High-voltage quasi-rectangular nanosecond pulses hold significant promise for pulsed power applications. Nevertheless, research on generating such pulses via gas-insulated compact Marx generators (CMGs) remains challenging. This study presents a modified CMG incorporating an auxiliary waveform modulation structure that enables flexible control of stray parameters to achieve time-separated discharge phases. Without additional peaking circuits or complex pulse-forming networks (PFNs), the generator is capable of producing fast-rising quasi-rectangular pulses through optimized parameters of circuit elements. A ten-stage prototype is designed and implemented, housed in a compact pressure vessel of 216-mm diameter and 700-mm length. Experimental results demonstrate that the generator delivers bipolar pulses with amplitudes exceeding +/- 200 kV, rise times below 6.5 ns, plateau durations over 50 ns, and trigger jitter under 1 ns across an open-terminated 50-Omega coaxial cable. The trigger-erection time delay jitter remains below 1 ns throughout the full operating voltage range, ensuring precise temporal control. When used to trigger a 3.5-MV gas-insulated linear large-scale Marx generator, it achieves a trigger delay jitter of 2.5 ns, exhibiting high stability and precision. The proposed system offers a compact, reliable solution for high-voltage nanosecond pulse generation with superior waveform quality and timing control.

In the simulation of electric fields and coupling responses within a bounded wave electromagnetic pulse simulator, obtaining accurate waveforms for high-voltage source excitation proves challenging. Field test waveforms are usually used to replace the source excitation waveforms in the simulation environment, which can cause significant errors between actual simulation results and experimental results. A method to quickly derive excitation signal waveform parameters for the electromagnetic pulse simulator by utilizing the measured coupling current of the antenna in the bounded wave simulator is introduced in this study. The proposed approach involves treating the test environment, including the antenna, as a two-port network in which the input side of this network represents the high-voltage pulse source of the electromagnetic pulse simulator, while the output is connected to the standard load. Subsequently, the transfer function between the voltage of the high-voltage pulse source and the coupling current on the load is calculated. By considering the waveform parameters of the high-voltage pulse source as optimization variables and comparing coupling current waveforms obtained through actual measurement with those derived from the transfer function, the waveform parameters of the high-voltage pulse source are determined through optimization. To validate the effectiveness of the proposed method, tests and simulations were implemented on a shortwave antenna. The obtained test data aligned with simulation results, confirming the effectiveness of the approach.

The transfer capacitor is an essential component in pulse compression circuits, and its capacitance can significantly impact the output pulse waveform. During the transportation and operation of transfer capacitors, changes in their internal mechanical structure may occur, affecting the capacitance of the transfer capacitor and disrupting the operation of the devices. In this research, the variation of capacitance of transfer capacitors under mechanical stress is investigated. The results indicate that under minor stable vibrations, the capacitance of the transfer capacitor remains relatively constant. However, when exposed to impact stress, the capacitance undergoes discernible changes. This phenomenon arises from the influence of impact stress on the mechanical structure of the transfer capacitor, leading to alterations in the distance between the two electrodes of transfer capacitor.

The breakdown process of a self-triggered preionized switch with the four-electrode structure under nanosecond pulses is studied based on optical diagnosis. The breakdown of both subgaps initiates from the main electrode surface in the form of uniform discharge. A uniform-like low-impedance discharge channel forms first in the cathode side subgap. Then several streamer-like high-impedance discharge channels form later in the anode side subgap. Therefore, the anode side is under overvoltage after the cathode side discharge channel forms. The intensity of discharge channels indicates that the anode side subgap first breaks down. When the injection time of preionization is postponed, the uniform discharge in the cathode side converts into streamer. It is also proved that the trigger discharge can preionize each subgap separately, indicating that both photoemission and photoionization mechanisms contribute to the preionization process.

Analyzing the effect of temperature on the capacitance value of the film capacitors used in the electromagnetic pulse (EMP) simulator helps to ensure the stability of the output pulse. In this paper, a temperature loading test platform for the film capacitors is constructed, and the variation of capacitance values of film capacitors at different temperatures, heating times and alternating temperatures are tested and compared. The capacitance values of film capacitors are positively correlated with the temperature. At the same temperature and after applying the alternating-temperature stress to return to the ambient temperature, the capacitance values fluctuate up and down with the heating times, but the variation is not significant.

Coaxial peaking capacitor is a key component in high-altitude electromagnetic pulse (EMP) simulators with fast front pulse output. It poses significant technical and engineering challenges in limiting radiation field amplitude and test space. This paper presents the design and testing of a 180 pF, 3 MV coaxial peaking capacitor with improved insulation performance. In the insulation design, the length of the dielectric film is extended to reduce the background electric field on the flashover path. The electric field threshold obtained from image diagnosis is used as a reference. During capacitor testing, the insulation characteristics are diagnosed using both direct and indirect methods. The voltage measured by a D-dot probe, the output waveform of the Marx generator in the primary source, and the radiation field waveform are analyzed to understand the flashover characteristics of the capacitor and to improve the reliability of the test results. The experimental results demonstrate that the peaking capacitor can operate stably at 3.0 MV. If flashover occurring on the dropping edge of the pulse is permitted, the operating voltage can be greater than 3.7 MV without significantly affecting the radiation field waveform. The analysis on the surface flashover morphology of the peaking capacitor reveals that the flashover mainly occurs at the dropping edge of the capacitor's waveform, indicating that the damage to the film is not serious. This research significantly increases the working voltage of coaxial peaking capacitors and contributes to the development of high-altitude EMP simulation technology.

High-altitude electromagnetic pulse (HEMP) simulators, consisting of fast risetime pulsed drivers and large antennas, are utilized to produce intense transient electromagnetic environments in a large test volume greater than tens of meters. Key parameters of the electric fields of concern are limited in some special ranges, including pulse risetime of a few nanoseconds, full widths at half maximum of tens of nanoseconds, and peak intensities greater than 50 kV/m. These simulators can be used to test and evaluate the hardening performances of full-scale military equipment, such as large ground vehicles, aircraft, and ships. Globally, researchers target a fast risetime of 1-2 ns, high peak strength of 70-100 kV/m, and large test volume in the range of 30-300 m. Recently, both single-side and two-side driven fast pulsers were developed, with peak output voltages of 3 and 5 MV, respectively. New pulsers with improved antennas can help obtain a test volume greater than 70 m. However, it is a challenge to further improve the output voltage of single-side driven fast pulsers to more than 10 MV. Chinese researchers have developed several large HEMP simulators, with output voltages of single-side-driven fast pulsers greater than 3 MV. Given this background, the present study reviews the progress at Northwest Institute of Nuclear Technology and other research institutions, with respect to designs of critical components, such as compact primary high-voltage pulsers, pulsed megavolt capacitors with multilayer films, low-jitter gas switches, and large radiating antennas.

Fields of large size split vertically polarized bounded-wave electromagnetic pulse（EMP） simulator（called “Split simulator”） are simulated by using parallel finite-difference time-domain（FDTD） method. The influences of split simulator’s size parameters on the time-domain characteristics of the vertical component electric fields at the testing points on the symmetrical plane in the Split simulator are discussed and analyzed, the study of the transient distribution characteristics of the electric fields’ peak-values is given, and the comparison between the Split simulator and normal simulator is also given. The simulation results show that the higher the maxium height of the simulator, the higher the starting height of the testing point where the FWHM of the vertical component of the measuring point field is abnormal, and the peak-value of the vertical component electric field at the same testing point decreases with the increase of the maximum height of the Split simulator. The peak-values of the vertical component electric field at the testing points near the source and the lower plate decrease obviously, as the total upper plate’s projection length increases; The rise-time of the vertical component electric field at the same testing point becomes quickly, and the peak-value increases, as the trapezoid plate’s projection length increases; The FWHM of the vertical component electric field becomes narrower, and the peak-value decreases, as the maximum width of the upper plate’s void increases. Both the rise-time and FWHM of the electric field in the Split simulator meet the IEC standard, and the difference between the peak-values of electric fields in Split simulator and those in normal simulator is little, as appropriate size parameters and appropriate working-space height of the Split simulator are chosen.

Self-triggered preionized switches with four electrode structures are widely used in large pulsed power generators. Reducing its time delay jitter is an essential requirement. To understand the time delay, this article investigates its breakdown processes and mechanisms. The switch is divided into two symmetrical sub-gaps with a sphere-plate structure. The breakdown processes of the sub-gaps in nitrogen are photographed by an intensified charge coupled device (ICCD) camera. It is observed that two sub-gaps can be separately preionized by the spark discharge. Uniform discharge can be stimulated from the main electrodes of two sub-gaps. Increasing the gap distance or postponing the injection time of preionization makes the uniform discharge easier to transfer to streamer discharge. Combining the discharge phenomena and the spectra of trigger gap discharge, it is inferred that the initial electrons are generated through photoemission and cascade ionization. Furthermore, a large number of excited molecules generated by preionization will accelerate the breakdown process. Under the same voltage pulse, a uniform-like discharge channel with relatively low resistance (k Omega class) first forms in the cathode side sub gap. In a practical four-electrode switch whose grading resistors are in k Omega to 10 k Omega class, the anode side sub-gap will withstand overvoltage and break down first after the channel in the cathode side forms.

With the development of high voltage devices, various methods have been taken to avoid the insulation failure. Ceramic coating, as one of the methods, can be used as electrical insulation in demanding conditions, like high-temperature ambience or other harsh environment. However, the breakdown characteristics of ceramic coating switch have been little discussed, especially under pulse voltages. In this paper, to study the effect of alumina coating on breakdown characteristics, the high voltage experiments have been carried out in 0.3 MPa SF6 under nanosecond pulses. The electrode substrates were treated by a method of plasma electrolytic oxidation (PEO) to produce a layer of alumina coating. Four pairs of electrodes with varied thickness of alumina coating (20$\mu$m, 60$\mu$m, 100$\mu$m and 120$\mu$m), as well as one pair of electrodes without coating, were employed in this experiment. The applied voltage was produced by a compact Marx generator, and the rise time of it was about 40 ns. The breakdown voltage and breakdown time delay have been measured. The results showed that the thickness of coating layer have little effect on breakdown strength. And as the pulse steepness increased, the effect of coating will get strong at first and then go into a gently-changed period. Comparing the secondary electron emission coefficients of steel and alumina and the U-t curves of the four pairs of electrodes, the different process of discharge initiation and time delay might be the main causes for this phenomenon. Since the discharge initiates differently from cathode and anode, the coating electrode was utilized respectively as cathode and anode to investigate the effect of coating layer on initial process.

The flashover failure and damage of polypropylene (PP) and polyester (PET) films under nanosecond current pulses were experimentally investigated. A nanosecond pulse current test platform was established to study the flashover characteristics, performance degradation, and structural damage of dielectric films under repeated nanosecond current pulses. The accumulation and emission of surface charge influenced by field distortion and trap distribution were analyzed to clarify the mechanism by which the flashover voltage increases linearly with gas pressure at low pressure and saturates at high pressure. The significant effects of pulse energy, gas pressure, and discharge gap on film damage behavior were investigated. The damage pattern of PP and PET films includes surface deformation, reduced flash-voltage tolerance, molecular chain breakage, and group shedding. The damage of PP films is manifested as fine furrows, while PET corresponds to mountainous bumps. With the increasing number of discharges, the breakdown voltage of PP films decreased significantly, while the flash tolerance performance of PET films was more stable. Under the conditions of high pulse current amplitude, small discharge gap, and low air pressure, the damage degree of the film intensifies, which is characterized by severe destruction to the molecular structure and a significant decrease in flashover withstand voltage. Polymer films, flashover, nanosecond current pulses; performance degradation; microstructural damage.

Megavolt switches in large pulses typically use a cascade configuration to increase operating voltage and ensure electric field uniformity. We must reduce their jitter when we need multipulse superposition to obtain a higher voltage pulse. Moreover, in the high-altitude electromagnetic pulse simulator, the megavolt pulsed switch needs to adopt the self-triggering technique to simplify the structure and reduce weight for good mobility and hoisting requirements. However, it is difficult to carry out a large number of experiments and completely reveal the influence mechanisms of the characteristics of megavolt switches via experiments. Therefore, this paper derives a breakdown probability distribution model of a cascade switch based on the characteristic parameters of a single-stage switch, which can be obtained easier from experiments. The experiment of the cascade switch is then carried out to validate the model analysis. In the model, we assume that the rising rate of the pulse voltage is a constant and that the breakdown probability of a single-stage switch follows the Weibull distribution. It is common in cascade switches to allow one stage to break down in triggered mode while others are in self-breakdown mode due to overvoltage. When the cascade switch has n stages, and one stage is triggered, it is seen as three equivalent switches based on their breakdown sequence. This paper assumes that x self-breakdown stages first close, the triggered stage closes next, and finally, (n-x-1) self-breakdown stages close. The range of x is [0, n-1]. Then the breakdown probability distribution model of each equivalent switch can be derived using the existing calculation method. The influence mechanisms of the breakdown characteristics of a three-stage cascade switch are semiquantitatively analyzed based on the model. If the uniformity of breakdown characteristics of each triggered switch can be guaranteed and three stages are all triggered, the switch jitter is approximately equal to the jitter of a single-stage triggered switch because the latter two stages break down under high overvoltage. The mean breakdown voltage of each stage is lower than a single-stage triggered switch. When only one stage is triggered, most probably, the triggered stage will close first, followed by two self-breakdown stages under high overvoltage. The switch jitter is significantly lower than that under the self-breakdown mode, and the mean breakdown voltage of each stage is higher than that when all three stages are triggered. However, the possibility that one self-breakdown stage closes first results in the switch jitter being approximately twice when all three stages are triggered. The improvement path of the switch jitter characteristics can be inferred using the model analysis. The jitter of the triggered single-stage switch should be reduced, and it is preferable to trigger all stages if the uniformity of the breakdown characteristic of each single-stage switch can be guaranteed. In the experiment, the three-stage cascade switch operates on a pulsed voltage with a rise time of 300 ns and an operating voltage range of 0.8—2.0 MV. Self-triggered continuous preionization is adapted to eliminate the influence of trigger gap jitter on the switch jitter, ensuring the uniformity of breakdown characteristics of each single-stage switch. When one stage is triggered, the time delay jitter is 2.9—7.7 ns, and the breakdown voltage jitter is 0.51%—2.21%. When three stages are triggered, the time delay jitter is 2.3—3.6 ns, and the breakdown voltage jitter is 0.59%—0.91%. The preceding analysis is verified, and the switch characteristics are improved compared to the original one. Moreover, the self-breakdown characteristic is more stable when the gas pressure exceeds 0.3 MPa; thus, the switch jitter is approximately the same when we trigger one or three stages.

Surface flashover of dielectric materials often causes extensive damage to pulse power devices. Nowadays, this insulation issue is exacerbated by the demands for ever-higher operating voltages. To solve this issue, the flashover voltage and the streamer development characteristics were investigated under positive and negative nanosecond impulse, respectively, in pressured SF6 gas. First, the surface flashover strength was obtained and the Weibull distribution is introduced to describe the statistical characteristics. The results showed that the negative flashover voltage was much higher than the positive one for Al2O3-epoxy. Within the range of 0.1–0.3 MPa, the negative voltage increased with the gas pressure, while the positive one performed independent with pressure. Then, to investigate the streamer evolving process, a high-speed camera was used to get the discharge picture and the instant velocity of streamers at different conditions. Surface morphology change after the flashover was also analyzed. At last, the possible mechanisms for polarity and pressure effects were given.

In order to further improve the output voltage amplitude and repetitive operation reliability of nanosecond pulse generator (PG) based on avalanche transistor (AT), the method of power synthesis of modularized Marx circuits with transmission line transformer (TLT) is employed in this article. The modified positive $4\times11$ -stage Marx circuit introducing base-triggering method is designed and implemented to improve the operation reliability. The triggering synchronization and the consistence of output characteristics of multiple Marx modules with an identical trigger pulse are validated. The feasibility of combining pulses generated from four Marx modules with an optimized four-stage TLT is demonstrated. The influences of the type and quantity of magnetic core and the superimposed stage number of TLT on the output performance of the generator are investigated. The operation characteristics of the generator including output parameters, trigger time-delay, and loop efficiency in the whole working range are evaluated. The experimental results show that the working range of the overall device has been widened significantly with the injection of additional initial carrier. Accordingly, the generator prototype is capable of achieving the output performance with an adjustable voltage amplitude in the range of 7.8–26.7 kV on a matched 300- $\Omega $ resistive load and 12.2–38.9 kV on the high-resistance load, respectively, a 10%–90% rise time of 3.6 ns, and a full-width at half-maximum (FWHM) of 12.2 ns. A long-term test with the maximum repetition rate of 3 kHz is conducted to realize high-voltage output and high-frequency operation simultaneously.

A 4-MV self-triggered switch based on ultraviolet (UV) illumination technology is designed and tested under approximate 800-ns rise time pulse voltage. The switch is composed of five sections in series, and each section comprises an UV-illuminating gap (UV gap) and a main gap. The voltage charged on the UV gap is obtained by a resistance–capacitance coupling circuit, which enhances synchronization with the main gap voltage. The gap distance of the 4-MV switch is determined according to the performance of the same type of switch under a shorter pulse (about 300 ns), while the characteristics of the gap under 800-ns pulse voltage is still unclear. The experimental results show that low jitter breakdown of the switch can be realized effectively when two UV gaps are functioning. The breakdown jitter percentage achieves less than 1.55% within the range of 1.0–3.2 MV. When the working scope is narrowed down to 1.4–2.1 MV, the breakdown jitter is less than 1%.

The flashover process of the peaking capacitor in the electromagnetic pulse (EMP) simulator is studied based on theoretical analyses and experimental verification in this paper. There are deeper and denser ablation spots on the film surface near the inner core of the destroyed peaking capacitor, while the damage of the outer film is relatively slight, which indicates that the flashover current along the inner film is larger. Besides, the circuit simulation analyses show that the earlier the flashover occurs on the film of the peaking capacitor, the smaller the flashover current. The electric field on the non-conical surface of the capacitor is mainly occupied with normal component. Differently, for the conical surface, the electric field on the outer layer is dominated by the normal component and the parallel component is the main part on the inner layer. It is considered that the flashover on the conical surface originates from outer layers and develops gradually to the inner until the flashover penetrates through all layers. Furthermore, the images of flashover show that the flashover firstly occurs on the outermost layers and develops to the inner layers with the increase of the voltage. For such special structure of the peaking capacitor, the parallel component of electric field is more likely to facilitate the flashover under nanosecond pulse. These results may exhibit specific reference implication in the design of insulation for the peaking capacitor used in EMP simulator.

介绍了一种50?kV紧凑型自动化纳秒脉冲源,输出脉冲幅值1～50?kV连续可调,输出前沿约2?ns,脉宽约21?ns,搭配有界波导波天线,可建立满足IEC61000-2-9标准要求的电磁环境.该脉冲源采用电容直流充放电的方法实现输出电压连续可调,通过改变储能电容大小的方法实现输出脉冲脉宽可调.研制了一套远程光控的控制系统实现脉冲源的全自动化运行.该脉冲源可用于绝缘材料击穿特性试验,以及外接导波天线可产生特定的电磁环境等.通过设计和选用更高耐压的储能电容、充电绝缘子等部件,脉冲源可输出更高的电压.

In system-level high-altitude electromagnetic pulse (HEMP) illumination tests, it is common to perform the test in two orthogonal polarizations of the incident electric field. With the judgment standard of the electromagnetic norm, this paper evaluated, improved, generalized, and verified this method. The evaluated result shows that the maximum error of the maximum coupling in this method is less than 3 dB. Meanwhile, this method is improved by serving the 1.2 period of the larger coupling of the 2 illuminations as the maximum coupling. The maximum error can be controlled within 1.5 dB. Moreover, this method is generalized to non-orthogonal conditions. Expressions of the deviations of this method are strictly derived. Based on this method, the application and some extended thinking are discussed. At last, a current coupling test is designed and carried out to verify the methods and conclusions. The methods introduced in this paper can be applied to any linear system in the illumination test under approximate transverse electromagnetic (TEM) waves.

研究了一种自触发紫外预电离开关击穿时延抖动特性的影响因素,结果表明:触发间隙电容放电阶段起预电离作用时,预电离注入时刻开关电场是开关时延抖动的决定性因素,提高工作系数和采用逸出功更低的电极材料对降低开关在脉冲峰值附近击穿时的时延抖动效果有限.提出的改进方法为:减小开关均压电阻阻值,显著延长触发间隙的有效燃弧时间,消除预电离注入时间及抖动的影响.采用改进方法时可以使开关在工作电压300～800?kV、前沿100?ns、180?ns的脉冲峰值附近击穿时的时延抖动分别小于1.3?ns、2.8?ns.

The polymer film dielectric has been widely used in high voltage equipment for bulk insulation owing to the excellent insulation performance. Martin’s empirical formula for bulk breakdown is frequently used to predict the breakdown field strength of multilayer thin film with different electrode areas. In this article, based on the academic consensus that the cumulative breakdown probability of single-layer film obeys the Weibull distribution, the relationship between the cumulative breakdown probability of film dielectric under different electrode areas and the applied electrical field is deduced. Moreover, the relationship between the cumulative breakdown probability of film dielectric with various numbers of film layers and the applied electrical field is obtained and simplified. On these bases, combined with the experimental results of the average breakdown field of film dielectric with smaller electrode areas and fewer layers, a prediction model of the average breakdown field strength of film dielectric with different electrode areas and film layers is established. To obtain the solution of the prediction model concerning the number of film layer, an effective approximate solution method is proposed to solve the calculation problem with ultralarge and ultrasmall data, which significantly reduces the complexity of the solution process. A coaxial film capacitor is developed and tested to verify the prediction model with the “Chenguang” accelerator. The experimental breakdown field strength of 54 layers of 15 $\mu \text{m}$ biaxially oriented polypropylene (BOPP) film with an electrode area of 218.1 cm2 is 355.6 V/ $\mu \text{m}$ , which is comparable to the predicted result of 374 V/ $\mu \text{m}$ . The prediction model presented in this article is of certain reference significance for the insulation design of film dielectrics.