Created attachment 19 [details] see text Hi, My collegues and I discover a problem with neutron calculations at low energies (<10keV). We are using thermal scattering and the HP-physics. The results differ quite much between the versions 8.2 and 9.0. We have comparison data, calculated with MCNP which hints strongly that the results of the 8.2 version are the correct ones. What was changed in between??? Is there a connection to the error occuring in calculations sometimes: G4HadFinalState: fatal - negative energy (see my hadronics-forum question 770)? Do you do verify the consistency of new versions for low energy neutrons? We would appreciate very much if this problem ist solved soon, as all three of us depend on the neutron physics. Sylvia Studeny Attachment: 1)Response of a REM-counter for different versions of Geant4 and MCNP, direct respone calculations and foldung of neutron flux for version 8.2 2) Neutron flux inside a 6-inch sphere of PE at different Energies
Created attachment 20 [details] Response of bonner sphere to second attachment: Response of 6" bonner sphere, comaprsion between version 8.2 and 9.0 generally: My collegue also did calculations with version 9.1 for the REM-counter. These gave the same results as version 9.0 (which are wrong) Sylvia
Before version 8.3 the hadronic processes did not pay attention to user-defined isotope abundances for elements. Cross sections were calculated according to natural abundances only. This was fixed in 8.3 and some refinements made in version 9.0, so that you should expect to see some difference in your results.
