FLUKA: Re: copper cross sections

[Alfredo Ferrari <alfredo.ferrari@cern.ch>]

Hi all

Graham and Ian pointed out a problem in the copper self-shielded cross
sections for neutrons below 20 MeV at 293 K.
                  They are right, I investigated the problem which turned
out to be a bug in NJOY which was fixed in 1996, while the cross sections
for copper were generated in 1995. I was thinking I checked all data sets
prior to 1996 for the bug, but clearly I missed Copper because usually the
bug shows up in a much more gigantic way (among the other effects gamma
production was exponentially increasing in the lowest energy groups).

I reprocessed all Copper data, both 293 and 87, infinite dilution and
fully self-shielded (SS), just for consistency even though the problem
was only in the 293 SS ones.

I'll put updated tar files for all Fluka flavours within tomorrow
(hopefully, I'll send an e-mail just when completed).
They will not contain ANY OTHER DIFFERENCE but this fix in the cross
section file and an up-to-date routine in the library for reading the
cross section file since meanwhile we changed somewhat its format.

So, upgrade safely you should get exactly the same results (and crashes)
as now, of course if you are not using copper cross sections...

A real new release including most of the improvements shown at MC2000
will come possibly the next month. I was hoping to be able to release it
just after Christmas but I have overwhelming duties in a new proposal for
ICARUS which is due in few weeks.

                   Ciao!
                  Alfredo

P.S. I will stop to put the new releases on my AFS public areas since the
new official Fluka web page (www.cern.ch/fluka) is now tested and
everybody can easily download the same files from there.


On Tue, 6 Feb 2001, Ian Dawson wrote:

> hi Alfredo, there's a problem with 72 copper SS cross sections
> at 293K (CuSS293). I've summarised some comparisons below.
>
>
> 1) Cu87 and Cu293 are the same except for thermal bin (as expected!)
>
> 2) Cu87 and CuSS87 show reasonable differences in resonance region
>    (ie SIGT for CuSS87 < SIGT for Cu87). Same values for other bins,
>    except small difference in thermal bin that should have same value.
>    (=>OK?)
>
> 3) Cu293 and CuSS293 have identical values of SIGT but different
>    values of GAMGEN for all bins (=>wrong!).
>
> 4) CuSS87 and CuSS293 have many different values of SIGT and GAMGEN
>    when only thermal bin should be different (=>wrong!).
>
> Therefore the problem is in CuSS293. The CuSS87 cross sections look OK so
> I will use these until 293 ones are repaired? ... ?
>
>
> Cheers, Ian
>
>
>
>

-- 

+----------------------------------------------------------------------------+
|  Alfredo Ferrari                ||  Tel.: +41.22.767.6119                  |
|  CERN-SL                        ||  Fax.: +41.22.767.7555                  |
|  1211 Geneva 23                 ||  e-mail: Alfredo.Ferrari@cern.ch        |
|  Switzerland                    ||          Alfredo.Ferrari@mi.infn.it     |
+----------------------------------------------------------------------------+