Email alert
RSS2025 Vol. 37, No. 3
Recommend Articles
Display Method:
2025,
37: 1-2.
2025,
37: 035001.
doi: 10.11884/HPLPB202537.240406
Abstract:
With the development of fast switching characteristics of SiC MOSFET devices, they are widely used in circuit systems that require high-speed and flexible high-voltage pulse output. Studies have shown that the on-time of SiC MOSFETs is mainly affected by the gate drive technology and its implementation. Accordingly, related studies have mostly focused on the optimization of its gate drive method. In this study, a SiC MOSFET gate drive circuit was parametrically tested and optimized, and applied to ultra-fast conduction SiC MOSFET devices to achieve a significant reduction in on-time. To verify the optimization effect, the research team designed and prepared an optimized prototype of the gate boost driver for experimental testing. The test data show that the optimized gate drive voltage regulation method effectively improves the device performance, with a pulse voltage rising edge time of up to 27 ns under 10 kV voltage level and 50 A current conditions.
With the development of fast switching characteristics of SiC MOSFET devices, they are widely used in circuit systems that require high-speed and flexible high-voltage pulse output. Studies have shown that the on-time of SiC MOSFETs is mainly affected by the gate drive technology and its implementation. Accordingly, related studies have mostly focused on the optimization of its gate drive method. In this study, a SiC MOSFET gate drive circuit was parametrically tested and optimized, and applied to ultra-fast conduction SiC MOSFET devices to achieve a significant reduction in on-time. To verify the optimization effect, the research team designed and prepared an optimized prototype of the gate boost driver for experimental testing. The test data show that the optimized gate drive voltage regulation method effectively improves the device performance, with a pulse voltage rising edge time of up to 27 ns under 10 kV voltage level and 50 A current conditions.
2025,
37: 035002.
doi: 10.11884/HPLPB202537.240409
Abstract:
Vacuum arc neutron tube is a neutron source with high pulse neutron yield. It is widely used in oil logging, element activation analysis and other fields. One of the typical characteristics of vacuum arc neutron tube is large discharge arc current and short pulse width. A set of vacuum arc discharge and extraction power supply with large current is developed, including trigger module, main arc module and extraction module. The trigger module uses LRC discharge mode, and the peak output voltage is 5 kV. The arc module uses LRCD half-cycle discharge mode to generate the main arc current. The peak current is 600 A, and the half-height width is about 0.5 μs. The extraction module uses a multi-voltage rectifier circuit to rectify AC high-voltage into the required high-voltage DC, with an output voltage of 120 kV and a load of 2 A. The simulation model is established according to the design scheme, and the simulation results show that the power supply can output the set electrical parameters well. Finally, a high-current vacuum arc discharge and extraction power supply is fabricated. The test results show that the output parameters of the power supply are consistent with the design values, and the power supply can be used to drive a vacuum arc neutron tube.
Vacuum arc neutron tube is a neutron source with high pulse neutron yield. It is widely used in oil logging, element activation analysis and other fields. One of the typical characteristics of vacuum arc neutron tube is large discharge arc current and short pulse width. A set of vacuum arc discharge and extraction power supply with large current is developed, including trigger module, main arc module and extraction module. The trigger module uses LRC discharge mode, and the peak output voltage is 5 kV. The arc module uses LRCD half-cycle discharge mode to generate the main arc current. The peak current is 600 A, and the half-height width is about 0.5 μs. The extraction module uses a multi-voltage rectifier circuit to rectify AC high-voltage into the required high-voltage DC, with an output voltage of 120 kV and a load of 2 A. The simulation model is established according to the design scheme, and the simulation results show that the power supply can output the set electrical parameters well. Finally, a high-current vacuum arc discharge and extraction power supply is fabricated. The test results show that the output parameters of the power supply are consistent with the design values, and the power supply can be used to drive a vacuum arc neutron tube.
2025,
37: 035003.
doi: 10.11884/HPLPB202537.240377
Abstract:
In view of the application requirements of the repetitive Marx-type pulse power source, a technology combining solid-state modulator and pulse transformer inductive adder was adopted to conduct research on high-power repetition-rate and high-voltage pulse charging power supply technology. The constant current charging technology based on BUCK-BOOST topology circuit was analyzed, and designs of key components such as high-voltage pulse transformer, pulse modulation module, BUCK-BOOST charging module, and control unit were provided. Finally a high-power repetitive pulse charging power supply prototype was developed. The prototype design specifications include a maximum output average power of 250 kW and a maximum repetition rate of 100 Hz. Charging tests were conducted on a 0.5 μF capacitive load, and an output pulse with a voltage of about 100 kV and a pulse rise time of about 300 μs was obtained. The pulse charging power supply adopts the design concept of multi module parallel connection in the primary design of the pulse transformer, which reduces the device volume and high voltage insulation risk. Under the conditions of high output power and high voltage, it achieves compact and modular design.
In view of the application requirements of the repetitive Marx-type pulse power source, a technology combining solid-state modulator and pulse transformer inductive adder was adopted to conduct research on high-power repetition-rate and high-voltage pulse charging power supply technology. The constant current charging technology based on BUCK-BOOST topology circuit was analyzed, and designs of key components such as high-voltage pulse transformer, pulse modulation module, BUCK-BOOST charging module, and control unit were provided. Finally a high-power repetitive pulse charging power supply prototype was developed. The prototype design specifications include a maximum output average power of 250 kW and a maximum repetition rate of 100 Hz. Charging tests were conducted on a 0.5 μF capacitive load, and an output pulse with a voltage of about 100 kV and a pulse rise time of about 300 μs was obtained. The pulse charging power supply adopts the design concept of multi module parallel connection in the primary design of the pulse transformer, which reduces the device volume and high voltage insulation risk. Under the conditions of high output power and high voltage, it achieves compact and modular design.
2025,
37: 035004.
doi: 10.11884/HPLPB202537.240376
Abstract:
To meet the demand of modulator power supplies, such as klystrons with output pulses of up to milliseconds, this article describes a long pulse high-voltage pulse power supply based on the Marx circuit. It uses solid-state semiconductor switches as the main discharge switches. Introducing independent charging semiconductor switches at each stage of the Marx units, it solves the limitation of Marx structures that use resistors to charge energy storage capacitors and cannot operate at high repetition rates. In addition, adding an auxiliary power supply to power the driving circuit of each stage’s Marx unit semiconductor switch, it can output long pulses of up to 1ms or more when triggered by optical fibers. The verification device designed using this Marx circuit has 6 basic charging and discharging units, which can output long pulses with an amplitude of −10 kV/1 A and a width of 1 ms. When outputting short pulses, the maximum output frequency exceeds 50 kHz.
To meet the demand of modulator power supplies, such as klystrons with output pulses of up to milliseconds, this article describes a long pulse high-voltage pulse power supply based on the Marx circuit. It uses solid-state semiconductor switches as the main discharge switches. Introducing independent charging semiconductor switches at each stage of the Marx units, it solves the limitation of Marx structures that use resistors to charge energy storage capacitors and cannot operate at high repetition rates. In addition, adding an auxiliary power supply to power the driving circuit of each stage’s Marx unit semiconductor switch, it can output long pulses of up to 1ms or more when triggered by optical fibers. The verification device designed using this Marx circuit has 6 basic charging and discharging units, which can output long pulses with an amplitude of −10 kV/1 A and a width of 1 ms. When outputting short pulses, the maximum output frequency exceeds 50 kHz.
2025,
37: 035005.
doi: 10.11884/HPLPB202537.240254
Abstract:
To reduce the threshold of using pulse power source, a compact pulse power source based on Marx generator is designed and implemented. The Marx generator is a 7-stage unipolar charging coaxial structure with low inductance ceramic capacitor and ultraviolet preionization output narrow pulse. It uses adjustable primary high voltage power supply and 2-way synchronous trigger switch, and metal shell as grounding shield and discharge circuit; it is filled with high pressure N2. Using the power source based on the above design, when the charging voltage is 26 kV and the charging voltage is 0.3 MPa, the high voltage pulse of 33 ns rise time, 59 ns pulse width and 109.2 kV amplitude is obtained on the 60 Ω load. The power source is housed in an aluminum alloy cylinder with a diameter of 0.2 m and a length of 1.1m. This paper provides a design idea and a reference prototype for compact and modular pulse power source, which can be used as flash X-ray radiography drive source.
To reduce the threshold of using pulse power source, a compact pulse power source based on Marx generator is designed and implemented. The Marx generator is a 7-stage unipolar charging coaxial structure with low inductance ceramic capacitor and ultraviolet preionization output narrow pulse. It uses adjustable primary high voltage power supply and 2-way synchronous trigger switch, and metal shell as grounding shield and discharge circuit; it is filled with high pressure N2. Using the power source based on the above design, when the charging voltage is 26 kV and the charging voltage is 0.3 MPa, the high voltage pulse of 33 ns rise time, 59 ns pulse width and 109.2 kV amplitude is obtained on the 60 Ω load. The power source is housed in an aluminum alloy cylinder with a diameter of 0.2 m and a length of 1.1m. This paper provides a design idea and a reference prototype for compact and modular pulse power source, which can be used as flash X-ray radiography drive source.
2025,
37: 035006.
doi: 10.11884/HPLPB202537.240183
Abstract:
Neutral beam injection heating is an effective heating method in magnetic confinement fusion experiments. If the ion source ignites during experimental operation, the extraction of the ion beam is terminated, reducing the efficiency and power of the neutral beam ion source beam extraction. To prolong the extraction of the ion source beam in case of abnormal situations, research on high-voltage power fast recovery technology was carried out, in which the beam was re-extracted by running the high-voltage power supply again. A fast recovery control system was developed based on PXI Express technology, using the PXIe-8861 processor and PXIe-7820R programmable logic gate array hardware board for fast recovery technology. The control system adopts a heartbeat packet mechanism for board and communication status monitoring, with two parameter configuration methods for client and upper computer, realizing the functions of online/offline data viewing and analysis. Through the configuration of the upper computer mode, the control system supports voltage and quantity control, meeting multiple working modes such as modulation, fast recovery, and single operation. Test results on a megawatt-class strong ion source show that the control system interface is user-friendly, the logical structure is designed clearly, and it has various control modes. Furthermore, the system can restart the high-voltage power supply to improve the extraction power of the ion source beam during the experiment.
Neutral beam injection heating is an effective heating method in magnetic confinement fusion experiments. If the ion source ignites during experimental operation, the extraction of the ion beam is terminated, reducing the efficiency and power of the neutral beam ion source beam extraction. To prolong the extraction of the ion source beam in case of abnormal situations, research on high-voltage power fast recovery technology was carried out, in which the beam was re-extracted by running the high-voltage power supply again. A fast recovery control system was developed based on PXI Express technology, using the PXIe-8861 processor and PXIe-7820R programmable logic gate array hardware board for fast recovery technology. The control system adopts a heartbeat packet mechanism for board and communication status monitoring, with two parameter configuration methods for client and upper computer, realizing the functions of online/offline data viewing and analysis. Through the configuration of the upper computer mode, the control system supports voltage and quantity control, meeting multiple working modes such as modulation, fast recovery, and single operation. Test results on a megawatt-class strong ion source show that the control system interface is user-friendly, the logical structure is designed clearly, and it has various control modes. Furthermore, the system can restart the high-voltage power supply to improve the extraction power of the ion source beam during the experiment.
2025,
37: 035007.
doi: 10.11884/HPLPB202537.240253
Abstract:
This paper proposes a square wave pulse adder with truncation function based on the self-triggering technology. The N-type switch and the P-type switch are connected in series to form a special half-bridge structure, which only needs to provide an isolated bipolar signal to control the first-stage charging and discharging switches. All other switches are turned on and off step by step to generate high-voltage square wave pulses. This technology not only greatly simplifies the driving circuit of the pulse adder, but also realizes the truncation function to generate quasi-square wave pulses with fast front and back edges. By using the automatic conduction characteristics of the depleted N-type switch, the self-charging without control is realized, and the insulation level of the drive circuit is significantly improved. A 9-stage power supply prototype is built for experimental verification, and the the experimental results show that a stable repetitive positive square wave pulse is generated on a 10 kΩ resistive load. The voltage amplitude is adjustable from 2.3 kV to 3.6 kV, the pulse width is adjustable from 1 μs to 10 μs, and the frequency is adjustable from 0 kHz to 1 kHz. The front and rear edges are both about 100 ns, and accelerate with the increase of operating voltage. The waveform under 10 kΩ and 3 nF resistance-capacitance series load is still a good square wave pulse, and the front and rear edges of the pulse do not slow down significantly with the resistance load. The compact structure of the pulse adder can help to miniaturize solid-state pulse generators.
This paper proposes a square wave pulse adder with truncation function based on the self-triggering technology. The N-type switch and the P-type switch are connected in series to form a special half-bridge structure, which only needs to provide an isolated bipolar signal to control the first-stage charging and discharging switches. All other switches are turned on and off step by step to generate high-voltage square wave pulses. This technology not only greatly simplifies the driving circuit of the pulse adder, but also realizes the truncation function to generate quasi-square wave pulses with fast front and back edges. By using the automatic conduction characteristics of the depleted N-type switch, the self-charging without control is realized, and the insulation level of the drive circuit is significantly improved. A 9-stage power supply prototype is built for experimental verification, and the the experimental results show that a stable repetitive positive square wave pulse is generated on a 10 kΩ resistive load. The voltage amplitude is adjustable from 2.3 kV to 3.6 kV, the pulse width is adjustable from 1 μs to 10 μs, and the frequency is adjustable from 0 kHz to 1 kHz. The front and rear edges are both about 100 ns, and accelerate with the increase of operating voltage. The waveform under 10 kΩ and 3 nF resistance-capacitance series load is still a good square wave pulse, and the front and rear edges of the pulse do not slow down significantly with the resistance load. The compact structure of the pulse adder can help to miniaturize solid-state pulse generators.
2025,
37: 035008.
doi: 10.11884/HPLPB202537.240315
Abstract:
Pulsed power drive source is a key part of high power microwave technology. The quality of the output waveform of pulsed power drive source directly affects the output of high power microwave devices. Aiming at the oscillation problem of the flat-top output waveform of pulse power drive source, we designed and developed a compact pulsed power drive source based on PFN-Marx, and optimized the waveform. The parameters of PFN-Marx generators with different structures are analyzed by PSpice simulation, so as to determine the number of sections and levels of the PFN-Marx generator; the oscillation problem of the output waveform is converted into the degree of deviation of each extreme point from the reference value in the flat-top area of the waveform. The objective function is constructed with the root mean square error with the minimum flat-top ripple error as the goal, and the circuit model is established in Simulink. Combined with the MATLAB genetic algorithm, the inductance of the PFN is continuously iteratively optimized. Finally, a set of optimal values is determined, and the inductance structure is redesigned to adjust the inductance value conveniently and to achieve quick waveform optimization. The optimized single-stage PFN outputs a waveform with a leading edge of 24.4 ns and a pulse width of 93.6 ns on a 10 Ω load, and it has good flat-top performance. The assembled 7-stage PFN-Marx generator has an output quasi-square wave under a charging voltage of 53.8 kV and a load impedance of 75 Ω. The pulse peak amplitude is 189.2 kV, pulse width is 93.2 ns, rise time is 8.4 ns, decrease time is 33.6 ns, and the ripple coefficient is 3.5%.
Pulsed power drive source is a key part of high power microwave technology. The quality of the output waveform of pulsed power drive source directly affects the output of high power microwave devices. Aiming at the oscillation problem of the flat-top output waveform of pulse power drive source, we designed and developed a compact pulsed power drive source based on PFN-Marx, and optimized the waveform. The parameters of PFN-Marx generators with different structures are analyzed by PSpice simulation, so as to determine the number of sections and levels of the PFN-Marx generator; the oscillation problem of the output waveform is converted into the degree of deviation of each extreme point from the reference value in the flat-top area of the waveform. The objective function is constructed with the root mean square error with the minimum flat-top ripple error as the goal, and the circuit model is established in Simulink. Combined with the MATLAB genetic algorithm, the inductance of the PFN is continuously iteratively optimized. Finally, a set of optimal values is determined, and the inductance structure is redesigned to adjust the inductance value conveniently and to achieve quick waveform optimization. The optimized single-stage PFN outputs a waveform with a leading edge of 24.4 ns and a pulse width of 93.6 ns on a 10 Ω load, and it has good flat-top performance. The assembled 7-stage PFN-Marx generator has an output quasi-square wave under a charging voltage of 53.8 kV and a load impedance of 75 Ω. The pulse peak amplitude is 189.2 kV, pulse width is 93.2 ns, rise time is 8.4 ns, decrease time is 33.6 ns, and the ripple coefficient is 3.5%.
2025,
37: 035009.
doi: 10.11884/HPLPB202537.240355
Abstract:
Inductive energy storage is one of the main energy storage methods of pulse power technology, and the problem of limiting voltage of open circuit switch is its inherent problem. To explore the voltage limiting method of superconducting pulse power supply, this paper introduces two traditional voltage limiting methods, namely ZnO varistor voltage limiting and pulse capacitor voltage limiting, and puts forward the combination of pulse capacitor and varistor to limit the open switching voltage. The working principle of superconducting pulse power supply under three voltage limiting modes is analyzed, and the output characteristics under different voltage limiting parameters are simulated based on the same superconducting pulse transformer parameters. Aiming at the combined voltage limiting mode, a small experimental platform was built to verify the voltage limiting performance of the device. Simulation and experimental results show that the combined voltage limiting mode has higher transmission efficiency and better turn-off performance than varistor, and it has faster discharge and lower voltage amplitude than pulse capacitance limiting mode.
Inductive energy storage is one of the main energy storage methods of pulse power technology, and the problem of limiting voltage of open circuit switch is its inherent problem. To explore the voltage limiting method of superconducting pulse power supply, this paper introduces two traditional voltage limiting methods, namely ZnO varistor voltage limiting and pulse capacitor voltage limiting, and puts forward the combination of pulse capacitor and varistor to limit the open switching voltage. The working principle of superconducting pulse power supply under three voltage limiting modes is analyzed, and the output characteristics under different voltage limiting parameters are simulated based on the same superconducting pulse transformer parameters. Aiming at the combined voltage limiting mode, a small experimental platform was built to verify the voltage limiting performance of the device. Simulation and experimental results show that the combined voltage limiting mode has higher transmission efficiency and better turn-off performance than varistor, and it has faster discharge and lower voltage amplitude than pulse capacitance limiting mode.
2025,
37: 035010.
doi: 10.11884/HPLPB202537.240344
Abstract:
A fast pulsed power supply is developed to meet the requirements of High Intensity heavy-ion Accelerator Facility (HIAF) ion source. Based on front-stage voltage fast switching method, we proposed a circuit topology ensuring fast rising of current and flat top stability. In this paper the circuit is analyzed, the essential parameters of the circuit are calculated, and the control method is established for adapting abruptly-changed voltage. A 16 kW power supply based on the principle of modular manufacturing was tested on real magnet load. The results show the value of flat top current reaches 570 A, the error is less than ±1×10−4, the repetition frequency reaches 3 Hz, the rise time and fall time is less than 30 ms, the flat top time can be set within 900 ms.
A fast pulsed power supply is developed to meet the requirements of High Intensity heavy-ion Accelerator Facility (HIAF) ion source. Based on front-stage voltage fast switching method, we proposed a circuit topology ensuring fast rising of current and flat top stability. In this paper the circuit is analyzed, the essential parameters of the circuit are calculated, and the control method is established for adapting abruptly-changed voltage. A 16 kW power supply based on the principle of modular manufacturing was tested on real magnet load. The results show the value of flat top current reaches 570 A, the error is less than ±1×10−4, the repetition frequency reaches 3 Hz, the rise time and fall time is less than 30 ms, the flat top time can be set within 900 ms.
2025,
37: 015011.
doi: 10.11884/HPLPB202537.240302
Abstract:
To satisfy high reliability requirements of scintillation detectors in neutron testing devices, an all-hardware automatic detection and control circuit had been developed. We proposed to use a multivibrator to generate self-checking pulse signal with a certain period of time, and use comparator, monostable flip-flop and other circuits to monitor and judge the working state of the main channel of scintillation detector in real time. The automatic switching control between the main channel signal and the standby channel signal was realized by analog switch. We have solved the problems such as complex output circuit, large volume and poor adaptability to radiation environment in the current automatic detection and control circuit, realized redundant design of scintillation detector channels, and improved the working reliability of scintillation detector. The circuit was successfully used in general telemetry probe and met the experimental requirements well.
To satisfy high reliability requirements of scintillation detectors in neutron testing devices, an all-hardware automatic detection and control circuit had been developed. We proposed to use a multivibrator to generate self-checking pulse signal with a certain period of time, and use comparator, monostable flip-flop and other circuits to monitor and judge the working state of the main channel of scintillation detector in real time. The automatic switching control between the main channel signal and the standby channel signal was realized by analog switch. We have solved the problems such as complex output circuit, large volume and poor adaptability to radiation environment in the current automatic detection and control circuit, realized redundant design of scintillation detector channels, and improved the working reliability of scintillation detector. The circuit was successfully used in general telemetry probe and met the experimental requirements well.
2025,
37: 035012.
doi: 10.11884/HPLPB202537.240259
Abstract:
The magnetic compression power supply, which is an important part of the magnetic compression system, is mainly used to supply power to the compressed magnet to form the magnetic field shape required for the field reversed configuration (FRC) plasma compression. The large number of compressed magnet coils and the complex coupling relationship between the coils make it difficult to solve the power supply parameters required to form the target magnetic field. A method is proposed for designing the parameters of the magnetic compression power supply based on genetic algorithm with high efficiency. According to the topology of the magnetic compression power supply and the coupling relationship between the coils, the physical model of the power supply system is derived. On the basis of the physical model, the design method of compressed power supply parameters based on genetic algorithm is proposed, and the basic principle of the method is clarified. The algorithm code and the MATLAB simulation model are established. Under the ideal situation and practical engineering design conditions, the power supply parameters of the HFRC magnetic compression system were optimized, the results of the optimized magnetic field are basically consistent with the target magnetic field shape. Meanwhile, a MAXWELL simulation model is established for comparative analysis, the results of these two modes fits well, which verifies the effectiveness and accuracy of the method in power supply design.
The magnetic compression power supply, which is an important part of the magnetic compression system, is mainly used to supply power to the compressed magnet to form the magnetic field shape required for the field reversed configuration (FRC) plasma compression. The large number of compressed magnet coils and the complex coupling relationship between the coils make it difficult to solve the power supply parameters required to form the target magnetic field. A method is proposed for designing the parameters of the magnetic compression power supply based on genetic algorithm with high efficiency. According to the topology of the magnetic compression power supply and the coupling relationship between the coils, the physical model of the power supply system is derived. On the basis of the physical model, the design method of compressed power supply parameters based on genetic algorithm is proposed, and the basic principle of the method is clarified. The algorithm code and the MATLAB simulation model are established. Under the ideal situation and practical engineering design conditions, the power supply parameters of the HFRC magnetic compression system were optimized, the results of the optimized magnetic field are basically consistent with the target magnetic field shape. Meanwhile, a MAXWELL simulation model is established for comparative analysis, the results of these two modes fits well, which verifies the effectiveness and accuracy of the method in power supply design.
2025,
37: 035013.
doi: 10.11884/HPLPB202537.240431
Abstract:
The high voltage power supply is an important part of the neutral beam injection heating system. It determines the beam energy and the quality of the extraction beam current. To meet the power demand of the neutral beam injection system of the HL-3 device, a modular high voltage power supply system based on pulse step modulation technology is designed, and it can achieve the rated voltage output by connecting power modules in series. Each power module uses a single-phase bridge rectifier and half-bridge output control, and a soft start circuit and an energy discharge circuit are designed so that the power supply module can carry out function tests under the condition of utility power supply. The regulator module circuit is developed to realize the precise control of power supply output voltage by controlling the output voltage level of the regulator module. The power supply module and system are simulated using the Simulink toolbox in MATLAB, and tests of the soft start function, rated output and short circuit protection performance of the power supply are carried out. By building a1600 V/50 A/5 s test prototype and a 80 kV/50 A/5 s test platform, the electrical performance of the power system module hasbeen experimentally verified. The simulation and experimental results show that the power module test prototype meets the rated output requirements and can stably realize 80 kV voltage output through series connection. The short circuit protection time is less than 6 μs, which can realize fast protection shutdown of the power supply and meet the performance requirements of the HL-3 device 5 MW neutral beam injection system for the high voltage power supply.
The high voltage power supply is an important part of the neutral beam injection heating system. It determines the beam energy and the quality of the extraction beam current. To meet the power demand of the neutral beam injection system of the HL-3 device, a modular high voltage power supply system based on pulse step modulation technology is designed, and it can achieve the rated voltage output by connecting power modules in series. Each power module uses a single-phase bridge rectifier and half-bridge output control, and a soft start circuit and an energy discharge circuit are designed so that the power supply module can carry out function tests under the condition of utility power supply. The regulator module circuit is developed to realize the precise control of power supply output voltage by controlling the output voltage level of the regulator module. The power supply module and system are simulated using the Simulink toolbox in MATLAB, and tests of the soft start function, rated output and short circuit protection performance of the power supply are carried out. By building a
2025,
37: 035014.
doi: 10.11884/HPLPB202537.240273
Abstract:
Shenzhen’s medium-energy high-repetition-rate X-ray free electron laser (Shenzhen Superconducting Soft X-ray Free Electron Laser, S3FEL) requires 1 MHz high-repetition-rate and high-stability kicker. Transmission line structure kicker is an effective way to achieve high repetition rate. However, the insufficient waveform stability of the transmission line structure kicker limits the application of this type of kicker in large particle accelerators. To solve the above problem, this paper studies the input waveform and circuit structure parameters of the transmission line structure kicker. It analyzes the main factors affecting the stability of kicker’s working waveform using mathematical tools such as Fourier analysis, and reveals the relationship between the harmonic order of kicker ideal waveform and the cut-off frequency of kicker transmission line structure. On this basis, this paper proposes a method to reduce the deviation between the actual waveform and the ideal waveform of kicker. This method can obtain the ideal working waveform of kicker within a certain range by adjusting the input waveform parameters or the cut-off frequency of kicker. To verify the above relationship, this paper uses circuit simulation software to simulate different waveforms and different circuit parameters of kicker. The simulation results verify that the above relationship revealed and confirm the effectiveness of the method mentioned.
Shenzhen’s medium-energy high-repetition-rate X-ray free electron laser (Shenzhen Superconducting Soft X-ray Free Electron Laser, S3FEL) requires 1 MHz high-repetition-rate and high-stability kicker. Transmission line structure kicker is an effective way to achieve high repetition rate. However, the insufficient waveform stability of the transmission line structure kicker limits the application of this type of kicker in large particle accelerators. To solve the above problem, this paper studies the input waveform and circuit structure parameters of the transmission line structure kicker. It analyzes the main factors affecting the stability of kicker’s working waveform using mathematical tools such as Fourier analysis, and reveals the relationship between the harmonic order of kicker ideal waveform and the cut-off frequency of kicker transmission line structure. On this basis, this paper proposes a method to reduce the deviation between the actual waveform and the ideal waveform of kicker. This method can obtain the ideal working waveform of kicker within a certain range by adjusting the input waveform parameters or the cut-off frequency of kicker. To verify the above relationship, this paper uses circuit simulation software to simulate different waveforms and different circuit parameters of kicker. The simulation results verify that the above relationship revealed and confirm the effectiveness of the method mentioned.
2025,
37: 035015.
doi: 10.11884/HPLPB202537.240318
Abstract:
Hybrid integrated DC-DC converters are widely used in harsh environments and applications with high-reliability requirements due to their wide operating temperature range and long-term reliability. This paper presents the design of a hybrid integrated DC-DC converter with an input voltage of 28 V and an output of 5 V/20 A, leveraging the excellent high-frequency, low-loss characteristics of gallium nitride (GaN) devices. The design incorporates an active clamp soft-switching topology, hybrid integrated micro-assembly technology, and high-current, low thermal resistance hermetic packaging techniques. The converter operates at a switching frequency of 800 kHz and achieves a peak efficiency of 92%. The paper elaborates in detail on the design methods and technical details of the active clamp power circuit, the control of parasitic parameters and oscillating voltage in the GaN HEMT drive circuit, the optimization of synchronous rectification timing and dead-time, the thick-film hybrid integration process, and heat dissipation design. Through simulations and prototype experiments, it verifies and demonstrates the advantages of GaN HEMTS and hybrid integrated circuits in high power density and high efficiency aspects.
Hybrid integrated DC-DC converters are widely used in harsh environments and applications with high-reliability requirements due to their wide operating temperature range and long-term reliability. This paper presents the design of a hybrid integrated DC-DC converter with an input voltage of 28 V and an output of 5 V/20 A, leveraging the excellent high-frequency, low-loss characteristics of gallium nitride (GaN) devices. The design incorporates an active clamp soft-switching topology, hybrid integrated micro-assembly technology, and high-current, low thermal resistance hermetic packaging techniques. The converter operates at a switching frequency of 800 kHz and achieves a peak efficiency of 92%. The paper elaborates in detail on the design methods and technical details of the active clamp power circuit, the control of parasitic parameters and oscillating voltage in the GaN HEMT drive circuit, the optimization of synchronous rectification timing and dead-time, the thick-film hybrid integration process, and heat dissipation design. Through simulations and prototype experiments, it verifies and demonstrates the advantages of GaN HEMTS and hybrid integrated circuits in high power density and high efficiency aspects.
2025,
37: 035016.
doi: 10.11884/HPLPB202537.240258
Abstract:
Shields are widely used in EMC applications because they can realize electromagnetic shielding, but few studies have focused on the effect of shields on electromagnetic crosstalk in a double differential mode loop, consisting of independent power loop and signal loop. A crosstalk analysis model based on the theory of multi-conductor transmission lines is proposed. In this method, the per-unit-length equivalent circuit model of the system is established, then the transmission line equations of the system are written by using the idea of finite difference and Kirchhoff’s voltage law and current law, and the crosstalk results are obtained finally. The crosstalk results are compared with the simulation results obtained by CST software to verify the feasibility and validity of the model and calculation method. Separate calculations of inductive coupling and capacitive coupling are carried out to study the effect of shielding layer on electromagnetic crosstalk between cables, and crosstalk suppression methods are obtained, which can provide guidance for the selection of cables and the selection of shield grounding method in practical engineering applications.
Shields are widely used in EMC applications because they can realize electromagnetic shielding, but few studies have focused on the effect of shields on electromagnetic crosstalk in a double differential mode loop, consisting of independent power loop and signal loop. A crosstalk analysis model based on the theory of multi-conductor transmission lines is proposed. In this method, the per-unit-length equivalent circuit model of the system is established, then the transmission line equations of the system are written by using the idea of finite difference and Kirchhoff’s voltage law and current law, and the crosstalk results are obtained finally. The crosstalk results are compared with the simulation results obtained by CST software to verify the feasibility and validity of the model and calculation method. Separate calculations of inductive coupling and capacitive coupling are carried out to study the effect of shielding layer on electromagnetic crosstalk between cables, and crosstalk suppression methods are obtained, which can provide guidance for the selection of cables and the selection of shield grounding method in practical engineering applications.
2025,
37: 035017.
doi: 10.11884/HPLPB202537.240403
Abstract:
To address issues of low output voltage ripple, low ignition energy, rapid shutdown, and to reduce the risk of high-voltage ignition damaging high-frequency switching devices in neutral beam injection systems, this paper proposes a modular inverter-type power supply design. A 120 kV/80 A high-power modular inverter is developed. It incorporates 84 high-frequency inverter modules (1600 V/80 A each) in a staggered-phase cascade. Transformer isolation of each module safeguards switching devices during ignition, and 20 kHz staggered phase control minimizes filter capacitor requirements while maintaining voltage ripple within specifications. Performance tests on the prototype show a rise time of 67 μs, turn-off time of 3.3 μs, ignition energy release of 1.3 J, and voltage ripple of 3.3%, with the ripple reduced to less than 1% after interleaving. The results demonstrate that the design achieves low voltage ripple, fast shutdown, and controlled energy release, confirming its feasibility
To address issues of low output voltage ripple, low ignition energy, rapid shutdown, and to reduce the risk of high-voltage ignition damaging high-frequency switching devices in neutral beam injection systems, this paper proposes a modular inverter-type power supply design. A 120 kV/80 A high-power modular inverter is developed. It incorporates 84 high-frequency inverter modules (
2025,
37: 035018.
doi: 10.11884/HPLPB202537.240233
Abstract:
During the charging period of capacitor charging power supply, the input power gradually increases with the increase of output voltage, which not only requires the power grid to provide a larger peak power, but also causes larger current harmonics. This paper proposes a new control algorithm based on a single-stage charging scheme with buffered energy storage capacitor that realizes the function of constant power input and constant current output only by using single-stage energy conversion. The algorithm not only achieves better constant power characteristics, but also lifts the restriction on the auxiliary energy storage capacitor. It can improve the equivalent excitation voltage by controlling the voltage of the energy storage capacitor. Matlab/Simulink was used to build a simulation model of capacitor charging with constant power input. The simulation results show that the constant power at the input end is realized during the 400−2 000 V stage of capacitor charging, the initial voltage of auxiliary capacitor is increased, and the voltage of load capacitor is increased at the same time. This shows that the algorithm can achieve better input constant power characteristics and circuit boost function.
During the charging period of capacitor charging power supply, the input power gradually increases with the increase of output voltage, which not only requires the power grid to provide a larger peak power, but also causes larger current harmonics. This paper proposes a new control algorithm based on a single-stage charging scheme with buffered energy storage capacitor that realizes the function of constant power input and constant current output only by using single-stage energy conversion. The algorithm not only achieves better constant power characteristics, but also lifts the restriction on the auxiliary energy storage capacitor. It can improve the equivalent excitation voltage by controlling the voltage of the energy storage capacitor. Matlab/Simulink was used to build a simulation model of capacitor charging with constant power input. The simulation results show that the constant power at the input end is realized during the 400−2 000 V stage of capacitor charging, the initial voltage of auxiliary capacitor is increased, and the voltage of load capacitor is increased at the same time. This shows that the algorithm can achieve better input constant power characteristics and circuit boost function.
2025,
37: 035019.
doi: 10.11884/HPLPB202537.240267
Abstract:
To obtain the actual bandwidth and RC time constant of the passive integrator, the frequency characteristics of the integrator under 1 MΩ load are measured by using an active high impedance probe and a network analyzer. The simulation results show that the low frequency inflection point of integrator with 1 MΩ load is much lower than that of integrator with 50 Ω load, and the measured results are consistent with the simulation results. The measured results show that the high frequency band of the integrator frequency characteristic deviates from the theoretical trend of linear attenuation inversely proportional to the frequency, which is related to the spurious parameters of the actual circuit. The upper frequency limit of the integrator is determined by transforming the ordinate, and the working bandwidth of the integrator can be obtained by combining the calculated low frequency results. The RC constant is calculated by using the amplitude-frequency characteristic formula of integrator combined with the frequency sweep measurement results, which is simple and has less uncertainty.
To obtain the actual bandwidth and RC time constant of the passive integrator, the frequency characteristics of the integrator under 1 MΩ load are measured by using an active high impedance probe and a network analyzer. The simulation results show that the low frequency inflection point of integrator with 1 MΩ load is much lower than that of integrator with 50 Ω load, and the measured results are consistent with the simulation results. The measured results show that the high frequency band of the integrator frequency characteristic deviates from the theoretical trend of linear attenuation inversely proportional to the frequency, which is related to the spurious parameters of the actual circuit. The upper frequency limit of the integrator is determined by transforming the ordinate, and the working bandwidth of the integrator can be obtained by combining the calculated low frequency results. The RC constant is calculated by using the amplitude-frequency characteristic formula of integrator combined with the frequency sweep measurement results, which is simple and has less uncertainty.
2025,
37: 035020.
doi: 10.11884/HPLPB202537.240423
Abstract:
This paper presents a control system based on EPICS (Experimental Physics and Industrial Control System) for multiple low-energy and high-energy power supplies configured along the injector prototype of a high-repetition-rate XFEL (X-ray Free Electron Laser) facility. Due to the global applicability of EPICS in accelerator technology and its established technical foundation in China, the system is designed based on the EPICS architecture. It employs a hierarchical distributed network structure, establishes Channel Access (CA) based on the TCP/IP protocol, and offers application interface subroutine libraries for both clients and servers. The driver development of all underlying power devices is completed through a unique device communication driver software module based on data flow. And the client control software is developed based on CSS (Control System Studio). This system fulfills users’ requirements for real-time centralized remote monitoring and control of power supplies, as well as data sharing with other related subsystems, which improves the operating efficiency and reliability of the device.
This paper presents a control system based on EPICS (Experimental Physics and Industrial Control System) for multiple low-energy and high-energy power supplies configured along the injector prototype of a high-repetition-rate XFEL (X-ray Free Electron Laser) facility. Due to the global applicability of EPICS in accelerator technology and its established technical foundation in China, the system is designed based on the EPICS architecture. It employs a hierarchical distributed network structure, establishes Channel Access (CA) based on the TCP/IP protocol, and offers application interface subroutine libraries for both clients and servers. The driver development of all underlying power devices is completed through a unique device communication driver software module based on data flow. And the client control software is developed based on CSS (Control System Studio). This system fulfills users’ requirements for real-time centralized remote monitoring and control of power supplies, as well as data sharing with other related subsystems, which improves the operating efficiency and reliability of the device.
2025,
37: 035021.
doi: 10.11884/HPLPB202537.240303
Abstract:
To conduct high parameter physics research, a long pulse switching power supply module and control system were developed to meet the power supply needs of pulse generator sets and power grids. By changing the wiring method of the AC terminal, the module power supply type can be switched. The built-in soft start circuit can reduce the impact of surge current on the power grid and switches during the module charging process. The module meets the design requirements for protection, driving, and status monitoring under two operating modes. The controller adopts a microcontroller and programmable logic gate array architecture, with both local and remote control functions. It implements soft start control and monitoring functions in microcontrollers by using integrated circuit bus to expand IO, reducing the complexity of data transmission links and systems; The control algorithm of the power supply is implemented in the programmable logic gate array, and the interface design of the photoelectric conversion box is unified through level signal conversion processing. The test results under the established testing conditions show that the module and control system design meet the requirements, and the power supply achieves 80 kV/100 s/20 A output parameter testing and protection experiment testing.
To conduct high parameter physics research, a long pulse switching power supply module and control system were developed to meet the power supply needs of pulse generator sets and power grids. By changing the wiring method of the AC terminal, the module power supply type can be switched. The built-in soft start circuit can reduce the impact of surge current on the power grid and switches during the module charging process. The module meets the design requirements for protection, driving, and status monitoring under two operating modes. The controller adopts a microcontroller and programmable logic gate array architecture, with both local and remote control functions. It implements soft start control and monitoring functions in microcontrollers by using integrated circuit bus to expand IO, reducing the complexity of data transmission links and systems; The control algorithm of the power supply is implemented in the programmable logic gate array, and the interface design of the photoelectric conversion box is unified through level signal conversion processing. The test results under the established testing conditions show that the module and control system design meet the requirements, and the power supply achieves 80 kV/100 s/20 A output parameter testing and protection experiment testing.
2025,
37: 035022.
doi: 10.11884/HPLPB202537.240272
Abstract:
Spacecraft have to be exposed to complex and harsh space radiation environments for a long time during their in-orbit service. III-V compound solar cells, represented by GaAs, are widely used in the aerospace field due to their high photoelectric conversion efficiency and radiation resistance. The spatial radiation damage effect of GaAs solar cells was studied using finite element method and technology computer-aided design (TCAD). Based on the electrical parameters of GaAs solar cells under AM0 spectral irradiation, a single-junction solar cell structure model and irradiation damage model were established. The volt ampere characteristic curves of the cells under different electron irradiation conditions were obtained, and the simulation results in this paper were verified with existing experimental results. The degradation law of GaAs solar cell electrical performance under space environment irradiation was analyzed. The results indicate that irradiation damage defects reduce the diffusion length of minority carriers and decrease the collection efficiency of photo generated carriers. At a certain electron energy, the degradation amplitude of the electrical performance of solar cells increases with the increase of irradiation dose level.
Spacecraft have to be exposed to complex and harsh space radiation environments for a long time during their in-orbit service. III-V compound solar cells, represented by GaAs, are widely used in the aerospace field due to their high photoelectric conversion efficiency and radiation resistance. The spatial radiation damage effect of GaAs solar cells was studied using finite element method and technology computer-aided design (TCAD). Based on the electrical parameters of GaAs solar cells under AM0 spectral irradiation, a single-junction solar cell structure model and irradiation damage model were established. The volt ampere characteristic curves of the cells under different electron irradiation conditions were obtained, and the simulation results in this paper were verified with existing experimental results. The degradation law of GaAs solar cell electrical performance under space environment irradiation was analyzed. The results indicate that irradiation damage defects reduce the diffusion length of minority carriers and decrease the collection efficiency of photo generated carriers. At a certain electron energy, the degradation amplitude of the electrical performance of solar cells increases with the increase of irradiation dose level.
2025,
37: 035023.
doi: 10.11884/HPLPB202537.240371
Abstract:
Multiple switches need to be triggered in parallel in a linear transformer driver (LTD) cavity where tens of bricks connect in parallel and switch-to-switch isolation is necessary. The lack of transient isolation will result in poor tolerance to switch jitter which manifests as increased probability of late-firing switches and further affect the output performance of LTD cavity. To optimize the type and parameter of the isolation element, the resistive and inductive isolation modes were compared in circuit simulation and experiments. The results show that the inductive isolation has better effect than resistive isolation both for synchronization of switches and the load voltage of LTD cavity. Among the inductors with different inductance, the 5 μH inductor has the best output data.
Multiple switches need to be triggered in parallel in a linear transformer driver (LTD) cavity where tens of bricks connect in parallel and switch-to-switch isolation is necessary. The lack of transient isolation will result in poor tolerance to switch jitter which manifests as increased probability of late-firing switches and further affect the output performance of LTD cavity. To optimize the type and parameter of the isolation element, the resistive and inductive isolation modes were compared in circuit simulation and experiments. The results show that the inductive isolation has better effect than resistive isolation both for synchronization of switches and the load voltage of LTD cavity. Among the inductors with different inductance, the 5 μH inductor has the best output data.
2025,
37: 035024.
doi: 10.11884/HPLPB202537.240263
Abstract:
To solve the problems of the contradiction between response speed and overshoot and the poor anti-interference ability in the complex electromagnetic environment, the traditional PI controller is used in the resonant magnetic perturbations (RMP) coil power supply in tokamak. This paper adopts the linear active disturbance rejection control strategy to extract the differential quality of the reference signal in the transition differentiator to achieve fast current output without overshoot. Based on partially known mathematical model parameters of the power system, a fourth-order linear extended state observer is designed to estimate the disturbance of the system as a new state quantity to compensate the internal uncertainties and external disturbances of the power system. Finally, the simulation results show that compared with the traditional PI control strategy, the active disturbance rejection control strategy can effectively improve the dynamic characteristics of the output current signal. At the same time, it has stronger robustness and anti-interference characteristics in the case of complex environmental disturbance.
To solve the problems of the contradiction between response speed and overshoot and the poor anti-interference ability in the complex electromagnetic environment, the traditional PI controller is used in the resonant magnetic perturbations (RMP) coil power supply in tokamak. This paper adopts the linear active disturbance rejection control strategy to extract the differential quality of the reference signal in the transition differentiator to achieve fast current output without overshoot. Based on partially known mathematical model parameters of the power system, a fourth-order linear extended state observer is designed to estimate the disturbance of the system as a new state quantity to compensate the internal uncertainties and external disturbances of the power system. Finally, the simulation results show that compared with the traditional PI control strategy, the active disturbance rejection control strategy can effectively improve the dynamic characteristics of the output current signal. At the same time, it has stronger robustness and anti-interference characteristics in the case of complex environmental disturbance.
2025,
37: 035025.
doi: 10.11884/HPLPB202537.240417
Abstract:
A discharge strategy for a battery-supercapacitor hybrid energy storage system is designed based on model predictive control theory to match the power and energy requirement of the magnet coil of HL-3. Using toroidal field coil as the load of the energy storage system, the mathematical model of the system and the objective function based on battery/supercapacitor characteristics and energy demands of the load is established. The optimal switching sequence is solved in real time. Long cycle control is applied on battery energy storage system to achieve stable discharge of the battery, while short cycle control is applied on supercapacitor energy storage system to achieve transient response of the supercapacitor. Simulation experiments are conducted using MATLAB/Simulink. The hybrid energy storage system stably outputs a flat top current that meets the load demand, with a current ripple of 0.22%. The simulation results verify the effectiveness of the proposed control method.
A discharge strategy for a battery-supercapacitor hybrid energy storage system is designed based on model predictive control theory to match the power and energy requirement of the magnet coil of HL-3. Using toroidal field coil as the load of the energy storage system, the mathematical model of the system and the objective function based on battery/supercapacitor characteristics and energy demands of the load is established. The optimal switching sequence is solved in real time. Long cycle control is applied on battery energy storage system to achieve stable discharge of the battery, while short cycle control is applied on supercapacitor energy storage system to achieve transient response of the supercapacitor. Simulation experiments are conducted using MATLAB/Simulink. The hybrid energy storage system stably outputs a flat top current that meets the load demand, with a current ripple of 0.22%. The simulation results verify the effectiveness of the proposed control method.
