I have realized that in my SeensDet class, whenever I ask (the aStep object) for the G4VPhysicalVolume I am in, I get the same pointer! Still, the aStep object knows the correct position, while seemingly pointing to the last G4PhysicalVolume created. Moreover, each time I ask this (same!!) pointer for its mother, it always gives the correct (different!!) answer! It seems as if some 'Optimisation' might cause this, but this sure does not take into account that all these G4VPhysicalVolumes, though sharing the same logical volume relative transformation, might have different copyNo!! I think I can anyway expect that if I ask for a G4VPhysicalVolume, I get the exact thing I have created at that point in space. If the tracking engine is smarter and translates my definition into something else, where all these physical volumes are shared, then it should not pretend to be able to hand me the old object (type).
This behabior is correct and as designed. For getting volume information from the preStepPoint, you have to get it through the touchable object. G4TouchableHistory* aTH = (G4TouchableHistory*)(aStep->GetPreStepPoint()->GetTouchable()); For example, if you want to get the copy numbers of the volume and its mother volume, copyNo = aTH->GetReplicaNumber(); motherCopyNo = aTH->GetReplicaNumber(1); These features are documented in both user's guide and examples.
