2D ellipsoid space charge calculator #56

Activity

1. austin-hoover

   added

   enhancementNew feature or requestNew feature or request

   on Apr 2, 2025
2. shishlo commented on Apr 15, 2025

   

   shishlo

   on Apr 15, 2025

   Contributor

   More actions

   I agree. That is a great idea. The set of 2D Ellipsoids with/or without boundaries will give us a fast tool for ring dynamics calculations. Parallelization scheme should be more flexible and faster then FFT based algorithms.
3. austin-hoover

   self-assigned this

   on Aug 22, 2025
4. austin-hoover commented on Sep 5, 2026

   

   austin-hoover

   on Sep 5, 2026

   ContributorAuthor

   More actions

   Implemented in PR #143, mostly by AI. Looks like it's giving correct results, comparing to 2D PIC with KV distribution (1 ellipse) or Gaussian distribution (5-10 ellipses) for long coasting bunches.

   • No boundaries conditions were implemented.
   • Have not tested with MPI.
   • Do we want to weight transverse kicks by longitudinal density, like in SC2p5D?
   • Similarly, do we want to add a node under SCSliceBySlice2D? This would run a separate ellipse solver in each longitudinal slice.
   • I renamed classes to follow same convention as other space charge nodes.