I am sorry that I am not exactly sure which category this should be in.
I am modeling a fast neutron spectrometer using Geant4 10.6.1. It is an ionization chamber which consists of a cylinder containing high pressure He3 and Ar. It is illuminated with monochromatic neutrons (25meV [thermal], 500keV, 1000keV, 1500keV, 2000keV, 2500 keV) uniformly across the face either at 0 or 90 degrees to the symmetry axis. The spectrometer sits in air. I then calculate total energy deposited in the volume per event and output the results. The resulting spectra look good.
However, I am occasionally getting the following warning message. The rate of occurrence is about 0 to 3 per 10^8 events.
-------- WWWW ------- G4Exception-START -------- WWWW -------
*** G4Exception : hadEla001
issued by : G4HadronElastic::ApplyYourself
hElasticCHIPS wrong sampling t= 11275.2 tmax= 9953.46 for proton ekin=2.35936 MeV off (Z,A)=(2,3) - will be resampled
*** This is just a warning message. ***
-------- WWWW -------- G4Exception-END --------- WWWW -------
The ekin, t and tmax values vary from warning to warning. I can find no documentation on it.
In some of the simulations, I have surrounded the cylinder with a thin cylindrical shield of boron nitride and cadmium to suppress thermal neutrons. Again resulting spectra look good. However, 4 of the 7 runs output the following warning message just before the end of run (run is ended successfully, however).:
" Loop-counter exceeded the threshold value at 378th line of /path-to/geant4/geant4.10.06.p01/source/processes/hadronic/models/particle_hp/src/G4ParticleHPInelasticCompFS.cc. "
I looked in the Geant4 source and found the line of code, but I do not understand what is going on there.
Has anybody come across either of these messages and know what they mean?
hello, John!
I have encounter the the same problem of “Loop-counter exceeded the threshold value at 246th line of /public/backup/geant4/geant4-11.0.2/source/processes/hadronic/models/particle_hp/src/G4ParticleHPLegendreStore.cc.” when I’m doing a simulation about proton beams bombarding a spallation target. The program antomatically stops when this eorror occurs.
Have you solved the problem? Or do you have any clue? Thank you in advance.
I never received any replies regarding the original post. I never pursued the problem further because it was not causing any problems, that I could find, with either the results or execution of the problem and so was only a mild annoyance. You say that your loop counter error causes your program to automatically stop. That was not the case for me. (But note that the errors are generated from different modules.)
My problem occurred for Geant4 10.6.1 and 10.7.0. The warning messages were absent for 11.0.0, but I only tried one run for that version, so I’m not sure if the problem was fixed.
I just completed a longer run using Geant4 11.1.2 with parameters that had previously given me the “–WWWW—” and “Loop-counter” warning errors using Geant4 10.7.0. The two errors are still present (2 in 10^9 and 1 in 10^9 events respectively).
Note that my “Loop-counter” error:
Loop-counter exceeded the threshold value at 355th line of /path-to/geant4/geant4-v11.1.2/source/processes/hadronic/models/particle_hp/src/G4ParticleHPInelasticCompFS.cc
is different from yours.
Loop-counter exceeded the threshold value at 246th line of /public/backup/geant4/geant4-11.0.2/source/processes/hadronic/models/particle_hp/src/G4ParticleHPLegendreStore.cc
So the problem has not yet been fixed in the latest stable release.
Thank you for your reply. I rechecked the output file carefully and noticed that it was not the “Loop-counter” error that caused the program stop. Like you said, it is just a warning message and does not affect the correctness of the results. My simulation stoped because of the HP model. I already sended a bug report on Geant4 websit. Thanks again!
inside Geant4 hadronic model sampling of final state not always ideal. In this case, sampled value of t (relativistic invariant momentum transfer) is unphysical. This value is resampled but before this warning message is printed. This is needed in order to show problematic places in Geant4 codes. In newer versions of Geant4 number of warnings should be reduced or completely gone.