Re: The error

Dear all,
I see no reason why the LAM-BIAS could generate geometry errors. In this
input, the LAM-BIAS is essential in order to obtain interactions. The
messages that are printed in the output should not lead to a crash. Maybe
the problem disappears when running without LAM_BIAS simply becouse there
are almost no interactions. I'll try to have a look
Paola
> Dear Andrzej,
>
> I have two observations about the input files, that you have written
> to calculate differential cross sections of a 1 GeV/u carbon ion beam
> on a gold target (cau8.inp and cau6.inp).
>
> 1) You heavily use the card LAM-BIAS, to reduce the interaction length
> for nuclear inelastic interacions of heavy ions by a factor 10^4.
> A such very large reduction factor is NOT recommended for hadrons,
> as explicitly written in the manual (ch. 7.37).
> As consequence of this, you get a too large (in principle not fatal)
> error messages from the GEOFAR routine, reporting a not correct
> normalization of the director cosines. This happens always close to
> the zero of your coordinate system (the center of your gold target).
> You get about 1 error every 10 events.... so finally you get a crash.
>
> I removed your bias and I could run 10^5 events without problems.
> Obviously without bias you have to run more primary events, but this
> is
> compensated by the fact that the cpu time per event is reduced
> by a very large factor.
>
> 2) You use the STEPSIZE card, setting the maximum step = 1 cm and the
> minumum step = 0.01 cm. This does not make sense in my opinion,
> manly concerning the minimum step size: in your first input file
> you have a gold target 22 micron thick... I would comment this
> card, or I would set only the maximum step size.
>
>
> Finally, a comment about the second plot of the results, that you have
> sent in attachment in your first email. Did you have applied the
> correction by the amount of pions, that are already decayed at your
> boundary crossing?
> This is not a large correction (you used a detector at 30 cm from your
> target), but I would think that it has to be done... I would correct
> bin by bin, because the decay length in the lab system depends on the
> pion energy (you will get a correction of about 2% at low energies
> and of about 0.7% at 700 MeV...).
>
> Hope it helps,
>
> regards,
>
> Anna F.
>
>
>
>
>
>> Dear FLUKA users
>> I tray to calculate the diff.cross sections in heavy ion colision C+Au.
>> and have any error. I can't understand what is reason.
>> In attachment are input file (cau8.inp) and cau8001.err and
>> cau80001.out
>> What I have done badly?
>> Kind regards
>> Andrzej
>>
>

Paola Sala
INFN Milano
tel. Milano +39-0250317374
tel. CERN +41-227679148
Received on Wed Jul 29 2009 - 14:45:57 CEST