Losses studies

Moderators: cyao, michael_borland

Post Reply
simona.bettoni
Posts: 45
Joined: 06 Jul 2011, 03:14

Losses studies

Post by simona.bettoni » 31 Jul 2020, 08:54

Hi to all,
I am doing some losses simulation studies in a ring.
I followed what I read in the manual, and I generate the kind of losses (from the %.loss file) shown in this entry. I report also the xHalfAperture in the same area. They make sense when I use ALTER_ELEMENT applied to the beam at the injection to add an offset or an angle, and I do the tracking, but I see some discretization in s. How can I reduce it?
Another probably also more important question is the following: why some particles are lost at a position outside of the aperture? I attach a second image for this. Again, I think I am discretizing something not properly.
I attach also the files I used for the simulations.
Thanks for any comment/help.
Attachments
Files.zip
Files used for the simulation
(300.74 KiB) Downloaded 187 times
Capture2.PNG
Case of 90 degs at the location of the losses
Capture1.PNG
Case of angle and offset at the injection

michael_borland
Posts: 1927
Joined: 19 May 2008, 09:33
Location: Argonne National Laboratory
Contact:

Re: Losses studies

Post by michael_borland » 01 Aug 2020, 10:39

Simona,

elegant checks the aperture limits at the ends of all elements, and also inside certain elements, such as CSBEND, KQUAD, and KSEXT. In your lattice, you use CSBEND, QUAD, and KSEXT, but you don't give the N_KICKS parameter on CSBEND or KSEXT.

My suggestions
  • Replace QUAD by KQUAD, which is symplectic and thus best for ring tracking.
  • Set the N_KICKS parameter on the CSBEND, KQUAD, and KSEXT elements to ensure that the modeling is accurate. I suggest trying N_KICKS=40 for CSBEND, 30 for KQUAD, and 15 for KSEXT, but you'll need to explore this to ensure that the orbit, tunes, and chromaticities obtained in tracking are reasonable.
  • You can subdivide the elements using the &divide_elements command to make the discretization finer, at the expense of possibly longer run times.
--Michael

simona.bettoni
Posts: 45
Joined: 06 Jul 2011, 03:14

Re: Losses studies

Post by simona.bettoni » 01 Aug 2020, 13:30

Thank you very much for your help.
Kind regards,
Simona.

Post Reply