Semi-analytical algorithms to study longitudinal beam instabilities in double rf systems

Double rf systems are critical for achieving the parameters of 4th-generation light sources. These systems, comprising both main and harmonic rf cavities, relax statistical collective effects but also introduce instabilities, such as Robinson and periodic transient beam loading (PTBL) instabilities. In this paper, we provide semi-analytical algorithms designed to predict and analyze these instabilities with improved accuracy and robustness. The algorithms leverage recent advancements in the field, offering a computationally efficient and accurate complement to multibunch tracking simulations. Using the SOLEIL II project as a case study, we demonstrate how these algorithms can optimize rf cavity parameters in high-dimensional parameter spaces, thereby maximizing the Touschek lifetime. An open-source Python package, ALBuMS (Algorithms for Longitudinal Multibunch Beam Stability), is provided as an accessible tool for double rf system stability analysis.

💡 Research Summary

The paper presents a set of semi‑analytical algorithms for predicting and analysing longitudinal beam instabilities that arise in double‑rf systems, i.e., storage‑ring light sources equipped with both a main accelerating cavity (MC) and a harmonic cavity (HC). While double‑rf operation is now a de‑facto standard for fourth‑generation synchrotron light sources because it lengthens the bunch, reduces the Touschek effect and mitigates intra‑beam scattering, the presence of a harmonic cavity also introduces new collective‑beam instabilities. The most relevant are Robinson‑type coupled‑bunch instabilities (ℓ = 0), the periodic transient beam‑loading (PTBL) instability (ℓ = ±1), and conventional coupled‑bunch instabilities driven by higher‑order modes (HOMs).

Historically, the community has relied on multi‑bunch macro‑particle tracking codes (e.g., mbtrack, mbtrack2) to evaluate these effects. Although accurate, such simulations are computationally heavy, often requiring days on a cluster for a single set of rf parameters, and they provide limited physical insight without complementary analytical models. The classic Bosch algorithm (1995) offered a faster, semi‑analytical alternative, but it suffers from several limitations: convergence strongly depends on the integration limits used for the bunch‑length calculation; it assumes symmetric Gaussian‑like bunches and a quadratic rf potential; it treats Landau damping through a simple proportionality to the synchrotron‑frequency spread; it cannot handle active harmonic cavities or non‑optimal detuning; and it does not model PTBL correctly.

The authors therefore develop a modernised algorithm that addresses these shortcomings. The key innovations are:

1. Full longitudinal equilibrium via coupled Haïssinski equations – instead of using scalar form factors and a Taylor‑expanded potential, the algorithm solves a system of coupled Haïssinski equations for the MC and HC simultaneously. This yields the true longitudinal distribution, allowing asymmetric bunch shapes, double‑humped profiles (ξ > 1), and arbitrary impedance contributions (both short‑range broadband and long‑range wakefields).

2. Direct extraction of statistical moments – from the equilibrium distribution the algorithm computes the standard deviation (bunch length), complex form factors, and higher‑order moments. These quantities replace the heuristic scalar form factors used in the Bosch method.

3. Rigorous stability assessment – the growth rates of Robinson, fast‑mode‑coupling, and HOM‑driven coupled‑bunch instabilities are obtained from the linearised Vlasov equation using the actual dipole coherent frequency ω_R derived from the equilibrium distribution. Landau damping is no longer approximated by a constant factor; instead a stability diagram is constructed and the complex coherent frequency is checked against the Landau contour, ensuring that damping is only claimed when the incoherent synchrotron‑frequency spread truly overlaps the mode.

4. Explicit PTBL modelling – two complementary PTBL models are implemented. The first follows the reactive‑impedance‑dominant picture, evaluating the low‑frequency growth rate for ℓ = ±1 modes. The second incorporates recent theoretical work showing that PTBL can involve multiple azimuthal (m) and coupled‑bunch (ℓ) modes simultaneously; the algorithm therefore evaluates a matrix of coupled equations to capture possible mode‑mixing effects.

5. Improved convergence strategy – integration limits for the bunch‑length calculation are updated adaptively, and form‑factor updates use a weighted averaging scheme that guarantees monotonic convergence even for highly non‑linear potentials.

The modified workflow (illustrated in the right‑hand side of Fig. 1) proceeds as follows: user supplies MC and HC voltages, phases and detuning (or the harmonic‑to‑main voltage ratio ξ for active HCs); the coupled Haïssinski system is solved; statistical moments are extracted; each instability class is evaluated using the appropriate growth‑rate formula; Landau damping is assessed via the stability diagram; and a final stability verdict is returned.

To validate the approach, the authors apply the algorithm to the SOLEIL II upgrade project, a 2.75 GeV, 500 mA storage ring that will employ a passive harmonic cavity to increase the bunch length by a factor of ≈2.5. They perform a five‑dimensional scan over MC voltage, HC voltage, MC phase, HC phase, and HC detuning, comparing the semi‑analytical predictions with full mbtrack2 multi‑bunch simulations for a subset of points. The agreement is excellent: predicted growth rates, stability boundaries, and Touschek‑lifetime optimisation points match the tracking results within a few percent, while the semi‑analytical scans require only minutes on a standard workstation.

Furthermore, the authors demonstrate how short‑range broadband wakefields (modeled as an impedance Z ∝ √ω) and a single narrow‑band HOM can be incorporated into the algorithm. They show that the fast‑mode‑coupling instability is particularly sensitive to the broadband component, whereas PTBL is driven mainly by the reactive part of the HOM. This combined treatment is, to the authors’ knowledge, the first systematic comparison of these effects using a semi‑analytical tool.

All the methods are packaged in an open‑source Python library called ALBuMS (Algorithms for Longitudinal Multi‑Bunch Beam Stability). ALBuMS provides a clean API, automatic unit handling, and example notebooks that reproduce the SOLEIL II case study. The code is released under a permissive BSD‑3 license, encouraging adoption by other laboratories and integration with existing accelerator‑physics toolchains.

In conclusion, the paper delivers a robust, fast, and physically transparent framework for assessing longitudinal stability in double‑rf storage rings. By overcoming the convergence and symmetry limitations of earlier semi‑analytical methods, incorporating rigorous Landau‑damping checks, and explicitly modelling PTBL, the authors enable accelerator designers to explore high‑dimensional rf‑parameter spaces early in the design cycle, dramatically reducing reliance on costly multi‑bunch tracking. Future work is suggested on extending the method to non‑uniform filling patterns, time‑dependent cavity detuning (e.g., fast‑feedback systems), and coupling with transverse dynamics for a fully 3‑D stability analysis.