Obvious and Non-Obvious Aspects of Digital Self-Excited-Loops for SRF Cavity Control

Basic Information

• Paper ID: 2510.14835
• Title: Obvious and non-obvious aspects of digital Self-Excited-Loops for SRF cavity control
• Authors: Larry Doolittle (LBNL), Shreeharshini Murthy (LBNL), Matei Guran (FNAL), Lennon Reyes (FNAL), Shrividhyaa Sankar Raman (FNAL), Philip Varghese (FNAL)
• Classification: physics.acc-ph (Accelerator Physics)
• Publication Date: October 16, 2025
• Paper Link: https://arxiv.org/abs/2510.14835v1

Abstract

This paper reviews the application of Self-Excited Loops (SEL) in superconducting radiofrequency (SRF) cavity control. Delayen first demonstrated the advantages of SEL in controlling narrowband SRF cavities in 1978, with the key capability of establishing closed-loop amplitude control early in the setup process, stabilizing Lorentz forces to allow cavity tuning and phase loop configuration in a stable environment. With the development of modern FPGA DSP technology, this fundamental concept has been implemented globally with multiple variations and operational scenarios, which have both obvious and non-obvious effects on latency, feedback stability, and resilience.

Research Background and Motivation

1. Problem to be Addressed: Amplitude and phase stability issues in superconducting radiofrequency cavity control systems, particularly instabilities caused by Lorentz forces during cavity tuning.
2. Problem Significance:
   • The metallic structure of SRF cavities deforms due to acoustic environment and Lorentz forces, causing real-time changes in the imaginary part of detuning
   • Traditional control methods struggle to maintain stability during the tuning process
   • Precise amplitude and phase control must be maintained during beam operation
3. Limitations of Existing Methods:
   • Adaptability issues of traditional analog SEL circuits in modern digital environments
   • Lack of unified standards among different laboratory implementations
   • Insufficiently understood issues regarding latency, stability, and resilience
4. Research Motivation: Systematically analyze and compare various digital SEL implementation schemes to provide theoretical foundations and practical guidance for SRF cavity control system design.

Core Contributions

1. Theoretical Analysis: Provides critical characteristic analysis of SEL systems in open-loop, amplitude-stabilized, and phase-stabilized modes
2. Technical Comparison: Systematically compares original analog circuits with CORDIC-based digital implementation schemes
3. Topology Summary: Organizes digital SEL implementation topologies from LBNL, JLab, S-DALINAC, BARC and other laboratories
4. Stability Analysis: Provides theoretical framework and experimental verification for system stability
5. Practical Guidance: Offers specific recommendations for off-resonance operation and phase detector design

Detailed Methodology

Task Definition

Design and implement a digitalized self-excited loop system for controlling the amplitude and phase of superconducting radiofrequency cavities, ensuring stability and robustness under various operating conditions.

Theoretical Foundation

The state-space representation of cavity voltage is:

dV/dt = aV + bγ + cI_b

where V, γ, I_b, a, b, c are all complex numbers, varying slowly relative to the RF resonance timescale.

Delayen's key contribution was to derive from the standard Lorentz equation:

V/γ = 1/(1 + jψ)

where ψ = ℑ(a)/ℜ(a), the required driving signal:

γ = V(1 + jψ)

CORDIC Technology Core

The CORDIC structure produces:

z_out = A · z_in ∏(i=0 to n-1) exp(jσ_i θ_i)
θ_out = θ_in - Σ(i=0 to n-1) σ_i θ_i

where:

• θ_i = tan^(-1) 2^(-i)
• A = ∏(i=0 to n-1) √(1 + 2^(-2i)) ≈ 1.64676

Two standard modes:

1. Rotation Mode: θ_out ≈ 0, used to obtain z_out = A · z_in · exp(jθ_in)
2. Vector Mode: ℑ(z_out) ≈ 0, used to obtain θ_out = ∠z_in and ℜ(z_out) = A · |z_in|

PI Controller Design

Implemented with zero integral saturation, including:

• Proportional gain term KP
• Integral gain term KI
• Configurable saturation limits
• Zero integral saturation to prevent integral windup

Off-Resonance Operation Handling

Employs a state machine to handle phase detector issues during off-resonance operation:

• Register the condition when actual phase error transitions from positive to negative values
• Phase detector output remains at maximum positive saturation
• Until frequency error is eliminated by the tuner control loop

Digital SEL Topology Structures

LBNL Topology (2015)

• Uses two cascaded CORDICs: one for vector mode detection, one for rotation mode generation
• Independent amplitude and phase PI controllers
• Phase offset compensation

JLab Topology (2008)

• Introduces serialization of amplitude and phase feedback paths
• Uses complex multiplier
• Higher latency but relatively simple structure

Improved Proposed Topology

• Re-orthogonalize amplitude and phase loops
• Neither PI output routed through second CORDIC
• Reduced overall latency

S-DALINAC Topology (2011)

• Dedicated amplitude ADC
• P-term control only
• I-term routed to tuner loop

BARC Topology (2014)

• Does not use CORDIC blocks
• Uses squaring circuits and local feedback loops to construct limiters
• Lower phase feedback latency
• Higher multiplier resource requirements

Experimental Setup

Stability Analysis

• Cavity bandwidth: 20 Hz
• Zero-dB crossover: 20 kHz
• Controller zero: 31.4 kHz
• Latency: 2.39 μs

Test Conditions

• Detuning range: -200 Hz to +200 Hz
• Reactance limit: 0.6
• Half-bandwidth: 20 Hz

Experimental Results

Stability Verification

Experiments demonstrate that when the condition ℑ(a) << ω_0dB is satisfied, system stability is not significantly affected. Nyquist stability plots show open-loop and closed-loop gain characteristics under different detuning conditions.

Actual Measurement Results

• Phase-locked SEL limiting on Q component operates as expected
• Measured waveforms show phase and amplitude response of cavity and forward signals
• Validates theoretically predicted system behavior

Simulation Results

Idealized controller simulation waveforms display:

• Normalized detuning frequency response
• Phase tracking of cavity and forward signals
• Dynamic behavior of reactance and total forward amplitude

Related Work

The paper cites key literature in SEL technology development:

1. Delayen 1978: Original SEL theory and phase-amplitude stabilization of superconducting resonators
2. Volder 1959: Invention of CORDIC computation technique
3. AD9901 (1996): Reference for phase detector design
4. Specific implementations from various laboratories: JLab (2008), S-DALINAC (2011), BARC (2014), LBNL (2015)

Conclusions and Discussion

Main Conclusions

1. Digital SEL implementations preserve the core characteristics of Delayen's 1978 design: detecting phase error and using feedback controllers to generate driving signals of the form 1+jψ
2. Different topology structures have respective advantages and disadvantages in latency, resource usage, and stability
3. Phase loop limiting is an important component of the SEL process
4. Appropriate phase resolver is critical for off-resonance operation

Limitations

The paper explicitly identifies unconsidered aspects:

• Units and calibration of cavity state values
• System identification and drift correction
• Analog and digital up/down conversion
• Mitigation of instabilities caused by high P gain and adjacent cavity passband modes
• Transmit channel linearization
• Timing-based feedforward beam load correction

Future Directions

1. Latency Reduction: Use higher DSP clock speeds, add direct P-term bypass around CORDIC steps
2. Leader/Follower CORDIC Technique: Reduce cascaded dual-CORDIC latency by half
3. Code Sharing: Promote mixed-mode testing platforms and implementation scheme compatibility among global laboratories
4. Standardization: Overcome licensing and different programming language barriers

In-Depth Evaluation

Strengths

1. Strong Systematicity: Comprehensively compares multiple digital SEL implementation schemes, providing a unified analytical framework
2. Solid Theoretical Foundation: Derives from physical principles, combining mathematical analysis with engineering practice
3. High Practical Value: Provides specific technical guidance for SRF cavity control system design
4. Excellent Visualization: Uses clear block diagrams and symbol systems, facilitating understanding of complex digital signal processing workflows

Weaknesses

1. Limited Experimental Data: Primarily based on simulations and limited actual measurements, lacking large-scale system validation
2. Insufficient Performance Comparison: Quantitative performance comparisons among various topology structures lack sufficient detail
3. Inadequate Optimization Guidance: Lacks clear guidance on selecting the most suitable topology for specific applications

Impact

1. Field Contribution: Provides important technical summary and analysis for accelerator physics and RF control fields
2. Practical Value: Directly guides LLRF system design for major facilities such as LCLS-II
3. Standardization Promotion: Facilitates standardization and best practices for digital SEL implementations

Applicable Scenarios

• Low-level radiofrequency control systems for superconducting accelerators
• Precision control of high-Q narrowband resonators
• RF systems requiring rapid amplitude stabilization
• FPGA-based digital signal processing applications

References

The paper cites 12 key references, covering the complete technical development trajectory from SEL theoretical foundations to specific implementations at various laboratories, providing readers with pathways for in-depth research. Particularly noteworthy are Delayen's pioneering work and original CORDIC literature, as well as technical reports from major laboratories.