Go to the Indico Home Page Oldest event Older event Up to category Newer event Newest event Printable version
  • Compact style
  • Indico style
  • Indico style - inline minutes
  • Indico style - numbered
  • Indico style - numbered + minutes
  • Indico Weeks View
    • Switch to the management area of this event
    Login

    SKEKB MADX model

    → Europe/Zurich
    SKEKB MADX model
    Zoom Meeting ID
    65122721377
    Host
    Rogelio Tomas Garcia
    Useful links
    Join via phone
    Zoom URL
    Hide

    Solenoid region is the most 

    HER and LER have an independent definition of solenoid as the field is expanded around a straight line trough the solenoid (Bz + Bx components). See screenshot in indico (Oide-san sketch).

    SAD computes reference orbit by tracking from IP to the solenoid edges. 

    BMAD could be used to translate among all codes. Maybe improvements for MADX style are needed in the BMAD->MAD conversion. Felix mentions that misalignments with GEO=1 are not implemented. To be checked.

    In SAD Mult element takes Bz component from overlapping solenoid (defined in previous element). SK0 (bend) in mult element comes from solenoid field expansion.

    (typo on slide Bad model -> Bmad model)

    Mult element in SAD has a length and contains alignment features. SAD mult is the standard 2D (dBz/ds=0) multipole kick also available in MAD. SAD uses a criteria to determine the number of drift-kick expansions for a single mult with length, depending on strength.

    Fringe attribute F1 of mult in sad is a length to apply a linear tapering of the strentgh versus s (effectively  the slices), see sketch in screenshot.

    Fringe=3 means linear tapering applies on both sides of elements.

    If Disfrin is zero Maxwellian fringes are applied.

    Beam orbit (wrt solenoid straight line reference) is given in SOL element (CHI1 and CHI2 are angles) but this is ignored and recalculated by SAD.

    MAchine geometry is defined by angle in bends and solenoids  (considering elements tilt = rotate in SAD). In BMAD.

    Leon proposes to do matching in MADX of the misalignment parametets to reproduce SAD misalignments, tbc.

    Factors and signs of mults are to be checked, as BMAD and SAD differ.

    Patch element does not work in MAD but we can use rotation and translation, this is why matching is convenient. Longitudinal offsets are unclear between SAD and BMAD.

    BMAD to MADX uses matrix element. Some tests showed some small discrepancies, with very close emittances. Demin will generate a SKEKB MADX file from BMAD. The limitation of this approach is loss of control of individual elements. 

    Maxwellian fringe fields in SKEKB are important for DA. We do have them in MAD, this can only be a medium-term effort. Expansion of the drift should be more easily done. 

     

    Actual short term tasks:

    -Start exploring in MADX a simple SKEKB IR based on SAD description and implementing drift-kick approximations for multipole and overlapping elements. Leon 

    -Produce MADX SKEKB lattice from BMAD, Demin in coming days to upload to repository (indico). 

    -Implement in MADX a drift with analytical dependencies on misalignments. Tobias

    Longer term:

    -Does MADX need to find orbit from exact tracking? Tobias to start discussions.

    -Maxwellian fringe fields. These were under discussion, maybe quadrupole case is easy?

     

     

     

    There are minutes attached to this event. Show them.