From: Glenn Horton-Smith (gahs@phys.ksu.edu)
Date: Wed Sep 21 2005 - 13:48:32 CDT
Hi all,
Tim's explanation is a good summary:
>This is a feature, not a bug, that I will let Glenn explain. (Actually a good feature).
>
>/generator/event_window 1000000000
>
>should, I believe, create more "conventional" events. The argument is time in ns, with default=1000. So something happening 1001 ns by default after the muon triggers a new "event".
>
>
Just to elaborate a bit more: there's a custom "physics process" added
to GLG4sim that suspends tracks when their time goes past the
event_window. The process stacks them for use as parent particles in a
later event. This is to allow you to have some approximation of "singles
events" in a self triggering detector. It can be quite useful, but also
quite confusing, particularly if you are used to well-behaved,
beam-triggered detectors.
On a related note, I should warn everyone about a "shortcut" inherited
from KLG4sim that might mess up your neutron simulations.
"GLG4NeutronDiffusionAndCapture" is a process that takes thermal
neutrons “away” from Geant4 and skips over all the tracking of thermal
neutron diffusion until the neutron captures. Basically, it just adds an
exponential time step and a Gaussian spatial jump to the neutron, then
forces an immediate capture in whatever material it started in. *It does
_not_ know how to account for edge effects where there are different
diffusion lengths or capture times in adjacent volumes.*
You can deactivate this process using "/process/inactivate
NeutronDiffusionAndCapture neutron".
Deactivating this process will cause the neutron to be tracked
continuously from the time of its creation until its eventual capture.
This will profoundly alter the effect of the "event window". The effect
is much easier to understand than to explain, so I'll leave it as an
exercise to the reader. :-) See also pages 10-11 and 23 in the "Useful
GLG4sim tricks" talk
[http://neutrino.phys.ksu.edu/MAND-sim/MAND-sim%20talks/GLG4sim%20tricks.pdf]
.
Cheers,
Glenn.
This archive was generated by hypermail 2.1.6 : Thu Sep 22 2005 - 12:01:02 CDT