Re: [fluka-discuss]: FLUKA vs GEANT4 B4C efficiency

From: Santana, Mario <msantana_at_slac.stanford.edu>
Date: Fri, 18 Jul 2014 12:31:49 -0700

Hi Georgios,

This would require someone acquainted with both FLUKA and GEANT4 in this type of applications, but here's my two cents:
You are using 'NEW-DEFA' for your DEFAULTS card. I'd think you would be better off using 'PRECISION' specially if your goal is to pinpoint fine discrepancies between results of two codes. With 'PRECISION' you have a "fully analogue absorption for low-energy neutrons", lower thresholds for several reactions and several other more precise default options.
You are requesting the code to compute the splitting of ions into nucleons through the PHYSICS card with IONSPLIT sdum, but you are leaving the default minimum energy for this to be accounted for at 0.1 MeV/nucleon, which is higher than your beam energies, so this card is useless.
Also, you are using the LOW-MAT card for materials like Aluminum, Carbon, etc. for which I am not sure you need it. Finally, if you open your file with flair you will get a number of red flags in the definition of your EVENTBIN detectors (e.g. You don't define the first region to which they apply).

In my opinion the results match rather well, but if you are worried for the 13% difference in the efficiency, maybe you should also look at the statistical error for each of the calculations and see whether the difference is relevant (I just see the RMS error, which is quite high, but I think it applies to the full spectrum distribution).

Mario



From: Georgios Tsiledakis <Georgios.Tsiledakis_at_cern.ch<mailto:Georgios.Tsiledakis_at_cern.ch>>
Date: Tuesday, July 15, 2014 1:04 AM
To: fluka-discuss <fluka-discuss_at_fluka.org<mailto:fluka-discuss_at_fluka.org>>
Subject: [fluka-discuss]: FLUKA vs GEANT4 B4C efficiency

Dear experts,

Please have a look at the attached pdf&input file...
A perpendicular thermal neutron beam hits a 2 micrometer B4C layer (B10)
placed in front of a gaseous volume of CF4 (100k primaries).
-The slide 1 shows the material used in Fluka/Geant4 respectively.
-Slides 2-3==> The energy deposition in the gas with/without energy detection thr.
 The He4&Li7 contributions as well as the response when 100k g of 500 keV hit the
 B4C layer - FLUKA MC.

-Slides 4-5==> The same spectra BUT obtained by Geant4.

-Slide 6==>Comparison of the 2 MC codes at Ethr > 10 keV.

My questions are the following:

1) Why there is this 13% difference in the efficiencies?

2) Why we cannot obtain in FLUKA the electrons or gammas distribution superimposed together
     with Li and He?
     The alphas in Geant4 close to E = 0 are also 0 contrary to Fluka.

Does this difference comes due to the :
- Difference cross section in B10 for the 2 codes?
- Does the dE/dX for for very small energies+very small thickness for e,g can give
   such discepancies?
- What's code is the right estimation for this study?

I am looking forward to hearing from you

Best regards

Georgios
Received on Fri Jul 18 2014 - 23:00:34 CEST

This archive was generated by hypermail 2.3.0 : Fri Jul 18 2014 - 23:00:37 CEST