Problem 2443

Summary: Peculiarity with kinetic energy for charged fragments in G4Evaporation
Product: Geant4 Reporter: Aleksandr Svetlichnyi <aleksandr.svetlichnyy>
Component: processes/hadronic/models/de_excitation/evaporationAssignee: Vladimir.Ivantchenko
Status: RESOLVED FIXED    
Severity: normal CC: Alberto.Ribon
Priority: P4    
Version: 10.4   
Hardware: PC   
OS: Linux   
Attachments: Evaporation spectra for neutrons, protons and alpha-particles

Description Aleksandr Svetlichnyi 2021-11-10 12:37:50 CET 
Created attachment 738 [details]
Evaporation spectra for neutrons, protons and alpha-particles

Proton and alpha particles spectra possess a long "tail" of the distribution that is not typical for an evaporation spectra, see an attachment. Default method for kinetic energy sampling in G4Evaporation is an inherited method from G4VEmissionProbability G4VEmissionProbability::SampleEnergy() (red histogram in the attached picture), probably some bugs can be there.  

For the details see: https://geant4-forum.web.cern.ch/t/strange-distribution-of-kinetic-energy-of-evaporated-fragments-in-g4evaporation/6505
Comment 1 Vladimir.Ivantchenko 2021-12-15 17:04:44 CET 
Hello Aleksandr,

thank you for this report. We need to improve our evaporation code. 

However, we cannot do anything about Geant4 10.4. With the current manpower and LHC schedule we can introduce improvements to Geant4 11.0 and/or back-port fixes to Geant4 10.7. Geant4 has no possibility to make official patches even to Geant4 10.6. 

After 10.4 there were several updates of the code, so it is difficult make assumptions what to change or improve. If you strongly coupled with Geant4 10.4 I would think about possibility to back-port de-excitation directory from 11.0 to 10.4. In that case we can improve de-excitation.

VI
Comment 2 Vladimir.Ivantchenko 2022-09-29 19:53:05 CEST 
Hello Aleksandr,

the problem is fixed and the fix will be available with the next public release in December 2022. The problem was in low limit on number of attempts to sample energy of the emitted fragment. The limit is increased and more smart regression method is used.

Vladimir