Problem 565

Summary:	Wrong readout number
Product:	Geant4	Reporter:	farnea
Component:	readout	Assignee:	Makoto.Asai
Status:	RESOLVED INVALID
Severity:	major	CC:	farnea
Priority:	P1
Version:	5.2
Hardware:	PC
OS:	Linux

Description farnea 2004-01-19 03:10:29 CET

I am writing a program to describe an array of segmented germanium detectors.
The "physical" detectors are generated as unsegmented and the actual segment
number is obtained using a readout geometry. I noticed that often an interaction
point is assigned to the wrong segment, or that it is assigned to a segment
even if it is clearly outside all of them. I have checked with a very simple
program (which I can give to anyone who would like to check) that there is
the same problem even with a very simple physical/readout geometry. Most of these
points with mistake are quite far from the segment border (a few mm, which should
be far larger than the tolerances). With the test program, I have used versions
4.5.2.p02 and 4.6.0

Comment 1 asai 2004-01-27 11:35:59 CET

Additional information is being requested.

Comment 2 asai 2004-01-27 15:46:59 CET

The user examined the readout volume with the position of PostStepPoint.
G4TouchableHistory of ReadOutGeometry corresponds to the position of
PreStepPoint. Given readout geometry is a virtual and artificial geometry
which does not affect to the Geant4 tracking, boudary of readout geometry
does not limit a step. Thus, PostStepPoint may not belong to the volume of
readout geometry where PreStepPoint belongs to.