Skip to content

Minutes of #5 Remote E-Beam Meeting

The meeting was held on Vidyo on 20/05/2020 - See indico

Actions

  • A. Latina, A. Rossi: verify if gas-jet monitor can be simulated with RF-Track.

Status of Betacool Source code (D. Gamba)

The source code of BETACOOL is now publicly available on github, after having received the green light from A. Sidorin. A mirror of the repository is kept on the CERN GitLab. The choice of using github is to make it easier for CERN externals to collaborate on the development/maintenance of the code.

An alternative code to BETACOOL for cooling studies is also being developed at JLab, and is called JSPEC. Radiasoft LLC, an US company, also provides an online interface to such a code. This code is meant to be benchmarked against betacool, so in principle is a good alternative for benchmarking RF-Track. JSPEC seems very simple to use, and requires very few, well defined, input parameters. See for example NAPAC2019-TUPLO04.

Status on e-cooling simulations with BETACOOL and RF-Track (B. Veglia)

A short introduction on e-cooling and the considered simulation tools was given. In particular:

  • BETACOOL: two of its main algorithm have been considered: RMS Dynamics, which provides quick estimates of the evolution of second order momenta of a Gaussian beam, and Model Beam, which solves Fokker-Plank equation for quasi-Gaussian beam distribution represented by a few thousand particles. The Tracking algorithm provided by BETACOOL was not considered at this stage.

  • RF-Track: it tracks the ion beam as an ensemble of particles with arbitrary distribution, while electrons are represented as a fluid on Cartesian mesh.

One of the key effect to model e-cooling is the cooling force, which depending on the boundary conditions can become complex to express. The two models considered by Bianca are:

  • semi-empirical formula by Parkhomchuk in BETACOOL: it introduces and effective velocity spread that is normally adjusted ad-hoc to account for imperfections on the straightness of the magnetic field, power converter ripples, misalignments, etc.
  • cooling force description by H. B. Nersisyan implemented in RF-Track (see for example PhysRevSTAB.16.074201) with no approximations, provided that one knows and inputs all beam parameters and instrument imperfections.

Preliminary results shown by Bianca are a brave attempt to compare at first order both simulations in BETACOOL and RF-Track and some experimental data.

  • The comparison of available data on the longitudinal cooling force from ESR and simulation with RF-Track were already shown in the past by A. Latina (E-BEAM meeting #1) are remarkably good, as well as with BETACOOL assuming a reasonable effective electron temperature of 5 meV.
  • Data obtained on LEIR can be matched with RF-Track (E-BEAM meeting #1), while BETACOOL requires un-reasonably high value for the electron effective temperature to match the data. The reason of this discrepancy is still under investigation.
  • The time evolution of the transverse emittance in LEIR appears to cool faster than what predicted by both BETACOOL and RF-Track.
  • The momentum spread evolution is also showing different behaviors between codes, different algorithms implemented in BETACOOL and actual data from LEIR, the latter still showing the fastest cooling behavior.

The general comment is that there are many unknowns and hidden parameters, especially in BETACOOL, that require careful investigation before drawing any conclusion from the results presented.

Discussion

  • Andrea: the speed of cooling is extremely sensitive to delta velocity between ions and electrons, so this could be used as fitting parameters. The momentum distribution of ions in LEIR is also very far from Gaussian: A stack of Gaussians is assumed in RF-Track, but this is not the case for BETACOOL, which might have a strong impact on the final result.

  • Roberto: a different number of electrons is used RF-Track and BETACOOL: what is the most realistic value? Andrea replied that since the force is linear with electron current, even a variation of a factor of a few would be hardly visible on the logarithmic plots presented.

  • Andrea: the cooling force obtained from RF-Track is estimated by tracking the force on the single ion from all electrons, taking into account the electron temperature, and then folded with the ion distribution. This is not necessarily the case in BETACOOL.

  • Christian: the first data point from LEIR transverse emittance evolution might be an artifact. If one would start from the second point, the comparison between measurements and simulations could change considerably.

  • Christian: what are all simulation parameters? e.g. what are the assumed electron temperatures? Were they changed to try fitting the experimental data? Bianca: electron temperatures are the ones assumed from cooler device. In BETACOOL mainly the effective temperature and electron density was varied. In RF-Track there is no corresponding parameter, only the angular spread between ions and electrons. Ideally, one would need to find an expression to translate the angular spread used in the RF-Track with the effective temperature of BETACOOL, but it is not obvious how to do it.

  • Christian: electron temperature values have big uncertainties.

  • Adriana: The cooling force seems to be better matched by RF-track, while cooling time seems to be better fit by beta-cool. Is there a known reason? General comment is that this might be due to way integrals are implemented, as well as the provided initial conditions.

  • Christian: Were possible betatron oscillations considered in simulations? Transverse velocities have an influence on longitudinal cooling, and that could partially explain differences. In general, also the ion distribution should be included.

  • Andrea: space charge could affect also the ion beam evolution. In general, at this stage, we should concentrate first on agreement between ideal cases in simulation codes rather than trying to compare with experimental data.

AoB: Modeling of HEL Gas-Jet Monitor with RF-Track

A gas-jet monitor is planned to be used in HEL for electron-ion (see E-Beam #8). One of the issue of this kind of monitor is that beam-gas interactions can generate ions that remain trapped in the magnetic field of the HEL, and those could potentially degrade the performance of the electron beam. Adriana asked if such a monitor could be simulated using RF-Track. Andrea replied that it seems possible, and it could be studied. Marton Ady mentioned that a technical student is being hired to do vacuum calculation for HEL. He/She could potentially perform also those kind of simulation.