Digital Low Level RF System Research Articles

Development of a digital low-level RF system for the STCF

With the development of particle physics, higher requirements are put forward for the brightness and energy of the collider. A new collider — the Super Tau-Charm Facility (STCF) — is under study in China. To meet the high acquirement of the quality of the beam, a digital low-level RF (LLRF) system that achieves high stability plays a critical role as a part of the microwave system of Linacs. This paper introduces the design of a digital LLRF system that works in the S-band (2856 MHz) as required by Linacs of the STCF. The system architecture and hardware design as well as the control algorithms and software design are discussed and tested. The system includes a signal source, a frequency synthesis system and an IF signal processor. The phase noise of the signal source is as low as 20.3 fs, and the low-level IF signal processor adopts amplitude and phase control by a PI-type controller. They jointly realize ultra-low phase noise signal output. The table experiments show that the short-term amplitude and phase stability of the system is better than 0.01% and 0.01°, respectively. The beam experiments show that the closed-loop phase stability in the cavity is about 0.1°. The results prove that the LLRF system meets the requirement of the STCF.

Tuner Loop Based on FPGA for Petra Cavity at TPS Booster Ring

The low-level radio frequency system of the Taiwan Photon Source, TPS, booster ring was replaced by the digital low-level RF system, DLLRF, in 2018. After that, the phase drift compensation loop for energy-saving operation and the tuner loop were also implemented in the DLLRF system sequentially. We used Altera-DE3 to build the core of DLLRF and handle the high-speed ADC/DAC procedure for RF signal sampling. As we con-sider the response time of the tuner, we choose an-other low-cost board, Altera-DE0-Nano, to develop the tuner loop for 5-Cell-Petra-Cavity. It has an eight-channel 12-bits-ADC, 150 ksps, ADC128S022, to detect the positions of two tuners and two transmitted powers sampled from two cells of the 5-Cell-Petra-Cavity for a power balance function. The phase information of for-ward power and cavity gap voltage will come from the Altera-DE3 to tell the tuner loop in the Altera-DE0-Nano whether the cavity is on resonance or not. The tuner loop controls the cavity to work not only at the resonance frequency but also with a balanced electric field distribution. This study presents the tuner loop’s architecture, including the locking resonance frequency and field balance functions. The performance of the field balance function is observed by the archive data of the two tuners positions and two transmit power signals.

Digital low-level RF system for the infrared free-electron laser facility

The infrared free-electron laser facility (IR-FEL) is a new free-electron oscillator user facility in the University of Science and Technology of China (USTC), which has come into user operation in September 2021. To get a low energy spread electron beam, we designed and fabricated a digital low-level radio frequency system (DLLRF) based on the MTCA.4 platform to monitor and control the microwave field. The simulation for RF system is implemented to estimate the feedback coefficients and analyze the influence of driving chain noise and measurement noise in this case. Beam tests show that the pulse-to-pulse PI feedback eliminates the slow drift effectively and improves the stability of the phase from 3.4° to 0.19° (RMS). The application of the PI-type feed-forward decreases intra-pulse ripples of phase from 5° to 1° (peak to peak).

Superconducting radio-frequency cavity fault classification using machine learning at Jefferson Laboratory

We report on the development of machine learning models for classifying C100 superconducting radio-frequency (SRF) cavity faults in the Continuous Electron Beam Accelerator Facility (CEBAF) at Jefferson Lab. CEBAF is a continuous-wave recirculating linac utilizing 418 SRF cavities to accelerate electrons up to 12 GeV through 5-passes. Of these, 96 cavities (12 cryomodules) are designed with a digital low-level RF system configured such that a cavity fault triggers waveform recordings of 17 RF signals for each of the 8 cavities in the cryomodule. Subject matter experts (SME) are able to analyze the collected time-series data and identify which of the eight cavities faulted first and classify the type of fault. This information is used to find trends and strategically deploy mitigations to problematic cryomodules. However manually labeling the data is laborious and time-consuming. By leveraging machine learning, near real-time (rather than post-mortem) identification of the offending cavity and classification of the fault type has been implemented. We discuss performance of the ML models during a recent physics run. Results show the cavity identification and fault classification models have accuracies of 84.9% and 78.2%, respectively.

Upgrade of low level RF system based on Micro Telecom Computing Architecture (MTCA) for HLS-II LINAC

The performance of Hefei Light Source II (HLS-II) has improved a lot after major maintenance and reconstruction. To further provide continuous and stable light, the RF system of the HLS-II LINAC needs to be upgraded for top-off mode. It is required that the RF power source have long-time stability and reliability, but the old analog low level RF system(LLRF) can’t meet the requirement. Hence a digital low level RF control system based on Micro Telecom Computing Architecture(MTCA) is designed and implemented to control the amplitude and phase of the RF power source. This system is composed of digital board cards based on FPGA, RF board cards, MTCA chassis and a frequency synthesis system. It works at 2856 MHz of S band, with phase and amplitude stability up to 0.2° and 0.04% respectively, which meets the top-off mode requirement of 0.25° RMS phase jitter of the digital low-level RF system in the HLS-II LINAC.

Upgrade of CERN’s PSB Digital Low-Level RF System

Upgrade of CERN’s PSB Digital Low-Level RF System

Novel auto-startup technology for two cavities of a medical accelerator with one RF source

At Ibaraki Neutron Medical Research Center, an 8-MeV proton linear accelerator is being developed for iBNCT (Ibaraki - Boron Neutron Capture Therapy), with the aim of achieving the highest possible performance at a low cost. In this accelerator, there are two RF cavities driven by only one high-power klystron as the RF source. A digital low-level RF system has been successfully implemented for the iBNCT accelerator, with very good RF field regulation. An effective multipurpose and smart tuner controller was also developed. Finally, a novel auto-startup technology for two cavities with one RF source was successfully developed. The time required for the RF startup process is greatly reduced from the conventional time of 30 min, to less than 1 min. The good stability and fast response achieved by using this technology make it suitable for the medical use of iBNCT accelerators. This technology will be extremely useful in the development of similar accelerators.

A 166.6 MHz superconducting rf system for the HEPS storage ring

A superconducting 166.6 MHz quarter-wave beta=1 cavity was recently proposed for the High Energy Photon Source (HEPS), a 6 GeV kilometer-scale light source. Four 166.6 MHz cavities will be used for main acceleration in the newly planned on-axis beam injection scheme realized by a double-frequency RF system. The fundamental frequency, 166.6 MHz, was dictated by the fast injection kicker technology and the preference of using 499.8 MHz SC RF cavity as the third harmonic. Each 166.6 MHz cavity will be operated at 4.2 K providing 1.2 MV accelerating voltage and 150 kW of power to the electron beam. The input coupler will use single-window coaxial type graded up to 200 kW CW power. Each cavity will be equipped with a 200 kW solid-state amplifier and digital low-level RF system. This paper describes the 166.6 MHz RF system with a focus on the design and optimization of the RF cavity and its ancillaries, the LLRF system and the status of the solid-state amplifiers.

New digital low-level rf system for heavy-ion synchrotrons

In the scope of the Facility for Antiproton and Ion Research (FAIR) project, several new synchrotrons and storage rings will be built. The existing heavy-ion synchrotron SIS18 has to be upgraded to serve as an injector for the FAIR accelerators. All this imposes new requirements on the low-level rf (LLRF) systems. These requirements include fast ramping modes, arbitrary ion species, and complex beam manipulations such as dual-harmonic operation, bunch merging/splitting, barrier bucket operation, or bunch compression. In order to fulfill these tasks, a completely new and unique system architecture has been developed since 2002, and the system is now used in SIS18 operation. The presentation of this novel system architecture is the purpose of this paper. We first describe the requirements and the design of the LLRF system. Afterwards, some key components and key interfaces of the system are summarized followed by a discussion of technological aspects. Finally, we present some beam experiment results that were obtained using the new LLRF system.

Design approaches for the X band LLRF system

Abstract The low-level RF (LLRF) system regulates disturbances over a limited bandwidth in accordance with its capabilities and the RF loop parameters. The disturbances usually originate in the RF system or can be coupled to the RF system from the environment. In this paper a general overview of the possible design approaches for a digital LLRF system operating in X band is presented. Firstly, the possible design approaches of the RF front/back ends are presented and reviewed. We also define the main design parameters for the RF front/back ends. Parameters like isolation between channels, noise, gain, linearity and number of IF stages are put into the perspective of machines using RF components in the X band. An important part of the LLRF system is the local RF timing generation and distribution, which is also treated in the paper. In the second part of the paper the main design approaches in the digital signal processing part of the LLRF system are presented. The emphasis is on the algorithms that are machine specific. Some standard processing algorithms like adaptive feed-forward and arbitrary shaping of feed-forward pulses are presented. Finally, a suggestion for the X band LLRF design is given.