Email alert
RSS2012 Vol. 24, No. 03
Recommend Articles
Display Method:
2012,
24.
2012,
24: 505-510.
Abstract:
The energy deposition of underwater electrical explosion of a copper wire was investigated with pulsed voltage in the time scale of a few microseconds. A self-integrating Rogowski coil and a voltage divider were used for the measurements of the current and voltage on the wire load, respectively. Three stages of melting, liquid state and vaporization were defined by the definition of four critical time points, i.e. the beginning of melting, the end of melting, the beginning of vaporization and the beginning of breakdown. The deposited energy in the three stages and before breakdown was calculated by mathematical method. The effects of circuit parameters, including applied voltage and circuit inductance, and the properties of copper wire consisting of the length and diameter on deposited energy in the three stages and before breakdown were analyzed by experiment and calculation. The influence of circuit parameters and properties of copper wire is in significant in the melting stage, and obvious in the stages of liquid state and vaporization. By adjusting circuit parameters, the deposited energy in the stage of liquid state and before breakdown obviously increases with the increase of the current rise rate.
The energy deposition of underwater electrical explosion of a copper wire was investigated with pulsed voltage in the time scale of a few microseconds. A self-integrating Rogowski coil and a voltage divider were used for the measurements of the current and voltage on the wire load, respectively. Three stages of melting, liquid state and vaporization were defined by the definition of four critical time points, i.e. the beginning of melting, the end of melting, the beginning of vaporization and the beginning of breakdown. The deposited energy in the three stages and before breakdown was calculated by mathematical method. The effects of circuit parameters, including applied voltage and circuit inductance, and the properties of copper wire consisting of the length and diameter on deposited energy in the three stages and before breakdown were analyzed by experiment and calculation. The influence of circuit parameters and properties of copper wire is in significant in the melting stage, and obvious in the stages of liquid state and vaporization. By adjusting circuit parameters, the deposited energy in the stage of liquid state and before breakdown obviously increases with the increase of the current rise rate.
2012,
24: 511-514.
Abstract:
Effects of magnetic reconnection on coronal plasma acceleration and energy balance have been discussed. Acceleration of coronal plasma to the array axis can be divided into two stages. Firstly, coronal plasma is pushed radially inward the array axis mainly by global magnetic force or thermal force, depending on the wire number. Secondly, plasma jets are accelerated to Alfven speed by magnetic reconnection, eventually reaching the array axis as precursor pinch. The thickness of the reconnection layer that is comparable to the ion inertial length indicates the motions of electrons and ions are decoupled in the current sheet. Strong radial electric field produced by charge separation converts magnetic energy to axial kinetic energy of plasmas, and thermalization of radial and axial kinetic energy accounts for radiation yield. For facilities with 1 MA drive level, the energy of electromagnetic pulse produced by magnetic reconnection can reach 1 kJ.
Effects of magnetic reconnection on coronal plasma acceleration and energy balance have been discussed. Acceleration of coronal plasma to the array axis can be divided into two stages. Firstly, coronal plasma is pushed radially inward the array axis mainly by global magnetic force or thermal force, depending on the wire number. Secondly, plasma jets are accelerated to Alfven speed by magnetic reconnection, eventually reaching the array axis as precursor pinch. The thickness of the reconnection layer that is comparable to the ion inertial length indicates the motions of electrons and ions are decoupled in the current sheet. Strong radial electric field produced by charge separation converts magnetic energy to axial kinetic energy of plasmas, and thermalization of radial and axial kinetic energy accounts for radiation yield. For facilities with 1 MA drive level, the energy of electromagnetic pulse produced by magnetic reconnection can reach 1 kJ.
2012,
24: 515-518.
Abstract:
Based on the ZORK model describing the Saturn facility, a zero dimensional load model of the wire array Z-pinch on Yang accelerator is designed using Pspice to simulate the implosion process. Comparisons between the calculated results and experimental data prove the load model to be correct. The applicability and shortcomings of the load model are presented. One-dimensional magnetohydrodynamic calculations are performed by using the current curve obtained from calculated results of experiment Yang 1050#, and the parameters such as implosion time and radiation X-ray power are obtained.
Based on the ZORK model describing the Saturn facility, a zero dimensional load model of the wire array Z-pinch on Yang accelerator is designed using Pspice to simulate the implosion process. Comparisons between the calculated results and experimental data prove the load model to be correct. The applicability and shortcomings of the load model are presented. One-dimensional magnetohydrodynamic calculations are performed by using the current curve obtained from calculated results of experiment Yang 1050#, and the parameters such as implosion time and radiation X-ray power are obtained.
2012,
24: 519-523.
doi: 10.3788/HPLPB20122403.0519
Abstract:
The Qiangguang-Ⅰ (QG-Ⅰ) accelerator can produce high current pulses with rise time of about 100 ns and amplitude of about 2 MA. The load current is usually monitored by self-integrating Rogowski coils. In order to verify the obtained results of the coils, a differential loop with fast response time, relatively simple design, and appropriate performance against electromagnetic noise, has been developed. Calibration experiments show that the differential loop has a response time of about 1.2 ns, a response frequency range from 10 kHz to 100 MHz, and a sensitivity of 6.1310-11(Vs)/A. The fast response may help to monitor transient variation of the load current related to some rapid physical processes. The differential loop has been tested within short-circuit diode of QG-I accelerator and compared with self-integrating Rogowski coils. The current waveforms obtained through numerical integration of the loop signal exhibit good agreements with those from the coils, with an amplitude deviation less than 10%. Comparison of the results confirms the availability of the loop, and also verifies the reliability of the load current measurements on the accelerator.
The Qiangguang-Ⅰ (QG-Ⅰ) accelerator can produce high current pulses with rise time of about 100 ns and amplitude of about 2 MA. The load current is usually monitored by self-integrating Rogowski coils. In order to verify the obtained results of the coils, a differential loop with fast response time, relatively simple design, and appropriate performance against electromagnetic noise, has been developed. Calibration experiments show that the differential loop has a response time of about 1.2 ns, a response frequency range from 10 kHz to 100 MHz, and a sensitivity of 6.1310-11(Vs)/A. The fast response may help to monitor transient variation of the load current related to some rapid physical processes. The differential loop has been tested within short-circuit diode of QG-I accelerator and compared with self-integrating Rogowski coils. The current waveforms obtained through numerical integration of the loop signal exhibit good agreements with those from the coils, with an amplitude deviation less than 10%. Comparison of the results confirms the availability of the loop, and also verifies the reliability of the load current measurements on the accelerator.
2012,
24: 524-527.
Abstract:
The test platform has been developed to generate exciting pulsed voltages with the rise time less than 30 ns. The loss characteristics of cores of 25 m 2605TCA Metglas and 50 m DG6 electrical steel were then studied. A characteristic parameter, the gradient of the voltage pulse applied per unit core area, is proposed to describe the exciting condition applied on magnetic cores. The loss of the DG6 core is about 4 times that of the 2605TCA core. Most loss of the DG6 core, about 75%, is due to eddy current. For the 2605TCA core, the percentage is about 28%.
The test platform has been developed to generate exciting pulsed voltages with the rise time less than 30 ns. The loss characteristics of cores of 25 m 2605TCA Metglas and 50 m DG6 electrical steel were then studied. A characteristic parameter, the gradient of the voltage pulse applied per unit core area, is proposed to describe the exciting condition applied on magnetic cores. The loss of the DG6 core is about 4 times that of the 2605TCA core. Most loss of the DG6 core, about 75%, is due to eddy current. For the 2605TCA core, the percentage is about 28%.
2012,
24: 529-534.
Abstract:
The principle and structure of a hybrid voltage divider are proposed and the calibration and experimental testing are carried out. The ideal response conditions of the divider are analyzed by using the simplified divider model, the attenuation characteristic of two stage voltage dividing is discussed, and then an error control method is presented. Through load calibration, the pulsed voltage frequency response of the divider can be greater than 2.9 MHz with the largest pulse width of 40 s, the attenuation ratio (or the attenuation coefficient) of 2.60 kV/V and the measurement error of less than 5%. The divider has the merits of low cost, easy fabrication, and being capable of ns to s level pulsed voltage measurement, and it can be applied to laboratory scale pulsed power measurement.
The principle and structure of a hybrid voltage divider are proposed and the calibration and experimental testing are carried out. The ideal response conditions of the divider are analyzed by using the simplified divider model, the attenuation characteristic of two stage voltage dividing is discussed, and then an error control method is presented. Through load calibration, the pulsed voltage frequency response of the divider can be greater than 2.9 MHz with the largest pulse width of 40 s, the attenuation ratio (or the attenuation coefficient) of 2.60 kV/V and the measurement error of less than 5%. The divider has the merits of low cost, easy fabrication, and being capable of ns to s level pulsed voltage measurement, and it can be applied to laboratory scale pulsed power measurement.
2012,
24: 535-538.
doi: 10.3788/HPLPB20122403.0535
Abstract:
A monitorng system for 1 200 kV high-power microwave source driver is designed for automatic control requirements. In the system， a high-power voltage regulator produces continuous adjustable voltage with a multifunction DAQ card， which is transformed into direct current high voltage by high voltage epoxy transformer and silicon diodes and so on. A high-speed I/O card makes the trigger system work according to timing. The master computer can accomplish real-time monitoring of DC high voltage，thyratron anode voltage， ground current and other devices in the driver via serial interface of RS-485， and the central host computer communicates with the control computer via Ethernet. The driver can work individually or harmonically with the central computer. LabVIEW is adopted as the development tool of the software system. Moreover， in order to decrease the influences of high voltage and intense current produced in high pulsed power experiments on operational site， reliability design is blent in the system hardware and software. The experimental results show that the system is an easily-manipulated， real-time system with user-friendly interface， performing reliably， and stably by rule and line， and it has good expansibility and portability in intelligent control for high-power microwave source driver.
A monitorng system for 1 200 kV high-power microwave source driver is designed for automatic control requirements. In the system， a high-power voltage regulator produces continuous adjustable voltage with a multifunction DAQ card， which is transformed into direct current high voltage by high voltage epoxy transformer and silicon diodes and so on. A high-speed I/O card makes the trigger system work according to timing. The master computer can accomplish real-time monitoring of DC high voltage，thyratron anode voltage， ground current and other devices in the driver via serial interface of RS-485， and the central host computer communicates with the control computer via Ethernet. The driver can work individually or harmonically with the central computer. LabVIEW is adopted as the development tool of the software system. Moreover， in order to decrease the influences of high voltage and intense current produced in high pulsed power experiments on operational site， reliability design is blent in the system hardware and software. The experimental results show that the system is an easily-manipulated， real-time system with user-friendly interface， performing reliably， and stably by rule and line， and it has good expansibility and portability in intelligent control for high-power microwave source driver.
2012,
24: 539-543.
doi: 10.3788/HPLPB20122403.0539
Abstract:
The development of single-wire and multi-wire Z-pinches was investigated by X-ray backlighting using X-pinch as soft X-ray source. The experiments were carried out on the pulsed power generator PPG-Ⅰ（400 kA/500 kV/100 ns）developed by the Department of Electrical Engineering of Tsinghua University. The X-pinch acting as the X-ray source and the single-wire or multi-wire Z-pinch acting as the object were installed in the place of a current-return rod or the center between the anode and the cathode. The X-ray films of high resolution and high sensitivity were used to record the results. The current sensor and Rogowski coil of our own design were used to monitor the current. In order to measure the mass ablation rate of the thin wire， the step wedge of micron-level thickness was designed. Through a large number of imaging experiments， the physical images of the plasma merging, the coronal plasma formation and the instabilities development of Z-pinch and some important parameters like mass ablation rate and core expansion rate were obtained.
The development of single-wire and multi-wire Z-pinches was investigated by X-ray backlighting using X-pinch as soft X-ray source. The experiments were carried out on the pulsed power generator PPG-Ⅰ（400 kA/500 kV/100 ns）developed by the Department of Electrical Engineering of Tsinghua University. The X-pinch acting as the X-ray source and the single-wire or multi-wire Z-pinch acting as the object were installed in the place of a current-return rod or the center between the anode and the cathode. The X-ray films of high resolution and high sensitivity were used to record the results. The current sensor and Rogowski coil of our own design were used to monitor the current. In order to measure the mass ablation rate of the thin wire， the step wedge of micron-level thickness was designed. Through a large number of imaging experiments， the physical images of the plasma merging, the coronal plasma formation and the instabilities development of Z-pinch and some important parameters like mass ablation rate and core expansion rate were obtained.
2012,
24: 544-548.
doi: 10.3788/HPLPB20122403.0544
Abstract:
The discharge characteristic and plasma parameters of electrical explosion of aluminum thin wire were studied by framing camera and spectrograph. The experiments were carried out under several different discharge voltages with aluminum wires of 40 m and 100 m in diameter. The experiments show that there are two types of secondary breakdown processes in the metal vapor: inner breakdown and surface breakdown. The inner breakdown, which generates a plasma with better spatial uniformity and symmetry and consequently leads to more steady and repeatable discharge, takes place more easily in thinner wire under the same conditions. The electron temperature and density of the plasma have the order of magnitude of 104 K and 1018 cm-3, respectively.
The discharge characteristic and plasma parameters of electrical explosion of aluminum thin wire were studied by framing camera and spectrograph. The experiments were carried out under several different discharge voltages with aluminum wires of 40 m and 100 m in diameter. The experiments show that there are two types of secondary breakdown processes in the metal vapor: inner breakdown and surface breakdown. The inner breakdown, which generates a plasma with better spatial uniformity and symmetry and consequently leads to more steady and repeatable discharge, takes place more easily in thinner wire under the same conditions. The electron temperature and density of the plasma have the order of magnitude of 104 K and 1018 cm-3, respectively.
2012,
24: 549-553.
doi: 10.3788/HPLPB20122403.0549
Abstract:
The paper presents the experimental investigation of laser triggered gas switch on Qiangguang-Ⅰ accelerator. In the experiments, the laser was transmitted along the axes of the water line through the diode. The results of self-breakdown experiments and trigger experiments are presented and analyzed. At the undervoltage ratio of 88.3%, the average delay of the laser triggered gas switch is 34.2 ns, and the time jitter is less than 3.86 ns, when the maximal operating voltage is 2 MV and the maximal current is over 600 kA. The problems occurring during experiments are summarized and analyzed, and corresponding improvement measures are proposed.
The paper presents the experimental investigation of laser triggered gas switch on Qiangguang-Ⅰ accelerator. In the experiments, the laser was transmitted along the axes of the water line through the diode. The results of self-breakdown experiments and trigger experiments are presented and analyzed. At the undervoltage ratio of 88.3%, the average delay of the laser triggered gas switch is 34.2 ns, and the time jitter is less than 3.86 ns, when the maximal operating voltage is 2 MV and the maximal current is over 600 kA. The problems occurring during experiments are summarized and analyzed, and corresponding improvement measures are proposed.
2012,
24: 554-558.
doi: 10.3788/HPLPB20122403.0554
Abstract:
This paper studies the main performance parameters and measurements of metallized polypropylene film capacitors， such as energy density， voltage maintaining performance and insulation resistance. Through current and insulation resistance tests are carried out for metallized film capacitor windings， and parameter optimization of the windings is discussed. According to the requirement of high current， long lifetime and vacuum applications， a 50 kV/20 F metallized film capacitor with 120 kA/80s discharge current is developed， which can work at -50 to 60 ℃. After impregnation techniques optimization， the lifetime of the capacitor can be increased up to 2 to 3 times that of the dry metallized film capacitor. An impregnated metallized film capacitor with energy storage density of 2.0 kJ/L is designed and tested with a lifetime over 1 000 shots. It can achieve a lifetime of over 10 000 shots with energy storage density of 1.3 kJ/L. Another capacitor with 1.4 kJ/L energy storage density is designed and tested with discharge current storage up to 100 kA in vacuum applications (pressure lower than 10-3 Pa).
This paper studies the main performance parameters and measurements of metallized polypropylene film capacitors， such as energy density， voltage maintaining performance and insulation resistance. Through current and insulation resistance tests are carried out for metallized film capacitor windings， and parameter optimization of the windings is discussed. According to the requirement of high current， long lifetime and vacuum applications， a 50 kV/20 F metallized film capacitor with 120 kA/80s discharge current is developed， which can work at -50 to 60 ℃. After impregnation techniques optimization， the lifetime of the capacitor can be increased up to 2 to 3 times that of the dry metallized film capacitor. An impregnated metallized film capacitor with energy storage density of 2.0 kJ/L is designed and tested with a lifetime over 1 000 shots. It can achieve a lifetime of over 10 000 shots with energy storage density of 1.3 kJ/L. Another capacitor with 1.4 kJ/L energy storage density is designed and tested with discharge current storage up to 100 kA in vacuum applications (pressure lower than 10-3 Pa).
2012,
24: 559-562.
doi: 10.3788/HPLPB20122403.0559
Abstract:
Surface flashover in vacuum is a penetrability discharge phenomenon which occurs at the vacuum-insulator interface. One of the effective ways to increase the surface flashover voltage is magnetic flashover inhibition. In this work， a series of experiments were carried out to find out the relationships between applied magnetic field and surface flashover voltage. The experimental results show that the surface flashover voltage of PMMA can be increased up to 1.8 times by applying a magnetic field of 1.1 T which is perpendicular to the electric field and parallel to the insulator surface.
Surface flashover in vacuum is a penetrability discharge phenomenon which occurs at the vacuum-insulator interface. One of the effective ways to increase the surface flashover voltage is magnetic flashover inhibition. In this work， a series of experiments were carried out to find out the relationships between applied magnetic field and surface flashover voltage. The experimental results show that the surface flashover voltage of PMMA can be increased up to 1.8 times by applying a magnetic field of 1.1 T which is perpendicular to the electric field and parallel to the insulator surface.
2012,
24: 563-566.
doi: 10.3788/HPLPB20122403.0563
Abstract:
High gradient insulators (HGIs) were developed to solve the vacuum surface flashover problem in pulsed power system. The vacuum out-gassing rate and desorption gas components of HGIs were measured using an all-metal ultrahigh-vacuum analysis system. The high frequency dielectric constant was measured by impedance analyzer. Based on the nanosecond pulsed experimental platform， the vacuum surface flashover characteristics of HGIs were tested. The results indicate that HGIs have good high frequency dielectric performance with the inflection frequency up to 200 MHz and the flashover field strength is about 190 kV/cm after 30 discharges. The flashover performance of HGIs has direct relation to the insulation layer materials and the ratio of insulation layer and metal layer.
High gradient insulators (HGIs) were developed to solve the vacuum surface flashover problem in pulsed power system. The vacuum out-gassing rate and desorption gas components of HGIs were measured using an all-metal ultrahigh-vacuum analysis system. The high frequency dielectric constant was measured by impedance analyzer. Based on the nanosecond pulsed experimental platform， the vacuum surface flashover characteristics of HGIs were tested. The results indicate that HGIs have good high frequency dielectric performance with the inflection frequency up to 200 MHz and the flashover field strength is about 190 kV/cm after 30 discharges. The flashover performance of HGIs has direct relation to the insulation layer materials and the ratio of insulation layer and metal layer.
2012,
24: 567-570.
doi: 10.3788/HPLPB20122403.0567
Abstract:
The lead zirconate titanate PbZr0.95Ti0.05O3(PZT95/5) ferroelectric ceramics are generally used in the explosive-electric transducer, as multilayer PZT95/5 thin films arrangement to realize high voltage output. However, the dielectric breakdown of the ceramics needs to be taken seriously. A quasi-static electric field model has been developed for voltage distribution analysis of multilayer PZT95/5 ceramics compressed by shock wave in the normal mode. The calculation results reveal that the electric potential difference of PZT95/5 films increases nonlinearly from ground electrode to output electrode. Since the thickness of ceramic materials has an effect upon dielectric strength, it is a suitable way to avoid device dielectric breakdown by designing carefully the nonuniform layout of PZT95/5 thin films. By using inequality restriction, anonuniform layout optimization of PZT95/5 films arrangement is obtained, which reduces the layer number to 19 and makes the voltage of each layer of PZT95/5 element being lower than its breakdown voltage.
The lead zirconate titanate PbZr0.95Ti0.05O3(PZT95/5) ferroelectric ceramics are generally used in the explosive-electric transducer, as multilayer PZT95/5 thin films arrangement to realize high voltage output. However, the dielectric breakdown of the ceramics needs to be taken seriously. A quasi-static electric field model has been developed for voltage distribution analysis of multilayer PZT95/5 ceramics compressed by shock wave in the normal mode. The calculation results reveal that the electric potential difference of PZT95/5 films increases nonlinearly from ground electrode to output electrode. Since the thickness of ceramic materials has an effect upon dielectric strength, it is a suitable way to avoid device dielectric breakdown by designing carefully the nonuniform layout of PZT95/5 thin films. By using inequality restriction, anonuniform layout optimization of PZT95/5 films arrangement is obtained, which reduces the layer number to 19 and makes the voltage of each layer of PZT95/5 element being lower than its breakdown voltage.
2012,
24: 571-575.
doi: 10.3788/HPLPB20122403.0571
Abstract:
This paper summarizes an experimental study on lifetime of the insulator used for multi-gap gas switch. The mechanism of the insulator lifetime being reduced because of pollution is analyzed. The distribution of static electric field on insulator surface is simulated. The change of self-breakdown voltage and surface resistance due to pollution is experimentally obtained. After 13 000 shots under the discharge current of 32 kA， the self-breakdown voltage reduces from 171.5 to 130.8 kV and the surface resistance reduces from 200 to 22.6 G， which means the insulator can not operate normally. The measures for prolonging lifetime of the insulator are put forward.
This paper summarizes an experimental study on lifetime of the insulator used for multi-gap gas switch. The mechanism of the insulator lifetime being reduced because of pollution is analyzed. The distribution of static electric field on insulator surface is simulated. The change of self-breakdown voltage and surface resistance due to pollution is experimentally obtained. After 13 000 shots under the discharge current of 32 kA， the self-breakdown voltage reduces from 171.5 to 130.8 kV and the surface resistance reduces from 200 to 22.6 G， which means the insulator can not operate normally. The measures for prolonging lifetime of the insulator are put forward.
2012,
24: 576-580.
doi: 10.3788/HPLPB20122403.0576
Abstract:
The scheme of three spiral transmission lines is presented, according to the concept that multiple electron beams produced by single pulsed power generator drive multiple HPM sources. Since the three pulses output by the scheme are produced by single main pulse, their time synchronization accuracy is high. Based on the Tesla-type pulsed power generator TPG2000, design results of three transmission lines are presented by theoretical simulation. Through simulation of pulse transmission process and optimization of structure, three pulses of the same parameters are obtained, and two of them have a 20 ns delay compared with the other one.
The scheme of three spiral transmission lines is presented, according to the concept that multiple electron beams produced by single pulsed power generator drive multiple HPM sources. Since the three pulses output by the scheme are produced by single main pulse, their time synchronization accuracy is high. Based on the Tesla-type pulsed power generator TPG2000, design results of three transmission lines are presented by theoretical simulation. Through simulation of pulse transmission process and optimization of structure, three pulses of the same parameters are obtained, and two of them have a 20 ns delay compared with the other one.
2012,
24: 581-586.
doi: 10.3788/HPLPB20122403.0581
Abstract:
Considering operation features of magnetically insulated transmission line(MITL), the circuit simulation method of MITL is explained, and the MITL circuit simulation code based on wave process is developed. The physical descriptions of the criterion of MITL operation status, operation impedance and loss current are discussed. The effects on simulation results due to different physical models are analyzed. The analysis indicates that loss resistance due to overmatched impedance has significant influences on MITL circuit simulation. The circuit simulation based on existing physical model is posterior, and it is useful in wire arrays configuration design, load current and implosion characteristic estimation. The MITL design, especially for future higher current MITLs, needs deep insight into magnetic insulation phenomena and operation processes.
Considering operation features of magnetically insulated transmission line(MITL), the circuit simulation method of MITL is explained, and the MITL circuit simulation code based on wave process is developed. The physical descriptions of the criterion of MITL operation status, operation impedance and loss current are discussed. The effects on simulation results due to different physical models are analyzed. The analysis indicates that loss resistance due to overmatched impedance has significant influences on MITL circuit simulation. The circuit simulation based on existing physical model is posterior, and it is useful in wire arrays configuration design, load current and implosion characteristic estimation. The MITL design, especially for future higher current MITLs, needs deep insight into magnetic insulation phenomena and operation processes.
2012,
24: 587-591.
doi: 10.3788/HPLPB20122403.0587
Abstract:
The electrostatic field near the trigger pin of multi-rod triggered vacuum switch (TVS) was simulated with ANSOFT. Several experiments were carried out under different working polarities. The relationship between the working polarities and the time delay was obtained， and the influence of electrostatic field on the dissipation of initiate plasma was analyzed. The simulation results show that， when the trigger voltage is positive， the electric field strength near the cathode surface is enhanced， which facilitates the formation of cathode spots， and reduces the turn-on delay and jitter. The experimental results are in accordance with the simulation results，proving that the TVS works reliably with small delay and jitter in the positive working polarity.
The electrostatic field near the trigger pin of multi-rod triggered vacuum switch (TVS) was simulated with ANSOFT. Several experiments were carried out under different working polarities. The relationship between the working polarities and the time delay was obtained， and the influence of electrostatic field on the dissipation of initiate plasma was analyzed. The simulation results show that， when the trigger voltage is positive， the electric field strength near the cathode surface is enhanced， which facilitates the formation of cathode spots， and reduces the turn-on delay and jitter. The experimental results are in accordance with the simulation results，proving that the TVS works reliably with small delay and jitter in the positive working polarity.
2012,
24: 592-596.
doi: 10.3788/HPLPB20122403.0592
Abstract:
The characteristics of nanosecond-pulse surface dielectric barrier discharge (DBD)are investigated in air at atmospheric pressure with the domestic repetitive unipolar nanosecond-pulse generator， the characteristic of surface DBD is investigated in this paper. It is found that nanosecond-pulse surface DBD is a filamentary discharge in essence. The discharge happens during the rising time of the applied voltage pulse. Higher applied voltage and Higher repetitive frequency lead to intenser and more uniform discharge. Voltage amplitudeaffects the discharge uniformity more， while pulse repetition frequency affects the discharge intensity more. There is an optimum electrode gap width to get the best electrical characteristics of surface DBD. Moreover， surface flashover is prone to happen when glass is chosen as dielectric barrier.
The characteristics of nanosecond-pulse surface dielectric barrier discharge (DBD)are investigated in air at atmospheric pressure with the domestic repetitive unipolar nanosecond-pulse generator， the characteristic of surface DBD is investigated in this paper. It is found that nanosecond-pulse surface DBD is a filamentary discharge in essence. The discharge happens during the rising time of the applied voltage pulse. Higher applied voltage and Higher repetitive frequency lead to intenser and more uniform discharge. Voltage amplitudeaffects the discharge uniformity more， while pulse repetition frequency affects the discharge intensity more. There is an optimum electrode gap width to get the best electrical characteristics of surface DBD. Moreover， surface flashover is prone to happen when glass is chosen as dielectric barrier.
2012,
24: 597-601.
doi: 10.3788/HPLPB20122403.0597
Abstract:
With an excitation of negative repetitive pulses of 15 ns rise time and 30 to 40 ns duration, corona discharge in nanosecond-pulse regime at 120 kV in atmospheric-pressure air is experimentally investigated, and the characteristic of nanosecond-pulsed corona discharge is analyzed by the measurement of electrical discharge parameters, images and X-ray emission. The results show that X-ray emission occurs in nanosecond-pulsed corona discharge, but the intensity is weak. The counts of X rays decrease with the air gap spacing but increase with the pulse repetition frequency. In addition, owing to the fact that the residual particles in the gap enhance the local electric field when the next pulse is applied, separated corona channels are easily obtained at high pulse repetition frequency.
With an excitation of negative repetitive pulses of 15 ns rise time and 30 to 40 ns duration, corona discharge in nanosecond-pulse regime at 120 kV in atmospheric-pressure air is experimentally investigated, and the characteristic of nanosecond-pulsed corona discharge is analyzed by the measurement of electrical discharge parameters, images and X-ray emission. The results show that X-ray emission occurs in nanosecond-pulsed corona discharge, but the intensity is weak. The counts of X rays decrease with the air gap spacing but increase with the pulse repetition frequency. In addition, owing to the fact that the residual particles in the gap enhance the local electric field when the next pulse is applied, separated corona channels are easily obtained at high pulse repetition frequency.
2012,
24: 602-606.
doi: 10.3788/HPLPB20122403.0602
Abstract:
In order to optimize the structure of multi-gap gas switch used in fast linear transformer driver(FLTD) module, the electric field distribution of gas switches with different structured electrodes before and after trigger were simulated. The effect of pressure balance measures on the electric field distribution of the gaps in the gas switch before trigger was studied by applying the new circuit model of the gas switch. Moreover, the relationship between the stray capacitances of the middle electrodes and the electric field distribution in the gas switch after trigger was studied. The results show that the ring electrode of circular section is conducive to the formation of a stable multi-channel discharge, and it can reduce the switch inductance and jitter; the DC voltage characteristics of the switch can be effectively improved by using the resistor of the same resistance in parallel between the gaps. At the trigger moment, the voltage distribution of each gap is mainly related to the trigger voltage rise time, the resistance and stray capacitance, and the shorter the trigger voltage rise time is, the more obvious the effect of the stray capacitance on the gap voltage distribution is.
In order to optimize the structure of multi-gap gas switch used in fast linear transformer driver(FLTD) module, the electric field distribution of gas switches with different structured electrodes before and after trigger were simulated. The effect of pressure balance measures on the electric field distribution of the gaps in the gas switch before trigger was studied by applying the new circuit model of the gas switch. Moreover, the relationship between the stray capacitances of the middle electrodes and the electric field distribution in the gas switch after trigger was studied. The results show that the ring electrode of circular section is conducive to the formation of a stable multi-channel discharge, and it can reduce the switch inductance and jitter; the DC voltage characteristics of the switch can be effectively improved by using the resistor of the same resistance in parallel between the gaps. At the trigger moment, the voltage distribution of each gap is mainly related to the trigger voltage rise time, the resistance and stray capacitance, and the shorter the trigger voltage rise time is, the more obvious the effect of the stray capacitance on the gap voltage distribution is.
2012,
24: 607-611.
doi: 10.3788/HPLPB20122403.0607
Abstract:
Two photoconductive semiconductor switches(PCSSs) were fabricated using V-doped semi-insulating 6H-SiC transverse-cut wafer with different dark resistivities. A series of photoconductive experiments with different bias voltages and different triggering laser energies were done on the two samples. The experimental results indicate that, the sample with higher dark resistivity can endure much higher bias voltage, and the maximum bias voltage are elevated from 4.2 kV for the sample with lower dark resistivity to 32 kV for the sample with higher dark resistivity. However, the sample with higher dark resistivity has higher on-state resistance of k level, at least one order-of-magnitude higher than that of the sample with lower dark resistivity, less than 100 . Carrier mobilities and lifetimes were estimated by comparing laser waveforms and photo-current waveforms. In comparison with GaAs-PCSSs, the extremely low carriers lifetime and mobility are found to be the reasons for the high on-state resistance of SiC-PCSSs. Then an improved V:6H-SiC PCSS was fabricated and measured. The maximum working voltage is 10 kV, the photocurrent is 90 A, and the on-state performanceis elevated greatly.
Two photoconductive semiconductor switches(PCSSs) were fabricated using V-doped semi-insulating 6H-SiC transverse-cut wafer with different dark resistivities. A series of photoconductive experiments with different bias voltages and different triggering laser energies were done on the two samples. The experimental results indicate that, the sample with higher dark resistivity can endure much higher bias voltage, and the maximum bias voltage are elevated from 4.2 kV for the sample with lower dark resistivity to 32 kV for the sample with higher dark resistivity. However, the sample with higher dark resistivity has higher on-state resistance of k level, at least one order-of-magnitude higher than that of the sample with lower dark resistivity, less than 100 . Carrier mobilities and lifetimes were estimated by comparing laser waveforms and photo-current waveforms. In comparison with GaAs-PCSSs, the extremely low carriers lifetime and mobility are found to be the reasons for the high on-state resistance of SiC-PCSSs. Then an improved V:6H-SiC PCSS was fabricated and measured. The maximum working voltage is 10 kV, the photocurrent is 90 A, and the on-state performanceis elevated greatly.
2012,
24: 612-616.
doi: 10.3788/HPLPB20122403.0612
Abstract:
The dielectric barrier discharge (DBD) at one atmospheric pressure is excitated in a configuration of two water electrodes by a repetitive nanosecond-pulse generator with 40 ns rise time and 70 ns duration. The DBD characteristic is studied by the measurement of DBD voltage and current, DBD images of long and nanosecond exposures. Especially, the discharge images taken by ICCD with 2 ns exposure time are used to study the temporal behaviour of the discharge development. The results indicate that the nanosecond-pulse discharge is uniform and stable with the water electrodes in atmospheric air and there is a secondary discharge. The homogeneous discharge starts from the central part of the discharge area and spreads radially outward.
The dielectric barrier discharge (DBD) at one atmospheric pressure is excitated in a configuration of two water electrodes by a repetitive nanosecond-pulse generator with 40 ns rise time and 70 ns duration. The DBD characteristic is studied by the measurement of DBD voltage and current, DBD images of long and nanosecond exposures. Especially, the discharge images taken by ICCD with 2 ns exposure time are used to study the temporal behaviour of the discharge development. The results indicate that the nanosecond-pulse discharge is uniform and stable with the water electrodes in atmospheric air and there is a secondary discharge. The homogeneous discharge starts from the central part of the discharge area and spreads radially outward.
2012,
24: 617-620.
Abstract:
In the experiments， with the use of a single needle electrode， atmospheric pressure plasma jets are excited by a repetitive unipolar nanosecond-pulse generator， with working gases such as helium， argon， nitrogen and air. The results show that the jet length increases with the rise of applied voltage； as the flow rate of working gas increases， the length of jet becomes gradually longer and then reduces gradually to saturation after a certain flow rate due to turbulence. In addition， the plasma jets of different working gases have very different appearances. Helium and argon gas jets are needle-like， and the longest jet length is over 7 cm； whereas in nitrogen and in air， the jet is no more than 2 cm long， of brush-like mode.
In the experiments， with the use of a single needle electrode， atmospheric pressure plasma jets are excited by a repetitive unipolar nanosecond-pulse generator， with working gases such as helium， argon， nitrogen and air. The results show that the jet length increases with the rise of applied voltage； as the flow rate of working gas increases， the length of jet becomes gradually longer and then reduces gradually to saturation after a certain flow rate due to turbulence. In addition， the plasma jets of different working gases have very different appearances. Helium and argon gas jets are needle-like， and the longest jet length is over 7 cm； whereas in nitrogen and in air， the jet is no more than 2 cm long， of brush-like mode.
2012,
24: 621-624.
Abstract:
Triggered by array microhollow cathode structure， nanosecond pulsed switches can realize multichannel discharge at atmosphere with low trigger voltage. Double sided printed circuit board drilled by means of laser was used as array microhollow cathode， and the influence of switch duty factor， trigger current， the number of microhollow cathode holes and the diameter of the microhollow cathode holes on the trigger voltage and the delay and jitter of the switch is discussed. The results show that more microhollow cathode holes and 100 m diameter of the microhollow cathode hole result in lower trigger voltage， and higher trigger current， larger duty factor of the switch and more microhollow cathode holes result in shorter switch delay time with decreased jitter. In designing nanosecond pulsed switches， switch performance will be improved with more consideration of the factors above.
Triggered by array microhollow cathode structure， nanosecond pulsed switches can realize multichannel discharge at atmosphere with low trigger voltage. Double sided printed circuit board drilled by means of laser was used as array microhollow cathode， and the influence of switch duty factor， trigger current， the number of microhollow cathode holes and the diameter of the microhollow cathode holes on the trigger voltage and the delay and jitter of the switch is discussed. The results show that more microhollow cathode holes and 100 m diameter of the microhollow cathode hole result in lower trigger voltage， and higher trigger current， larger duty factor of the switch and more microhollow cathode holes result in shorter switch delay time with decreased jitter. In designing nanosecond pulsed switches， switch performance will be improved with more consideration of the factors above.
2012,
24: 625-629.
doi: 10.3788/HPLPB20122403.0625
Abstract:
Numerical representations of switch delay and jitter are given according to previous experimental data and theories. Random arrays are created by Matlab to simulate switch delay and jitter. A simplified model is established to calculate the working characteristics of linear transformer driver (LTD) module with multiple bricks in parallel. M-file is used to control the time sequence of the multiple switches in module to simulate switch jitters. Then effects of switch jitters on peak power and pulse front of 40-brick LTD module with 1 MA peak current and 100 ns rise time are studied. The simulation result shows that， in a certain range， with the increase of jitter， the peak power and system stability decrease， and pulse front increases remarkably. The decrease rate of peak power increases with the increase of jitter，and the pulse front increases almost linearly. Since the switch jitter has important influences on system output， there is a jitter threshold to satisfy certain system stability requirement. As for 1 MA LTD module with 40 bricks in parallel， the jitter threshold is about 17 ns when the system stability requirement is 5%.
Numerical representations of switch delay and jitter are given according to previous experimental data and theories. Random arrays are created by Matlab to simulate switch delay and jitter. A simplified model is established to calculate the working characteristics of linear transformer driver (LTD) module with multiple bricks in parallel. M-file is used to control the time sequence of the multiple switches in module to simulate switch jitters. Then effects of switch jitters on peak power and pulse front of 40-brick LTD module with 1 MA peak current and 100 ns rise time are studied. The simulation result shows that， in a certain range， with the increase of jitter， the peak power and system stability decrease， and pulse front increases remarkably. The decrease rate of peak power increases with the increase of jitter，and the pulse front increases almost linearly. Since the switch jitter has important influences on system output， there is a jitter threshold to satisfy certain system stability requirement. As for 1 MA LTD module with 40 bricks in parallel， the jitter threshold is about 17 ns when the system stability requirement is 5%.
2012,
24: 630-634.
doi: 10.3788/HPLPB20122403.0630
Abstract:
The high power switch is a key assembly in pulse power generators. The paper introduces the designof a 2 MV laser triggered multi-stage gas switch， and presents the results of its static and triggered experiments on Qiangguang-Ⅰ accelerator. The results show that the design goals are achieved. However， the switch has to face the problem of large transmission electric charge when it works in the transmission line of low impedance. Larger transmission electric charge will cause more discharge products， reducing the insulation capability of the switch and thus directly influencing its lifetime.
The high power switch is a key assembly in pulse power generators. The paper introduces the designof a 2 MV laser triggered multi-stage gas switch， and presents the results of its static and triggered experiments on Qiangguang-Ⅰ accelerator. The results show that the design goals are achieved. However， the switch has to face the problem of large transmission electric charge when it works in the transmission line of low impedance. Larger transmission electric charge will cause more discharge products， reducing the insulation capability of the switch and thus directly influencing its lifetime.
2012,
24: 635-638.
doi: 10.3788/HPLPB20122403.0635
Abstract:
This paper introduces an experiment of 3 mm-gap GaAs photoconductive semiconductor switch (PCSS) triggered by laser diode，which produces nonlinear output on 600 load. The peak power of laser diode is 70 W，and its output rise-time and duration (FWHM) are 20 and 40 ns， respectively. Along with the bias electric field enhancement， the output voltage increases linearly； when the bias electric field exceeds the threshold， about 2.53 kV/mm， the output voltage increases rapidly after a small peak and delay， and the avalanche occurs. Experiment results indicate that the production of nonlinear output has relation to carrier accumulation and impact ionization in GaAs chip； the enhancement of bias electric field increases the number of collective carriers， the degree of impact ionization， and thus accelerate the switchs turn into avalanche mode.
This paper introduces an experiment of 3 mm-gap GaAs photoconductive semiconductor switch (PCSS) triggered by laser diode，which produces nonlinear output on 600 load. The peak power of laser diode is 70 W，and its output rise-time and duration (FWHM) are 20 and 40 ns， respectively. Along with the bias electric field enhancement， the output voltage increases linearly； when the bias electric field exceeds the threshold， about 2.53 kV/mm， the output voltage increases rapidly after a small peak and delay， and the avalanche occurs. Experiment results indicate that the production of nonlinear output has relation to carrier accumulation and impact ionization in GaAs chip； the enhancement of bias electric field increases the number of collective carriers， the degree of impact ionization， and thus accelerate the switchs turn into avalanche mode.
2012,
24: 639-642.
doi: 10.3788/HPLPB20122403.0639
Abstract:
The static characteristics of 3 MV multi-stage multi-channel closing switch(MMCS) should be defined. The self-breakdown characteristics of the MMCS with different number of gap stages and the trigger gap were studied by reducing the number of electric gap stages. The results indicate that the self-breakdown voltage varies nonlinearly with the gap distance， which deviates from the theoretical value obtained by Bradley empirical formula. The static electric field strengths of multiple gaps in series were calculated by using Ansoft program. By analyzing the experimental data， electric field uniformity coefficient and Bradley empirical formula， a new Bradley extrapolation formula was obtained， which can better reflect the self-breakdown voltage of MMCS and reduce the discrepancy between experiment and theory. The relative error was reduced to 5% by using the Bradley extrapolation formula.
The static characteristics of 3 MV multi-stage multi-channel closing switch(MMCS) should be defined. The self-breakdown characteristics of the MMCS with different number of gap stages and the trigger gap were studied by reducing the number of electric gap stages. The results indicate that the self-breakdown voltage varies nonlinearly with the gap distance， which deviates from the theoretical value obtained by Bradley empirical formula. The static electric field strengths of multiple gaps in series were calculated by using Ansoft program. By analyzing the experimental data， electric field uniformity coefficient and Bradley empirical formula， a new Bradley extrapolation formula was obtained， which can better reflect the self-breakdown voltage of MMCS and reduce the discrepancy between experiment and theory. The relative error was reduced to 5% by using the Bradley extrapolation formula.
2012,
24: 643-646.
doi: 10.3788/HPLPB20122403.0643
Abstract:
The paper introduces the design and application of a trigger system for the 2 MV laser triggered gas switch(LTGS)， which was designed as the main switch for Qiangguang-Ⅰ accelerator. In this system， a 266 nm Nd：YAG laser was used， and a beam with a pulse energy of 80mJ， a pulse width of 7 ns and a divergence angle of 0.5 mrad was achieved when the laser was triggered. The two original trigger signals of this trigger system were generated from the trigger system of Qiangguang-Ⅰ， and they were delayed by two DG535 generators， respectively. A laser beam was focused in the LTGS through the crossover tube at 200 ns before the maximum of the self-breakdown voltage. Inexperiments， the trigger system worked well when SF6 was filled in switch and the gas pressure was 0.2 to 0.3 MPa， and the trigger jitter was 3.86 ns at the gas pressure of 0.3 MPa.
The paper introduces the design and application of a trigger system for the 2 MV laser triggered gas switch(LTGS)， which was designed as the main switch for Qiangguang-Ⅰ accelerator. In this system， a 266 nm Nd：YAG laser was used， and a beam with a pulse energy of 80mJ， a pulse width of 7 ns and a divergence angle of 0.5 mrad was achieved when the laser was triggered. The two original trigger signals of this trigger system were generated from the trigger system of Qiangguang-Ⅰ， and they were delayed by two DG535 generators， respectively. A laser beam was focused in the LTGS through the crossover tube at 200 ns before the maximum of the self-breakdown voltage. Inexperiments， the trigger system worked well when SF6 was filled in switch and the gas pressure was 0.2 to 0.3 MPa， and the trigger jitter was 3.86 ns at the gas pressure of 0.3 MPa.
2012,
24: 647-650.
doi: 10.3788/HPLPB20122403.0647
Abstract:
With a low-inductance experimental apparatus， the arc radius is obtained by photographing the discharge processes of the spark gap switch with high-speed framing camera， and a formula is proposed for calculating the arc radius. The time-dependent arc resistance and conductivity of spark switch gap in the low-inductance circuit are obtained by solving the circuit equation using measured values of current and voltage. The arc radius increases and gradually reaches a saturation value while the discharge channel develops. The radius is 0.6 mm at the peak discharge current of 2.8 kA. The saturation resistance and the time to reach saturation decrease with increasing the peak current of discharge. In addition， the saturation value of arc resistance can be minimized to 0.08 .
With a low-inductance experimental apparatus， the arc radius is obtained by photographing the discharge processes of the spark gap switch with high-speed framing camera， and a formula is proposed for calculating the arc radius. The time-dependent arc resistance and conductivity of spark switch gap in the low-inductance circuit are obtained by solving the circuit equation using measured values of current and voltage. The arc radius increases and gradually reaches a saturation value while the discharge channel develops. The radius is 0.6 mm at the peak discharge current of 2.8 kA. The saturation resistance and the time to reach saturation decrease with increasing the peak current of discharge. In addition， the saturation value of arc resistance can be minimized to 0.08 .
2012,
24: 651-654.
doi: 10.3788/HPLPB20122403.0651
Abstract:
The 100 GW fast linear transformer driver(LTD)generator has been constructed and tested successfully based on the 100 kA-LTD stage prototype constructed in 2008. The generator consists of 10 LTD stages connected in series， and the stored energy is 20 kJ. The outer diameter of the generator is about 1.5 m， and the length is about 2.2 m. At the charge voltage of 85 kV， the generator can deliver 1.1 MV fast pulses with the rise time of 53 ns and the full-width at half-maximum(FWHM) of 146 ns into a 9.4 diode load.
The 100 GW fast linear transformer driver(LTD)generator has been constructed and tested successfully based on the 100 kA-LTD stage prototype constructed in 2008. The generator consists of 10 LTD stages connected in series， and the stored energy is 20 kJ. The outer diameter of the generator is about 1.5 m， and the length is about 2.2 m. At the charge voltage of 85 kV， the generator can deliver 1.1 MV fast pulses with the rise time of 53 ns and the full-width at half-maximum(FWHM) of 146 ns into a 9.4 diode load.
2012,
24: 655-658.
doi: 10.3788/HPLPB20122403.0655
Abstract:
Modular linear transformer driver(LTD) is a novel fast discharge high pulsed power source basedon primary storage energy. It can realize the connection of multiple LTD modules in series or parallel conveniently， and can directly acquire 100 to 200 ns high power pulses. This paper introduces an LTD stage with the rise time of 100 ns and the current of 300 kA. The stage of LTD consists of 14 bricks in parallel. The 200 kV multi-gap multi-channel gas spark switch with the diameter of 98 mm and the height of 127 mm has been developed. The influence on the load output current of the switch working voltage ratio was investigated. The result indicates that the 14 bricks can discharge synchronously， when the switch working voltage ratio is 0.7.When the charging voltage of the stage is 90 kV and the switch gas pressure is 0.32 MPa， the output current is about 302 kA， the risetime is about 100 ns (10% to 90%)， and the power is about 23 GW on a matched load.
Modular linear transformer driver(LTD) is a novel fast discharge high pulsed power source basedon primary storage energy. It can realize the connection of multiple LTD modules in series or parallel conveniently， and can directly acquire 100 to 200 ns high power pulses. This paper introduces an LTD stage with the rise time of 100 ns and the current of 300 kA. The stage of LTD consists of 14 bricks in parallel. The 200 kV multi-gap multi-channel gas spark switch with the diameter of 98 mm and the height of 127 mm has been developed. The influence on the load output current of the switch working voltage ratio was investigated. The result indicates that the 14 bricks can discharge synchronously， when the switch working voltage ratio is 0.7.When the charging voltage of the stage is 90 kV and the switch gas pressure is 0.32 MPa， the output current is about 302 kA， the risetime is about 100 ns (10% to 90%)， and the power is about 23 GW on a matched load.
2012,
24: 659-662.
doi: 10.3788/HPLPB20122403.0659
Abstract:
A modularized layer-wound spiral stripline was designed using polypropylene as insulation material. The performance of the insulation film and the electrode foil was analyzed. A four-stage stacked Blumlein line based on the stripline operating with four gas switches and a resistance load delivers a quasi-rectangular output waveform at the repetitive frequency of 10 Hz. The stages are stacked in series at one end， and charged in parallel and synchronously commuted with the switches at the other end. In this way， relatively low charging voltages are multiplied to give a high discharge voltage across an arbitrary load. The Blumlein line can output 220 kV pulses with durations of 182 ns， and rising times of 50 ns at the repetitive frequency of 10 Hz.
A modularized layer-wound spiral stripline was designed using polypropylene as insulation material. The performance of the insulation film and the electrode foil was analyzed. A four-stage stacked Blumlein line based on the stripline operating with four gas switches and a resistance load delivers a quasi-rectangular output waveform at the repetitive frequency of 10 Hz. The stages are stacked in series at one end， and charged in parallel and synchronously commuted with the switches at the other end. In this way， relatively low charging voltages are multiplied to give a high discharge voltage across an arbitrary load. The Blumlein line can output 220 kV pulses with durations of 182 ns， and rising times of 50 ns at the repetitive frequency of 10 Hz.
2012,
24: 663-667.
doi: 10.3788/HPLPB20122403.0663
Abstract:
The paper introduces a compact pulsed power generator for X-pinch experiments. The design and preliminary experiments of the device are presented. The results show that when each capacitor in the Marx generator is charged with 60 kV voltage, the output current can reach 117 kA, the output voltage pulse width is 60 ns, and the output voltage rise time is no more than 30 ns. The device has already radiated X-rays successfully.
The paper introduces a compact pulsed power generator for X-pinch experiments. The design and preliminary experiments of the device are presented. The results show that when each capacitor in the Marx generator is charged with 60 kV voltage, the output current can reach 117 kA, the output voltage pulse width is 60 ns, and the output voltage rise time is no more than 30 ns. The device has already radiated X-rays successfully.
2012,
24: 668-672.
doi: 10.3788/HPLPB20122403.0668
Abstract:
The high speed sliding contact is between the armature and the rails in the electromagnetic launcher experiments. The contact performance can be influenced by the frictional heating and current joule heating, and the influence is more evident in the rapid fire mode due to the rail temperature rise. This paper discusses the contact performance with the sliding contact resistance. In the experiments, two 5-g armatures are launched with the speed of 1 000 m/s, and frequency of 166 Hz. The effect on the contact resistance of the temperature rise is calculated approximately. The results show that the contact resistance of the second armature is higher than that of the first armature with the temperature rise. In addition, thermal management should be adopted in the multi-armature launching in rapid fire mode.
The high speed sliding contact is between the armature and the rails in the electromagnetic launcher experiments. The contact performance can be influenced by the frictional heating and current joule heating, and the influence is more evident in the rapid fire mode due to the rail temperature rise. This paper discusses the contact performance with the sliding contact resistance. In the experiments, two 5-g armatures are launched with the speed of 1 000 m/s, and frequency of 166 Hz. The effect on the contact resistance of the temperature rise is calculated approximately. The results show that the contact resistance of the second armature is higher than that of the first armature with the temperature rise. In addition, thermal management should be adopted in the multi-armature launching in rapid fire mode.
2012,
24: 673-677.
doi: 10.3788/HPLPB20122403.0673
Abstract:
The paper presents a design of all-solid-state pulsed power generator based on Marx generator with MOSFET semiconductor switch devices. Charging resistance is replaced by fast recovery diode in the circuit, which reduces power loss in charging process. Drive circuit is integrated in and gets power supply from main circuit. Optical fiber is used to deliver the drive signal so as to protect it from the interference of discharging. A clockwise/counter-clockwise ring-shaped compact configuration is applied, which reduces the loop inductance, and makes the generator modularized and compact. The proposed generator is composed of 180 Marx unites in series, each of which is charged at a low voltage of 900 V, and generates a fast output pulse of voltage 150 kV, rise time less than 500 ns and duration adjustable between 1 and 5 s. Some experimental results of the pulsed power generator are listed, with a load of 50 k resistance and 5 pF capacitance in parallel. Through comparison and analysis, some factors affecting the rise time of output pulses are summarized, including loop inductance, drive voltage of MOSFET and distributed capacitance of main circuit. Approaches for pulse rise time improvement are also discussed.
The paper presents a design of all-solid-state pulsed power generator based on Marx generator with MOSFET semiconductor switch devices. Charging resistance is replaced by fast recovery diode in the circuit, which reduces power loss in charging process. Drive circuit is integrated in and gets power supply from main circuit. Optical fiber is used to deliver the drive signal so as to protect it from the interference of discharging. A clockwise/counter-clockwise ring-shaped compact configuration is applied, which reduces the loop inductance, and makes the generator modularized and compact. The proposed generator is composed of 180 Marx unites in series, each of which is charged at a low voltage of 900 V, and generates a fast output pulse of voltage 150 kV, rise time less than 500 ns and duration adjustable between 1 and 5 s. Some experimental results of the pulsed power generator are listed, with a load of 50 k resistance and 5 pF capacitance in parallel. Through comparison and analysis, some factors affecting the rise time of output pulses are summarized, including loop inductance, drive voltage of MOSFET and distributed capacitance of main circuit. Approaches for pulse rise time improvement are also discussed.
2012,
24: 678-680.
doi: 10.3788/HPLPB20122403.0678
Abstract:
The paper introduces a TTL digital circuit， which is designed as the trigger control system of high power pulse source， and presents the development of a high-voltage pulse source. In the source， the TTL circuit operated after the optical isolation circuit and power driver， and outputs pulse signals to drive MOSFET， which controls the pulse signal output from high-voltage switch. The source can outputs pulses with voltage of 5 kV， pulse-width larger than 200 s and fall-time less than 30 ns. Besides long-distance control and local control mode， it also includes inner trigger and external trigger (one electric signal and one optical signal as the trigger). A series of experiments have been conducted to validate the design， and the one-shot trigger results are presented.
The paper introduces a TTL digital circuit， which is designed as the trigger control system of high power pulse source， and presents the development of a high-voltage pulse source. In the source， the TTL circuit operated after the optical isolation circuit and power driver， and outputs pulse signals to drive MOSFET， which controls the pulse signal output from high-voltage switch. The source can outputs pulses with voltage of 5 kV， pulse-width larger than 200 s and fall-time less than 30 ns. Besides long-distance control and local control mode， it also includes inner trigger and external trigger (one electric signal and one optical signal as the trigger). A series of experiments have been conducted to validate the design， and the one-shot trigger results are presented.
2012,
24: 681-684.
doi: 10.3788/HPLPB20122403.0681
Abstract:
According to the trigger method of high-power switch device hydrogen thyratron used in the repetitive pulsed power system， a trigger device was designed for hydrogen thyratron VE4147， which can output -150 V DC bias， and pulses with voltage 2 000 V， current 10 A， width 800 ns and repetition-rate up to 10 kHz. The design takes into account interference protection， low power consumption， thermal management and so on， to make sure that such trigger device can work uninterruptedly at the repetition-rate of 10 kHz. So far the trigger device has been used in sorts of repetitive pulsed power systems， such as 100 kV/2 kHz high-voltage pulsed power supply， 70 kV/10 kHz hydrogen thyratron aging platform， and 150 kV/1 kHz corona plasma generator of adjustable pulse width.
According to the trigger method of high-power switch device hydrogen thyratron used in the repetitive pulsed power system， a trigger device was designed for hydrogen thyratron VE4147， which can output -150 V DC bias， and pulses with voltage 2 000 V， current 10 A， width 800 ns and repetition-rate up to 10 kHz. The design takes into account interference protection， low power consumption， thermal management and so on， to make sure that such trigger device can work uninterruptedly at the repetition-rate of 10 kHz. So far the trigger device has been used in sorts of repetitive pulsed power systems， such as 100 kV/2 kHz high-voltage pulsed power supply， 70 kV/10 kHz hydrogen thyratron aging platform， and 150 kV/1 kHz corona plasma generator of adjustable pulse width.
2012,
24: 685-688.
doi: 10.3788/HPLPB20122403.0685
Abstract:
The tests of 24 Marx generators on the primary test stand(PTS) have been carried out. When the charging voltage is 60 kV and the working ratio is 62%， the delay time of output voltage for 24 Marx generatorsranges from 245 to 285 ns with a total spread of 40 ns and individual jitters less than 10 ns. The timing diagram of the PTS control system and the charging waveform of the intermediate storage capacitor are analyzed. When the spread of delay time for output voltage of 24 Marx generators is 50 ns， the voltage loss of intermediate storage capacitor is less than 1% and it has no effect on the operation state of the main switch.
The tests of 24 Marx generators on the primary test stand(PTS) have been carried out. When the charging voltage is 60 kV and the working ratio is 62%， the delay time of output voltage for 24 Marx generatorsranges from 245 to 285 ns with a total spread of 40 ns and individual jitters less than 10 ns. The timing diagram of the PTS control system and the charging waveform of the intermediate storage capacitor are analyzed. When the spread of delay time for output voltage of 24 Marx generators is 50 ns， the voltage loss of intermediate storage capacitor is less than 1% and it has no effect on the operation state of the main switch.
2012,
24: 689-692.
doi: 10.3788/HPLPB20122403.0689
Abstract:
The establishing method of Pspice subcircuit was introduced， and self-breakdown switch and high current diode subcircuits were created based on their principles of work， which can describe their working processes. The circuit of Flash Ⅱ accelerator was then built using the subcircuits created as its switches and diode. The simulation results of diode voltage，current and impedance in normal operation and short-circuit state accord with the experimental results. The method of subcircuit establishment and parameters control can be used to simulate the components with nonlinear impedance and complicated working process， and it is favorable for parameters control of circuits with a large number of identical components.
The establishing method of Pspice subcircuit was introduced， and self-breakdown switch and high current diode subcircuits were created based on their principles of work， which can describe their working processes. The circuit of Flash Ⅱ accelerator was then built using the subcircuits created as its switches and diode. The simulation results of diode voltage，current and impedance in normal operation and short-circuit state accord with the experimental results. The method of subcircuit establishment and parameters control can be used to simulate the components with nonlinear impedance and complicated working process， and it is favorable for parameters control of circuits with a large number of identical components.
2012,
24: 693-696.
doi: 10.3788/HPLPB20122403.0693
Abstract:
A 10 kJ/s， 100 kV supply has been presented to meet high-quality power requirements in capacitor charging power supply(CCPS) applications. The power supply is designed based on a parallel resonant inverter. It may operate with pulse width modulation mode at a fixed switching frequency up to 40 kHz， and adopts digital signal processing controller to realize its feedback control. This paper also presents the design experience and experimental results of the CCPS and its main component high-frequency high-voltage transformer.
A 10 kJ/s， 100 kV supply has been presented to meet high-quality power requirements in capacitor charging power supply(CCPS) applications. The power supply is designed based on a parallel resonant inverter. It may operate with pulse width modulation mode at a fixed switching frequency up to 40 kHz， and adopts digital signal processing controller to realize its feedback control. This paper also presents the design experience and experimental results of the CCPS and its main component high-frequency high-voltage transformer.
2012,
24: 697-702.
doi: 10.3788/HPLPB20122403.0697
Abstract:
The triple resonance pulse transformer is a new type of pulse transformer, which is based on air-cored transformer. The voltage across the pulse transformer is significantly less than the output voltage; instead, the full output voltage appears across the tuning inductor. By analyzing the pulse transformers lossless circuit, the analytical expression for the output voltage and the relation of the triple resonance circuit parameters are presented for the eigen frequency ratio of the circuit being 1∶2∶3, and the maximal ratio of peak load voltage to peak transformer voltage is 2.77. Then a design method for the triple resonance pulse transformer, iterative simulation method, is presented. A pulse power generator based on the triple resonance pulse transformer developed from an air-cored transformer is developed accordingly. The experimental results indicate that the ratio of the peak output voltage to the peak voltage on high voltage winding is 2 and the peak output voltage of the triple resonance pulse transformer is 600 kV. The theoretical and experimental results agree well with each other. Therefore, any double resonance pulse transformer may be developed into a triple resonance pulse transformer.
The triple resonance pulse transformer is a new type of pulse transformer, which is based on air-cored transformer. The voltage across the pulse transformer is significantly less than the output voltage; instead, the full output voltage appears across the tuning inductor. By analyzing the pulse transformers lossless circuit, the analytical expression for the output voltage and the relation of the triple resonance circuit parameters are presented for the eigen frequency ratio of the circuit being 1∶2∶3, and the maximal ratio of peak load voltage to peak transformer voltage is 2.77. Then a design method for the triple resonance pulse transformer, iterative simulation method, is presented. A pulse power generator based on the triple resonance pulse transformer developed from an air-cored transformer is developed accordingly. The experimental results indicate that the ratio of the peak output voltage to the peak voltage on high voltage winding is 2 and the peak output voltage of the triple resonance pulse transformer is 600 kV. The theoretical and experimental results agree well with each other. Therefore, any double resonance pulse transformer may be developed into a triple resonance pulse transformer.
2012,
24: 703-705.
doi: 10.3788/HPLPB20122403.0703
Abstract:
Some common operation failures are introduced for Marx generator, water dielectric transmission line, diode during the running of the Flash-Ⅱ accelerator. The 200 kV gas switch in Marx generator, the 3 MV main switch and the pre-pulse switch in transmission line, the cathode surface electric field and the gap between cathode and anode are key factors for the common failures. Some maintenance methods and precautions are introduced accordingly to realize stable operation of the accelerator.
Some common operation failures are introduced for Marx generator, water dielectric transmission line, diode during the running of the Flash-Ⅱ accelerator. The 200 kV gas switch in Marx generator, the 3 MV main switch and the pre-pulse switch in transmission line, the cathode surface electric field and the gap between cathode and anode are key factors for the common failures. Some maintenance methods and precautions are introduced accordingly to realize stable operation of the accelerator.
2012,
24: 706-710.
doi: 10.3788/HPLPB20122403.0706
Abstract:
A multi-pulsed electron source is introduced, including the design and the first experimental results. The multi-pulsed electron source can generate multi-pulsed intense electron beams with energy of 2 to 3 MeV and beam intensity of 2.5 kA. The pulse interval is adjustable from 100 ns to 1 000 ns without special step. An inductive adder with multi-pulsed driving source is chosen to generate pulsed vacuum diode voltage and both velvet cathode and large-emission-area thermionic dispenser cathode are adopted to generate multi-pulsed intense electron beams. In order to adopt the two different cathodes, two different diodes with the same interface are adopted. The first experimental results indicate that the source runs stably. The multi-pulsed diode voltage is now up to 2.7 MV and the beam intensity is more than 1.6 kA both near the anode hole with Faraday cup and near the exit of the source with B-dot. The experimental results also indicate that using velvet as cathode the cathode plasma is inevitable and beam intensities become much bigger from pulse to pulse when the largr-emission-area thermionic dispenser cathode is used, there is few cathode plasma and beam intensities are almost the same form pulse to pulse.
A multi-pulsed electron source is introduced, including the design and the first experimental results. The multi-pulsed electron source can generate multi-pulsed intense electron beams with energy of 2 to 3 MeV and beam intensity of 2.5 kA. The pulse interval is adjustable from 100 ns to 1 000 ns without special step. An inductive adder with multi-pulsed driving source is chosen to generate pulsed vacuum diode voltage and both velvet cathode and large-emission-area thermionic dispenser cathode are adopted to generate multi-pulsed intense electron beams. In order to adopt the two different cathodes, two different diodes with the same interface are adopted. The first experimental results indicate that the source runs stably. The multi-pulsed diode voltage is now up to 2.7 MV and the beam intensity is more than 1.6 kA both near the anode hole with Faraday cup and near the exit of the source with B-dot. The experimental results also indicate that using velvet as cathode the cathode plasma is inevitable and beam intensities become much bigger from pulse to pulse when the largr-emission-area thermionic dispenser cathode is used, there is few cathode plasma and beam intensities are almost the same form pulse to pulse.
2012,
24: 711-714.
doi: 10.3788/HPLPB20122403.0711
Abstract:
A 100 kV nanosecond pulse generator has been designed and tested, which is based on Tesla transformer and single-cylinder pulse forming line. The design of main parameters of the pulse generator is described. A high voltage nanosecond pulse generator is achieved with the highest output voltage of 100 kV, peak power up to 250 MW, pulse duration of 4 ns, rise time of 1 ns, and repetition rates ranging from 1 to 100 Hz. The pulse generator can be used as a trigger source to reduce the jitter of larger nanosecond pulse generator. Acting as the trigger source, the pulse generator reduces the jitter of GW-level Tesla pulser from 500 to 150 ns.
A 100 kV nanosecond pulse generator has been designed and tested, which is based on Tesla transformer and single-cylinder pulse forming line. The design of main parameters of the pulse generator is described. A high voltage nanosecond pulse generator is achieved with the highest output voltage of 100 kV, peak power up to 250 MW, pulse duration of 4 ns, rise time of 1 ns, and repetition rates ranging from 1 to 100 Hz. The pulse generator can be used as a trigger source to reduce the jitter of larger nanosecond pulse generator. Acting as the trigger source, the pulse generator reduces the jitter of GW-level Tesla pulser from 500 to 150 ns.
2012,
24: 715-718.
doi: 10.3788/HPLPB20122403.0715
Abstract:
A 300 kV high voltage pulse generator designed according to the principle of bipolar charged Marx generator is introduced, which serves as the driving of X-ray system. Low jitter spark gap switches and low inductance ceramic capacitors are used in the pulse generator to shorten the risetime and pulse width of the output pulse. The generator can get a pulse with amplitude of 200 to 300 kV and pulse width less than 100 ns on the 75 resistance load. It features stable performance, compact structure, and is easy to operate.
A 300 kV high voltage pulse generator designed according to the principle of bipolar charged Marx generator is introduced, which serves as the driving of X-ray system. Low jitter spark gap switches and low inductance ceramic capacitors are used in the pulse generator to shorten the risetime and pulse width of the output pulse. The generator can get a pulse with amplitude of 200 to 300 kV and pulse width less than 100 ns on the 75 resistance load. It features stable performance, compact structure, and is easy to operate.
2012,
24: 719-722.
doi: 10.3788/HPLPB20122403.0719
Abstract:
This paper describes the design and implementation of a 60 kJ/s capacitor charging power supply(CCPS)with output voltage of 50 kV, based on the theoretical, simulative and experimental research of high frequency(HF)AC-link technology. Design details of the power supply are reported, which consists of EMI filter, matrix switch， LC resonant bank and HF transformer. A high power density of above 0.6 W/cm3 is achieved with the CCPS, by utilizing various technologies on HF AC-Link topology and proper packaging. The experimental results demonstrate that the power factor and efficiency of the CCPS are higher than 0.95 and 0.90， respectively. In a word, the CCPS features high power density, high power factor and high efficiency.
This paper describes the design and implementation of a 60 kJ/s capacitor charging power supply(CCPS)with output voltage of 50 kV, based on the theoretical, simulative and experimental research of high frequency(HF)AC-link technology. Design details of the power supply are reported, which consists of EMI filter, matrix switch， LC resonant bank and HF transformer. A high power density of above 0.6 W/cm3 is achieved with the CCPS, by utilizing various technologies on HF AC-Link topology and proper packaging. The experimental results demonstrate that the power factor and efficiency of the CCPS are higher than 0.95 and 0.90， respectively. In a word, the CCPS features high power density, high power factor and high efficiency.
2012,
24: 723-726.
doi: 10.3788/HPLPB20122403.0723
Abstract:
According to the impedance-matched pulse-forming-line technology, the theoretical model of an ultra-wideband high power microwave generator is established, and corresponding simulations are carried out. The results show that, the generator can generate a nanosecond pulse and a front picosecond pulse, with the impedance-matched technology. The switching delay of picosecond switch has an influence on the picosecond pulse voltage, and smaller capacitance of the picosecond pulse-forming-line leads to higher radiated pulse voltage and narrower picosecond pulse-width. The transfer efficiency of stored energy in pulse-forming-lines to microwave is 100% in theory.
According to the impedance-matched pulse-forming-line technology, the theoretical model of an ultra-wideband high power microwave generator is established, and corresponding simulations are carried out. The results show that, the generator can generate a nanosecond pulse and a front picosecond pulse, with the impedance-matched technology. The switching delay of picosecond switch has an influence on the picosecond pulse voltage, and smaller capacitance of the picosecond pulse-forming-line leads to higher radiated pulse voltage and narrower picosecond pulse-width. The transfer efficiency of stored energy in pulse-forming-lines to microwave is 100% in theory.
2012,
24: 727-731.
doi: 10.3788/HPLPB20122403.0727
Abstract:
An output cavity with double-rows of metal posts for a wide-gap klystron amplifier (WKA) is built and simulated using the three-dimension fully electromagnetic particle-in-cell (PIC) code and high frequency analysis software. The results show that both the first row and the second row of posts affect the high frequency characteristics of the output cavity, thus the double-rows of posts must be designed together with the output cavity, and the design purpose is to obtain proper external quality factor and gap electric field through controlling the posts scattering characteristic. Under the electric power of 2.9 GW and the electron beam modulation depth of 90%, a microwave power of 1.06 GW with an average power efficiency of about 36.5% is obtained at 3.6 GHz from an output cavity with double rows of posts designed according to the rules mentioned above.
An output cavity with double-rows of metal posts for a wide-gap klystron amplifier (WKA) is built and simulated using the three-dimension fully electromagnetic particle-in-cell (PIC) code and high frequency analysis software. The results show that both the first row and the second row of posts affect the high frequency characteristics of the output cavity, thus the double-rows of posts must be designed together with the output cavity, and the design purpose is to obtain proper external quality factor and gap electric field through controlling the posts scattering characteristic. Under the electric power of 2.9 GW and the electron beam modulation depth of 90%, a microwave power of 1.06 GW with an average power efficiency of about 36.5% is obtained at 3.6 GHz from an output cavity with double rows of posts designed according to the rules mentioned above.
Numerical simulation on gas ionization course in magnetically insulated transmission line oscillator
2012,
24: 732-738.
doi: 10.3788/HPLPB20122403.0732
Abstract:
Monte-Carlo collision (MMC) method was introduced for description of the gas collision ionization course. The MCC gas ionization module based on the method was developed for 3D fully electromagnetic and PIC code NEPTUNE. The magnetically insulated transmission line oscillator(MILO) model with He gas was simulated by the developed NEPTUNE code. The simulated results indicate that, in the condition of lower He gas density, He+ ions produced by ionization collision can not move freely because of their weight, and form channels which can neutralize space-charge field perfectly and then improve the bunching and beam-wave interaction efficiency obviously, so the output power of microwave increases, and the oscillation forming time shortens markedly. When the density of He gas increases to a higher level, the numbers of electrons and He+ ions increase quickly because of stronger ionization collision, the energy of electrons is separated due to collision courses, which goes against the bunching and beam-wave interaction improvement, and then the output power of microwave is decreased or cut off. The shortening of oscillation forming time is induced by the initial ionization stage with lower electron and ion densities. With the increasing gas density, the output power pulse will be shortened, and the diode of MILO will be shut down. In addition, the MILO models with air, vapor and carbon dioxide gas were simulated, respectively. The simulated results indicate that, in these conditions, the shortening of output microwave pulse is more severe compared with that for the MILO filled with He gas at the same pressure.
Monte-Carlo collision (MMC) method was introduced for description of the gas collision ionization course. The MCC gas ionization module based on the method was developed for 3D fully electromagnetic and PIC code NEPTUNE. The magnetically insulated transmission line oscillator(MILO) model with He gas was simulated by the developed NEPTUNE code. The simulated results indicate that, in the condition of lower He gas density, He+ ions produced by ionization collision can not move freely because of their weight, and form channels which can neutralize space-charge field perfectly and then improve the bunching and beam-wave interaction efficiency obviously, so the output power of microwave increases, and the oscillation forming time shortens markedly. When the density of He gas increases to a higher level, the numbers of electrons and He+ ions increase quickly because of stronger ionization collision, the energy of electrons is separated due to collision courses, which goes against the bunching and beam-wave interaction improvement, and then the output power of microwave is decreased or cut off. The shortening of oscillation forming time is induced by the initial ionization stage with lower electron and ion densities. With the increasing gas density, the output power pulse will be shortened, and the diode of MILO will be shut down. In addition, the MILO models with air, vapor and carbon dioxide gas were simulated, respectively. The simulated results indicate that, in these conditions, the shortening of output microwave pulse is more severe compared with that for the MILO filled with He gas at the same pressure.
2012,
24: 739-742.
doi: 10.3788/HPLPB20122403.0739
Abstract:
A Cassegrain phase-correcting Fresnel zone plate antenna has been designed based on the Fresnel principle and the Cassegrain antenna design method. It consists of a suite of Fresnel zone continuous phase correcting concentric rings, and its diameter of aperture is 200 mm. Compared with the traditional paraboloid antenna, the Cassegrain Fresnel zone plate antenna has a flat and light structure. It precedes the discrete phase diffraction antenna in radiation performance. The simulation has been performed based on the physical optics method, and the test has been performed with the near-field scanning system. The test shows that, at 95 GHz the antenna has an efficiency of 65%. Its 3 dB beam widths are 0.95 and 1.05, and its gain is 44.1 dB.
A Cassegrain phase-correcting Fresnel zone plate antenna has been designed based on the Fresnel principle and the Cassegrain antenna design method. It consists of a suite of Fresnel zone continuous phase correcting concentric rings, and its diameter of aperture is 200 mm. Compared with the traditional paraboloid antenna, the Cassegrain Fresnel zone plate antenna has a flat and light structure. It precedes the discrete phase diffraction antenna in radiation performance. The simulation has been performed based on the physical optics method, and the test has been performed with the near-field scanning system. The test shows that, at 95 GHz the antenna has an efficiency of 65%. Its 3 dB beam widths are 0.95 and 1.05, and its gain is 44.1 dB.
2012,
24: 743-746.
doi: 10.3788/HPLPB20122403.0743
Abstract:
This paper analyzes an S-band relativistic multiple-beam klystron amplifier with three-dimension electromagnetic codes, including the high frequency characteristics analysis and beam-wave interaction studies. The simulation results show that the adoption of coaxial resonant cavity can reduce the adverse effect of electric field asymmetryin the cavity, for the beams do not have to concentrate at the center to pass through the cavity, the use of multiple-beam can reduce the magnetic field strength for beam transport to about 0.4 T. Witha 700 kV, 5.8 kA electron beam under 0.4 T magnetic field, the amplifier with three coaxial resonant cavities has a peak pulsed output power of 1.4 GW with a corresponding electronic efficiency of 35%.
This paper analyzes an S-band relativistic multiple-beam klystron amplifier with three-dimension electromagnetic codes, including the high frequency characteristics analysis and beam-wave interaction studies. The simulation results show that the adoption of coaxial resonant cavity can reduce the adverse effect of electric field asymmetryin the cavity, for the beams do not have to concentrate at the center to pass through the cavity, the use of multiple-beam can reduce the magnetic field strength for beam transport to about 0.4 T. Witha 700 kV, 5.8 kA electron beam under 0.4 T magnetic field, the amplifier with three coaxial resonant cavities has a peak pulsed output power of 1.4 GW with a corresponding electronic efficiency of 35%.
2012,
24: 747-751.
doi: 10.3788/HPLPB20122403.0747
Abstract:
The dispersion characteristics of slow wave structure(SWS) and the beam-wave interaction theory are developed for klystron-like relativistic backward wave oscillator. The dispersion characteristics show that the operation mode of the device is TM01 mode, near point, and the maximum growth rate is relatively small, which is partly due to the larger span between the electron beam and SWS and will result in a larger saturation time. The frequency predicated by the dispersion characteristics is very close to that obtained in previous particle-in-cell simulation. In the beam-wave interaction theory, the 0th space harmonic of forward wave, the -1st space harmonic of backward wave, and the 1st-order beam space-charge wave are taken into account. The resonant reflector provides pre-modulation for beam current and electron energy, and the modulation cavity and extraction cavity introduce larger changes in coupling impedance and phase velocity. Both steady state and unsteady state calculations show that beam-wave conversion efficiency of above 40% can be obtained.
The dispersion characteristics of slow wave structure(SWS) and the beam-wave interaction theory are developed for klystron-like relativistic backward wave oscillator. The dispersion characteristics show that the operation mode of the device is TM01 mode, near point, and the maximum growth rate is relatively small, which is partly due to the larger span between the electron beam and SWS and will result in a larger saturation time. The frequency predicated by the dispersion characteristics is very close to that obtained in previous particle-in-cell simulation. In the beam-wave interaction theory, the 0th space harmonic of forward wave, the -1st space harmonic of backward wave, and the 1st-order beam space-charge wave are taken into account. The resonant reflector provides pre-modulation for beam current and electron energy, and the modulation cavity and extraction cavity introduce larger changes in coupling impedance and phase velocity. Both steady state and unsteady state calculations show that beam-wave conversion efficiency of above 40% can be obtained.
2012,
24: 752-756.
doi: 10.3788/HPLPB20122403.0752
Abstract:
A method for passive compression of a frequency-swept wave packet propagating through helically corrugated waveguide as the dispersive medium has been developed. By 3D fully electromagnetic particle-in-cell (PIC) code， the dispersion of a helically corrugated waveguide and the ways for improving the pulse compression ratio were investigated. The power compression ratio was 6.65 higher than previous simulation results after optimization. The simulation study demonstrates that a better compression result can be obtained by optimizing the frequency sweep mode of input microwave pulse， using improved elliptical waveguide and extending the length of the helical tapers.
A method for passive compression of a frequency-swept wave packet propagating through helically corrugated waveguide as the dispersive medium has been developed. By 3D fully electromagnetic particle-in-cell (PIC) code， the dispersion of a helically corrugated waveguide and the ways for improving the pulse compression ratio were investigated. The power compression ratio was 6.65 higher than previous simulation results after optimization. The simulation study demonstrates that a better compression result can be obtained by optimizing the frequency sweep mode of input microwave pulse， using improved elliptical waveguide and extending the length of the helical tapers.
