Two other things regarding the Si material/medium definition:
1) Considering as how the Si recoil detectors are so thin (about 300
microns), it would probably be of merit to change the maximum step
size variable, STEMAX, in the Si medium definition to something much
smaller than the default BLAST step size of 10cm. In my recoil
analysis that I am doing, I use a step size of 100 microns, for
example.
2) However, even if 1) is done, GEANT will still almost always pass
right through the recoil detectors, oblivious to their existance,
due to the large step size in the volumes right before the recoil
detectors (the default GEANT step size currently in CVS is 10cm!!).
One way around this is to (as Tim suggested at the last software
meeting) make a large volume enclosing each recoil detector volume
such that the entrance planes of both this enclosing volume and the
recoil detector volume overlap. As such, whenever a particle enters
the enclosing volume, GEANT shifts the particle back to the entrance
plane of the volume. At this point, as long as the recoil volume is
declared as an 'ONLY' volume, it will supercede the enclosing volume
(assuming the enclosing volume is defined as a 'MANY' volume) and
GEANT will always "see" the recoil detectors, provided the enclosing
volume is at least 10cm thick. I use such enclosing volumes in my
analysis. I do not think they are a part of the file rc_geom.f,
though. I think they should be.
Both of these changes, in my opinion, should be added to CVS, else the
recoil detectors will not be seen correctly by GEANT.
Thanks,
Aaron
This archive was generated by hypermail 2.1.2 : Mon Feb 24 2014 - 14:07:28 EST