|Message: general problem for optical photons at surfaces shared by different volumes||Not Logged In (login)|
|1||0||1||no text / full text of posting|
|2||1||All||text for level 1 only / text for All postings|
|1||2||1||2 levels / 1 level (original posting)|
|2||3||2||3 levels / 2 levels|
|3||3||All||3 levels / all levels (all postings)|
I came across a problem in my simulation, that seems to be a general, severe problem to me:
I created a WLS fibre with two claddings by just putting the outer cladding into the world volume, the inner cladding into the outer cladding volume and the fibre core into the inner cladding volume, just like it is done in the example http://geant4.cern.ch/support/source/geant4/examples/extended/optical/wls/. The code I use can be found in the attachment (unfortunately, it was not possible to me, to upload a tar-ball). One can switch between the 3 setups (which I will explain below) by commending out lines in the constructors of DetectorConstruction.cc and ParticleSource.cc.
Then, I fired optical photons into it (parallel to the fibre axis, coming from the lower edge of the pictures). As the photons fly parallel to the fibre axis and hit the front and back surface vertically (if they are not absorbed or emitted by the WLS fibre), no refraction should occur and the photons should leave the fibre like they entered it.
In the first setup, the fibre was placed parallel without rotating it. in this case, no problem is visible (http://hypernews.slac.stanford.edu/HyperNews/geant4/get/AUX/2012/08/22/01.29-84866-e1_no_VISIBLE_problem.png). But if I rotate the fibre when placing it (case 2), the problem becomes visible (http://hypernews.slac.stanford.edu/HyperNews/geant4/get/AUX/2012/08/22/01.30-82930-oblem_becomes_visible.png): The photons enter the fibre parallel to its axis but when leaving it, they are diffracted by 90° to the left. This problem can be solved by avoiding the different fibre volumes to share surfaces: If the volumes are all minimal shorter than the volume they are placed in (case 3), this strange behavior of the optical photons disappears (http://hypernews.slac.stanford.edu/HyperNews/geant4/get/AUX/2012/08/22/01.30-93174-case3_problem_solved.png).
I am not an expert, but these findings lead me to the conclusion, that there is a general problem with optical photons at surfaces which are shared by different (rotated) volumes (just like these volumes cause problems when being used in G4UnionSolids or G4SubtractionSolids). If this is the case, this would severely restrict the possible optical setups, that can be simulated with Geant4 without mistakes.
Therefore, I would like to ask you: Is this a bug? Can it be fixed? How? Or do I make any mistake or am I missing something?
|Inline Depth:||Outline Depth:||Add message:|