LNO section meeting on MADX and Xsuite for FCC-ee

→ Europe/Zurich
18/3-008 - CLIC Meeting room (CERN)

18/3-008 - CLIC Meeting room

CERN

20
Show room on map
Zoom Meeting ID
63950404142
Host
Rogelio Tomas Garcia
Useful links
Join via phone
Zoom URL

LNO section meeting on MADX and Xsuite for FCC-ee

 

MAD-X module review for FCC-ee, Guillaume Simon, slides

As part of his thesis studies, Guillaume is reviewing modeling of the FCC with MAD-X. He shows work in progress in understanding small tune differences observed with tapering and tracking and plans further comparison with TWISS and implementation of the exact R matrix. Specially for the quadrupole (that is the current bottleneck for FCCee) a 2 week period for this implementation is reasonable.

 

Xsuite recent developments, Giovanni Iadarola, slides

Giovanni recalls that Xsuite is primarily a multi-particle simulation (tracking) code developed to perform both single particle and collective effects studies. It was found to be convenient to also compute TWISS parameters with Xsuite, in particular for initial checks before tracking, tuning and matching. Xsuite is based on Python which makes it straightforward to import and profit from many powerful packages. The standard Python root finder (numpy.fsolve) can be used to find the closed orbit, and transfer matrices can be obtained using differences in tracking around the closed orbit. He illustrates this with examples including LHC, and finds very good agreement with MADX for TWISS parameters, including coupling. More recently, Xsuite has been enhanced for dealing with radiation including tapering. The Twiss with radiation is implemented following E. Forest recipe. This is part of a wider effort to make Xsuite ready for FCC-ee studies and to provide a common basis to including many packages dealing with quantum effects, interfacing with Geant4 for collimation, beam-beam, Beamstrahlung etc.

 

Magnet polarity checks for LHC and HL-LHC, Riccardo De Maria, slides

Riccardo reviews the LHC polarity conventions and illustrates how the MAD-X variable values and signs are translated into normalized strength as visible in TWISS tables, strength and currents in LSA and finally actual currents at magnet leads. There are differences between conventions at the different levels and in addition exceptions that make the situation rather complicated. For HL-LHC he recommends to use conventions similar to LHC to minimize extra work. On the MAD-X side the conventions would stay as presently. On the LSA side he proposes to follow the MAD-X conventions by using the polarity change variable in LSA for all skew elements, and to introduce these changes in LS3 or profiting from a big optics change as could be the polarity change of the triplet (proposed by Stephane, maybe in 2024 or 2025, to mitigate energy deposition).

We agree with his proposals, that will be presented and discussed in a future LBOC meeting.

 

Python packages for particle accelerator beam optics and analysis (boa) initiative, Felix Soubelet

Various beam optics tools have been regrouped and listed on https://abpcomputing.web.cern.ch. These contain a list of softwares with different stage of maturity and support with links to the codes, with for example Frequency Analysis Tools for harmonic analysis of tracking data, Data Format Tools (tfs-pandas, sdds) or some LHC-specific tools. Currently, most packages are on GitHub, some on GitLab.

A new initiative to curate and maintain Python tools for optics analysis was started, and a GitHub organisation was created that can be found at https://github.com/beamopticsanalysis. It will curate and regroup the official maintained packages for optics analysis. It includes so far boa, and empty package akin to Xsuite for easy installation of the various tools and cpymadtools, a python wrapper around cpymad for easy, efficient optics studies with MAD-X through Python. They are already available on PyPI.

Helmut reminds that GitLab was chosen for the FCC-ee-lattice and is also the main developer platform of GEANT4 that is also much used outside High-Energy physics and CERN.

 

First optics measurements in LEIR, Felix Carlier, slides

As reported on the Nov 23 meeting, it became possible this year to acquire turn-by-turn BPM data in LEIR to perform optics measurements. The  extraction kicker can now also be triggered at flat bottom and produce large oscillations in the horizontal plane with some leakage into the vertical plane. The beta-beating measured in the horizontal plane using the extraction kicker agrees well with AC-dipole excitation results. Instead, the smaller vertical component of the extraction kicks was not yet sufficient to obtain reliable measurements in the vertical plane using the extraction kicker. The measurements of the optics in the vertical plane have still to rely mainly on AC-dipole excitation close to the resonance with poorer reproducibility. Further improvements should be possible using more extended averaging.

There are minutes attached to this event. Show them.