Re: Particle biasing

From: Stefan Roesler <sroesler_at_mail.cern.ch>
Date: Sat, 18 Oct 2008 18:50:13 +0200

Hi David,

Default thresholds for electrons and gammas are mentioned in the Manual
in a Note under Emfcut:

    1) Default values are available for the electron and photon production
       thresholds in electromagnetic interactions, but they are generated
       by a complex logic based on possible other settings (transport
       cut-offs, delta-ray production thresholds, DEFAULTS card, other
       defaults). It is not always guaranteed that the resulting values be
       appropriate to the problem of interest. Therefore, in order to have
       good control of the physics, the user is recommended to provide
       explicit threshold values, or at least to check them on the main
       output.

In order to be sure what is applied it is always useful to check the
output file. Production cutoffs are listed here, e.g.:

 STEEL
      Rho = 8.00000 g/cm**3 Rlc= 1.71175 cm
      Ae = 1.51100 MeV Ue = 5083.24 MeV
      Ap = 0.333333 MeV Up = 5082.73 MeV
      dE/dx fluctuations activated for this medium, level 1
      below the threshold for explicit secondary electron production
     (up to 2I discrete levels, up to 2 K-edges)

and transport cutoffs just below

          6 2 STEEL 28 STEEL
     Ecut = 1.5110E+00 MeV, Pcut = 3.3333E-01 MeV, BIAS = F,
Ray. = F, S(q,Z) = T, Pz(q,Z) = F

As you can see the default values in your case are 1MeV kinetic energy for
electrons/positrons (the above energies are total energy) and 333keV for
gammas.

What to use for shielding calculations depends on the problem (beam
particle, energy, geometry). Since you go as low as 10keV in your source
routine your thesholds should probably be set to the lower limit in FLUKA,
i.e. 1keV.

Cheers
Stefan

On Sat, 18 Oct 2008, David Horvath wrote:

>
> Dear Stefan and Alfredo,
>
> thank you for looking into my problem.
>
> I made also a run with BIASING WHAT(3) = 1.0, but with this setting the
> usimbs.f is ignored, and no biasing is applied.
>
> Stefan:
>
> And thank you for pointing out the inappropiate production and transport
> cut-off settings.
> Can you tell me what are the default cut-off enegries, I couldn't find
> them in the manual.
>
> I tried out the following settings:
> EMFCUT -1D-6 1D-6 0.0 AIR @LASTMAT
> PROD-CUT
>
> EMFCUT -1D-3 1D-6 0.0 BlckHole @LASTREG
>
> Am I right to set a high electron transport cut-off energy to a
> shielding calculation?
>
> Cheers,
> David
>
>
> Stefan Roesler =EDrta:
> > Hi Alfredo,
> >
> > I looked into the problem of David and found some strange things:
> >
> > If I set Biasing/What(3)=3D'blank' and Sdum=3D'USER', e.g.
> >
> > BIASING 0.0 1. BlckHole @LASTREG USER
> >
> > I get a stop
> >
> > Abort called from DOEIMB reason .NOT.LIMPRE Run stopped!
> > STOP .NOT.LIMPRE
> >
> > I have the impression that the code thinks there is no importance biasing
> > and is missing some initialization(?)
> >
> > Furthermore, if I set What(3)=3D1 and use David's usimbs I get consistent
> > results with the other two cases while if I set What(3)=3D2 I can confirm
> > Davids observations. Taking into accound that usimbs sets FIMP either 1 or 0
> > 2 it looks stange to me that this should be due to biasing-artifacts. What
> > do you think?
> >
> > btw, to David: the default thresholds for transport and production of
> > gamma and e+/e- might not be suitable (too high) as you have primaries
>
> > down to 10keV(?)
> >
> >
> > Cheers
> > Stefan
>
>

-- 
___________________________________
Stefan Roesler
CERN, SC/RP
CH-1211 Geneva 23
Switzerland
Phone:  +41-22-7679891
Fax:    +41-22-7669639
E-mail: Stefan.Roesler_at_cern.ch
Received on Sat Oct 18 2008 - 19:54:03 CEST

This archive was generated by hypermail 2.2.0 : Sat Oct 18 2008 - 19:54:04 CEST