Re: A question on scintillation handling

Dear Filimon,

It is difficult to see what gets wrong just from the back trace; could you run with /tracking/verbose 1 and send me the end of file with a few tracks?
Also, what are the versions of everything (root, geant4, geant4_vmc) and what is your physics list selection?

Best regards,

Ivana

Filimon Roukoutakis wrote:
> Dear Ivana, thanks for all this information. I have started my
> implementation and at the moment I get the following output.
> I assume this happens right when starting to track optical photons
> because by enabling more G4-native verbosity I can see that I have some
> optical photons created (and presumably put in the stack, so the
> scintillation mechanism part seems to work...). Any hint on where to
> look for the problem? The problem occurs in my case even if I comment
> out all my optical geometry code, the thing that triggers it is the
> addition of "+optical" to my TG4Runconfiguration... The N06 example runs
> properly even after enabling scintillation, so I guess I miss something
> from the example in my code.
> filimon
>
> *** Break *** floating point exception
>
>
>
> ===========================================================
> There was a crash (#7 0x00a7136d in SigHandler(ESignals) () from
> /opt/root/v5-27-04/lib/libCore.so.5.27).
> This is the entire stack trace of all threads:
> ===========================================================
> #0 0x006cd422 in __kernel_vsyscall ()
> #1 0x05e997d3 in __waitpid_nocancel () at
> ../sysdeps/unix/syscall-template.S:82
> #2 0x05e3ade3 in do_system (line=<value optimized out>) at
> ../sysdeps/posix/system.c:149
> #3 0x0039327d in system (line=0xbce79c0
> "/opt/root/v5-27-04/etc/gdb-backtrace.sh 16138 1>&2") at pt-system.c:29
> #4 0x00a6981d in TUnixSystem::Exec(char const*) () from
> /opt/root/v5-27-04/lib/libCore.so.5.27
> #5 0x00a6fed5 in TUnixSystem::StackTrace() () from
> /opt/root/v5-27-04/lib/libCore.so.5.27
> #6 0x00a71265 in TUnixSystem::DispatchSignals(ESignals) () from
> /opt/root/v5-27-04/lib/libCore.so.5.27
> #7 0x00a7136d in SigHandler(ESignals) () from
> /opt/root/v5-27-04/lib/libCore.so.5.27
> #8 0x00a66a32 in sighandler(int) () from
> /opt/root/v5-27-04/lib/libCore.so.5.27
> #9 <signal handler called>
> #10 0x0595718a in G4VProcess::AtRestGPIL (this=0xba37360, track=...,
> condition=0xa394118) at
> /home/filimon/geant4.9.3/source/processes/management/include/G4VProcess.hh:456
>
> #11 0x05955cd2 in G4SteppingManager::InvokeAtRestDoItProcs
> (this=0xa394010) at src/G4SteppingManager2.cc:298
> #12 0x05952d5b in G4SteppingManager::Stepping (this=0xa394010) at
> src/G4SteppingManager.cc:160
> #13 0x0595eff3 in G4TrackingManager::ProcessOneTrack (this=0xa393fe8,
> apValueG4Track=0xbce69c8) at src/G4TrackingManager.cc:126
> #14 0x028695f0 in G4EventManager::DoProcessing (this=0xa393fa0,
> anEvent=0xb7bcf00) at src/G4EventManager.cc:185
> #15 0x02869d18 in G4EventManager::ProcessOneEvent (this=0xa393fa0,
> anEvent=0xb7bcf00) at src/G4EventManager.cc:335
> #16 0x052349c0 in G4RunManager::DoEventLoop (this=0xa393e68, n_event=10,
> macroFile=0x0, n_select=-1) at src/G4RunManager.cc:235
> #17 0x052343a8 in G4RunManager::BeamOn (this=0xa393e68, n_event=10,
> macroFile=0x0, n_select=-1) at src/G4RunManager.cc:140
> #18 0x05d96677 in TG4RunManager::ProcessRun (this=0xa3d2ad8,
> nofEvents=10) at run/src/TG4RunManager.cxx:397
> #19 0x05d9af17 in TGeant4::ProcessRun (this=0xa3ca390, nofEvents=10) at
> run/src/TGeant4.cxx:1179
> #20 0x058d7f46 in TMCApplication::RunMC (this=0xa35cca0, nEvents=10) at
> TMCApplication.cpp:126
>
>
> On 09/20/2010 11:05 AM, Ivana Hrivnacova wrote:

>>
>> Dear Filimon,
>>
>> The examples in VMC implement the same setting (the same as it was at 
>> the time when we did this work) as the novice examples in Geant4.
>>
>> The parameters which are in geant4_vmc preceded with the comment:
>> // Added (for Geant3)
>> are not needed in Geant4 (and they are not present in Geant4 native 
>> example N06) and were added in order to get the example working with 
>> Geant3, where all three sets of parameters for Cerenkov process are 
>> always requested (see TVirtualMC::SetCerenkov(...)). Note, that they 
>> have special values (DBL_MAX for absorption and 0. for efficiency).
>>
>> The scintillation is inactivated in the macro as it is also 
>> inactivated in Geant4 native example in the default macro.
>>
>> The Birks correction is implemented in G4Scintillation.cc since G4.9.2
>> (see: 
>> http://hypernews.slac.stanford.edu/HyperNews/geant4/get/opticalphotons/311/1.html) 
>>
>> what was after the implementation of the examples in VMC,
>> that's why the setting of the Birk constant to the material, which you 
>> can find in Geant4 native example is missing in the VMC example.
>> As this setting is done in a different way than the setting of other 
>> material properties, it will need a special support in the VMC 
>> interface. I will have a look at it and let you know when this is 
>> available.
>>
>> Geant3 does not include scintillation process at all.
>>
>> Best regards,
>>
>> Ivana
>>
>>
>>
>> Filimon Roukoutakis wrote:
>>> Dear all,
>>> I am going through example E06 of geant4_vmc tarball. Trying to 
>>> simulate a lead/scint fiber calorimeter (now moving to the optical 
>>> part as the energy deposition seems correct according to data 
>>> comparison) I have an initial question.
>>> In the Ex06DetectorConstruction.cxx file in the ConstructOpProperties 
>>> method there are some comments like "Added (for Geant3)". Since I do 
>>> not know to which set of properties they apply (only for 
>>> efficiency?), my question is, are these only used by G3 back-end, in 
>>> the sense that G4 handles them differently? If yes, by which 
>>> mechanism? I see that in the provided g4config.in file the 
>>> scintillation is deactivated. In my use-case should I then enable 
>>> this (or simply delete the lines) and provide all the G3 information 
>>> that is in the ConstructOpProperties in a G4 format in the g4config 
>>> file or are the fields under "Added (for Geant3)" also applicable for 
>>> G4 (I am obviously interested for the yield/MeV and the slow/fast 
>>> component arrays and ratio/settings). To summarize, if I just remove 
>>> the line about disabling scintillation in the g4.in file, will 
>>> scintillation work correctly for both G3 _and G4 with something 
>>> similar to the example code (provided ofcourse that I use the proper 
>>> values for my scintillating material where applicable...)?
>>> A more advanced but related question is whether there is some 
>>> provision for birk's law under G3 or G4. Any provision/experience on 
>>> this?
>>> Thanks,
>>> filimon
>>

-- 
==============================================================
e-mail:  Ivana.Hrivnacova_at_cern.ch
address: Institut de Physique Nucléaire, 91406 Orsay, France
phone:   +33 169156594
==============================================================