2021 Vol. 33, No. 7
Based on our findings published in 2016, we have carried out further studies on the performance improvements of the relativistic magnetron using all-cavity output and semi-transparent cathode. By modifying the semi-transparent cathode, that is, rotating it in azimuthal direction and using uneven height of the electron emission surfaces, and partially optimizing of the cathode parameters, the mode competitions in the initial stage are prohibited effectively in relatively wider range of the operating parameters. At the same time, the starting up time is shortened and the output efficiency is improved obviously. After optimization, the three dimensional particle-in-cell simulations show that with the beam voltage of 518 kV and beam current of 4.1 kA, 1.42 GW output microwave with efficiency about 66% can be obtained at S-band. The corresponding applied magnetic field is 0.575 T. The effects of the beam voltage and applied magnetic field on the output characteristics are also obtained.
To achieve long-term steady operation of high power microwave source, relativistic magnetron (RM) with diffraction output has been investigated for long-term operation. Experimental investigation was carried out based on an L-band relativistic magnetron with diffraction output. This paper introduces the structure of L-band RM with diffraction output as well as experimental setup and presents test method for the parameters. Operation time longer than 55 min has been obtained with output power of more than 1 GW and rep-rate of 10 Hz. The frequency is 1.57 GHz. The tube works steadily in the long-term operation and no mode competition has been observed.
A relativistic magnetron with online switchable rotation direction of a circularly polarized TE11 output mode is proposed. In the device, the same cavity magnetron is adopted as the beam-wave interaction structure, the all-cavity extraction structure is adopted as the output structure, and the Helmholtz coils system is adopted as the magnetic system. In this paper, the output mode components of the device are theoretically analyzed by the mode excitation theory of all-cavity extraction structure, and the performance of the device is investigated by Particle in Cell simulation. Simulation results show that when the applied voltage is 770 kV and the applied magnetic field that has the same direction with the output microwave is 0.2 T, the device that operates at 5π/6 mode can output a right circularly polarized TE11 mode with the mode purity of more than 99%, the operating frequency of 2.35 GHz and the output power of 3.86 GW, corresponding to the power conversion efficiencyof 55.5%. When the direction of the applied magnetic field is reversed, the rotation direction of the right circularly polarized TE11 mode can be online switched to the left in the condition of keeping other performance of the device.
The commutation phenomena produced by anode outgassing ionization and secondary electron emission are a possible factor to limit the performance of the transparent cathode relativistic magnetron with all cavity axial extraction (TCMAC). In this paper, physical modeling technology of the high-power microwave device TCMAC anode outgassing ionization phenomena and three-dimensional particle simulation are analyzed. The outgassing ionization and secondary electron emission can reduce the output power. When the number of positive ions is greater than that of electrons, the plasma caused by the ionization will make the microwave output power decline rapidly and cause system breakdown.
As the mode competition of magnetrons becomes more intense, to better study the mode of magnetron, a mode decomposition method is proposed in this paper, and the A6 relativistic magnetron is selected as the research object. The results of mode decomposition show that when the relativistic magnetron oscillates, there will be a phenomenon of multi-mode coexistence, and the magnetron will oscillate at the frequency of the eigenmode with the strongest field strength. At the same time, the results show that the degenerate mode coexists and oscillates in the same trend. Combined with PIC simulations, it can be inferred that the relativistic magnetron with radial output structure can operate at the degenerate mode and obtain stable output signal.
An L-band relativistic magnetron with cathode endcaps is presented and investigated numerically. Cathode endcaps are introduced in the relativistic magnetron to decrease the axial leakage current and enhance the power efficiency. Three-dimensional particle-in-cell simulations are carried out to investigate the effects of the cathode endcaps. The simulation results indicate that when adding cathode endcaps at both upstream and downstream of the interaction space and extending cathode beyond the anode block into the diffraction output (DO), electron leakage current is reduced from above 1 kA to 72 A and the power efficiency increases obviously. The endcaps can not only decrease the leakage current but also bring microwave reflection. Thus, the radius and the location of the endcaps have significant impact on power efficiency, and they have optimum values to make the efficiency highest. Typical optimized simulation results are as follows: working at an applied voltage of 563 kV and a magnetic field of 0.34 T, the relativistic magnetron with diffraction output (MDO) radiates microwave of 2.13 GW at 1.59 GHz, and the corresponding power conversion efficiency is 75.5%.
It is confirmed that transparent cathode technology has a significant influence on oscillation start-up process in relativistic magnetron, while the fast start-up mechanism is under further evaluation. Electromagnetic field distribution and charged particle motion were simulated in an L-band relativistic magnetron with transparent cathode. Results show a great difference between solid cathode and transparent cathode. Both electrostatic field distribution and electron spoke are modulated by transparent cathode structure. To obtain an optimal operating state, transparent cathode and relativistic magnetron should be matched in a proper configuration.
We demonstrated a novel fiber laser system with oscillating-amplifying integrated configuration and a 5 kW output power was achieved experimentally for the first time with optical-to-optical efficiency of 80%. The reliability and the boot order of the laser system were studied intensively at high power level. When the laser system is deliberately turn on in a wrong order, such as the pump of the amplifying section is switched on before the oscillating section, or the pump of the oscillating section is switched off before the amplifying section, which is normally a disaster in a classic main oscillator power amplifier (MOPA) system, the laser system can trouble-free operate at full power level. For the first time, it is verified that there is no priority between the amplifier and the seed in the oscillating-amplifying integrated fiber laser system.
The SiO2 antireflective (AR) coating was prepared by sol-gel method and then modified by plasma combined with hexamethyldisilazane (HMDS). The effects of treatment on the surface morphology, microstructure, optical properties, and laser induced damage performance are investigated to obtain the SiO2 antireflective coating with good resistance to vacuum organic contaminants. The results show that the antireflection coating shrinks, the roughness decreases and the content of polar hydroxyl group decreases after plasma combined with HMDS surface modification. Plasma and HMDS surface treatment optimizes the structure and optical properties of the film. The hydrophobic capability of the film and the resistance performance to contaminants under vacuum conditions are significantly enhanced without bad effect on the high threshold of sol-gel silica antireflection coating.
Broadband laser can effectively reduce the nonlinear effect in the process of laser plasma interaction. This paper proposes a noncollinear matching broadband third-harmonic generation based on angular dispersion, which uses the noncollinear sum frequency of the broadband fundamental wave and the narrowband second harmonic to generate broadband third harmonic, and the sum frequency process is realized by a specially designed gradient grating. The fundamental laser beams of different frequencies are incident at a specific angle, which compensates for the phase mismatch caused by the wavelength difference, so that the whole waveband meets the match condition of phase. Theoretical simulation shows that using KDP crystal type II phase matching, high efficiency broadband third-harmonic generation can be achieved by combining the broad-band fundamental wave (center wavelength 1058 nm, bandwidth 10 nm) and the second harmonic (526.5 nm).
According to the demand of high power semiconductor laser for high power, low ripple and high reliability constant current driver, a constant current power supply based on four BUCK modules in parallel is designed. The maximum output power is 37.5 kW (250 V×150 A). The conduction time interval of the four modules is T/4 in turn, which weakens the ripple by cancellation between the modules and achieves the output current ripple rate of 0.066%. Using the parallel advantage of FPGA, it can quickly respond to the output protection, and turn off the output within 12.2 μs, so the protection is excellent. The experimental results verify the design, which has been successfully applied in a project.
The traditional method of measuring temporal coherence is realized by means of mechanical scanning. However, this method can not be used for single-shot measurement, and the measurement error is larger for broadband light. In this paper, a new method is proposed for measuring temporal coherence properties through one shot. By tilting one mirror of Michelson interferometer, the delay difference of beam wavefront varying with spatial position can be obtained. The temporal coherence information of light field can be calculated and extracted from a single interferogram. The temporal coherence of different broadband light is measured in the experiment, which is in good agreement with the theoretical results. The method will provide a more convenient measurement to temporal coherence for high-power broadband laser device, and improve the measurement efficiency.
To study the starting characteristics of the variable cross-section supersonic-supersonic ejector with different structural and flow parameters, the two-dimensional Reynolds averaged Naiver-Stokes equations were employed to examine the different total pressure and total temperature ratios of primary and secondary flows in the mixing chamber of the ejector. The “start-up coefficient” was defined to test whether the supersonic-supersonic field was established. The results show that the critical total pressure ratio and total temperature ratio of ejector start-up first decreased and then increase with the increase of the contraction ratio (range 0.7−0.9). When the contraction ratio is 0.8, the optimal total pressure ratio is 5.88 and optimal total temperature ratio is 0.21. With fixed structural parameters, it is more difficult to start the supersonic-supersonic ejector when the total temperature ratio is higher. In the starting state of the supersonic-supersonic ejector, the chamber pressure is constant as the total temperature ratio and total pressure ratio vary. Also, compared with the ejector coefficient, the defined start-up coefficient can directly determine whether the supersonic-supersonic ejector has started or not, regardless of the specific working conditions.
Large-scale space simulation of high field intensity electromagnetic environment is difficult to test radiation effects, and there are gaps in the application of existing BCI (bulk current injection) technology to nonlinear system tests. The research on the equivalent alternative radiation test method of the shielded cable coupling channel is carried out. Taking the equal response of the tested equipment as the equivalent basis, an analysis model of the response of the EUT (equipment under test) under the two conditions of radiation method and injection method is established. The equivalent corresponding relationship between the injection excitation source voltage and the radiation field intensity is deduced, and the conditions and test methods for the equivalent replacement of radiation with BCI are proposed and verified by experiments. The research results show that the BCI method can be accurately equivalent to the radiation effect test of the EUT, and the test error does not exceed 2 dB, which can meet the actual needs of the project.
Starting from reducing the peak power and average power on the collector surface of a gyrotron oscillator, we have designed a magnetic field scanning system composed of 12 transverse elliptical AC coils and 2 longitudinal DC coils. According to the design parameters of the collector electrode of 140 GHz, TE28, 8 mode, 1 MW gyrotron oscillator, and combining electromagnetic field theory with particle-in-cell (PIC) simulation program, the scanning system can realize that the peak and average power density of the electron beam on the collector surface are not more than 404.91 W/cm2 and 244.01 W/cm2 respectively with a scanning length of 443.33 mm, which effectively alleviates the pressure of power dissipation and cooling of the collector.
In this work, a 2.5-dimensional discontinuous Galerkin time-domain(2.5D-DGTD) method with perfectly matched layer is proposed as a flexible tool to solve accurately electromagnetic problems in which media are homogeneous in one direction. Two numerical examples are simulated to demonstrate the advantages of the proposed method, which are the coupling between an electric dipole and optical fiber, and the analysis of dispersion characteristics of a photonic crystal fiber. The method is compare with the traditional 2.5-dimensional finite-difference time-domain method. The results show that the 2.5D-DGTD method is more realistic, especially for the simulation of curved shapes, where compared the calculation memory is reduced by 10.4%, the calculation accuracy differs by 0.011%, the calculation time is shortened, and the calculation efficiency is increased by 74.9%.
Computed tomography is a major technique for nondestructive detection of internal defects of dense materials and large-size devices. It is widely used in material science, railway, aerospace, national defense, military industry and other industries. At present, conventional high-energy industrial computed tomography system uses the X-ray source based on traditional thermionic RF electron gun, which can only provide millimeter source size, thus limiting its imaging spatial resolution. A high-energy micro-focus X-ray source is the key means to realize high-resolution high-energy industry computed tomography. As an emerging accelerator technology, the laser wakefield accelerator is a promising candidate for the micro-focus high-energy industrial computed tomography. This article reports experimental results of a micro-focus X-ray source based on laser wakefield acceleration and a computed tomography for a turbine blade. Using a 20 TW Ti: sapphire laser system, an electron beam with a charge of (140±44) pC is generated through ionization-induced injection, and then an all-optical bremsstrahlung X-ray source with an accumulated source size of 25 μm is obtained by using a 1.5 mm tungsten target. Using this source, a preliminary compressed-sensing-based computed tomography for a turbine blade is performed.
Power supply prototype in High Intensity heavy ion Accelerator Facility-Booster Ring (HIAF-BRing) adopts the scheme of full energy and fast cycle storage pulse power supply topology. Its multi modules are connected in series and parallel pattern, and the power reaches megawatt level. Due to the high power and large scale of the power supply, a module fault interlock protection system based on Programmable Logic Controller (PLC), interlock boards and Field Programmable Gate Array (FPGA) is designed and implemented to protect the power supply in operation. In this paper, first, a design of double redundant module fault interlock is introduced. Second, the logic implemented in PLC is described. Third, the work about FPGA is given. Finally, the system is tested in three aspects: the responsive time of the power supply interlock loop, the total time from controller error occurrence to interlock finish, and the equipment fault response. The result shows that the module fault interlock system can action sensitively, timely and reliably in case of fault occurrence, which meets the requirements of the power supply prototype in HIAF-BRing.
In the construction of Hefei advanced light facility (HALF), vacuum components of superconducting materials are widely used, especially superconducting radio frequency cavity. The superconducting cavity with the characteristics of high accelerating gradient, low beam impedance, high unloaded quality factor and low operating cost has become the first choice for large accelerators in the 21st century. However, electron cloud (EC) phenomenon is generated by the secondary electron emission on the surface of the superconducting cavity and cryogenic vacuum chamber. Deposition of extra dose secondary electron multiplicative power can induce thermal load increase in the cryogenic zone, superconducting cavity quench and so on. Therefore, reduction of the secondary electron emission in the superconducting radio frequency cavity is a great challenge for the design of HALF. On the basis of secondary electron yield (SEY) measurement system for room temperature materials, the authors independently designed the cryogenic sample rack to allow liquid helium to flow through the sample stage and cool sample by heat conduction. It was significant to calculate heat leakage for deciding refrigerating capacity and liquid helium consumption rate. After the system integration and debugging, the cooling performance test was carried out. The results show that cryogenic SEY measurement system was established.
Post mortem system is one of the key analysis tools for the accelerators to achieve high reliability and high availability. The precise time correlated data with high resolution that archived during the failure process is the base for the post mortem system. A software based post mortem system is designed, developed and implemented. By using this software based post mortem system in the accelerator of the China Spallation Neutron Source, most of the fast beam loss processes whose root cause could not be accurately analyzed with normal tools are well determined. This software based post mortem system is a general purpose analysis tool for most of the accelerators.
Based on conservation equation of energy, channel characteristics of the cylindrical geometry of the plasma were described under different conductivity models. The variation of the channel radius, temperature, resistance, current and dissipated energy with time is obtained. The variation of shock wave pressure at a certain distance from the center of the discharge gap is also given. The results are compared with those calculated based on the spherical geometry of the plasma channel. The purpose of this paper is to provide a reference for further study of physical and chemical characteristics and shock wave characteristics of the discharge in liquid. The results show that there is significant difference in the channel pressure and radius when the plasma channel is respectively regarded as a sphere and a cylinder, but there is little difference in other physical properties. When the physical characteristics except the shock wave characteristic are described by using three conductivity models, the change trend is almost the same, while the shock wave characteristic is described more accurately by using the conductivity model σ2. By comparing the changes of electrical parameters and pressure parameters, the applicability of the model can be selected according to the experimental data or specific research problems, which also provides a reference for further study of the physicochemical characteristics and shock wave characteristics of discharge plasma in liquid.
In the process of depressurization during severe accidents for the second generation nuclear power plant, high temperature fluid flowing through the severe accident depressurization valves may cause deformation of the valve body or even cause the valve stem to fall down because of being heated by high temperature fluid, resulting in depressurization failure. In this paper, the spectrum analysis of the high temperature fluid state that the depressurization valve may experience in the process of pressure relief is carried out, and the high temperature fluid state under different pressure relief capacities is obtained. The influence of the valve channel deformation caused by high temperature on the depressurization effect is studied. Although the second type valve has the possibility of valve channel deformation, it can obtain a longer time window for severe accident mitigation. From the perspective of optimizing the severe accident measures, the discharge capacity range of the second type valve 450-600 t/h is recommended.
The size and space of droplets produced by the droplet generator affect the radiation and evaporation characteristics of the droplet layer, and the controllability of the size and space of the droplets deserves special attention. According to Weber’s jet instability correction equation, the dimensionless wave number and disturbance frequency range of uniform droplet flow were determined. Combined with the mass conservation of jet, the relationship between the dimensionless wave number and the size and space of droplets in uniform droplet flow were analyzed. The theoretical and experimental results were compared under different nozzle diameters and jet pressures, which verified the theoretical calculation results of droplet size and spacing, and provided a basis for the study of droplet layer radiation evaporation characteristics.
To realize effective recognition of far-region nuclear electromagnetic pulse and lightning electromagnetic pulse, a recognition algorithm based on Hilbert-Huang transform (HHT) and least squares support vector machine was proposed. According to the difference in the distribution of the Hilbert spectrum resulted by applying HHT to NEMP and LEMP, the energy ratio of two different normalized frequency bands was selected as features of signals. In addition, least square support vector machine (LSSVM) was chosen as the classifier. A test using measured data was performed to verify the algorithm, which can effectively identify NEMP and LEMP data through the energy ratio in Hilbert spectrum with 98.59% recognition rate.