Re: Bring computation time down for low energy neutron transport

Hi Ruan
A complete answer needs more details about your problem. However:
- use the NEUTRONS default
- If you are not interested in energy deposition, set EMF-OFF with the EMF
card.
Then: the number of primaries you need depends on what are you trying to
calculate. I mean, the important thing is the statistical error on the
quantity of interest. In Fluka, there are methods to improve the
statistical accuracy without having to simulate millions of primaries. The
way to do this depends on the specific geometry/scoring etc. You can have
a look to
section 2.2.18 of the Fluka manual, and to the "biasing" leture in the
last FLUKA course ( available from the fuka web site --> courses-->
materials)
ciao
Paola

> Hi Fluka experts
> I am a fairly new user to Fluka, so excuse me if this is a common / easy
> problem to solve.
> I am doing a basic simulation on low energy neutrons transport (the
> energy range is between 0.01 MeV to 10 Mev, with a average energy of 2
> MeV). At this stage my code is not very complicated. I just have a point
> source emitting neutrons, that move through various materials e.g. steel
> & concrete, before I detect it.
>
> The problem is that the simulation is very slow, I calculated that the
> simulation for 1000 000 particle, will take up to 5 days.
>
> Can anyone help me with this problem, please! Except for suggesting to
> run it on a faster computer (the computer that I am currently using is
> quite fast in any case)
>
> Thanks
> Ruan
> South Africa
>

Paola Sala
INFN Milano
tel. Milano +39-0250317374
tel. CERN +41-227679148
Received on Mon Oct 13 2008 - 09:53:25 CEST