close orbit and response matrix when random error exist?

Moderators: cyao, michael_borland

Post Reply
li.chao
Posts: 45
Joined: 18 Aug 2021, 08:59

close orbit and response matrix when random error exist?

Post by li.chao » 02 Aug 2022, 06:38

Hi Michael,

Recently, I am trying to evaluate the influence from random errors in our pertra4 lattice. The purpose of this study is to select lattices, with errors, which can insure beta-beating within 5%, coupling factor within 10%, and further used these selected lattices for DA and LMA tracking.

In the attached *.ele file, you can how I set the position offset errors for magnets and tilt errors for quadrupole. The steering element are set as combined horizontal-vertical kicker element.

In my *.ele file, there are several things I am not sure, hopefully you can help me to figure it out.

1) As I set in *.ele file, after the &correct &end command. I assume this command only take care of the closed orbit correction without coupling correction, is it?

2) With &correction_matrix_output &end command, I also print out the response matrix. What I expect is non-zero cross plane response matrix in *.vhrm or *.hvrm, whereas, the matrix elements in *.vhrm or *.hvrm are zero. Since the cross plane response matrix is necessary for the coupling factor calculation, how should I do to get the non-zero cross plane response matrix?

Great thanks for your answers.

yours Chao
Attachments
P4_H6BA_v4_0_cwig_ap.sdds
(359.18 KiB) Downloaded 104 times
track_nom.ele
(5.5 KiB) Downloaded 116 times
P4_H6BA_v4_0_cwig_chrom5.lte
(111.88 KiB) Downloaded 124 times

michael_borland
Posts: 1927
Joined: 19 May 2008, 09:33
Location: Argonne National Laboratory
Contact:

Re: close orbit and response matrix when random error exist?

Post by michael_borland » 02 Aug 2022, 09:43

Chao,

The coupled trajectory correction requires that only HMON and VMON elements are present in the beamline, so I converted your lattice. The attached files work for trajectory correction, but the tune correction isn't stable.

Is this a ring that is strongly coupled, or a transport line? The coupled correction feature is intended only for transport lines.

In a ring, if we want round beams, typically the x-y coupling is fairly low, but the tunes are close together. The steering and lattice functions are largely the same as for an uncoupled case, and no special steps are required for correction of the orbit and tunes.

--Michael
Attachments
P4_H6BA_v4_0_cwig_chrom5.lte
(112.6 KiB) Downloaded 127 times
track_nom.ele
(5.5 KiB) Downloaded 125 times

li.chao
Posts: 45
Joined: 18 Aug 2021, 08:59

Re: close orbit and response matrix when random error exist?

Post by li.chao » 02 Aug 2022, 12:32

michael_borland wrote:
02 Aug 2022, 09:43
Chao,

The coupled trajectory correction requires that only HMON and VMON elements are present in the beamline, so I converted your lattice. The attached files work for trajectory correction, but the tune correction isn't stable.

Is this a ring that is strongly coupled, or a transport line? The coupled correction feature is intended only for transport lines.

In a ring, if we want round beams, typically the x-y coupling is fairly low, but the tunes are close together. The steering and lattice functions are largely the same as for an uncoupled case, and no special steps are required for correction of the orbit and tunes.

--Michael


Hi Michael,

Thanks for you quick reply.

"The coupled trajectory correction requires that only HMON and VMON elements are present in the beamline, so I converted your lattice. The attached files work for trajectory correction, but the tune correction isn't stable.
Is this a ring that is strongly coupled, or a transport line? The coupled correction feature is intended only for transport lines.
"

The lattice is the petra4 storage ring lattice. In my previous lattice all steering magnet are set as KICKER element which is assumed to be able to give kick in both x and y plane. All BPMs are MONI element, which is assumed to be able to record x and y position. At the the moment, the coupling is only due to higher order magnet (KSEXT and KOCT) and errors.

As shown in my previous email, I am trying to do error analysis in the ring with Elegant. And also I am trying to make the lattice as close to reality as possible. That is the reason why I do orbit and tune correction before finally selecting lattice meet certain requirement (within 5% beta-beating for example) for further analysis. According to your replay, it seems like that the &correct &end command only use the diagonal response matrix *.hrm and *.vrm, to get the close orbit, and ignoring the off-diagonal response matrix *.hvrm, *.hvrm, right?

A further question is how can I set the *ele file to ensure non-zero cross plane response matrix *.hvrm can be printed out by Elegant

yours Chao

Post Reply