Problem 1344

Summary: aTrack->GetGlobalTime() for secondaries in hadElastic processes
Product: Geant4 Reporter: Sergey <morozovs>
Component: trackAssignee: kurasige
Status: RESOLVED DUPLICATE    
Severity: normal    
Priority: P5    
Version: other   
Hardware: PC   
OS: Linux   

Description Sergey 2012-08-13 12:52:43 CEST 
Dear GEANT4 developers. 

I'm currently working with Mokka (http://polzope.in2p3.fr:8081/MOKKA) but the problem I have found is more general and concerning GEANT4 itself. First of all I would like to say that I have a differences it timing of secondaries generated in hadElastic processes in GEANT4 versions: 9.4.p03 and 9.5.p01. Detailed analysis of tracks and steps shows the 9.4.p03 gives _correct_ GetGlobalTime() of secondaries from hadElastic but 9.5.p01 put as global time something like the local time of a given secondary from a time at postStepPoint of parent track where the particle (neutron mostly) has been created. To illustrate this there is an example of /track/verbose printout from 9.5.p01:

*********************************************************************************************************
* G4Track Information:   Particle = neutron,   Track ID = 425,   Parent ID = 409
*********************************************************************************************************

Step#    X(mm)    Y(mm)    Z(mm) KinE(MeV)  dE(MeV) StepLeng TrackLeng  NextVolume ProcName
    0    -1.73     46.1 1.91e+03      1.01        0        0         0 AbsorberLayerPhys7 initStep
Track global time 9.328911271832824 ns
    1      7.4     62.9 1.91e+03      1.01        0     19.6      19.6 WholeScinCassettePhys6 Transportation
    2     10.2       68  1.9e+03      1.01        0     5.94      25.5 HcalScinHousPhys Rear Transportation
    3     14.6     76.2  1.9e+03      1.01        0      9.5        35 HcalSensLayerPhys Transportation
    4     14.8     76.7  1.9e+03      1.01        0    0.546      35.6 HcalSensLayerPhys6 Transportation
    5     25.9     97.1  1.9e+03      1.01        0     23.7      59.3 HcalSensLayerPhys Transportation
    6     26.2     97.5  1.9e+03      1.01        0    0.546      59.9 HcalSensLayerPhys Transportation
    7     27.7      100  1.9e+03     0.562        0     3.39      63.3 HcalSensLayerPhys hadElastic
    8     28.3      101  1.9e+03     0.562        0    0.643      63.9 HcalCFmix_LayerPhys Transportation
    9     31.1      102  1.9e+03     0.562        0     3.37      67.3 HcalScinHousPhys Front Transportation
   10     34.9      103 1.89e+03     0.562        0      4.5      71.8 WholeScinCassettePhys6 Transportation
   11     37.3      104 1.89e+03     0.562        0     2.81      74.6 AbsorberLayerPhys6 Transportation
   12     42.4      106 1.89e+03      0.55        0     6.08      80.7 AbsorberLayerPhys6 hadElastic
   13       45      107 1.89e+03      0.55        0     4.11      84.8 WholeScinCassettePhys6 Transportation
   14     46.2      108 1.89e+03      0.55        0      1.9      86.7 HcalScinHousPhys Front Transportation
   15     48.1      110  1.9e+03      0.55        0     3.03      89.7 HcalCFmix_LayerPhys Transportation
   16     49.5      110  1.9e+03      0.55        0     2.28        92 HcalSensLayerPhys Transportation
   17     50.4      111  1.9e+03      0.55        0     1.52      93.5 HcalSensLayerPhys Transportation
   18     50.5      111  1.9e+03      0.55        0    0.175      93.7 HcalSensLayerPhys6 Transportation
   19     52.1      112  1.9e+03     0.527        0     2.57      96.2 HcalSensLayerPhys6 hadElastic
   20       53      112  1.9e+03     0.527        0     3.42      99.7 HcalSensLayerPhys Transportation
   21       53      112  1.9e+03     0.527        0    0.119      99.8 HcalScinHousPhys Rear Transportation
   22     53.5      112  1.9e+03     0.527        0     2.07       102 WholeScinCassettePhys6 Transportation
   23     53.8      112 1.91e+03     0.527        0     1.29       103 AbsorberLayerPhys7 Transportation
   24     54.6      111 1.91e+03         0        0     3.32       106 AbsorberLayerPhys7 NeutronInelastic

...
then after 
...

*********************************************************************************************************
* G4Track Information:   Particle = C12[0.0],   Track ID = 432,   Parent ID = 425
*********************************************************************************************************

Step#    X(mm)    Y(mm)    Z(mm) KinE(MeV)  dE(MeV) StepLeng TrackLeng  NextVolume ProcName
    0     52.1      112  1.9e+03    0.0222        0        0         0 HcalSensLayerPhys6 initStep
    1     52.1      112  1.9e+03         0   0.0222 0.000212  0.000212 HcalSensLayerPhys6 ionIoni
Step in HcalWholeSensLayerLogical6  neutron from trackID = 425,  Energy registered: 2.386724340070652 keV
,  Time postStep: 7.323298874313955 ns
,  Time trk: 7.323298874313955 ns
,  Time parentTrk: 9.328911271832824 ns

The last printout is from UserSteppingAction, parentTrk time has been kept on the previous track (exactly the same as for neutron Track ID = 425 at initial step - see a printout above made from PreUserTrackingAction).

So something has been changed from 9.4.p03 to 9.5.p01 in hadElastic timing of secondaries.
Comment 1 Sergey 2012-08-13 14:56:14 CEST 
May be this can help. I have found a change here:

source/processes/hadronic/management/src/G4HadronicProcess.cc

in 9.4.p03:
-------------------------------------------------------------------------------
  for(G4int i=0; i<aR->GetNumberOfSecondaries(); ++i)
  {
    G4LorentzVector theM = aR->GetSecondary(i)->GetParticle()->Get4Momentum();
    theM.rotate(rotation, it);
    theM*=aR->GetTrafoToLab();
    aR->GetSecondary(i)->GetParticle()->Set4Momentum(theM);
    G4double time = aR->GetSecondary(i)->GetTime();
    if(time<0) time = aT.GetGlobalTime();

    G4Track* track = new G4Track(aR->GetSecondary(i)->GetParticle(),
				 time,
				 aT.GetPosition());
------------------------------------------------------------------------------


in 9.5.p01:
------------------------------------------------------------------------------
  // check secondaries: apply rotation and Lorentz transformation
  G4int nSec = aR->GetNumberOfSecondaries();
  theTotalResult->SetNumberOfSecondaries(nSec);
 
  if(nSec > 0) {
    G4double time0 = aT.GetGlobalTime();
    for(G4int i=0; i<nSec; ++i)
      {
	G4LorentzVector theM = aR->GetSecondary(i)->GetParticle()->Get4Momentum();
	theM.rotate(rotation, it);
	theM *= aR->GetTrafoToLab();
	aR->GetSecondary(i)->GetParticle()->Set4Momentum(theM);
	G4double time = aR->GetSecondary(i)->GetTime();
	if(time<time0) { time = time0; }

	G4Track* track = new G4Track(aR->GetSecondary(i)->GetParticle(),
				     time,
				     aT.GetPosition());
------------------------------------------------------------------------------
Comment 2 Sergey 2012-08-13 15:31:53 CEST 
Well, as I see from 9.6.b01 (beta version) this piece of code has been improved:

-----------------------------------------------------------------------------

      // time of interaction starts from zero
      G4double time = aR->GetSecondary(i)->GetTime();
      if (time < 0.0) { time = 0.0; }

      // take into account global time
      time += time0;

      G4Track* track = new G4Track(aR->GetSecondary(i)->GetParticle(),
                                   time, aT.GetPosition());
-----------------------------------------------------------------------------

So GEANT4 knows about this problem and fixed it for the future releases..
Comment 3 kurasige 2012-08-14 10:48:33 CEST 
Hi

The problem has been reported #1305
http://bugzilla-geant4.kek.jp/show_bug.cgi?id=1305

The fix will be included in following patch and  future release.

Thanks

*** This problem has been marked as a duplicate of problem 1305 ***