Setting nominal arrival time in sdds_beam

[Björklund]

Hello, and happy new year,

I am looking into some simulations with two bunches separated by only a couple of degrees in the RF phase, and I want to vary this spacing to look at the effects.

When I send in my "artificial" double-bunch structure and keep center_arrival_time = 1, the driver bunch is of course shifted wrt the nominal (average) time of the two bunches, which will be right in between them because they have equally many particles. I would like to have the nominal arrival time in the center of the first bunch when I send it in, so that I can have some control over where the two bunches actually are wrt the RF, and not only each other.

I'm not really sure how things work in elegant, but would it help me to set center_arrival_time = 0 and give my bunches the appropriate time values wrt the RF sine wave before I send them into elegant? I expect this to give me problems, since I am also scanning the off-crest phase in two different linac sections, and I would like to tweak the phase parameter inside elegant and not elsewhere. It would be neat with something like the time_start parameter in bunched_beam, which, if I understood correctly, sets the nominal time of the bunch.

Do you have any good advice on this?
Best regards
Jonas

[michael_borland]

Jonas,

The best way to handle this is

• Ensure that in your input beam, there is a particle at the center of the first bunch and that it is the first particle in the file. It is probably easiest if t=0 and delta=0 for this particle.
• On the RFCA elements, set FIDUCIAL="first"), which forces elegant to use the first particle as the reference for phasing.

--Michael