Yasunori Tanaka (Kanazawa University)
• 39
Simulation of the post-arc sheath expansion process of vacuum circuit breaker

The topic of this paper is about ‘Computer modeling and computer aided design’ and ‘Vacuum arc physics’ of the Vacuum Arc.
The post-arc dielectric recovery process plays a vital role in the successful breaking process of VCBs. The electrons and ions of residual plasma separate from each electrode under the effect of a transient recovery voltage(TRV), and then a post-arc sheath forms in front of the post-arc cathode.
The post-arc sheath expansion process is the first stage of the dielectric recovery process. In order to investigate the microscopic mechanism of the this process, the Particle-in-Cell(PIC) method was adopted and a two-dimentional simulation model was developed in this paper.
The simulation will run by VSim. The dynamic particle distribution and the decay time of charged particles were obtained under different initial conditions. The influence of the residual plasma density, the rising rate of transient recovery voltage (TRV) and the gap distance at current zero is predicted. When other parameters are constant, the sheath will expand slower with higher initial plasma density as well as longer gap distance. The lager the rising rate of TRV is , the faster the sheath develops.

Speaker: Mr Shuang Leng (Dalian University of Technology)
• 40
Numerical Simulation of Transient Vacuum Arc with Active Anode Modes

This work investigates the transient arc behaviors during the anode spot mode with CuCr25 electrodes. In this simulation, a 2D transient magneto-hydro-dynamic (MHD) model considering multi-components is used to calculate the arc parameters. Then the energy flux is obtained to calculate the anode temperature which is set as the temperature boundary condition of the anode side in the MHD model in next time step. Simulation results show that most conclusions are similar with those given in our previous steady model. As for the transient process, when the anode temperature is lower than 1500K, the anode vapor can hardly be seen since its density is very lower. When the anode temperature is higher, the anode vapor density increases significantly. As a result, ions from the cathode are decelerated by the anode vapor when they arrive at the ionization layer, leading to high ion temperature here. Then the arc current are more likely to flow to the edge of the atom vapor area instead of the anode center. Thus, the energy flux to the anode begins to heat the area around the anode spot, leading to a uniform anode temperature distribution and a larger anode spot. The maximum anode temperature occurs after the peak current due to the thermal inertia. Then the anode temperature begins to decrease and remains uniform. The simulation results are consistent with the experimental results.

Speaker: Ze Yang
• 1:00 PM
Lunch break
• PISC meeting
• POSTER 1: Session #3
• 4:00 PM
Coffe break
• ORAL: SESSION 011

Topic C: Applications

• 41
Quality Assurance of Vacuum Interrupters Intended for Transmission Voltage Level Applications

ABSTRACT
Vacuum Interrupters (VI’s) have since their first commercial introduction in the late 1950s proven to be an extremely reliable component for use in all Low- and Medium-Voltage power switching applications. Today vacuum technology dominates in the distribution voltage level markets. One of the unknowns in the early days was the expected life of these sealed devices. Sealing techniques where in their beginnings and knowledge of gasses penetrating through (tiny) leaks, out-gassing of materials and permeation of gases through the enclosure were barely understood. In those days measuring even the degree of vacuum in pumped systems was still in its infancy and limited in range and accuracy. Under these circumstances it was virtually impossible to determine the vacuum life of hermetically sealed devices which were intended to be applied in switchgear systems with a service life of 30, 40 or 50 years. However, materials and components, design, processing and quality control as also measurement techniques have greatly improved over the years and vacuum life for factory new interrupters can nowadays be determined well beyond the longest expected service life. Modern VI’s for use up to and including distribution voltage levels have a design life of at least hundred years surpassing the service life of any practical switchgear system.
Resent developments however, where single-break VI’s are being introduced in the lower transmission voltage levels starting from 72.5kV and being applied in (high pressure) gasses other than air, raises the question if the current measurement techniques can be applied one-on-one on these substantial larger devices and maintain a sealed for life assurance for interrupters leaving the factory. The overall impact of an interrupter failure during its service life due to a poor degree of vacuum in the transmission voltage level domain could be substantial. Therefore, it is essential to have a proper vacuum measurement system in place for these new interrupter styles to ensure shipment of sealed for life devices.
This paper presents the current vacuum measurement technique for VI’s as also some basic design considerations like the use, selection and placement of getter materials. In the second part we will take a closer look at the current measurement technique to see if it is suitable or could be made suitable for VI’s with a large free internal volume and intended for use in other gases than air, like e.g.: N2, SF6, CO2 or NovecTM and gas mixtures under elevated pressures up to about 0.8MPa. Finally, the outcome of the investigation and recommendations are being discussed.

Speaker: Martin Leusenkamp (Eaton Corporation)
• 42
Prestrike Characteristics of Vacuum Circuit Breakers with Cup-type AMF Contacts and Spiral-type TMF Contacts under Capacitive Making Operations

The prestrike gap of vacuum interrupters is one of the most important characteristics for the investigation of the injected energy into the contacts from the prestrike arcs during the making operations and of the dielectric strength between the electrodes in vacuum interrupters. The objective of this paper is to analyze the characteristics of the prestrike gap distribution under the influence of different amplitudes of inrush current during the capacitive current prestrike process. A commercial double-break circuit breaker with two 12 kV vacuum interrupters is used during the experimental tests. The tests are operated for 30 cycles of making operations under the inrush current of 10 kA and 20 kA separately. The displacement of the moving contact is measured with a high precision linear displacement transducer from which the prestrike gaps can be computed. Weibull distribution is used to fit the prestrike gaps and the prestrike electric field strength. The experimental results show that with the increment of the inrush current between the contacts, the prestrike gap tends to increase, indicating that the possibility of breakdown between the contacts increases. Moreover, the scattering effect of the prestrike gaps becomes stronger. Compared with the single-break circuit breaker, the double-break circuit breaker can reduce the scattering effect of the prestrike gaps under the same amplitude of the inrush current.

Speaker: Yun Geng (Xi'an Jiaotong University)
• 43
Research on the Insulation Characteristics of the Shields Gaps in Vacuum Interrupters

The objective of this paper is to understand the insulation characteristics of gaps between shields in vacuum interrupter for standard lightning impulse voltage. The materials of the experimental shield were copper and stainless steel. The curvature radius at the edge of the shield was 3 mm. We designed two vacuum interrupters in experiment, and the gaps between shields could adjust manually from 4 mm to 8 mm. In experiment, the external insulation of the vacuum interrupter was insulated by oil, and the negative polarity lightning impulse voltage (12/50μs) was applied by an up-and-down method. The experimental results reveal that the insulation characteristic was severe influence by the times of imposed lightning impulse voltage. By continuously imposed the lightning impulse voltage on vacuum interrupter, the insulation strength increased to high voltages range and maintain stable in a definite time. Then the insulation strength reduced to low voltage range and maintained stable, and the insulation strength increased very low with the gap distance increased from 4 to 8 mm. The 50% breakdown voltage was calculated and used to evaluate the insulation strength. After many times of imposed lightning impulse voltage on vacuum interrupter, the ceramic was coated by metal vapor from the shield material. The metal layer on the ceramic surface have changed the electric field distribution in vacuum interrupters and destroyed the insulation performance.

Speaker: Weigang Feng (Xi‘an Jiaotong University)
• 44
The Impact of Particles on Post-arc breakdowns in Capacitive Interruptions

The advantage of long electrical life make vacuum circuit breakers suitable for clearing the capacitive loads. However, the high probability of post-arc breakdowns threatens the safety and reliability of the power system potentially. It is believed that particles generated in the interruption processes are responsible for triggering the breakdowns. The objective of this paper is experimentally observing the generation and dynamics of particles in capacitive interruptions as well as their resultant effects on post-arc breakdown, to further reveal possible mechanisms for particle-induced breakdown. An optical diagnostic platform based on a laser-shadow technique was constructed to detect the particles generated by different electrodes materials. Meanwhile, a 12 kV synthetic test circuit which contained an inrush current source was set up. Particles generation was observed to be closely related to the contact materials and inrush Currents. Additionally, under the influence of the recovery voltage, particles showed complex and diverse dynamic behaviors, such as bouncing, oscillating, levitating, lifting, and rotating. Besides, particles generated in capacitive interruption were inclined to wander for a longer time in the vacuum gap, lasting 200 ms after arc extinction. More importantly, it was found that post-arc re-breakdowns were more likely to occur when particles were close to the electrode at higher recovery voltages. Based on particles behaviors, breakdown moment and recovery voltage, particles were primarily responsible for the long delayed restrikes.
Key words: Vacuum circuit breakers, free particles, capacitive interruptions, restrikes, inrush current.

Speakers: Wenlong Yan (Xi’an Jiaotong University) , Zhenxing Wang (Xi'an Jiaotong University)
• 7:00 PM
BUS TO VILLA FOSCARINI
• 7:30 PM
CONFERENCE DINNER @ VILLA FOSCARINI ROSSI Villa Foscarini Rossi

#### Villa Foscarini Rossi

Via Doge Pisani, 1/2, 30039 Stra VE
• 11:30 PM
BUS TO ORTO BOTANICO
• Friday, 1 October
• Registration

Attendees Registration at the conference desk at Orto Botanico

• ORAL: Session #10
• 10:30 AM
Coffe break
• ORAL: Session #11
• 1:00 PM
CLOSING SESSION