The paper describes the results of a direct experimental observation of the development of an electrohydrodynamic instability at an explosive-emission liquid-metal cathode along with a theoretical analysis of the growth and destruction of the protrusion formed on the cathode surface.

An experimental set-up allowing measurements of DC field emission currents in the range of 10^-18 A to 10^-6 A in ultra high vacuum (P < 10^-10 Torr) at room temperature has been developed. The detection of currents in the range of 10^-18 A to 10^-12 A was carried out with an electron multiplier located behind a grid anode. Higher currents were measured using a massive anode and a picoammeter. The whole system was designed to study niobium samples from sheets used to build electron superconducting accelerator cavities. Comparisons of the electron field emission from samples prepared by different surface treatments as chemical etching, electropolishing, oxide coating by anodization, were done. Some of the obtained results are presented below.

A fast-video imaging technique was used to observe the pulsed breakdown behavior of an alumina tube, having two concentric planar electrodes on its end-face. Voltage-pulses, typically of 5 kV amplitude (5 kV microsecond(s) ^-1 time rate-of-rise), were applied to the radial M-I-M insulator-electrode regime under ultra-high vacuum (pressure < 10^-8 Torr), with video recordings made at 1000 frames per second. Images of the observed breakdown phenomena are presented, viewing along the center-axis and also in the plane of the M-I-M structure. These images are discussed in relation to plasma-jets associated with vacuum arcs, and the nature of the ion species within such jets.

The results of the investigation of amplitude and spatial-time characteristics of the initial conduction current at pulse voltage of a microsecond duration in the centimeter gaps in commercial vacuum are presented and discussed.

The electrodes used in this investigation to experimentally evaluate the effect of diamond turning on electrical breakdown characteristics in vacuum were first roughly machined by turning, and then machined to a mirror finish by diamond turning. Residual stresses of the electrode surfaces were measured by an X-ray diffraction method before and after the diamond turning. Several electrodes were annealed in vacuum for one hour at 400 degree(s)C or 700 degree(s)C before the diamond turning. Vacuum breakdown experiments using impulse voltages with the wave form of 64/700 microsecond(s) revealed that the diamond turning improves breakdown strength, that the conditioning of vacuum gaps is achieved by only few tens of breakdowns, and that annealed electrodes have a better hold-off voltage capability.

Stability of output parameters of power high-voltage acuuin devices is connected with the probles of interelectrode vacuua gap electric strength and ectrode tihera1 kO&dS increac a WOli as dveiopent of tab1e forsi electrodes. One of the way to solve this problea is using graphite being a well as known in eIectronis material or swe other attrials on the base of carbon with their unique aechanical and thermophysical properties as soon as their essential defect, the law electricstrength is eliminated.

The paper deals with ectons, emission centers produced on the surface of a cathode as a result of heating and explosion of its material due to large specific energy accumulated in cathode microvolumes. The criteria for the initiation and self-sustaining of ectons are given and their parameters and types are discussed.

For production of ionized atom fluxes a method of evaporating materials in vacuum by means of non-self-maintained discharge was developed. The conditions of the initiation discharge were examined. The necessary conditions are attainment of fixed vapor or gas pressure between the electrodes and presence of pre-breakdown current. The characteristics of the discharge in the voltage range from 30 to 700 V with currents of 1 to 50 A were investigated. The discharge plasma was studied with the assistance of electrostatic probes. The considered discharge region occupies the transition region from glow to arc. The cathode temperature is one of the fundamental parameters which defines the type of discharge and its burning. The regulation of temperature and discharge power influences on the processes which proceed on the cathode and in the space between the electrodes. The operation in different regimes results in the ability to control of ionization degree from 2 to 100%.

Using three kinds of copper electrodes with different surface areas, experiments were performed to investigate the influence of electrode area on conditioning effect, which is a characteristic of dielectric breakdowns in vacuum gaps. The conditioning process varied with electrode area: the smaller the electrode area, the sooner conditioning ended. Breakdown voltages after completion of conditioning also depended on electrode area: the smaller the electrode area, the higher the breakdown voltage.

Results are presented of studies of pulsed electric strength and threshold current densities for niobium cathodes in the superconducting and normal states. The process of heating and failure of cathode microemitters under the action of high density field emission current is analyzed. Data of the work function, limiting current densities and tensile strength of YBa₂Cu₃O_7-x cathode tips, effect of emission current on the superconducting properties of YBa₂Cu₃O_7-x thin films and peculiarities of prebreakdown phenomena in vacuum gap with YBa₂Cu₃O_7-x electrodes are given.

The cause of pre-breakdown currents appearing between metal electrodes copper soldered with ceramics after bakeout in vacuum up to 200...440 degree(s)C was established. Quest for it was conducted from the viewpoint of a hypothesis suggested by an author earlier on vacuum insulation violation. `Triple point'--metal-insulator contact in vacuum--is referred to as a pore in the gas medium of which ionization process resulting in various kinds of vacuum insulation violation develop. Experimental results confirm the hypothesis suggested.

`The total voltage effect'--being a main characteristic of vacuum insulation is explained from an earlier offered mechanism of vacuum insulation deterioration point of view. When an external electric field sagging in the negative electrode cracks and pores in their volume ionization processes occur. Gas medium in the pores appears as a result of the resonance gas desorption from the pores surface. `The total current effect' is explained by the same processes.

The effect of injection processes on the pulse discharges propagation mechanism in solid dielectrics has been shown. It has been established that during a cathode discharge these processes are analogous to laser action on solids, while in the case of an anode discharge they are similar to detonations of explosives.

A discharge initiation system is developed which is intended for discharge gap breakdown in an electrothermal launcher. It consists of a high voltage transformer, a pulse sharpening discharge element and a nonlinear inductance or magnetic switch. The high voltage transformer with air as dielectric produces output voltage up to 130 kV. It permits to realize with certainty the breakdown of gaps with length up to 3 cm. The breakdown occurs along the surface of dielectric dividing the gap electrodes, polyethylene, fluoric plastic and ceramics being used as a material between electrodes.

A `transparent anode' imaging technique has been used to display the spatial distribution of prebreakdown emission sites and the location of subsequent breakdown events. Using a real time video recording technique, it has been possible to address the following issues: (1) whether or not a spatial correlation exists between a DC prebreakdown emission site and the location of a subsequent DC or pulsed-field (PF) breakdown event, and (2) the evolutionary character of both DC and PF breakdown processes.

In nanosecond vacuum arc discharges, the size of cathode plasmas has been found to have the order of 10 micrometers . This outcome has decisive consequences for particle density, current density, time constants, and radiation properties. The present paper explores the question whether there occur principal changes in longer discharges. Experimental methods were: imaging the spots of microsecond(s) -discharges by absorption and emission techniques, and monitoring the shape of spectral lines in the range of nanoseconds through milliseconds. It has been found that in the light of atomic or ionic lines the diameter of emission pictures have the order of 100 micrometers , irrespective of the discharge duration in the range 10 ns through 100 microsecond(s) . Pictures obtained by differential laser absorption in the time range 100 ns through 50 microsecond(s) yielded a size <EQ 30 micrometers . Time-integrated analysis of emitted spectral lines yielded broadened profiles for discharges of 800 ns duration with a FWHM (full width at half maximum) up to 0.5 nm. This indicates the effect of high densities and magnetic fields, possible only with small spots. The same has been found for discharge durations up to 100 ms. Thus, it is suggested that the spot size remains small up to discharge durations of milliseconds. This outcome is associated with a principal non-stationary spot nature.

The influence of the Buneman instability on the parameters of a current carrying plasma is investigated. By using the quasi-linear approximation the analytical expressions which connect the plasma parameter changes and the turbulent electrical conductivity with the energy of the unstable electrostatic field fluctuations were obtained. The latter is expressed by the plasma supercritically. The experimental data of the arc discharge in an extended gap are analyzed. It is shown that the changes of the electron temperature and potential drop can be explained by this instability.

In this paper the processes in the low arc current (less than 1 kA) on a cathode with moderate melting temperature are discussed. In this arc an anode typically plays a passive part, acting only as a collector of the microparticles, metal vapor, ion and electron flux, which are ejected into a gap from the cathode spot region.

In this paper we shall discuss the peculiarities of the arc discharge in large interelectrode gaps and the breakdown of the near-electrode plasma sheath depending on the surface condition.

The cathode spot vacuum produces a jet of highly ionized plasma plus a spray of liquid droplets, both consisting of cathode material. The droplets are filtered from the plasma by passing the plasma through a curved, magnetic duct. A radial magnetic field may be applied to the face of the cathode to rotate and distribute the cathode spots in order to obtain even erosion and avoid local overheating.

The paper discusses experimental data which indicate that type 2 cathode spots originate only on oxide-cleaned metal surfaces. The suggestion is made that an instant at which an ion resonance recharging starts to play an active role in a near-cathode region can be taken as a criterion of type 2 spot origination.

This paper reports studies of contact admixtures effects on recovery voltages of contact pairs when admixtures have a form of multiple inclusions of some metal oxides introduced into standard contact material CuCr 50/50. The breakdown and recovery voltage studies of these materials were carried out at a simulating circuit with load current up to 300 A and frequencies 50 kHz - 120 kHz. Grains of alumina, and hafnium, lithium and lanthan oxides were used as inclusions. CuCr 50/50 with small tungsten chips was used as reference material. Statistical characteristics of recovery voltages were obtained for these materials and some promising compositions were selected for future studies.

Comparative characteristics of the cathode processes in the discharge with electrodes of pure metal and of CuCr composition are presented. The effect of the gap, electrode shape and other geometric factors on the cathode spots dynamics is investigated. The accomplished investigations allowed the elaborate methods to influence the cathode spot dynamics by geometric factors.

Plasma sheath near the substrate plays an important role in Vacuum Arc Deposition Technology (VADT). To analyze the sheath characteristics of the substrate in VADT, the unique ejection pattern of the cathode region must be taken into account. This paper describes a steady-state, two-dimension sheath model in which ions distribute specially according to a cosinoidal law. The model equations can be numerical soluted under specified conditions, to describe the sheath behavior of the substrate in the process of titanium film deposition using VADT. The model can be used for calculation of ion flux and heat flux flowing to the substrate in VADT, or in some other cases in which biased conductors were in vacuum arcs with cathode spots.

The paper presents a 2D hydrodynamic model of the plasma jet running out of the cathode spot of a vacuum arc, which is an outgrowth of the 1D problem that we have solved earlier. Also given are refined equations of state for a matter in a transitory phase, which conceivably might allow a through numerical simulation of a cathode spot from cold metal to hot plasma.

Hollow-cathode glow discharge, which operates at gaa pressures as low as those corresponding to a vacuum regime with \ d, where ) is the free path leugth of electron, d is the Bize of a cavity, 18used in electron and ion beam sources with plasma emitter.' However, a signilIcant difference in value between igniting and operating discharge voltages motivates the use of special systems for discharge ignition. Such systems, operating under increased preaurcs as compared to the preesure in the hollow cathode or using a high voltage ignition pulses, complicate source design and its electric power supplies.

A theoretical model is presented for transport of vacuum arc generated metal vapor plasma through a magnetized quarter-tours duct used for filtering out macroparticles in order to deposit high quality thin films. The model utilizes a two fluid approximation which takes into account collisions among the plasma particles. It is found that centrifugal forces must lead to a charge separation generated field, that determines plasma drift in the centrifugal force direction to the duct wall and give rise to ion loss. Another cause for plasma is the plasma pressure gradient. The plasma output flux is an increasing function of the magnetic field strength. The plasma flux in the output plane is asymmetrically skewed to favor the outside half. A further asymmetry in the flux distribution in the direction of the torroidal axis of symmetry is introduced if ions of different charge states are present in the plasma.

Multifluid equations describing the plasma of vacuum arcs expanding anisotropically from a cathode spot are given and discussed. Some first and preliminary results from an approximate analytical integration of such a system of equations are presented. Though the dependence of the plasma parameters on the direction is weak, the inclusion of angular terms and angular equations changes some results significantly, compared with the experiences from usual 1D models.

The present paper discusses the appearance of arc cathode spots observed in emission. Photographs yield a size of about 100 micrometers when using line radiation, whereas imaging by absorption techniques or emission pictures with continuum radiation result in values of about 10 micrometers , comparable to the crater traces left at the surface. Besides the size, the slope of the spot profiles is considered. Experimentally, it is found that the light intensity decreases with the distance r from the spot center proportional to r^-2. This slope is much less than the expected dependence proportional to r^-(beta), (beta) > 4, i.e. the observed spot edge is much less sharp than expected. The behavior of the radiation is theoretically studied by modeling the radial dependence of particle densities and temperature. It is shown that the line radiation stems from a region which is greatly influenced by the plasma expansion in competition with the finite transition probability of excited levels. It is concluded that emission pictures of cathode spot yield considerably higher spot diameters than corresponding to the spot core.

Voltage-current and other characteristics of the poorly explored type of discharge has been investigated and was found out to be quite different from the characteristics of other wide- known types of vacuum and gas discharges.

In this work the kinetics of macroparticle charging in the rarified part of the arc plasma jet are studied. The sheath in the vicinity of the macroparticle is collisionless and the problem with different Debye length to macroparticle radius ratio is analyzed. Maxwellian velocity distribution functions with different temperatures for the electrons and ions in an arbitrary ratio are allowed in the model. The influence of the plasma drift velocity on the macroparticle charging is discussed.

The paper deals with modeling processes at the electrode regions of pseudospark devices. We have analyzed the physical processes that could result in the initiation and development of explosive electron emission.

The results of investigation of high-current vacuum arc with Ti cathode filled by deuterium till x >= 1 (at.D/at.Ti) are presented. Electric and erosion characteristics were measured. The investigations proved this discharge to be very differing from the arc with degassed electrodes. The results obtained indicate the gas-filled electrodes have good prospects for some technical applications.

The paper presents the numerical implementation of the state-variable model of the high current arc interruption.

While one is trying to create the model of high speed charged participle flows from the cathode spot in vacuum arcs, attention should be paid to the non-contradictory choice of boundary conditions system.

Heat transfer to a thermally isolated graphite anode in a long duration vacuum arc was investigated. Anode bulk temperatures were measured as a function of time using two high temperature thermocouples. The anode surface temperature was optically determined. Surface temperatures of 2300 K were obtained in a 340 A arc. A 1D non-linear heat flow model for the anode was developed. A solution was obtained using a dynamic numerical method and the effective anode potential was determined to be approximately 6 V.

With the experiments presented in this paper applications of a retarding field analyzer for the measurement of ion potentials U_i in a vacuum arc plasma are discussed. The examined plasma was produced by a sinusoidal half-wave vacuum arc current. The experiments were concentrated on evaluating the plasma parameters at the last three milliseconds before current zero. In a current range from 300 A_rms to 10 kA_rms the ion potential distributions and their peak values were evaluated. With the increase of the arc current a decrease of the ion potential was found. By additional investigations of the angular distribution of the ion energies, a transition from a collision dominated interelectrode plasma to a freely expanding plasma was observed, depending on the arc current.

The use of reduced gaps in applications of high-current vacuum arc devices presents a number of interesting challenges. Specifically, standard contact designs have been developed over several decades to achieve controlled motion of high-current ac arcs in vacuum interrupters. For medium-voltage applications, the optimal maximum contact gap can typically range from about 6 mm to about 2 cm. However, the influence of the contact design may be gap dependent, so additional research may be appropriate if the contacts are to be used at smaller gaps. For example, the current through spiral contacts produces a magnetic field perpendicular to the arc column, but this will force the arc to move outward and run along the periphery of the petals only if a threshold separation is achieved. In this investigation, a framing camera was used to record the appearance and motion of drawn vacuum arcs between spiral-petal contacts with final gaps of 2 to 3 mm. After the rupture of the molten bridge, a high-pressure arc column formed and expanded across the width of the spiral arm. At the reduced gap, an intense anode spot formed if the peak current exceeded approximately 15 kA. Compared to results previously obtained at larger gaps, the arc motion was greatly reduced, and severe contact damage was observed at lower currents.

In this paper, the high current vacuum arc behavior with horse shoe electrode has been described. These are dynamic vacuum arc distribution, the maximum current density related to the interrupting limit and factors which influences the interrupting limits. Then based on the above analyses, several vacuum interrupters within (phi) 70 mm outside diameter have been tested. From the results, it shows that the small size vacuum interrupters can be made with the oblique, ellipse shaped horseshoe electrode. The interrupting ability of such type vacuum interrupters is further increased.

This paper analyzes the effect of the gap of electrodes on the state of vacuum arc by experiment and theory. And the model of vacuum arc is set up. The optimal gap can be deduced from controlling the vacuum arc to be diffusion state, in order to get the optimal moving characteristics of electrodes.

The main parameters and dimensions of cathode spots were under discussion for years. To solve these current questions, a new system was especially designs. The image converting High Speed Framing Camera, which combines a microscopical resolution of 5 micrometers with a nanosecond time resolution and a very high optical sensitivity. This camera was used to study the microscopical behavior of vacuum arc cathode spots in a pulsed high current arc discharge on copper. The direct observation of these spots with high resolution revealed the conclusions that one single cathode spot, as normally observed by optical means consists of a number of simultaneously existing microscopical sub-spots, each of them with a diameter of about 15 micrometers and a mean distance of 30...50 micrometers between them. The mean existence time of these sub-spots on copper was found to be about 3.2 microsecond(s) , where the position of a sub-spot remains unchanged (with an upper limit of about 2...3 micrometers ) during its existence time. An upper limit of the crater surface temperature was estimated by a comparison between the brightness of a cathode spot and of a black body radiation lamp to about 3000 K.

Micron-size macroparticles can be removed from vacuum arc plasmas using magnetic filters, permitting the deposition of high-quality thin films of the cathode material. The principle of magnetic filtering is explained, and ways of improving filter efficiency using additional electric and magnetic fields are explored. Under optimum conditions, about 25% of all ions entering the filter can be used for deposition. Other problems such as down or upscaling of deposition facilities are briefly discussed.

The present paper deals with the creation of a new mathematical model for 2D-computer simulation of the axisymmetrical radiative plasma magnitohydrodynamics (MHD) flows under vacuum spark discharge. In difference with the earlier works we use not only one but also two temperature approximation and take into consideration not only electrical but also magnetic fields in equation's system of the mathematical model. Created MHD-model is based on implicit full conservative difference scheme in combined Eulerian and Lagrangian approach. The system of difference equations is solved by using of Newton-like iteration procedure. Groups of equations are selected in accordance with the similar character of physical processes and the energy balance calculation is carried out for the convergence control.

After interruption of a high-frequency (hf) current by a vacuum arc, two distinct types of reignition can be observed. The first type follows immediately after interruption, the second type tends to allow a currentless pause of a few tenths of a microsecond. The post-arc current that flows after hf-current interruption has been measured and has a peak of several A and a decay time of a several hundreds of ns. A dynamic sheath model for the decaying plasma after hf current interruption has been used successfully to model the measured post-arc current waveshapes. The first type of reignition may be attributed to thermal stress during the period in which no ion sheath has been formed yet, the second type to high electric fields, either due to the ion sheath, or due to the TRV, the weight of each depending on gaplength and di/dt.

The cathode spot formation within first 22 ns was investigated by laser absorption photography and ps-pulse interferometry. The discharge was initiated between W-, Ag-, AuNi-, Pd- electrodes with cathode-anode distance below 100 micrometers , the arc duration was some milliseconds and arc current 5 A. A ps-pulse holographic interferometer and momentary absorption photography enabled us to determinate spatial-temporal density distributions in the ignition phase of the cathode spot. An absolute electron density value of order of 3 - 10²⁶ m^-3 has been determined indicating high conductivity values of the metal vapor plasma. Present measurements show that cathode spot plasma is essentially non- ideal and verify theoretical calculations resulting in an ionization potential decrease in dense cathode plasmas.

Computer calculation of pure metal film thickness distribution, using vacuum arc deposition technique, is presented in this paper. For random arc with a lower arc current, a model of single cathode spot with ion beam flux taking into account the cosine function spatial distribution is deduced. For arc steered by an external magnetic field, which is parallel to the cathode surface, a multiple cathode spots model is developed. And, in the case of random arc, calculation comparison between static substrate and rotating substrate is made. Results show that film thickness distribution is non-uniform when the arc is not controlled and tends to be uniform when the arc is steered by external parallel magnetic field with cathode geometry and substrate location being well chosen. Also, film thickness distribution is more uniform on a rotating substrate than on a static substrate.

Characterization of the cathode spot, which connect to a vacuum dc arc on a consumable cathode as a source of material for deposition, is studied using a high speed photograph technique. For a titanium cathode, external magnetic field effects on the behavior of the cathode spot is given with respect to number, size, and motion. Lastly, some discussion to the experimental results are presented.

The shield current distribution in a vacuum arc between CuCr40 contacts was studied by means of three cylindrical shields arrangement. It was found that the total shield current is affected by the arc mode. For a high-voltage oscillation sequence a significant effect on shield current distribution has an anode plasma jet. Then a great increase of the shield current is observed and the ratio of shield current to arc current was even more than 20%.

This report reviews research and development efforts within the last years for vacuum electron tubes, in particular power grid tubes for industrial applications. Physical and chemical effects are discussed that determine the performance of todays devices. Due to the progress made in the fundamental understanding of materials and newly developed processes the reliability and reproducibility of power grid tubes could be improved considerably. Modern computer controlled manufacturing methods ensure a high reproducibility of production and continuous quality certification according to ISO 9001 guarantees future high quality standards. Some typical applications of these tubes are given as an example.

Linear electric circuit with constant parameters becomes a nonlinear one with parameters variable in time after vacuum discharge gap is connected to it. Change of discharge parameters occurs by cycle. Full vacuum discharge cycle includes predischarge, spark, arc phases and current break phase (Fig. 1). Discharge development by incomplete cycle as well as development of one and the same discharge by several parallel channels with different cycles are possible. Every moment discharge current I is limited by one ofthe factors and namely - cathode emission ability L, plasma emission ability I, vacuum gap conductivity P. outer circuit resistance R -but it can not be limited by simultaneous infidence of several these factors. Fragment of the scheme (Fig.2) reflecting succession of turning on of current limiting factors may present how the process ofevery discharge goes on.

Switching characteristics of the triggered vacuum interrupter (TVI) combining the properties of vacuum interrupter and triggered vacuum switch are studied. In order to reduce the chopping current, electrode insertions made of a highly arcing-proof composite material are used. Along with the expected low chopping current less than 1 A the TVI has good switching capabilities both at anode and cathode modes of operation with a trigger current I_t > 100 A and trigger voltage U_t > 500 V. The effect of the load on the switching capability of the interrupter at the anode mode of operation is investigated. The obtained results make it possible to specify the requirements for trigger source output parameters with the aim to provide a reliable switching for different polarities of the main electrodes. Model tests of TVI in power frequency networks showed the possibility of its utilization under controlled switching with an accuracy to microsecond units and keeping all the properties of the vacuum interrupters.

The small-size sealed-off triggered vacuum switch TVS-7 designed to switch current pulse to 10 kA at high repetition rate to 1 kHz and maximum voltage to 25 kV is described. TVS switching characteristics are studied systematically versus the trigger current parameters for different polarities of the main electrodes. It is shown the TVS ensures a monotonous current rise of submicrosecond duration with a rate of current rise up to 10¹¹ A/s and a minimum switch-on delay time of 50 ns at a trigger current amplitude of I_t >= 100 A. In the cathode mode of operation the TVS is reliably triggered with reducing of trigger current up to I_t >= 1 A. But in this case current and voltage oscillations occur at an initial discharge development stage. This unstable stage of discharge burning is subjected to investigation. A minimum trigger energy, of about 10 mJ is required to ensure stable conditions of discharge burning with any polarity at the main electrodes. Under the condition power losses at switch-on stage of the TVS are substantially reduced in comparison with those of traditional vacuum switches. This became possible because of the use of a two-stage trigger system of the erosion type and a small main vacuum gap.

The paper considers pulsed power repetitive systems using plasma opening switches. The solutions of a row of technical problems connected with both repetitive operation of accelerator having high peak power and x-ray radiation converter are proposed. These machines are to be prospective for advanced radiation technologies.

We made some experiments on "Angara-5-l" (2-3 MA, 100 ns) [1], using low linear mass liner (10-40 iukg/cm ) as a part of double Z-pinch. The low mass liner surrounded coaxially a cylindrical load of higher mass (100-200 mkg/cin). Current instability and poor compression of the low mass liner were specific features of that inhoinogeneous Z-pinch. We registered '-0.5 TW pulse of soft X-ray radiation from inner load in the instability moment. We treat the effect as a Plasma Opening Switch (P05) process in external current shell of double pinch.

The current switching by means of a gas-filled electron switch has been studied experimentally and theoretically. Under deep volume charge compensation conditions the switch showed complete controllability. The switched current density was more than current density of a similar vacuum switch by an order.

In this report the results of the optimization of the multi-mode mcs-scale PEOS-based generator parameter's are presented. The PEOS parameters influence on the energy transfer efficiency was studied, the PEOS being considered as the part of the magnetically self- insulated line.