EMF Fluka

From: Alberto Fasso <fasso_at_mail.cern.ch>
Date: Wed, 16 Sep 2009 19:40:16 +0200 (CEST)

Dear Joerg,

you should have sent your question to the FLUKA discussion list. We
discourage sending questions personally to individual authors, for two
1) the answer can be of interest to other users
2) the person addressed might not be the best informed about the particular
    issue (as in the present case)

Anyway, I will try to answer as well as I can, but I take the liberty to
forward the answer also to the discussion list.
I have given some contribution to the EMF part of FLUKA, but the flag
you are asking about is related to Multiple Coulomb Scattering (MCS),
which doesn't concern only electrons and positrons but also charged hadrons
and muons (if you look at your output, you will see that a similar
statement is printed about hadrons and muons just above the one you have
reported). The FLUKA Multiple Coulomb Scattering algorithm is due to
Alfredo Ferrari and Paola Sala, and they could answer your question much
better. They have experimented very thoroughly with various ways to handle
MCS near boundaries, where analytical theories fail. One of these ways
(about which I know nothing about) can be activated by WHAT(2) in command
MULSOPT, with SDUM = GLOBEMF or GLOBAL. I copy from the manual:
    WHAT(2) : index of step stretching factor tabulation to be used
              by the electron/positron transport algorithm when
              approaching a boundary.
              The values of the index implemented for the moment are 1,2,3,4.
              Values 11,12,13,14 cause the sensing algorithm to multiply
              the range/mcs step rather than the current step.
              Values 101,111,102,112,103,113,104,114 have the additional
              effect of making the algorithm resample as unphysical any
              step cut at a boundary and "reflected" from the boundary.
              = 0.0 : ignored
              < 0.0 : the tabulation index is reset to default
              Default: WHAT(2) = 1 (maximum accuracy)
The flag you are asking about becomes T (true) if you set WHAT(2) = 111, or
any value in 101,111,102,112,103,113,104,114.

You can experiment with it, but I have the impression that most of the above
has been made more or less obsolete by the introduction in FLUKA of single
scattering in 1995 (see WHAT(4) and WHAT(6) of the same command, again with
SDUM = GLOBEMF or GLOBAL). That is probably the best way to handle MCS at
and near boundaries.

Unfortunately, as far as I know no documentation is available beyond the
above command description.

Best regards,


On Wed, 16 Sep 2009, Jörg Wulff wrote:

> Dear Alberto,
> since are one of the FLUKA EMF authors, I address an EMF related question
> right to you, hoping you can find the time to answer.
> As a FLUKA novice, I was playing around with some electron beams. In the
> output-files of any run I find:
> ...
> === Electromagnetic Showers: ===
> EM showers are treated by the EMF (A.Fasso`,A.Ferrari,P.R.Sala) code
> Electrons/positrons:
> Flag for MCS check with boundary normals: F
> ...
> I could not find any documentation about "Flag for MCS check with boundary
> normals". Actually I do not know what the check is used for. Can you tell
> me or give a reference? How do I turn it on?
> Best regards,
> Joerg

Alberto Fasso`
SLAC-RP, MS 48, 2575 Sand Hill Road, Menlo Park CA 94025
Phone: (1 650) 926 4762   Fax: (1 650) 926 3569
Received on Fri Sep 18 2009 - 17:44:24 CEST

This archive was generated by hypermail 2.2.0 : Fri Sep 18 2009 - 17:44:28 CEST