Problem 769

Summary: track in unusable state
Product: Geant4 Reporter: zh.zhang
Component: processes/hadronicAssignee: Hans-Peter.Wellisch
Status: RESOLVED FIXED    
Severity: normal CC: zh.zhang
Priority: P2    
Version: 7.0   
Hardware: PC   
OS: Linux   

Description zh.zhang 2005-07-12 13:04:37 CEST 
I use GEANT4.7.0 without patch to simulation antiproton annihilation with gold
and reconstruct the trajectory of charged measons. I use the physics list from
the copy of novice exampleN04. During running of the programme, a warning
message issued by GEANT as follows:

>>> Current event ID is 2900;    146 trajectories stored in
 this event.
>>> Current event ID is 3000;    108 trajectories stored in
 this event.
 G4HadronicProcess: track in unusable state - 1
 G4HadronicProcess: returning unchanged track

 *** G4Exception : 001
       issued by : G4HadronicProcess
 bailing out
 *** This is just a warning message.
  >>> Current event ID is 3100;    3 trajectories stored in this event.
Comment 1 dennis.herbert.wright 2005-07-20 12:40:59 CEST 
I will try to reproduce your problem.  Please send a tar file of your code to me:
dwright@slac.stanford.edu
I will compile it against Geant4 7.0 and go from there.
Comment 2 Hans-Peter.Wellisch 2005-08-08 03:42:59 CEST 
Hi, this one was assigned incorrectly.
Comment 3 Hans-Peter.Wellisch 2005-08-08 03:49:59 CEST 
Hi,

  this is a feature, rather than a bug. What happens is this: A low energy
charged particle is being tracked. Based on the inelastic cross-section,
hadronics decides that the particle should undergo an inelastic interation, and
limits the step. Due to fluctuations the particle comes to a rest before
reaching the designated locus of the interaction. At rest, the inelastic
process can of course not handle secondary production. It bails out.

The warning is there only, because if this is a frequent condition, then the
stepping approximation is not good for your problem, and you will want to use
user-defined step limits to make smaller steps by hand. Note also that I have
never seen such a case to date.

Many greetings,

Hans-Peter.