Page 1 of 1

Scan parameters with a sdds_bunch

Posted: 10 Apr 2018, 07:42
by PCaminal
Hi cyao, michael_borland, all,

I am working with a sdds_bunch generated with a particle in cell (PIC) code for plasma acceleration with external injection and I want to propagate it through a diagnostics beamline.

1st QUESTION: instead of using the bunched_beam namelist command I am using sdds_beam. Which are the initial Twiss parameters, emittance, chirp, etc... used for such a beam? Are they calculated automatically from the distribution? Are the alphas and betas going to be overwritten from the twiss_output parameters if introduced there? If yes, how does these effect the particle distribution given? What about the chirp? What about the central momentum?

2nd QUESTION (very related to the first): If want to vary some parameters to simulate some sort of jitter effects in the plasma-acceleration process, which are the parameters that I should scan?

I am bit confused, how elegant is managing this issues internally... I was first just introducing the following parameters in the sdds_bunch namelist command:
input = path_to_folder/filename.sdds
input_type = "elegant" !!! that is x, xp, y, yp, t, p, dt, particleID
sample_fraction = 1e-4
center_transversely = 1
and the initial (default?) Twiss parameters assumed by elegant were beta_x, beta_y = 1.0 m, and alpha_x, alpha_y = 0.0, which does not correspond to the intial parameters of the distribution given in sdds_bunch... No bunched_beam parameters were given, and no alphas or betas given in twiss_output.

Could you please help me make headway with this issue?
Thanks a lot in advance!

Best regards,
Pau González Caminal

Re: Scan parameters with a sdds_bunch

Posted: 10 Apr 2018, 09:54
by michael_borland

The &sdds_beam command directs elegant to load the particles and track them without any modifications (except for those performed by beamline elements of course). There's no direct connection to the &twiss_output command.

If you want to modify the twiss parameters of the beam you are tracking, you can insert a TWISS element in the beamline. You can also use the program sddsmatchtwiss to transform the beam before running elegant.

If you want to propagate the beam's twiss parameters through the lattice using &twiss_output, you can do that using the program sddsanalyzebeam (be sure to give the -correctedOnly option) and the reference_file parameter of &twiss_output.