Dear FLUKA experts,
in my simulation I have a neutron beam, which up to now has always been monoenergetic at 25meV. In the settings of my simulation I use a GLOBAL “Analogue” card, DEFAULT PRECISIO and two PHYSICS cards with COALESCE “On” and EVAPORAT “New Evap with heavy frag”, and I don’t have any problems running the simulation.
Now with the use of source_newgen.f I’m trying to implement this neutron energy distrbution:
The executable file seems fine but when I run the simulation I get this error:
Instead, if I deactivate the PHYSICS COALESCE card, the simulation runs but it gives questionable results, like these strange scattering-like lines in the neutron fluence scoring with USRBIN:
From what I understand, the limit of QME should be 0.1 GeV and since the maximum energy of my beam is around 300keV I can’t figure out what’s wrong.
I attached the .flair and the source_newgen.f of my simulation below.
It is crashing because you are not calling the DPMjet library, however there is no need to use the coalescence card in your problem so you can disable it.
Concerning the scattering-like lines in your scoring, it might be due to the group wise treatment of low energy neutrons that is implement in FLUKA. The scattering angle are discretized by group of energy. (you can see more info in the manual).
When activating the newly point wise treatment (using the card LOW-PWXS) I don’t observe any more of those lines. You will need the newest version FLUKA 4.3 (available since september).
Dear Jean-Baptiste,
thank you for your answer. It seems strange to me since up to now I did every simulation without the LOW-PWXS option (I had the older FLUKA), and with a monoenergetic neutron beam at 25meV there was no angle discretization.
Is there any way to get rid of the angle discretization problem without using a fully pointwise treatment with LOW-PWXS?
When simulating a monoenergetic neutron beam at 25meV I do observe the same line artifact even-though it is of lower intensity.
This is probably due to you enabling the free-gas approximation in Hydrogen to the lowest energy (which will give some mistakes in your thermal flux). This approximation treats the neutron interaction point wise in Hydrogen.
Still, a fraction of thermal neutron will experience a recoil in Boron (very low fraction indeed as it will compete against absorption), those ones will be treated group wise.
At higher energy this phenomenon will be more visible as more neutrons will inelastically interact with Boron.
The CPU time “lost” using fully pointwise treatment is only of around 30%. As it is the correct way to go, may I ask why you would like to use the group wise treatment ?
Dear Jean-Baptiste,
thank you for your answer. I asked about the group wise treatment just in case I wanted to compare my old simulations to the new ones without running them again, since they were done with FLUKA 4.2.x version.
Anyway, I thought that the point wise CPU penalty was much higher so I will definetly use the pointwise approach from now on.
