I am currently attempting to simulate the Frascati Neutron Generator (FNG) model using MCNP and compare the results with those obtained from FLUKA.
In this model, the FNG team provided an MCNP source definition file, which I used as a reference to create the source description in FLUKA’s source_newgen.f file. In FLUKA, I enabled point-by-point low-energy neutron handling for each material using the LOW-PWXS card and set the database to ENDF-VIII.0. I used the same database in MCNP as well. Afterward, I performed calculations with both MCNP and FLUKA using the same 175 energy groups and compared the neutron fluxes across different energy ranges.
The results show that FLUKA calculates a significantly higher neutron flux below 1 MeV compared to MCNP. Additionally, the material activation and shutdown dose rates calculated by FLUKA differ from results obtained by other software and experimental data, which has left me somewhat confused.
Could you please clarify whether FLUKA handles low-energy neutrons differently from MCNP? I would also appreciate your help in identifying the potential causes of the flux discrepancy. What additional files would be helpful for your review?
Please note that the “MCNP” in the figure legend refers to the results from MCNP calculations, while T-426 represents the experimental team’s calculations. The “MCNP” calculations use the same database as FLUKA, whereas the T-426 team used MCNP with a different database, which explains the differences between the “MCNP” and T-426 results.
Thank you for your clarification and guidance. I have gathered the requested files and attached them for your review. Please find the following files included:
Source data file:SPEC_1.TXT ->SPEC_19.TXT
source_newgen_updated.f file
Input file: fng_neutron_trans.inp
Should you require any additional information or have further questions, please feel free to let me know. I appreciate your time and assistance.
I hope this message finds you well. I would like to kindly inquire if there has been any progress regarding the issue I mentioned earlier. I have reviewed some similar questions on the FLUKA forum, and it seems that the treatment of neutron cross-sections below 20 MeV in FLUKA is indeed different from that in software like MCNP, which appears to be the cause of the discrepancy.
Thank you for your attention, and I look forward to your feedback.
thank you for the detailed questions and input files. I am currently looking into this issue that you see; apologies for any delay, your input and source are quite complicated. I have begun work in reproducing your results, and try to find possible issues with your inputs (if there are any).
Would you also be able to share in .txt form the results that are plotted in the figure in your initial post?
Second, let me understand the procedure correctly:
you have a simulation geometry from FNG
you use the source_newgen.f file to sample a diferential spectrum (energy-angle), which is split in 19 angular bins and 43 energy bins.
You then score neutrons in region C620, with a different energy interval and a single bin in energy. If I understand correctly, the plot you showed above is the result of each of these scoring USRTRACK, used as data points (x position I assume the arithmetic center of the bin)
I will let you know as soon as I have better clarity on this. If the error stems in fact from a difference in the treatment, I will keep you informed.
Thank you very much for your attention to this issue. I believe this will greatly help improve FLUKA’s handling of neutron transport problems for neutrons below 20 MeV, or assist others who encounter similar issues in understanding how to use FLUKA for such tasks, as there has been little relevant experience to refer to before now. I am more than happy to collaborate with you in investigating the cause of this discrepancy.
Attached, I have uploaded the raw data of the images mentioned in my initial post (if I understood correctly, the .txt format results refer to the raw data).
The FNG geometry in FLUKA has been converted from the MCNP geometry, and they are identical in terms of size and materials.
I replicated the source in MCNP using source_newgen.f and tested the difference between the source in FLUKA and MCNP. The small model is shown below. The neutron fluxes for the three cells, as computed in FLUKA, are: 8.3696E-01, 1.6538E+00, and 8.3956E-01, respectively. In MCNP, the results are: 8.1952E-01, 1.6484E+00, and 8.5863E-01. Based on these results, I believe that source_newgen.f is functioning correctly.
For the neutron transport problem in deuterium-tritium fusion reactions, the 175-group energy structure is commonly used. Therefore, when I performed the neutron statistics for the C620 region, I created 175 USRTRACK cards and ensured their energy limits corresponded to the bins used in MCNP. The flux data and the energy bin limits for each bin are provided in the neutron flux.txt file. I used this data to plot the horizontal staircase diagram above.
