Page 1 of 1

optimizing off-momentun and on close orbit

Posted: 06 Oct 2010, 09:49
by roblin
I have a lattice for which I compute the off-momentum closed orbit. When I subsequently track with
center_on_orbit=1 and center_momentum_also=1 with track everything works as expected.
In my closed orbit command i have start_from_centroid=1 and start_from_dp_centroid=1.


However, if instead of tracking, I want to do an optimization with the optimize command, it does not find the closed orbit when starting the optimization. It tells me:
"error: no beam present for closed-orbit calculation starting from centroid"

Re: optimizing off-momentun and on close orbit

Posted: 06 Oct 2010, 17:16
by roblin
This also happens with analyze_map. For some reason, I can not specify start_from_centroid=1 and/or
start_from_dp_centroid=1 in the closed orbit command. If i do that, then the analyze_map fails saying there is no beam.
Setting those options to zero makes it work but then of course, I am tracking on momentum.

Am i not understanding how closed_orbit is supposed to work?

Note that if i replace the analyze map by a track command
with center_momentum_also=1 and center_on_orbit=1, i can specify an off-momentum closed orbit and sucesfully track and produce the
correct closed orbit.




see attached .ele and .lte.

Re: optimizing off-momentun and on close orbit

Posted: 07 Oct 2010, 08:54
by michael_borland
Yves,

The problem is that elegant doesn't generate a beam when you use analyze_map, even if you give a bunched_beam command. I'm looking into a workaround or patch.

--Michael

Re: optimizing off-momentun and on close orbit

Posted: 07 Oct 2010, 09:03
by michael_borland
Yves,

The attached file will patch the analyze_map problem. Give it a try and let me know of any problems.

Can you post an example for the optimization problem?

--Michael