Radio Frequency Algorithms

Mastoridis, Themistoklis · 2025

Read Radio Frequency Algorithms by Mastoridis, Themistoklis (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/doe-osti-2998338-radio-frequency-algorithms.txt.

Introduction

California Polytechnic State University final technical report on award DE-SC0019287, by Themistoklis Mastoridis, dated 14 October 2025, covers radio-frequency control algorithms for CERN's Super Proton Synchrotron. The SPS project sought the best settings for the upgraded low-level radio-frequency system, balancing impedance reduction against hardware limits and noise. Beam quality estimates for the upgraded feedback systems were a second goal.

Models of the system were built in MATLAB and Simulink and validated against measurements at the SPS itself. Optimal configurations for maximum impedance reduction were determined for the present hardware. Students carried much of the work and co-authored two CERN technical notes, two Brookhaven technical notes, and two peer-reviewed papers. The report closes an Office of Science training-through-research award with its deliverables listed.

Catalog record

CREATORMastoridis, Themistoklis
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE21 KB
LICENSEPublic Domain
RIGHTSPUBLIC DOMAIN
SOURCEDOE OSTI

Source: https://www.osti.gov/biblio/2998338

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Final Technical Report Cover Page Recipient Organization: Cal Poly Corporation Project Title: Radio Frequency Algorithms Date of Report: 10/14/2025 Award Number DE- SC0019287 Sponsoring program o?ice: OGice of Science Project Director: Themis Mastoridis SPS The SPS project goal was to develop schemes for optimal configuration and operation of the upgraded SPS LLRF system considering maximum impedance reduction, hardware limitations, and noise issues. Additionally, we wanted to estimate the reduction of RF station impedance and transient beam loading eGects in the presence of the updated feedback systems, thus providing an estimate of the expected beam quality. Models of the SPS LLRF system have been developed in MATLAB and Simulink. These models were validated with measurements at the SPS. Optimal configurations for maximum impedance reduction have been determined for the present system. Multiple cavity filters were studied. In addition, future possible upgrades have been investigated including coupling the feedback systems of multiple cavities, thus creating a Multiple Input-Multiple Output (MIMO) control system. To optimize the coupling terms, a Linear-Quadratic Regulator (LQR) in conjunction with a Kalman filter was developed and their performance was compared to more traditional feedback system performance. An optimal design was proposed for the SPS LLRF. The performance of this system with HL-LHC beam intensities was estimated. A CERN technical note was published in 2020 presenting the models and algorithms used to determine the optimal LLRF design [9]. A peer-reviewed article presented an algorithm to optimally configure the Feedforward system in the SPS [14]. It also described models for the Feedback and Feedforward systems, as well as a simulation tool to optimize the system design using those models. The simulation was used to estimate the performance of the proposed design. The system was deployed, and its performance was presented in a conference paper [31]. This work has helped optimize the performance of the SPS, and as such, provide more protons and higher quality particle beams to the LHC, thus expanding its discovery potential. EIC The US EIC will be built at Brookhaven National Laboratory. The design includes an electron and an ion ring. Because of the dissimilarity of the two rings, the RF transients created by the clearing gaps will be very diGerent in the two rings and will impact the luminosity, will limit the RF power, and will aGect the performance of the RF loops. This makes the problem of managing the eGects of the gap transients crucial to the operation of the EIC. One of the project’s objectives was to develop LLRF algorithms to match the gap transients in the two dissimilar EIC rings, by applying a feed-forward algorithm to the RF reference. These algorithms should automatically adjust to beam intensity and pattern. A collaboration was initially established with Thomas JeGerson National Accelerator Facility (JLab). A LLRF design was proposed and modeled in Simulink. This model was then used to study possible mitigations techniques for the asymmetric gap transients in the EIC. Preliminary results were presented at the EIC collaboration meeting [5], at IPAC 2019 [3], and NAPAC 2019 [4]. The completed studies were included in the JLEIC pre-CDR and were published in Physical Review Accelerators and Beams [2]. Since the DOE selected BNL for the EIC site, the simulations and models were adapted appropriately to assist with design studies. A collaboration was established with Brookhaven National Lab (BNL) in 2020. Simulations were performed to estimate transient beam loading eGects and power limitations. Both the electron and hadron rings (all five RF systems) were studied. A Reversed Phasing RF scheme for the electron ring was investigated as well: some of the fundamental cavities are detuned in opposite ways, so that all cavities provide acceleration but some of the cavities provide focusing voltage and the rest provide defocusing voltage. This achieves the power per cavity as the normal scheme but reduces the detuning of individual cavities and reduces transient beam loading. The results have been included in the EIC Conceptual Design Report [13], as well as in the EIC Task Force for RF beam physics and critical component R&D – FY’20 Final Report. They have also been presented in two talks at the EIC Collaboration Meeting [6] and the APS 2020 meeting [7]. BNL technical notes are being prepared for publication summarizing these results. In addition, the collaboration has been expanded to include studies and simulations of the Crab Cavity LLRF. Of particular importance are RF noise eGects on transverse emittance, as well as the design of a LLRF system that would not only regulate the crabbing/uncrabbing stations independently but would also keep the crabbing and uncrabbing total voltage as close to zero as possible to avoid residual crabbing outside of the interaction region. A BNL technical note and a conference paper (with Cal Poly undergraduate students as co-authors) was published in February 2022 presenting estimated of the transverse emittance growth due to RF noise in the EIC, as well as proposing a mitigating feedback system and estimating its performance [15, 33]. The specifications for the required pickup were presented in [28]. The results were also presented in BNL talks in May [21] and July 2021 [20]. The crab cavity impedance eGect on transverse instabilities was studied. A model was developed and estimated presented in a peer-reviewed publication with a student co-author [24]. A simulation was developed to study the crab cavity beam interaction. Cal Poly students participated in the model development and were co-authors in the resulting BNL technical note [26]. The studies have been used to set requirements for the crab cavity LLRF [32]. Furthermore, a data-driven technique was developed to characterize the inter-bunch and intra-bunch beam dynamics in particle accelerators, with an emphasis on the EIC. This technique was presented at the 10th International Beam Instrumentation Conference [16]. CLIC The Compact Linear Collider (CLIC), is a proposed linear collider at CERN. CLIC will explore the energy frontier. It is the only mature design of a linear accelerator in the multi-TeV range. It will collide electrons and positrons and it aims to discover physics beyond the Standard Model. The CLIC damping rings need to generate ultra-low emittance bunches to achieve high luminosity. Therefore, big energy loss wigglers are required, and a high total voltage RF system is needed to compensate these losses. The resulting strong beam- loading transients aGect the bunch position and length. In addition, the RMS bunch position has to be regulated within 0.2 degrees at 2 GHz (280 fs) and the RMS bunch length should be less than 2% along the bunch train at extraction. These conflicting specifications lead to a challenging design for the LLRF system. Models and simulations of the proposed LLRF architecture were developed and the system performance was estimated in terms of the rms bunch position and the klystron power requirements. The completed studies were published in Nuclear Instruments and Methods in Physics Research Section A [8]. Two talks presented some of the results in September 2020 [11] and February 2021 [12]. A novel cavity design with ultra-low R/Q was investigated next and the promising results were published as a CERN note in December 2021 [17] and presented at a CERN talk in September 2021 [19]. Finally, we further optimized the cavity design, studied alternative repetition rates, and a

Read the complete text