RF-Track 2.7.2

RF-Track is a tracking code developed at CERN for the design and optimisation of particle accelerators, offering outstanding flexibility and rapid simulation speed.

RF-Track can simulate beams of particles of any energy, mass and charge, even mixed, solving fully relativistic equations of motion. It can simulate the effects of space-charge forces in both bunched and continuous-wave beams, synchrotron radiation emission, wakefield and beam-loading effects, multiple Coulomb scattering, inverse Compton scattering, and the list is still growing. It can transport beams through both conventional and special elements: 1D, 2D and 3D static or oscillating radio-frequency electromagnetic field maps (standing wave and traveling wave), flux concentrators and electron coolers. It allows element overlap and direct and indirect space charge calculations using fast parallel algorithms.

RF-Track is written in optimised and parallel C++ and comes with two distinct and independent user interfaces: one in Octave and one in Python. General knowledge of Octave or Python is recommended to get the most out of RF-Track.

Changelog

What is new in version 2.7.2

• Added Volume::distribute_screens_along_path() to simplify simulations of, for example, alpha magnets
• Added TM_mnp transverse-magnetic structures for simulating, for example, transverse deflecting cavities
• Added TE_mnp, the transverse-electric counterpart of TM_mnp, with magnetic mode coefficients in tesla
• Added set_scaling_factor(factor) to all field maps and analytic fields to simplify parametric scans
• UserElement now supports space charge, collective effects (a half kick before and a half kick after the element), and apertures
• UserElement and UserField are now time-dependent elements
• Various bug fixes and improvements

What was new in version 2.7.1

• Now Lattice::get_rf_elements() and Volume::get_rf_elements() return concrete element types
• Fixed race condition in parallel threads

What was new in version 2.7.0

• Added short- and long-range ResistiveWallWake effects
• Implemented the Sokolov–Ternov effect of spin flipping in magnetic fields
• Implemented SBend fringe fields in Volume tracking
• Made UserEffect, UserElement, UserField, and UserVisitor available in Octave
• Refined the EnergyStraggling model for more realistic simulations
• Improved cathode mirror-charge truncation using the full 3D bunch size and a configurable field-contribution tolerance
• Added Lattice::distribute_sc_nsteps() and Lattice::distribute_cfx_nsteps()
• Added Gaussian and ASTRA smoothing methods to set_smooth() for 1D field maps (see documentation)
• Added xmax, xmax90, xmax99, ymax, ymax90, and ymax99 to the transport table
• Various bug fixes and improvements

What was new in version 2.6.3

• Added Lattice_view::get_transport_table()
• Addressed memory leak in SWIG interface

What was new in version 2.6.2

• Volume::autophase() now accepts Bunch6d and not just Bunch6dT
• Volume::get_s0() and Volume::get_s1() now return a 3D oriented Frame
• Volume::accept() now accepts rft.UserVisitors like Lattice

What was new in version 2.6.1

• Refactored Lattice::track() to support set_tt_nsteps(0) in multi-turn simulations
• Lattice::set_sc_nsteps() and Lattice::set_cfx_nsteps() now support cross-element kicks, improving the performance and accuracy of multi-turn simulations
• Refactored Lattice::autophase() for improved execution speed
• Introduced Lattice_view for focused Lattice operations
• Exposed elements' offsets to Python and Octave
• Various bug fixes and improvements

What was new in version 2.5.6

• Added the Aperture_1d element, supporting arbitrary cylindrical aperture profiles
• Quasi-random particle distributions now start from zero
• Long-range wakefields now include the transverse defocusing effect by default
• Various bug fixes and improvements

What was new in version 2.5.5

• Implemented proper SBend element in Volume
• Insertion of Lattices in Volumes considers correctly bent reference trajectories
• Imporved long-range wakefield model allowing for the simulation of HOMs damping in detuned structures

What was new in version 2.5.4

• Bugfixes and general improvements (autophase, backtracking in Volume, generator)

What was new in version 2.5.3

• Lattice: lost particles’ longitudinal positions are now reported in meters (previously in millimeters)
• Added support for converting multipoles up to order 20 when importing MAD-X lattice files
• Changed default Volume::emission_range setting from 2 to 10 to improve simulation accuracy
• Fixed bug in transport table when multiple Volumes are inserted as elements into a Lattice

What was new in version 2.5.2

• Now screens in Volume preserve the same particle order as the incoming beam
• General bug fixes and performance improvements

What was new in version 2.5.1

• Fixed a bug introduced in version 2.5.0
• Added automatic update check on startup

What was new in version 2.5.0

• UserElement – Custom accelerator elements and transfer maps in Python or Octave.
• UserField – User‑defined electromagnetic field equations in Python or Octave.
• UserEffect – Prototype collective effects in Python or Octave.
• UserVisitor – Traverse lattices/volumes with custom Python functions.
• Secondary particle generation – Collective effects can now emit and track new particles (e.g., synchrotron‑radiation photons, BDSIM secondaries).
• Cathode emission – Improved accuracy in Volume‑based emission simulations.
• Minor improvements and bug fixes.

What was new in version 2.4.2

• Added BPM element to Volume
• Improved Volume–Lattice integration for Beam-Based Alignment in Volume
• Enabled partial polarization of individual macroparticles
• Bug fixes:
  • Fixed issue in PillBox cavity
  • Resolved problem with beam losses when all particles are lost
  • Corrected RF phase of structures imported from MAD-X

What was new in version 2.4.1

• Added spin polarization support to both Lattice and Volume tracking

What was new in version 2.4.0

• Improved speed and memory management in TW_Field
• Introduced new collective effect: IntraBeamScattering (developed by Paula Desiré Valdor)
• Updated Lattice transverse emittance definitions to use (x, x') and (y, y') instead of (x, px) and (y, py)
• Enabled arbitrary misalignment of cathodes in Volume, including mirror charge effects

What was new in version 2.3.4

• Fixed a memory problem in Python's numpy interface

What was new in version 2.3.3

• Added TW_Field element to create RF structures from shunt impedance, group velocity, and quality factor
• The new TW_Field element can account for steady input power, or time-dependent RF pulses
• Long-range wakefields can now have arbitrary linear polarization
• Improved speed of SW_Structure and TW_Structure elements
• Beam and BeamT can now be saved and loaded in binary format
• Elements can now be removed from and inserted into an existing Lattice
• Lattice and Volume transport tables now return transmission in number of particles (no longer in number of macroparticles)
• Improved error message handling to avoid redundant repetitions
• Various minor bug fixes

What was new in version 2.3.2

• Fixed a potential beam-laser synchronisation issue in the LaserBeam element
• Added multi-pulse laser trains in LaserBeam
• Fixed minor bug in Python interface

What was new in version 2.3.1

• ShortRangeWakefield allows the simulation of tapered irises and cell-to-cell misalignment
• Sped up Solenoid field calculation
• Now Screens can have a width, a height, and a time window
• Now SpaceCharge_Field() uses a fast Cartesian Multipole Expansion
• Added Undulator element to simulate planar undulators
• Minor bug fixes and improvements

What was new in version 2.3.0

• Introduced new Beam and BeamT models for multi-bunch simulations
• Optimized multi-bunch effects for multi-bunch beam simulations
• Modified the transport table to support multi-bunch beams
• Added the RF_TRACK_NOSPLASH environment variable
• Implemented element misalignment handling in Volume
• Added scaling error support to the Bpm element
• Enhanced Lattice::autophase() to automatically set the magnetic strength of magnets
• Improved detection of the solenoid fringe field extension in Volume
• Enhanced the 3d solenoid field modelling using multiple thin sheets of current
• Introduced a new on-the-fly BeamLoading implementation compatible with Beam (thanks to Javier Olivares Herrador)
• Enhanced long-range wakefields to accept tapered modes along the structure
• Added generalized Gaussian distribution to Generator
• Added the displaced() method to Bunch6d and Bunch6dT
• Added Screen element support to Lattice and Volume
• Implemented the fast multipole method to significantly speed up the SpaceCharge_Field() element
• Minor bug fixes and improvements

What was new in version 2.2.4

• Ensured compatibility with numpy v2.0.0

What was new in version 2.2.3

• Minor bugfixes and improvements
• Improved python packaging

What was new in version 2.2.2

• Lost particles are now stored in the Lattice or Volume in the laboratory reference frame
• Bug fixes and improvements in the Python interface
• Bug fix when appending a Volume into a Lattice
• Renamed some Lattice's methods ( vary_corrector_strengths -> vary_correctors_strengths )
• Added methods to facilitate beam-based alignment in Lattice
• Fixed problem with backtracking on screens in Volume
• Improved backtracking on screens in Volume

What was new in version 2.2.1

• Added particle ID
• Added name to elements
• Improved element accessibility in Lattice and Volume
• Exposed mean_Px and mean_Py in Bunch6d_info and Lattice transport table
• Volume::set_length() allows setting an arbitrary length
• Added element Pillbox_Cavity as generic TM_m1p cavity
• Bug fixes and improvements in Lattice and Volume

What was new in version 2.2.0

• Volume is fully integrated as an Element within Lattice
• Volume::s0 and Volume::s1 are now 3d reference frames that can connect Lattice beamlines with arbitrary orientations
• Renamed ParticleT.S as ParticleT.Z to avoid confusion
• Removed automatic projections from Bunch6dT to Bunch6d in get_phase_space()
• Lattice can now import MAD-X Twiss files directly
• Self-consistent BeamLoading effects in TW structures (many thanks to Javier Olivares Herrador)
• Added Energy Straggling effect in Absorber element (many thanks to Bernd Michael Stechauner)
• Volume::set_tt_nsteps(N) works in Lattice(), distributing N screens in Volume along the reference trajectory
• Added Bunch6d_QR, Quasi-Random bunch creation from Twiss parameters
• New hard-scattering model (many thanks to Bernd Michael Stechauner)
• Improved 'analytic' integration in Lattice
• Added Yoshida integrator 4th order (beware, rk2 is better)
• Improved beam <-> screens intersection in Volume tracking

What was new in version 2.1.6

• Implemented 'circular' and 'rectangular' shapes in the element Absorber
• Improved the way longitudinal emittance is handled in both Bunch6d and Bunch6dT
• Now Volume can track directly Bunch6d
• Now, Transport Table in Volume can also include future particles by default
• Implemented Response matrix calculation and Automatic Orbit correction
• Improved transport table in Volume to optionally ignore particles outside volume
• Updated autophase to ignore elements whose t0 is already set
• Added Multiple Coulomb Scattering and energy loss
• Introduced ASTRA-like Generator (many thanks to Avni Aksoy)
• Introduced quasi-random sequences generator
• Added to transport table rmax99 and rmax90, the envelope at 90% and 99% of rmax
• Fixed a bug in thin-lens Sbend
• Added disp_z to BunchInfo

What was new in version 2.1.5

• Added element Solenoid to both Volume and Lattice
• Added particle lifetime for muons simulation
• Added Incoherent synchrotron radiation in two versions: quantum and average
• Added new element SpaceCharge_Field()
• Fixed SWIG problems with typemaps
• Bug fix in Multipole element (many thanks to Ewa Oponowicz)

What was new in version 2.1.4

• Added dispersion information to bunch get_info() and transport_table
• Now Element::get_field() accepts vectors for X, Y, Z, and T
• Added velocity slicing in Space-charge PIC calculation
• Added Toroidal Harmonics element

What was new in version 2.1.3

• Added polarisation to Thomson scattering
• Implemented end fields in GenericFieldMap
• Implemented asymmetric laser beam in Compton Scattering module
• Added new element Multipole and new CollectiveEffect MultipoleKick

What was new in version 2.1.2

• Added autophasing of RF elements
• Added back-tracking
• Added Long-range wakefields

What was new in version 2.1

• Added the possibility to have collective effects in both Lattice() and Volume()
• Added short-range wakefield effects
• Added new element RF_FieldMap_2d() optimized for cylindrically symmetric fields
• Added new element LaserBeam() for the simulation of Compton backscattering

What was new in version 2.0.4

• Changed the way TrackingOptions.t_max_mm works, now indefinite tracking is achieved with t_max_mm = Inf, and no longer t_max_mm = 0
• Added element Static_Magnetic_FieldMap_2d for cylindrically symmetric fields
• Improved default field's Jacobian for Elements
• Added support for SDDS files (writing)
• Added element SBend for dS tracking
• Added external elements
• Added Python interface
• Reimplemented misalignments in Bunch6d tracking, full 3d rotations
• Added element Volume
• Added element Coil

What was new in version 2.0

• Reorganization of classes and files
• New FieldMap elements with extension _LINT and _CINT for linear and cubic interpolation
• Added element RF_Field_1d with a reconstruction of the 3d E and B fields from the on-axis electric field
• Added magnetic element Adiabatic Matching Device
• Bunch6d::get_phase_space and Bunch6dT::get_phase_space now use the first particle as the reference particle