Page 1 of 1
defining my own wiggler field
Posted: 28 Jul 2022, 08:19
by Runfeng_Luo
I wish to simulate the coherent betatron radiation with a plasma undulator, as described in 'Plasma Undulator Based on Laser Excitation of Wakefields in a Plasma Channel' by S. G. Rykovanov et al. I took a look at the manual, but it seems that I can one define the parameters for the wiggler/undulator. Is it possible for me to define my own wiggler field in ELEGANT simulation?
Re: defining my own wiggler field
Posted: 28 Jul 2022, 16:08
by michael_borland
There are several ways to define undulator/wiggler fields in elegant, starting from field maps
1. CWIGGLER accepts data defining the undulator field in terms of longitudinal and transverse harmonics.
2. BGGEXP accepts data defining the fields in terms of generalized gradient expansions.
3. BMXYZ accepts a 3D field map (Bx, By, Bz) vs (x, y, z).
CWIGGLER and BGGEXP model classical and quantum radiation emission in terms of the effect on the beam. However, this isn't going to give you the radiation field from an undulator.
--Michael
Re: defining my own wiggler field
Posted: 29 Jul 2022, 03:05
by Runfeng_Luo
ELEGANT is capable of simulating SASE FEL, right? Are you saying I can only obtain the effect the radiation does to the beam, but not the radiation spectrum?
Re: defining my own wiggler field
Posted: 01 Aug 2022, 10:02
by Seb_Wilkes
Do wait on Michael to confirm or deny, but yes to my understanding the synchrotron fields must be calculated ourselves, however the effects on a travelling bunch (particles) is modelled by Elegant. I had been looking at this for myself and for a classical wiggler I think the solution is to take the phase-space data before it goes into the wiggler and use postprocessing software to calculate the ensuing synchrotron fields.
If I may add, for plasma channels you may be better using a PIC code? While the answer may well depend on what the lifetime of the plasma structure is, and how confident you are that you can model the magnetic field without a full simulation. In any case, I do know that in PIC codes there is increasing work in postprocessing the simulation data to get synchrotron fields. Some even do it as part of the PIC cycle.
Re: defining my own wiggler field
Posted: 01 Aug 2022, 19:50
by michael_borland
Runfeng_Luo wrote: ↑29 Jul 2022, 03:05
ELEGANT is capable of simulating SASE FEL, right? Are you saying I can only obtain the effect the radiation does to the beam, but not the radiation spectrum?
It can compute SASE FEL output using Ming Xie's parametrization. It is not a real simulation, just an approximate estimate. You need to use a code like GINGER or GENESIS (etc) for the real FEL simulations.
--Michael
Re: defining my own wiggler field
Posted: 01 Aug 2022, 19:50
by michael_borland
Seb_Wilkes wrote: ↑01 Aug 2022, 10:02
Do wait on Michael to confirm or deny, but yes to my understanding the synchrotron fields must be calculated ourselves, however the effects on a travelling bunch (particles) is modelled by Elegant. I had been looking at this for myself and for a classical wiggler I think the solution is to take the phase-space data before it goes into the wiggler and use postprocessing software to calculate the ensuing synchrotron fields.
If I may add, for plasma channels you may be better using a PIC code? While the answer may well depend on what the lifetime of the plasma structure is, and how confident you are that you can model the magnetic field without a full simulation. In any case, I do know that in PIC codes there is increasing work in postprocessing the simulation data to get synchrotron fields. Some even do it as part of the PIC cycle.
Seb,
Yes, I think you have it right.
--Michael
Re: defining my own wiggler field
Posted: 03 Aug 2022, 03:12
by Runfeng_Luo
Cool, thanks for the clarification.