Third order dipole matrix

Moderators: cyao, michael_borland

Björklund
Posts: 84
Joined: 19 May 2016, 07:14

Re: Third order dipole matrix

Post by Björklund » 25 Oct 2018, 10:26

Hi Michael,

All right, I'll stick with setting the edge effects to 1, then! I'll try to rememeber to set FSE_CORRECTION to 1 if and when I go back to using the symplectic model at some point. It also only gives me issues on BC1 and not BC2, but this might be because of the difference in K1 for the dipoles, as you mention, which are larger in BC1. In a sense, it's good that there was actually some sort of issue, since I was starting to feel a bit pushy.

Just for reference, I'm using double-achromat compressors, which are anti-symmetric (at least dipole- and sextupole-wise), like a dogleg.

I tried looking for TRANSPORT, it seems like it's available from AccelSoft, but not for free. I'll try to solve the third-order matrix in some other way then, as I have a whole paper that's almost done, where I need the total compressor U1666 value for computing stuff. Is there any future possibility for getting this matrix in elegant, or the option to have EMATRIX to third order?

//Jonas

michael_borland
Posts: 1796
Joined: 19 May 2008, 09:33
Location: Argonne National Laboratory
Contact:

Re: Third order dipole matrix

Post by michael_borland » 25 Oct 2018, 13:26

Jonas,

I didn't realize your chicane was not symmetric, but it turns out not to make a difference. I tried a symmetric chicane and get the same slight issue with the EDGE[12]_EFFECTS=2 model.

If you use TRACKING_MATRIX=3 on the CSBENDs, the only part missing now should be the third-order fringe component. As for getting the analytical third-order matrix into elegant, including fringe effects, I'm looking into it more. L. Sagalovsky worked this out in the late 80s for his Ph. D. thesis, but apparently he's no longer in the field. Redoing his thesis seems to be quite a lot of work. I'm trying to get the TRANSPORT code from FNAL instead.

--Michael

Björklund
Posts: 84
Joined: 19 May 2016, 07:14

Re: Third order dipole matrix

Post by Björklund » 26 Oct 2018, 01:56

Hi,

I found his thesis on http://lss.fnal.gov/archive/thesis/1900 ... 989-53.pdf, but I don't have the insight to see if it can be more or less directly implemented from the form given in the thesis. Could be an interesting document to look through (for me, anyhow), nonetheless! Looking forward to see what you find.

I will try the tracking_matrix again on BC2 now that I have the edge effects set to 1. However, also the T-terms were very far off in that case, but I suppose we'll find out!

//Jonas

Björklund
Posts: 84
Joined: 19 May 2016, 07:14

Re: Third order dipole matrix

Post by Björklund » 28 Nov 2018, 03:56

Hi Michael!

Happy belated Thanksgiving!

I was just wondering if you managed to get a hold of TRANSPORT and what the status of a potential 3rd-order matrix implementation is. I have spoken to some other people who work and/or have worked on bunch compressors, who'd also be very interested in this feature, mainly for the U5666 values though.

Best regards
Jonas

michael_borland
Posts: 1796
Joined: 19 May 2008, 09:33
Location: Argonne National Laboratory
Contact:

Re: Third order dipole matrix

Post by michael_borland » 29 Nov 2018, 11:18

Jonas,

I was able to get an unofficial copy of the final FNAL version of TRANSPORT, which contains the code for the third-order dipole matrix. I have not had time to dig into it and understand what's there, but hope to at least start on that in the next month.

--Michael

Björklund
Posts: 84
Joined: 19 May 2016, 07:14

Re: Third order dipole matrix

Post by Björklund » 29 Nov 2018, 11:47

Hi Michael,

That's great news! I'm looking forward to seeing how this develops, and I hope it won't be too much work for you to implement. I am very excited about this feature!

//Jonas

tcharles
Posts: 10
Joined: 11 Aug 2014, 20:58

Re: Third order dipole matrix

Post by tcharles » 05 Nov 2019, 10:05

Hi Michael,

I have the same query as Jonas, and I was just wondering if this was followed up on. Has the third order dipole matrix been included in elegnat? I couldn't find a mention of it in the latest release details.

Many thanks,
Tessa

Post Reply