Message: Infinite loop in tracking when using v10.03.p01 from cvmfs on lxplus7 Not Logged In (login)
 Next-in-Thread Next-in-Thread
 Next-in-Forum Next-in-Forum

Question Infinite loop in tracking when using v10.03.p01 from cvmfs on lxplus7 

Forum: Event and Track Management
Date: 08 May, 2018
From: Gia Khoriauli <Gia Khoriauli>

Hello,

I have an issue of the infinite loops in tracking of different types of particles in the hadronic calorimeters of the NA62 experiment. The simulation of the calorimeters has no problem with the Geant4 version 10.01.p02 (cvmfs installation) under the slc6. But when using version 10.03.p01 under the CentOS7:

  /cvmfs/sft.cern.ch/lcg/releases/LCG_92/Geant4/10.03.p01/x86_64-centos7-gcc62-opt/

infinite loops in tracking starts to appear very often in the calorimeters. An infinite loop starts on a boundary of a daughter volume and the particle location starts alternating between the boundary of the daughter and the mother volume. A step size of the track on the boundary of a daughter volume is always zero and its very small in the mother volume. Bellow is a typical case of the infinite loop, where MUV1 is the mother volume and MUV1RubberPlate is its daughter volume.

*********************************************************************************************************
* G4Track Information:   Particle = gamma,   Track ID = 6240,   Parent ID = 2
*********************************************************************************************************

    1  -44.2 cm   53.2 cm    244 m     511 keV     0 eV   1.59 cm   1.59 cm   MUV1RubberPlate  Transportation
    2  -38.6 cm   55.2 cm    244 m     511 keV     0 eV   5.97 cm   7.55 cm         MUV1  Transportation
    3  -38.6 cm   55.2 cm    244 m     511 keV     0 eV      0 fm   7.55 cm   MUV1IronPlate  Transportation
    4  -38.6 cm   55.2 cm    244 m     511 keV     0 eV    767 fm   7.55 cm         MUV1  Transportation
    5  -38.6 cm   55.2 cm    244 m     511 keV     0 eV      0 fm   7.55 cm   MUV1IronPlate  Transportation
    6  -38.6 cm   55.2 cm    244 m     511 keV     0 eV    767 fm   7.55 cm         MUV1  Transportation
    7  -38.6 cm   55.2 cm    244 m     511 keV     0 eV      0 fm   7.55 cm   MUV1IronPlate  Transportation
    8  -38.6 cm   55.2 cm    244 m     511 keV     0 eV    767 fm   7.55 cm         MUV1  Transportation
    9  -38.6 cm   55.2 cm    244 m     511 keV     0 eV      0 fm   7.55 cm   MUV1IronPlate  Transportation
   10  -38.6 cm   55.2 cm    244 m     511 keV     0 eV    767 fm   7.55 cm         MUV1  Transportation
 ...

The 2nd, 3rd and 4th columns are the X,Y,Z coordinates of the step points; The 5th column is the energy of the photon; the 6th and 7th columns are the deposited energy and the step size; the track length is the 8th column. The MUV1 detector consists of alternating layers (along the beam-axis == z-axis) of iron and rubber plates and scintillator layers. In order to have a better idea about the detector geometry I attached a printout of the geantino scan of the detector, where geantino momentum was set to be parallel to the z-axis. There are no overlaps between the volumes and no extrusions. As I already mentioned in the beginning, there was no issue with the simulation of the calorimeters, when using the earlier version of Geant4.

Does this issue looks like an already known problem? If it's, then is it fixed in the latest version (10.4) of Geant4? Please, let me know if I need to provide more details. I would like to ask also for a suggestion what would be bes solution of this problem. Should one set, for example, some limits on the minimum length of tracks via G4UserLimits class?

Thank you very much for your time.

Best regards, Gia

   Attachment:
      http://hypernews.slac.stanford.edu/HyperNews/geant4/get/AUX/2018/05/08/10.50-38289-muv1_geantinoscan.txt

Inline Depth:
 1 1
 All All
Outline Depth:
 1 1
 2 2
 All All
Add message: (add)

1 More: Re: Infinite loop in tracking when using v10.03.p01 from cvmfs on lxplus7   (Gabriele Cosmo - 09 May, 2018)
(_ Feedback: Re: Infinite loop in tracking when using v10.03.p01 from cvmfs on lxplus7   (Gia Khoriauli - 25 May, 2018)
 Add Message Add Message
to: "Infinite loop in tracking when using v10.03.p01 from cvmfs on lxplus7"

 Subscribe Subscribe

This site runs SLAC HyperNews version 1.11-slac-98, derived from the original HyperNews


[ Geant 4 Home | Geant 4 HyperNews | Search | Request New Forum | Feedback ]