Closed rl3418 closed 11 months ago
By default, OSIRIS assumes that you are initializing your simulation with a self-consistent field distribution, which is not the case for your input file. This leads to the formation of non-physical EM fields, that will eventually break up your beam. I would recommend that you check the documentation on how to initialize the beam fields properly. You should also use the moving window along x, not y.
Osiris predicted unphysical divergence for an electron beam travelling in free space with gamma=200, n_e~1e20m-3. The momentum along the propagation direction drops rapidly to zero, but for a relativistic beam the longitudinal momentum shouldn't change by much. I performed the same simulations using Smilei. According to Smilei the longitudinal momentum doesn't drop by much, which is exactly what I expect.
I've attached the results I got from Osiris and Smilei. Under the Osiris folder you could also find information about the libraries I used to compile Osiris.
potential bug with osiris.zip