[WIP] ConstF: Allow kx<0 and ky<0 #1041
Conversation
|
@kyrsjo I don't see an element exactly analogous to this one in the MAD-X documentation. However, (at least for the transverse focusing), the closest analogy is the SOLENOID. If the rotation in x-y is ignored, then the solenoid provides equal focusing in x/y with strength k=Bz/Brho, following the convention defined here. In other words, the focusing field strength normalized by Brho. |
| m_cos_kyds = cos_kyds; | ||
| m_const_y = -m_ky * sin_kyds; | ||
| // Note: The convention here is m_kx[1/m] = sign(k)*sqrt(abs(k)) | ||
| // where k[1/m^2] is in the usual MADX convention |
There was a problem hiding this comment.
This statement is not clear, since there is not direct MADX analog. For solenoids, the focusing strength is k [1/m].
There was a problem hiding this comment.
I was mainly thinking that this is a strong analog to quadrupole mangets and plasma lenses, since the equations are exactly the same*, and there is no rotation implemented between the horizontal/vertical plane? So providing the conversion explicitly to the units used for quadrupoles makes sense IMO.
* Except for the convention of how k works.
| auto const [sin_kyds, cos_kyds] = amrex::Math::sincos(m_ky * slice_ds); | ||
| amrex::ParticleReal const_y = -m_ky * sin_kyds; | ||
| // Note: The convention here is m_kx[1/m] = sign(k)*sqrt(abs(k)) | ||
| // where k[1/m^2] is in the usual MADX convention |
|
This looks great; thanks a lot! I made only a minor comment. It looks ready to go once a simple test is added. Let us know if you would like some help with this. |
|
Hi, that's great! I'm on summer holiday right now, will be back in August. I can look at it then, or if you want to add the test and merge, please feel free! Thanks again for all the help - it makes it a lot more likely that I will contribute again, and now I know your setup and standards! |
|
Basically in |
|
In tests, also look at this for how I get the data out:
Relative to: |
3 participants
This pulls out the changes to ConstF from PR #1030 . It allows the ConstF elements to handle negative (defocusing) k values.
Question: Why is the unit of kx and ky in this element 1/m, i.e. the square root of the more common definition of k?