From: Alfredo Ferrari (alfredo.ferrari@cern.ch)
Date: Mon Dec 04 2006 - 09:00:29 CET
Hi Alfred
LOW-NEUT is on by default since several FLUKA versions. Statistics is
really an issue, I would strongly suggest people looks into LAM-BIAS,
it can change completely the speed of convergence.
I am not terribly surprised a monoenergetic beam is similar to a real
spectrum with the same average energy wrt produced hadrons, since
virtual photonuclear interactions scale with energy (that is the energy
going into the hadrons is proportional to the initial muon energy).
However I expect differences when the spectrum is down to low energies
and hence some mu-'s stop in the setup. They are going to be
absorbed producing additional muons. This contribution would
obviously be missed in a monoenergetic run. We had problems at CERN where
this was by far the dominating contribution, of course its relevance
is depending on the hardness and average energy of the specturm, as
well as the target material which influences the competition absorption vs
decay for stopping mu-'s.
Ciao
Alfredo
On Sun, 3 Dec 2006, Tang, Alfred wrote:
> Dear Alfredo,
> I appreciate the detailed explanation that you gave to Lindley. I have run a similiar set of simulation as Lindley's except that I use the real geometry and liquid scintillator (LS) chemical composition. What I have found in my simulations is that the energy distribtions for the same set of secondary productions as Lindley's qualitatively agree in shapes with those in publication (I am using a different normalization constant from that in publication). So I am assuming that I am doing something right. I have also compared my simulations using (1) the real primary muon- profile and (2) fixed muon- energy at 285GeV. To my surprise, the shapes of the two set of distribution are almost identical for E<=285GeV. Obviously the distributions of secondary particles for E>285GeV is zero in case (2) because of the conservation of energy. So I am assuming that using a monoenergetic beam of E=285GeV is not the problem in this case. However I think that your point of low statist!
ic!
> s (using only 10000 primary muons in the input file) can be the cause of the problem. I use about 100000 primary muons in my input file and run it for 5 cycles so that the total primary muons for each simulation is 500000. After hearing your point (f), I am also thinking that statistics is an important factor in this case. Lindley's input file does not define the LOW-NEUT card explicitly. I am not sure if LOW-NEUT is turned on by default whenever neutrons are simulated. Should she turn on LOW-NEUT explicitly?
> Regards,
> Alfred
>
>
>
> *****************************************************
> Alfred Tang
> Physics Department
> Kansas State University
> 13 Cardwell Hall
> Manhattan, KS 66506
>
> (785) 532-1699
>
> http://www.phys.ksu.edu/personal/atang
>
>
>
> -----Original Message-----
> From: owner-fluka-discuss@fisica.unimi.it on behalf of Alfredo Ferrari
> Sent: Sun 12/3/2006 11:10 AM
> To: Lindley Winslow
> Cc: Fluka Discussion
> Subject: Re: Neutron and Pion USRTRACK results.
>
> Dear Lindley
>
> a few comments, I hope useful:
>
> a) I checked that Fluka2006.3 and Fluka2005.6 give with your input
> results for neutrons and pions which are identical within the errors
> (as it should be since nothing changed in between the two versions
> about muon photonuclear interactions) -> if the results are different
> from expected or obtained by others the reason must be in your setup
> b) your setup has monoenergetic muons of 285 GeV. I suppose this is the
> <average> energy of a muon spectrum underground, which usually
> extends from "zero" to an energy higher than the average. Using a
> monoenergetic beam instead of the proper spectrum can ingenerate
> differences, particularly for those processes (ie muon- capture
> at rest) which don't scale at all with energy. If my supposition is
> wrong forget my comment
> c) you start your beam just in front of your target, again if the problem
> is referring to some underground situation, the rock before, aside,
> and after must be modelled for 1-2 metres minimum (particularly the
> one before), if not you miss all neutrons coming from interactions
> in the rock. The effect is more or less large depending on your
> detector size, in your case it should not be obviously negligible
> but it should make factor of a few as well
> d) pi+ *must* be higher than pi- for your problem. You have a LIQUIDS
> material which is mostly hydrogen, see below the output of FLUKA
> Material Number Atom content Partial Densities
>
> CARBON 6 0.41440E-01 0.26436
> HYDROGEN 3 0.95856 0.51318
> Is it possible you made it wrong, that is you wanted a 1:2
> C:H ratio in atom content and not in density?? Scintillators
> have typical C:H ratio of 1:1 to 1:2 in atoms usually! If
> so it could explain most of your problems, the correct COMPOUND card in
> that case would be (note the signs!)
> COMPOUND +0.34 6.0 +0.66 3.0 0.0 0. LIQUIDS
> This is very likely to be the culprit
>
> e) There is a contribution from real photon photonuclear interactions
> (that is photons coming from muon pair production, bremsstrahlung and
> delta rays can make (gamma,n) reactions). For a hadronic shower their
> contribution is usually negligible, for muons it can be of some
> relevance since thru bremss+pair+delta they expend much more energy
> than the one by virtual photonuclear interactions. To switch on real
> photon photonuclear interactions look at the PHOTONUC card in the
> manual (I would be surprised it makes more than a 10% contribution
> however...)
> f) Your input file is for 10000 muons. I needed 200000 to get n and
> pi statistics at 10% or worse level on totals!! Either you run for
> huge numbers of primaries, or you activate interaction biasing
> to improve statistics (see LAM-BIAS). In the latter case, please
> be careful in your mgdraw generated file that particles carry
> weights which will be different than 1 and in general different
> from one to the other (the FLUKA built-in scoring facilities like
> USRTRACK and USRBIN account automatically for this)
>
> Ciao
> Alfredo
>
> On Fri, 1 Dec 2006, Lindley Winslow wrote:
>
>>
>> --Apple-Mail-12--108175957
>> Content-Transfer-Encoding: 7bit
>> Content-Type: text/plain;
>> charset=US-ASCII;
>> delsp=yes;
>> format=flowed
>>
>> Hello all,
>>
>> I was comparing my results from the attached setup with results
>> others has obtained with an older version of FLUKA and a similar
>> setup. The most obvious difference is my neutrons are an order of
>> magnitude low and pi+ higher relative to pi-. I was wondering:
>>
>> 1) Had I correctly setup the PHYSICS model for a problem where
>> residual nuclei scoring was important?
>> 2) I thought I understood from the manual that LOW-MAT or LOW-NEUT
>> cards were not needed for this setup and that low energy neutrons
>> would be transported by default.
>>
>> I am running:
>> FLUKA2006 Version 3.0 Sep-06 by A. Ferrari DATE: 12/ 1/ 6 TIME:
>> 19: 5:41
>>
>> Thank you,
>> Lindley
>>
>>
>> --Apple-Mail-12--108175957
>> Content-Transfer-Encoding: 7bit
>> Content-Type: application/octet-stream;
>> x-unix-mode=0640;
>> name=cylinderLSExperiment.inp
>> Content-Disposition: attachment;
>> filename=cylinderLSExperiment.inp
>>
>> *...+....1....+....2....+....3....+....4....+....5....+....6....+....7....+....8
>> TITLE
>> Simple Muon Spallation Monte-Carlo in LS target.
>> *...+....1....+....2....+....3....+....4....+....5....+....6....+....7...+...8
>> BEAM 2.85E+02 MUON-
>> BEAMPOS 0.0 0.0 -50.0
>> *
>> GEOBEGIN COMBNAME
>> 0 0 A simple LS target inside vacuum
>> RPP body1 -5000000.0 +5000000.0 -5000000.0 +5000000.0 -5000000.0 +5000000.0
>> RPP body2 -1000000.0 +1000000.0 -1000000.0 +1000000.0 -100.0 +1000000.0
>> RCC body3 0.0 0.0 0.0 0.0 0.0 1000.0 500.0
>> * plane to separate the upstream and downstream part of the target
>> *XYP body4 2.5
>> END
>> * black hole
>> regBH1 5 +body1 -body2 -body3
>> * vacuum around
>> regVA2 5 +body2 -body3
>> * LS target 1st half
>> regBE3 5 +body3
>> * LS target 2nd half
>> *regBE4 5 +body3 -body4
>> END
>> * The next is an example of geometry debugging:
>> *GEOEND 150. 75. 220. 30. 0. -220.DEBUG
>> *GEOEND 120. 1. 110. 0. 0. 0. &
>> *STOP
>> * Normal End.
>> GEOEND
>> *...+....1....+....2....+....3....+....4....+....5....+....6....+....7....+....8
>> * Here defines materials and Compounds for LS
>> *...+....1....+....2....+....3....+....4....+....5....+....6....+....7....+...
>> *1) Hydrogen
>> MATERIAL 1.0 1.00794 0.0000837 3.0 1.0 HYDROGEN
>> *2) Carbon
>> MATERIAL 6.0 12.0107 2.00 6.0 CARBON
>> *3) Oxygen
>> MATERIAL 8.0 15.9994 0.00133 8.0 OXYGEN
>> * liquid scintillator
>> MATERIAL 0.0 0.0 0.77754 26.0 0.0 0. LIQUIDS
>> COMPOUND -0.34 6.0 -0.66 3.0 0.0 0. LIQUIDS
>> * LS target, 1st and 2nd half
>> ASSIGNMAT LIQUIDS regBE3
>> * regBE4
>> * External Black Hole
>> ASSIGNMAT BLCKHOLE regBH1
>> * Vacuum
>> ASSIGNMAT VACUUM regVA2
>> * Turn on the Physics Model, What is SCORE doing?
>> *...+....1....+....2....+....3....+....4....+....5....+....6....+....7....+....8
>> EVENTYPE 0.0 0.0 2.0 0.0 0.0 0.DPMJET
>> PHYSICS 3.0 EVAPORAT
>> PHYSICS 1.0 COALESCE
>> *
>> * Activates Residual Nuclei scoring
>> *...+....1....+....2....+....3....+....4....+....5....+....6....+....7....+...
>> RESNUCLEI 3.0 34.0 regBE3
>> RESNUCLEI 1.0 35.0 regBE3
>> RESNUCLEI 2.0 36.0 regBE3
>> *...+....1....+....2....+....3....+....4....+....5....+....6....+....7....+...
>> * Cartesian binning of the pion fluence inside and around the target
>> * R-Z binning of the neutron fluence inside and around the target
>> USRBIN 11.0 NEUTRON -50.0 500.0 0.0 1000.nuFluBin
>> USRBIN 0.0 0.0 0.0 50.0 1.0 1.0 &
>> * R-Z binning of the neutron fluence inside and around the target
>> USRBIN 11.0 NEUTRON 51.0 500.0 0.0 1000.nuFluBin
>> USRBIN 0.0 0.0 0.0 50.0 1.0 1.0 &
>> * R-Z binning of the neutron fluence inside and around the target
>> USRBIN 11.0 ELECTRON 52.0 500.0 0.0 1000.nuFluBin
>> USRBIN 0.0 0.0 0.0 50.0 1.0 1.0 &
>> * R-Z binning of the energy deposited inside and around the target
>> USRBIN 11.0 ENERGY -53.0 500.0 0.0 1000.nuFluBin
>> USRBIN 0.0 0.0 0.0 50.0 1.0 1.0 &
>> *...+....1....+....2....+....3....+....4....+....5....+....6....+....7...+...8
>> USRTRACK 1.0 NEUTRON 62.0 regBE3 1.0 1000. nuTrck
>> USRTRACK 1. 0. &
>> USRTRACK 1.0 ELECTRON 63.0 regBE3 1.0 1000.eminTrck
>> USRTRACK 1. 0. &
>> USRTRACK 1.0 POSITRON 64.0 regBE3 1.0 1000.eposTrck
>> USRTRACK 1. 0. &
>> USRTRACK 1.0 PHOTON 65.0 regBE3 1.0 1000.photTrck
>> USRTRACK 1. 0. &
>> USRTRACK 1.0 PROTON 66.0 regBE3 1.0 1000.protTrck
>> USRTRACK 1. 0. &
>> USRTRACK 1.0 PION+ 67.0 regBE3 1.0 1000.piplTrck
>> USRTRACK 1. 0. &
>> USRTRACK 1.0 PION- 68.0 regBE3 1.0 1000.pimnTrck
>> USRTRACK 1. 0. &
>> * Dump all particles.
>> * Notice slightly different bar to get formatting right.
>> USERDUMP 200.0 37.0 1.0 1.0 SRCEPART
>> * This one is from USDRAW
>> OPEN 71.0
>> usrdraw.txt
>> *...+....1....+....2....+....3....+....4....+....5....+....6....+....7....+....8
>> RANDOMIZE 1.0
>> *...+....1....+....2....+....3....+....4....+....5....+....6....+....7....+....8
>> START 10000.0
>> STOP
>>
>> --Apple-Mail-12--108175957--
>>
>
>
-- +----------------------------------------------------------------------------+ | Alfredo Ferrari || Tel.: +41.22.767.6119 | | CERN-AB || Fax.: +41.22.767.7555 | | 1211 Geneva 23 || e-mail: Alfredo.Ferrari@cern.ch | | Switzerland || Alfredo.Ferrari@mi.infn.it | +----------------------------------------------------------------------------+
This archive was generated by hypermail 2.1.6 : Mon Dec 04 2006 - 15:02:02 CET