The OPAL Discussion Forum

["Dragos Constantin" <dragos.constantin AT gmail.com>]

Hi Andy,
A quick update on this issue.

Obviously one can place an injected beam anyplace along the z-axis, and then a
properly set collimator will remove the macroparticles that back propagate
beyond the starting point of the simulation domain. Unfortunately, I have to
emit the macroparticles so this is not a good solution for my particular case.

After many trials, I figured I should run the simulation using multiple steps
if I want an emitted beam with no particles back propagating past the origin.

STEP 1: Start the simulation with a thin 'open' collimator in the origin, and
emit particles for a number of RF cycles (limit the number of MAXSTEPS of the
TRACK command).
STEP 2: Edit the output HDF5 file and remove any particles with negative z
coordinate from the last phase space.
STEP 3: Restart the simulation but with a 'closed' collimator and increase the
number of steps in the track command (the limiting tracking factor will be
ZSTOP). In this context, a 'closed' collimator has a small aperture and it is
positioned off-axis  so no particles will pass through. This will work only if
at there are particles in the region where the collimator is located,
otherwise this 'closed' collimator will not act on the back propagating
particles.

The last step gave me the most headaches. I was thinking once a beamline
element was declared, it will be considered part of the simulation no matter
what. That is definitely not the case for this situation.

One last thing, this works well with version 2.2.1. I could not run the
simulation with version 2022.1.0 as it has issues with the collimators. In
that case, particles are being removed in the central region where they should
freely propagate. I am not quite sure what to make out of it.

My plan is to download the latest version(s) sources and compile opal from
scratch as I did with v 2.2.1, and then redo some of these tests and confirm
if my findings regarding the collimators are true or not. I'll keep you
posted.

Thanks,
Dragos