- Contemporary messages sorted: [ by date ] [ by thread ] [ by subject ] [ by author ] [ by messages with attachments ]

From: Francesco Cerutti <Francesco.Cerutti_at_cern.ch>

Date: Wed, 14 Dec 2016 17:53:21 +0100

Dear Klaus,

actually what happens is that those neutrons get from the beginning a

different energy, being moved to neighbouring energy groups and

so distorting the spectrum. And the reason is as I previously wrote below.

Let's have three energy groups, from A to B, from B to C, and from C to D,

respectively. If you assign as energy value B, C, and D (which is itself a

bad idea, since a boundary value makes the group identification

ambiguous), and you input for them single precision numbers, it may happen

that your B value is interpreted as falling between A and B, your C value

as falling between C and D, and your D value as falling between C and D

too. So a given spectrum, expected for instance to get N counts in each of

the three groups, turns out to be polluted with N, zero, and 2N counts,

respectively.

Cheers

Francesco

**************************************************

Francesco Cerutti

CERN-EN/STI

CH-1211 Geneva 23

Switzerland

tel. ++41 22 7678962

fax ++41 22 7668854

On Wed, 14 Dec 2016, Ott, Klaus wrote:

*> Dear Francesco,
*

*>
*

*> now it works fine, the sampled spectrum looks like the original one. Thanks a lot!
*

*>
*

*> One last question: Why are neutrons of some energies not transported, if the source routine returns
*

*> energy values from table 10.4.1.1 (energy group structure of the 260- neutron data set)?
*

*> (these are the energy values in my EBIN array)
*

*>
*

*> Best regards
*

*>
*

*> Klaus
*

*>
*

*>
*

*> -----Ursprüngliche Nachricht-----
*

*> Von: Francesco Cerutti [mailto:Francesco.Cerutti_at_cern.ch]
*

*> Gesendet: Mittwoch, 14. Dezember 2016 12:17
*

*> An: Ott, Klaus <klaus.ott_at_helmholtz-berlin.de>
*

*> Cc: fluka-discuss_at_fluka.org
*

*> Betreff: Re: AW: [fluka-discuss]: Dips in sampled neutron spectrum
*

*>
*

*>
*

*> Dear Klaus,
*

*>
*

*> thanks. Now the picture is clear: you are sampling an energy value just at the boundary between two groups, which is not a good idea (since by definition a boundary value challenges the identification of the range it belongs to), and you are doing so with single precision numbers (as stored in your arrays, where the D specifier is missing). Apart from using double precision energy values, a robust solution - assuring your spectrum reproduction - is rather to take an energy value in the middle of the respective group
*

*>
*

*> ENERGY=(EBIN(K)+EBIN(K-1))/TWOTWO
*

*>
*

*> (To avoid the undesired case K=1, set SYCUM(1)=-ONEONE, i.e. to a negative value instead of zero, since FLRNDM(DUMMY) may return 0, never 1).
*

*>
*

*> Also for E>20 MeV, if the energy array is meant to store the histogram bin limits, it makes anyway more sense to sample a value in between.
*

*>
*

*> Kind regards
*

*>
*

*> Francesco
*

*>
*

*> **************************************************
*

*> Francesco Cerutti
*

*> CERN-EN/STI
*

*> CH-1211 Geneva 23
*

*> Switzerland
*

*> tel. ++41 22 7678962
*

*> fax ++41 22 7668854
*

*>
*

*> On Tue, 13 Dec 2016, Ott, Klaus wrote:
*

*>
*

*>>
*

*>> Dear Francesco,
*

*>>
*

*>>
*

*>>
*

*>> I send you the source file.
*

*>>
*

*>> Side note: It should be d2Phi/dOmega/dE
*

*>>
*

*>>
*

*>>
*

*>> Best regards
*

*>>
*

*>>
*

*>>
*

*>> Klaus
*

*>>
*

*>>
*

*>>
*

*>>
*

*>>
*

*>>
*

*>>
*

*>> ----Ursprüngliche Nachricht-----
*

*>> Von: Francesco Cerutti [mailto:Francesco.Cerutti_at_cern.ch]
*

*>> Gesendet: Dienstag, 13. Dezember 2016 15:22
*

*>> An: Ott, Klaus <klaus.ott_at_helmholtz-berlin.de>
*

*>> Cc: fluka-discuss_at_fluka.org
*

*>> Betreff: Re: AW: [fluka-discuss]: Dips in sampled neutron spectrum
*

*>>
*

*>>
*

*>>
*

*>>
*

*>>
*

*>> Dear Klaus,
*

*>>
*

*>>
*

*>>
*

*>> my first guess would be indeed related to the group structure. The
*

*>> actual group assignment resulting from the energy sampling in your
*

*>> source.f may be responsible for the dips. This would mean that the
*

*>> sampled energy value falls in the wrong group range. -
*

*>>
*

*>> Considering that your original spectrum already follows the group
*

*>> structure, this hypothesis looks surprising, but a final word cannot
*

*>> be said without looking into your source.f.
*

*>>
*

*>>
*

*>>
*

*>> On a side note, I understand that with dPhi you actually mean dE.
*

*>>
*

*>>
*

*>>
*

*>> Kind regards
*

*>>
*

*>>
*

*>>
*

*>> Francesco
*

*>>
*

*>>
*

*>>
*

*>> **************************************************
*

*>>
*

*>> Francesco Cerutti
*

*>>
*

*>> CERN-EN/STI
*

*>>
*

*>> CH-1211 Geneva 23
*

*>>
*

*>> Switzerland
*

*>>
*

*>> tel. ++41 22 7678962
*

*>>
*

*>> fax ++41 22 7668854
*

*>>
*

*>>
*

*>>
*

*>> On Mon, 12 Dec 2016, Ott, Klaus wrote:
*

*>>
*

*>>
*

*>>
*

*>>>
*

*>>
*

*>>> Dear Leonel,
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>> I used of course the low energy groups for sampling (the original
*

*>>
*

*>>> spectrum has been calculated by Fluka). The Fluka output is
*

*>>
*

*>>> d2/dPhi/dOmega and if you plot this, you see that this curve
*

*>>> decreases
*

*>>
*

*>>> by orders of magnitude. If these values are multiplied
*

*>>
*

*>>>
*

*>>
*

*>>> by the energy interval of the respective bin, the effect of the
*

*>>> small
*

*>>
*

*>>> low energy bins
*

*>>
*

*>>>
*

*>>
*

*>>> is (roughly) compensated. For neutron sampling I used the group
*

*>>
*

*>>> fluence
*

*>>
*

*>>> (d2/dPhi/dOmega*2pi*deltaEi) which I integrated (summarized) and
*

*>>
*

*>>> normalized
*

*>>
*

*>>>
*

*>>
*

*>>> and the energy values of the Fluka groups below 20 MeV (255) and
*

*>>
*

*>>> further 12 high energy bin values up to 0.5 GeV.
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>> Best regards
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>> Klaus
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>> Von: Leonel Morejon [mailto:leonel.morejon_at_eli-beams.eu]
*

*>>
*

*>>> Gesendet: Montag, 12. Dezember 2016 17:06
*

*>>
*

*>>> An: Ott, Klaus <klaus.ott_at_helmholtz-berlin.de>;
*

*>>
*

*>>> fluka-discuss_at_fluka.org
*

*>>
*

*>>> Betreff: Re: [fluka-discuss]: Dips in sampled neutron spectrum
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>> Dear Klaus Ott,
*

*>>
*

*>>>
*

*>>
*

*>>> It is probably is related to the way you scored and/or sampled.
*

*>>
*

*>>>
*

*>>
*

*>>> Did you take into account that in Fluka usually by default the
*

*>>
*

*>>> group approach is used for neutrons below 20 MeV?
*

*>>
*

*>>>
*

*>>
*

*>>> Is difficult to guess without your sampling routine, but if you
*

*>>
*

*>>> sampled linearly from all energies then the lower bins are extremely unlikely.
*

*>>
*

*>>>
*

*>>
*

*>>> Indeed if you sample linearly with energy, even with 10E10
*

*>>> sampled
*

*>>
*

*>>> particles, the particle density per energy in the low range (20MeV
*

*>>> to
*

*>>
*

*>>> 1e-5eV) is less than 1E3 part/eV, and this might be ok for bins
*

*>>
*

*>>> spawning 100-10meV but not for those thinner. Is not enough
*

*>>> statistics
*

*>>
*

*>>> especially because the groups are not linearly spaced, and some spawn 1e-2meV.
*

*>>
*

*>>>
*

*>>
*

*>>> To have them populated equally, you need to sample according to
*

*>>
*

*>>> the characteristics of the spectrum, Maxwell+1/v.
*

*>>
*

*>>>
*

*>>
*

*>>> But this are just conjectures, is hard to answer w/o seeing your
*

*>>
*

*>>> routine.
*

*>>
*

*>>>
*

*>>
*

*>>> Regards,
*

*>>
*

*>>>
*

*>>
*

*>>> Leonel
*

*>>
*

*>>>
*

*>>
*

*>>> On 12.12.2016 15:23, Ott, Klaus wrote:
*

*>>
*

*>>>
*

*>>
*

*>>> Dear Fluka team,
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>> I calculated a neutron spectrum outside the shielding of an
*

*>>
*

*>>> electron accelerator (1.7 GeV) and used that spectrum for
*

*>>
*

*>>> sampling in a simple geometry:
*

*>>
*

*>>>
*

*>>
*

*>>> an isotropic irradiating neutron source in the center of the
*

*>>
*

*>>> coordinate system, surrounded by vacuum and a sphere just used
*

*>>
*

*>>> for USRBDX
*

*>>
*

*>>>
*

*>>
*

*>>> counting. It is just a test to see, if the sampled spectrum is
*

*>>
*

*>>> the same as the original one.
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>> What I found is a similar spectrum, but it has several zero
*

*>>> bins
*

*>>
*

*>>> in the low energy region < 20 MeV. (This behaviour is not a
*

*>>
*

*>>> statistic effect, it occurs also
*

*>>
*

*>>>
*

*>>
*

*>>> when 7E9 neutrons are sampled and it is not an effect of a
*

*>>> wrong
*

*>>
*

*>>> integrated and normalized spectrum function used in source.f,
*

*>>
*

*>>> using source.f outside Fluka, an almost identical spectrum
*

*>>> is
*

*>>
*

*>>> returned).
*

*>>
*

*>>>
*

*>>
*

*>>> It seems to be an artefact of the neutron transportation
*

*>>> similar
*

*>>
*

*>>> to the one described in chapter 10 of the manual, which is in
*

*>>
*

*>>> that case a result of the interaction of neutrons with matter.
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>> Could you recommend optimized transport/threshold parameters
*

*>>> to
*

*>>
*

*>>> reduce/avoid this problem?
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>> Best regards
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>> Klaus Ott
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>> HZB/BESSYII Berlin Germany
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>> ____________________________________________________________________
*

*>>> __
*

*>>
*

*>>> ______
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>> Helmholtz-Zentrum Berlin für Materialien und Energie GmbH
*

*>>
*

*>>>
*

*>>
*

*>>> Mitglied der Hermann von Helmholtz-Gemeinschaft Deutscher
*

*>>
*

*>>> Forschungszentren e.V.
*

*>>
*

*>>>
*

*>>
*

*>>> Aufsichtsrat: Vorsitzender Dr. Karl Eugen Huthmacher, stv.
*

*>>
*

*>>> Vorsitzende Dr. Jutta Koch-Unterseher
*

*>>
*

*>>> Geschäftsführung: Prof. Dr. Anke Rita Kaysser-Pyzalla, Thomas
*

*>>
*

*>>> Frederking
*

*>>
*

*>>>
*

*>>
*

*>>> Sitz Berlin, AG Charlottenburg, 89 HRB 5583
*

*>>
*

*>>>
*

*>>
*

*>>> Postadresse:
*

*>>
*

*>>> Hahn-Meitner-Platz 1
*

*>>
*

*>>> D-14109 Berlin
*

*>>
*

*>>>
*

*>>
*

*>>> http://www.helmholtz-berlin.de
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>> --
*

*>>
*

*>>>
*

*>>
*

*>>> Leonel Morejon
*

*>>
*

*>>>
*

*>>
*

*>>> --------------------------------------
*

*>>
*

*>>>
*

*>>
*

*>>> ELI Beamlines
*

*>>
*

*>>>
*

*>>
*

*>>> Address:
*

*>>
*

*>>>
*

*>>
*

*>>> Za Radnicí 835
*

*>>
*

*>>>
*

*>>
*

*>>> 25241 Dolní Bøe¾any
*

*>>
*

*>>>
*

*>>
*

*>>> Czech Republic
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>> Office: +420 266 051 406
*

*>>
*

*>>>
*

*>>
*

*>>> Mobile: +420 773 666 453
*

*>>
*

*>>>
*

*>>
*

*>>> --------------------------------------
*

*>>
*

*>>>
*

*>>
*

*>>> ____________________________________________________________________
*

*>>> __
*

*>>
*

*>>> ______
*

*>>
*

*>>>
*

*>>
*

*>>> Helmholtz-Zentrum Berlin für Materialien und Energie GmbH
*

*>>
*

*>>>
*

*>>
*

*>>> Mitglied der Hermann von Helmholtz-Gemeinschaft Deutscher
*

*>>
*

*>>> Forschungszentren e.V.
*

*>>
*

*>>>
*

*>>
*

*>>> Aufsichtsrat: Vorsitzender Dr. Karl Eugen Huthmacher, stv. Vorsitzende Dr.
*

*>>
*

*>>> Jutta Koch-Unterseher
*

*>>
*

*>>> Geschäftsführung: Prof. Dr. Anke Rita Kaysser-Pyzalla, Thomas
*

*>>
*

*>>> Frederking
*

*>>
*

*>>>
*

*>>
*

*>>> Sitz Berlin, AG Charlottenburg, 89 HRB 5583
*

*>>
*

*>>>
*

*>>
*

*>>> Postadresse:
*

*>>
*

*>>> Hahn-Meitner-Platz 1
*

*>>
*

*>>> D-14109 Berlin
*

*>>
*

*>>>
*

*>>
*

*>>> http://www.helmholtz-berlin.de
*

*>>
*

*>>>
*

*>>
*

*>>>
*

*>>
*

*>>
*

*>>
*

*>>
*

*>> ______________________________________________________________________
*

*>> ___________
*

*>>
*

*>> Helmholtz-Zentrum Berlin für Materialien und Energie GmbH
*

*>>
*

*>> Mitglied der Hermann von Helmholtz-Gemeinschaft Deutscher Forschungszentren e.V.
*

*>>
*

*>> Aufsichtsrat: Vorsitzender Dr. Karl Eugen Huthmacher, stv. Vorsitzende
*

*>> Dr. Jutta Koch-Unterseher
*

*>> Geschäftsführung: Prof. Dr. Anke Rita Kaysser-Pyzalla, Thomas
*

*>> Frederking
*

*>>
*

*>> Sitz Berlin, AG Charlottenburg, 89 HRB 5583
*

*>>
*

*>> Postadresse:
*

*>> Hahn-Meitner-Platz 1
*

*>> D-14109 Berlin
*

*>>
*

*>> http://www.helmholtz-berlin.de
*

*>>
*

*>>
*

*>
*

*> ________________________________
*

*>
*

*> Helmholtz-Zentrum Berlin für Materialien und Energie GmbH
*

*>
*

*> Mitglied der Hermann von Helmholtz-Gemeinschaft Deutscher Forschungszentren e.V.
*

*>
*

*> Aufsichtsrat: Vorsitzender Dr. Karl Eugen Huthmacher, stv. Vorsitzende Dr. Jutta Koch-Unterseher
*

*> Geschäftsführung: Prof. Dr. Anke Rita Kaysser-Pyzalla, Thomas Frederking
*

*>
*

*> Sitz Berlin, AG Charlottenburg, 89 HRB 5583
*

*>
*

*> Postadresse:
*

*> Hahn-Meitner-Platz 1
*

*> D-14109 Berlin
*

*>
*

*> http://www.helmholtz-berlin.de
*

*>
*

__________________________________________________________________________

You can manage unsubscription from this mailing list at https://www.fluka.org/fluka.php?id=acc_info

Received on Wed Dec 14 2016 - 19:25:06 CET

Date: Wed, 14 Dec 2016 17:53:21 +0100

Dear Klaus,

actually what happens is that those neutrons get from the beginning a

different energy, being moved to neighbouring energy groups and

so distorting the spectrum. And the reason is as I previously wrote below.

Let's have three energy groups, from A to B, from B to C, and from C to D,

respectively. If you assign as energy value B, C, and D (which is itself a

bad idea, since a boundary value makes the group identification

ambiguous), and you input for them single precision numbers, it may happen

that your B value is interpreted as falling between A and B, your C value

as falling between C and D, and your D value as falling between C and D

too. So a given spectrum, expected for instance to get N counts in each of

the three groups, turns out to be polluted with N, zero, and 2N counts,

respectively.

Cheers

Francesco

**************************************************

Francesco Cerutti

CERN-EN/STI

CH-1211 Geneva 23

Switzerland

tel. ++41 22 7678962

fax ++41 22 7668854

On Wed, 14 Dec 2016, Ott, Klaus wrote:

__________________________________________________________________________

You can manage unsubscription from this mailing list at https://www.fluka.org/fluka.php?id=acc_info

Received on Wed Dec 14 2016 - 19:25:06 CET

*
This archive was generated by hypermail 2.3.0
: Wed Dec 14 2016 - 19:25:07 CET
*