BEPCⅡ直线加速器相控系统的建立

为了获得高品质束流, BEPCⅡ直线加速器采用相位系统来优化和控制16个速调管的RF相位. 相控系统分布: 低相位噪声射频信号源作为直线系统的主振荡器; 加以恒温水系统控制的稳定相位参考线用以提供相控PAD装置Local输入; 基于EPICS操作界面的控制软件用以连接16个控制环路的精确测量并对加在每个速调管前级输入的IΦA移相器加以调控; 基于能量最大法相位分析测试方法的引入对系统的测试进行了优化从而提高了系统的测试精度. 整个系统的建立最终控制了直线因温度缓慢变化引起束流相对加速微波相位偏移而造成的能散度变化.     

一种双模压缩微波制备的相位锁定方案

相位锁定是双模压缩微波制备的关键问题之一.针对基于超导180°混合环的制备方案相位稳定度不高且信息处理复杂等问题,提出一种相位锁定方案.对约瑟夫森参量放大器的信号输入进行相位调制,输出的单模压缩微波与另一未调制的同频单模压缩微波在超导180°混合环内干涉,实现双模压缩微波的制备与路径分离.将未调制的单模压缩微波与一路双模压缩微波混频,解调出相位调制信号可得到两路单模压缩微波的相对相位及误差,将相位误差反馈于约瑟夫森参量放大器的抽运实现相对相位的锁定,获得稳定的双模压缩输出.本研究对高性能纠缠微波源的设计提供了理论参考.     

基于矢量叠加实现差分编码的光子模数转换方案

提出并验证了一种基于矢量叠加原理对输入射频信号进行差分编码的光子模数转换方案.该方案采用两路不同波长脉冲源通过一个相位调制器对输入射频信号进行调制,调制输出接入延时线干涉仪从而产生两路具有固定相位差的差分调制信号.利用矢量叠加模块对两路差分调制信号的强度进行不同程度的衰减和组合,从而获得系统量化所需的各路传递函数之间的相移量.该方案能够对输入射频信号实现差分编码,有效提高了系统的比特精度.同时该方案结构简单,只需要一个相位调制器,一个延时线干涉仪和一个矢量叠加模块即可构建.另外,各路传递函数之间的相移量是通过衰减两路差分调制信号的强度来实现,避免了传统光子模数转换方案中调制器对温度、环境敏感而引入的相位漂移问题.本文对4比特模数转换系统进行了原理性验证,实验结果证实了基于矢量叠加原理的光子模数转换方案的可行性.     

基于MTCA的HLS-Ⅱ直线加速器低电平系统改造

合肥光源(HLS-Ⅱ)在重大维修改造之后,其光源性能有了很大的提升。为了进一步实现连续、平稳地供光,需要对其进行恒流改造。恒流运行要求直线加速器的微波功率源有长期的稳定性与可靠性,旧的模拟低电平控制系统满足不了要求。本文基于微型电信计算平台(MTCA)设计实现了数字低电平控制系统,控制微波功率源的幅度和相位,它由以FPGA为核心的数字板卡、射频板卡、 MTCA机箱以及频率合成系统组成。该数字低电平系统工作在2 856 MHz的S波段,在线运行幅度稳定度达到0.04%,相位稳定度达到0.2°,满足恒流改造对直线加速器数字低电平系统0.25°相位抖动RMS值的相位精度要求。     

快循环同步加速器射频加速电压幅度的数字化控制

 中国散裂中子源(CSNS)快循环同步加速器（RCS）中的射频低电平控制系统是基于FPGA的全数字控制系统,旨在完成对射频频率、加速电压和同步加速相位的控制。介绍了CSNS/RCS射频系统的低电平数字化控制设计方案,并着重对射频加速电压幅度控制回路进行了分析与讨论。电压幅度控制环路通过射频电压幅度信号与电压幅度设定值的比较,得到误差信号。误差信号经过控制器来控制输入到射频腔的功率,以达到稳定和改变腔压的目的。通过对控制对象的分析和建模,得到了满足系统要求的控制器。详细介绍了数字系统的实现,尤其是信号的解调和控制算法的实现。用ALTERA公司的DSP builder进行数字控制系统开发,系统仿真结果表明,环路误差信号大约于10 μs（400个系统时钟）后归于0,整个电压幅度控制环路能稳定运行。     

一种矩形腔式高功率合成器的设计

研究了一种基于矩形腔式功率合成的射频高功率合成器。该合成器可以实现功放模块与合成器的直接耦合,合成效率高,功率容量大,且功率容量可调,可以很好地满足目前CiADS中对固态发射机功率容量的梯度要求。12合1矩形腔式功率合成器仿真结果表明,合成器各输入端到输出端的幅度传输和相位传输具有很好的一致性,最大偏差分别在0.05 dB和0.5°以内,调节功放模块数量可以调节发射机的功率容量。     

Ka波段的反射型模拟电调移相器设计

为了能对射频信号进行连续相位调制,设计了Ka波段新型小型化的模拟电调移相器,该电路基于90°分支线电桥与并联结构变容二极管实现电路大范围移相,利用四分之一波长微带线实现电路匹配及阻抗变换,通过并联补偿电阻平衡其插入损耗波动,并对其进行了详细分析与探讨。仿真及实测结果表明:该移相器在29~31GHz频段范围内,可获得180°左右的相移量,插入损耗优于6.5dB,插入损耗波动在1dB内,相移误差小于10°。     

基于微波光子的可重构变频移相信号产生

提出一种基于偏振复用-双平行马赫-曾德尔调制器(PDM-DPMZM)的微波光子变频移相信号生成方案。通过改变调制器的直流偏置电压可以实现二倍频移相信号生成或者上/下变频移相信号生成,生成信号的相位仅通过控制检偏器偏振方向与调制器一个主轴之间的角度α就能实现-180°～180°的连续调谐。在光频梳的支持下,本方案可扩展为多通道独立相位调谐的系统。仿真结果表明,频率为5 GHz的射频信号在三种功能下可以分别转换为二倍频信号10 GHz、下变频信号1 GHz和上变频信号13 GHz,它们的相位可实现-180°～180°的全范围连续调谐,且不同相位下生成信号的功率响应相对平坦。     

矢量合成移相器

描述一种可靠性、价格、能耗和体积诸方面都优于机械相位调节器的新型电控移相器，它的工作原理基于矢量合成，故定名为矢量合成移相器。并已在兰州重离子加速器相位稳定系统０号、４号、６号机柜中代替机械相位调节器，取得了令人满意的结果。     

中国散裂中子源直线射频低电平远程控制软件研制

中国散裂中子源一期工程的直线加速器,共有8套数字化射频低电平控制单元,射频低电平的本地控制属于EPICS的异构系统,无法直接与EPICS客户端进行通信。通过在射频低电平本地控制上位机程序中嵌入一个C#类型的EPICS服务器程序,实现了使用EPICS客户端对射频低电平系统的远程控制, 从而把射频低电平控制系统接入基于EPICS框架的控制系统中。直线射频低电平远程控制投入在线运行以来,运行稳定可靠。     

FPGA-based amplitude and phase detection in DLLRF

The new generation particle accelerator requires a highly stable radio frequency (RF) system. The stability of the RF system is realized by the Low Level RF (LLRF) subsystem which controls the amplitude and phase of the RF signal. The detection of the RF signal's amplitude and phase is fundamental to LLRF controls. High-speed ADC (Analog to Digital Converter), DAC (Digital to Analog Converter) and FPGA (Field Programmable Gate Array) play very important roles in digital LLRF control systems. This paper describes the implementation of real-time amplitude and phase detection based of the FPGA with an analysis of the main factors that affect the detection accuracy such as jitter, algorithm's defects and non-linearity of devices, which is helpful for future work on high precision detection and control.     

kW级功率驱动的锁相高功率微波发生器

针对kW级微波驱动的锁相GW高功率微波,设计了一个高增益(大于50 dB)四腔相对论速调管放大器(RKA)。模拟表明,在此条件下高次模振荡严重影响器件的锁相实现。由此,将RKA结构与正反馈振荡电路结合起来,建立相应的等效电路来研究这种高次模激励的物理过程(即高次模的激励与中间腔之间耦合强度的相关性)。在高次模振荡的等效电路(即正反馈振荡电路)中,用衰减电阻代替结构中的微波吸收层来研究高次模振荡的抑制机理,衰减电阻通过对反馈过程的控制,提高了电路的自激振荡起振电流。在结构上按照衰减电阻要求设计了微波吸收层,将高次模振荡的起振电流提高到大于器件的工作电流,实现了高增益(约60 dB)条件下高次模激励的抑制。模拟获得了4 kW微波功率驱动的2.3 GW锁相高功率微波,增益接近60 dB。在LTD加速器平台的实验结果表明:注入微波由固态RF种子源提供(功率10 kW),输出功率达到1.8 GW,增益为52.6 dB,90 ns内输入和输出微波的相对相位差小于±10°,实验上实现了kW级注入微波对GW高功率微波的相位锁定。     

FPGA-based amplitude and phase detection in DLLRF

The new generation particle accelerator requires a highly stable radio frequency（RF） system. The stability of the RF system is realized by the Low Level RF（LLRF） subsystem which controls the amplitude and phase of the RF signal. The detection of the RF signal＇s amplitude and phase is fundamental to LLRF controls. High-speed ADC（Analog to Digital Converter） ,DAC（Digital to Analog Converter） and FPGA（Field Programmable Gate Array） play very important roles in digital LLRF control systems. This paper describes the implementation of real-time amplitude and phase detection based of the FPGA with an analysis of the main factors that affect the detection accuracy such as jitter,algorithm＇s defects and non-linearity of devices,which is helpful for future work on high precision detection and control.     

花瓣形辐照加速器NB100的高频系统设计

介绍了花瓣形辐照加速器NB100高频系统整体方案和设计过程,如输入组件采用磁耦合设计、功率源与加速腔间采用同轴线连接、隔离窗选用平板窗等;给出了耦合环的最终设计数据与图形。高频系统投入使用后,通过初步测试,得到了驻波比小于1.1情况下,入射功率连续100 kW、脉冲250 kW的结果,证实了高频系统设计的合理性。     

Some aspects in accelerator structure studies at SLAC

Recent progress in the accelerator structure studies at SLAC is reported. This paper covers the room temperature accelerator structures for the ILC e⁺/e^－ sources; RF structures for some photon science projects including RF deflectors and the LCLS RF gun; the high gradient accelerator R &; D in a global CLIC collaboration for the future multi-TeV linear colliders.     

汤姆逊硬X射线源中锁模脉冲激光稳频方法

 汤姆逊硬X射线源中，光阴极微波电子枪要求触发激光脉冲与腔中微波场相位的精确时间同步。理论分析了锁模脉冲激光器相位噪声，探讨了对激光脉冲相位探测和控制的方法，并通过实验构建了激光相位抖动测量系统和锁相环稳频回路，用基频鉴相信号驱动压电陶瓷晶体改变谐振腔长，实现了从开环3.42 ps到闭环1.46 ps的均方根相位抖动的测量和控制。     

Impact of the mobile terminal scheme on millimeter-wave radio over fiber systems based on photonic heterodyning techniques

The use of communication networks relying on millimeter-wave (mm-wave) wireless links promises a great capacity enhancement as well as improved security. However, given the high-directivity of mm-wave links, coverage requirements are difficult to meet unless the network is assisted by an infrastructure. Given its low-cost, power-efficiency, and high capacity, radio over fiber has emerged as a strong candidate for the implementation of such infrastructure. Among the different generation techniques, photonic heterodyning has attracted considerable attention due to its capacity to generate radio frequency (RF) signals in the entire microwave/mm-wave range without requiring broadband electro-optical modulator. However, the RF signals generated using these techniques suffer from significant phase noise, a major impairment that degrades the system performance. In this paper we study two approaches to overcome this limitation: (1) the use of optical sideband injection locking (OSBIL) to generate tones with highly correlated phase noise and (2) heterodyning independent lasers in combination with a mobile terminal (MT) that is insensitive to the phase of the RF signal. A qualitative comparison between the two techniques in terms of MT sensitivity to the RF phase noise, the power sensitivity, and base station power efficiency leads to the conclusion that OSBIL is more suitable for networks with medium-range node separation, whereas heterodyning of independent laser with phase-insensitive MT is a cost-efficient solution for networks where nodes are closer to each other.     

Collective effects for long bunches in dual harmonic RF systems

The storage of long bunches for large time intervals needs flattened stationary buckets with a large bucket height. Collective effects from the space charge and resistive impedance are studied by looking at the incoherent particle motion for the matched and mismatched bunches. Increasing the RF amplitude with particle number provides r.m.s wise matching for modest intensities. The incoherent motion of large amplitude particles depends on the details of the RF system. The resulting debunching process is a combination of the too small full RF acceptance together with the mismatch, enhanced by the collective effects. Irregular single particle motion is not associated with the coherent dipole instability. For the stationary phase space distribution of the Hofmann-Pedersen approach and for the dual harmonic RF system, stability limits are presented, which are too low if using realistic input distributions. For single and dual harmonic RF system with $d$=0.31, the tracking results are shown for intensities, by a factor of 3 above the threshold values. Small resistive impedances lead to coherent oscillations around the equilibrium phase value, as energy loss by resistive impedance is compensated by the energy gain of the RF system.     

Collective effects for long bunches in dual harmonic RF systems

The storage of long bunches for large time intervals needs flattened stationary buckets with a large bucket height.Collective effects from the space charge and resistive impedance are studied by looking at the incoherent particle motion for the matched and mismatched bunches.Increasing the RF amplitude with particle number provides r.m.s wise matching for modest intensities.The incoherent motion of large amplitude particles depends on the details of the RF system.The resulting debunching process is a combination of the too small full RF acceptance together with the mismatch,enhanced by the collective effects.Irregular single particle motion is not associated with the coherent dipole instability.For the stationary phase space distribution of the Hofmann-Pedersen approach and for the dual harmonic RF system,stability limits are presented,which are too low if using realistic input distributions.For single and dual harmonic RF system with d＝0.31,the tracking results are shown for intensities,by a factor of 3 above the threshold values.Small resistive impedances lead to coherent oscillations around the equilibrium phase value,as energy loss by resistive impedance is compensated by the energy gain of the RF system.     

Noisy fringe pattern demodulation by an iterative phase locked loop

The phase locked loop (PLL) technique applied to demodulate two-dimensional carrier-frequency fringe patterns has been developed recently. Here we present an extension to the basic PLL scheme to demodulate noisy fringe patterns. This modified technique estimates the phase in the fringe pattern iteratively; that is, the first wavefront estimation is done using a flat reference phase and the second iteration takes the demodulated phase found in the first iteration as the new reference. The third demodulating iteration uses the second phase estimation as the reference and so on, until further changes in the detected wavefront fall below a predefined threshold. During the iterative process the bandwidth of the iterative PLL system is gradually decreased to improve the signal-to-noise ratio of the detected phase as well as to resolve noise-generated phase inconsistencies.